C 2/06/93 405212046 MEMBER NAME EDITFILE (CLIST) M *********************************************************************** * * * PART 8 PART 8 * * H1 LUMI LOGBOOK * * * * WINTER SHUT DOWN 1993-1994 * * * * * * Use command: F ====== to find all messages, and then select * * those you are interested in * * * * The following commands allow to inspect earlier logbook parts: * * * * LL1 ........ list Part 1 of the Lumi Logbook: Jun'91 - Aug'91 * * LL2 ........ list Part 2 of the Lumi Logbook: Aug'91 - Dec'91 * * LL3 ........ list Part 3 of the Lumi Logbook: Jan'92 - May'92 * * LL4 ........ list Part 4 of the Lumi Logbook: Jun'92 - Aug'92 * * LL5 ........ list Part 5 of the Lumi Logbook: Aug'92 - Nov'92 * * LL6 ........ list Part 6 of the Lumi Logbook: Nov'92 - Jun'93 * * LL7 ........ list Part 7 of the Lumi Logbook: Jun'93 - Nov'93 * * * * S.Levonian Opened: 3/11/93 at 15:00 * * Closed: 21/05/94 at 21:00 * * * *********************************************************************** 03.11.1993 ==================== F11LEV === New Chapter of the logbook With this message a new part of the LUMI Logbook is openned. You can still inspect last part by the new command LL7 = LIST 'F11LEV.H1.LUMI.LOGBOOK.PART07' Store LL7 in your CLIST library for easy use (see lds 'F11LEV.CLIST' if not clear how to do this). 04.11.1993 ==================== H1KFOM === 100th tuples Reprocessing. It was found bug at filling of 100th tuples for H1Lumi's dedicated Runs. After correction all 100th tuples for H1lumi's dedicated Runs were reprocessed. Bug was connected with wrong filling of VCDEP- energy into tuple (all another variables were filled right). Reprocessed 100th tuples are kept at the same LOOK-data sets which were proclaimed earlier: standalone 26.7 GeV e-beam H1KFOM.ZLOOK.R65570 - standard H1Lumi's trigger setting H1KFOM.ZLOOK.R65572 - s86=PD, s93=PD2 H1KFOM.ZLOOK.R65574 - s86=PD, s93=PD2 H1KFOM.MLOOK.R65576 - s86=ET&PD, S93=PD1 standalone 26.7 GeV e-beam (after p-lost at epRun#353) H1KFOM.ZLOOK.R66202 - standard H1Lumi's trigger setting H1KFOM.ZLOOK.R66203 - standard H1KFOM.ZLOOK.R66204 - standard H1KFOM.ZLOOK.R66205 - standard ep-collisions (last 0.5 hour of epRun#367) H1KFOM.ZLOOK.R66552 - s86=PD with new low threshold H1KFOM.ZLOOK.R66553 - s86=PD with new low threshold. As seems all my previous remarks at LL7 about bad work of VC-trigger were connected with high mentioned bug (firstly mentioned at 26.10.93 message and repeated at two messages at 30.10.93). Now all is OK with checking of VC-trigger work. VC-trigger worked reasonable. Sorry for mistake. 04.11.1993 ==================== H1KFOM === Run 66553 Observation .... 115377 events were written on H1RAWD-cartridge at this Run. 114731 events from this Run have LREE and LRPE-banks. 77821 events from this Run have H1Lumi good reconstruction at both arms or at any single (Photon or Electron) arm. Good reconstruction means that: 1) there is 30GeV>Edep>0.5GeV at any single arm or at each; 2) no problems with cluster shape (topology of hitted cells); 3) Edep at selected with largest deposited energy 3x3 cells part of detectors is more then 0.5 GeV; 4) there are no another problems with coordinate and energy reconstruction. Statistics distribution with different subtriggers and trigger elements (some type of cross check(?)) Subtriggers: 77821-->77767-->36675(s86 only)-->36670(s86=1&PD=1) | | |-> 5(s86=1&PD=0) <--?? | |->22986(s93 only)-->22981(s93=1&VC=1) | | |-> 5(s93=1&VC=0) <--?? | |->17930(s94 only)-->17928(s94=1&(et=1&pd=1&vc=0)) | | |-> 2(s94=1¬.(et=1&pd=1&vc=0)) <-? | |-> 176(s95 only)--> 176(s95=1&(et=0&pd=1&vc=1) | |-> 0(s95=1¬.(et=0&pd=1&vc=1)) <-! |-> 54--> 16(s86=1&s93=1) |-> 12(s86=1&s94=1) |-> 15(s86=1&s95=1) |-> 11(s93=1&s95=1) At 54 events it was fixed double source of L2keep-event(both triggers were active according to existing prescale factors). At 77821 events there are 77875 combinations with active subtriggers: 36718 - s86=1 (36675+16+12+15) 23013 - s93=1 (22986+16+11) 17942 - s94=1 (17930+12) 202 - s95=1 ( 176+15+11) Trigger Elements: | | | Event | combination statistics VC=1 | PD=1 | ET=1 | Statistics | S86=1 | s93=1 | s94=1 | s95=1 ----------------------------------------------------------------- 0 | 0 | 0 | 12 | 5 | 5 | 2 | 0 1 | 0 | 0 | 5884 | 0 | 5884 | 0 | 0 0 | 1 | 0 | 3688 | 3688 | 0 | 0 | 0 1 | 1 | 0 | 31535 | 20285 |11087 | 0 | 202 0 | 0 | 1 | 0 | 0 | 0 | 0 | 0 1 | 0 | 1 | 81 | 0 | 81 | 0 | 0 0 | 1 | 1 | 19740 | 1812 | 0 |17940 | 0 1 | 1 | 1 | 16881 | 10928 | 5956 | 0 | 0 Data from this Run is good sample for: 1) attempt to make "classical" calibration (without using of LRPP and LREP-banks contents) - we have 19740 events with ET&PD&no(VC) trigger (for calibration of ET and PD-channels) and 16881 events with ET&PD&VC-trigger (for calibration of VC-channel. In principle we can get better energy resolution for this Run then it was produced with standard H1Lumi reconsrtruction package; 2) attempt to make of relation coefficients set (Run was collected during 1500 sec. and it were collected 19740 events - almost the same conditions at at so-called 10-Hz Runs) - as it was made at May'93 and compare with all measured relation coefficient sets; 3) attempt to make calibration of VC-channel with using of LREP- and LRPP-bank contents (as it was made at May'93) and define calibration coefficient for VCsp channel (may be with coorrection which was made by S.Levonian 07.06.93 - see message at LL7). Shortly about energy resolution digits which were produced with H1Lumi reconstruction package with using of LREP- LRPP-banks contents and relation coefficient set (May'93) and CCvc=3.230 a) ET&PD&no(VC)-sample: histos sample entries LOOK's peak finder Mean/Sigma Min.value Max.value all events 18666 26.92 0.91 26.45 1.96 8.13 49.40 |Xet|<6.6 15907 26.92 0.88 26.75 1.50 8.13 49.40 vcdep<0.3 13372 26.87 0.88 26.75 1.47 8.13 49.40 VCdep-spectrum(18666 entries) - 0.132 0.22 -1.04 1.58 VCdep spectrum has not any statistics after 1.5GeV. ETrec-spectrum(18666 entries) - 14.63 2.40 1.47 30.38 ETrec spectrum has maximum near 16.4 GeV and range of significant statistics from 5GeV up to 21 GeV; PDrec-spectrum(18666 entries) - 11.82 2.51 0.64 27.93 PDrec spectrum has maximum near 9.3 GeV and range of significant statistics from 6GeV up to 20 GeV; b) ET&PD&VC-sample: histos sample entries LOOK's peak finder Mean/Sigma Min.value Max.value all events 16235 26.35 1.35 25.80 2.31 7.03 53.40 |Xet|<6.6 13850 26.34 1.29 26.06 1.98 7.03 53.40 vcdep<0.3 160 26.40 1.06 26.31 1.49 12.80 28.90 VCdep-spectrum(16235 entries) - 3.03 1.88 -1.04 15.35 VCdep spectrum has maximum near 2.0 GeV and end of significant statistics near 10 GeV. ETrec-spectrum(16235 entries) - 14.47 2.45 1.74 28.89 ETrec spectrum has maximum near 16.3 GeV and range of significant statistics from 5GeV up to 21 GeV; PDrec-spectrum(16235 entries) - 11.32 2.70 ?.?? ??.?? PDrec spectrum has maximum near 9.7 GeV and range of significant statistics from 5GeV up to 20 GeV; Interesting information about Xpd and Ypd distributions at both samples: a) ET&PD&no(VC)-sample: mean/sigma values: histos Look's peak finder Xpd(cm) -0.24 / 1.41 ??? / ???? Ypd(cm) 0.22 / 0.51 0.296 / 0.331 b) ET&PD&VC-sample: Xpd(cm) -0.001/ 1.50 0.13 / 1.77 Ypd(cm) 0.23 / 0.87 0.42 / 0.73 05.11.1993 ==================== F11LEV === Explanation for Run 66553 The following information must be taken into account for the correct understanding of run 66553 (see H1KFOM summary on this run above). 1) 4 subtriggers only were used in this run: s86 = PD > 2.5 GeV s93 = VC s94 = ET & PD & noVC s95 = noET & PD & VC Important: first 3 are declared as monitor triggers, while s95 is physics trigger (see H1 TRIG). I believe, this was wrong decision, but since nobidy from rad.corr. group did not complain in July when it was discussed - too bad for them In addition there was 0.2 Hz of random trigger included. (The total rate was 90-100 Hz, so the relative amount of random triggers in the ran should be small: 2 permille level). The events triggered by the random trigger ONLY do NOT produce TLV1 bank! One can find the information about that in HEAR bank (artificial trigger words). 2) All monitor triggers passed through L4 (filter farm) without any rejection/modification. For the physics triggers L4 may do either of the following: o reject the event if it does not satisfy some conditions for this particular trigger and no other triggers were set o reset L1 trigger bit if the conditions to set this bit were not verified on the basis of the readout information (energies in our case) 3) As the result: o 30187 events taken by s95 only were rejected as empty events in H1 (Fomenko did not feel this, since they never appear after L4) o for 244 events which passed L4 the trigger bit s95 was reset by L4 after verification (on the basis of the real energies from LREE and LRPE banks and comparing them with some thresholds). 4) All the question marks in the Fomenko's message I can explain only as the situation, when the events taken by random trigger contained in TEL1 bank all lumi bits = 0 (naturally), and no TLV1 bank at all. Fomenko's software perhaps then took as subtriggers the information from previous event, which had TLV1 bank (A.F., please check how do you perform erasing of VEC for n-tuple for each event!) 05.11.1993 ==================== H1KFOM === TLV1-bits at 100th tuples. S.Levonian is right - I supposed that TLV1-banks were presented at each event and do not make some tool for labeling of events without TLV1-bank. It was possible really that at event without TLV1-bank it was used values of TLV1-bank from previous event. Similar tool was made for TEL1-bank (values et,pd,vc,pd1 and pd2 trigger elements are installed into 2-value (not 0 and not 1)). But TEL1 exists at all events. Today I inserted this tool for TLV1-bank too and submitted job for recreation of 100th tuple for Run 66553. It must appear LOOK-data set H1KFOM.CLOOK.R66553 with 100th tuple inside and with high mentioned convention - if TLV1-bank absent at H1 events then s86,s93,s94 and s95 will be installed into value 2. 05.11.1993 ==================== H1KFOM === HERA p-beam Large Current. Today it was fixed large current of HERA p-beam during "Machine Studies" announcement. Injection of p-beam started near 14:00 today and near 14:56 it was fixed Ip=24080 mkA with 180 p-bunches. Maximal bunch current was fixed as 204 mkA. Bunch structure was fixed as the next:10 filled + 1 empty + 10 filled + 1 empty +....10 filled (first 60 filled bunches) + 8 empty + the next 60 bunches with high mentioned structure(10+1+10+1+..+10) + 8 empty + the last 60 bunches with the same structure. Ramping was not started and near 15:10 Ip was killed (or lost). VCrate was fixed near 60 Hz. P.S. This is not too large current: there was already 32.5 mA in 180 p-bunches achieved this week, which is still only 20% of the design value. (S.Levonian) 06.11.1993 ==================== H1KFOM === Attempt of ep-collisions.. Yesterday at 22:00 HERA p-beam with energy 820 GeV and Ip=10.4mA was ready. Tonight near 04:00 it was started e-injection and very slowly Ie was incremented up to 5.6mA and ramping of e-ring was started near 06:00. e-beam was lost during ramping. At 06:10 p-beam was dumped. All high mentioned period it was looked for stable difference at e-filled bunches and rate-filled bunches (shift was fixed as 21 BC) - (may be it is needed some tuning at Central Trigger - may be reboot of H1 Central Trigger FIC, may be call experts, may be it was some mistake of HERA-people). 06.11.1993 ==================== H1KFOM === Relation Coefficients Set. As it was proclaimed earlier sample ET&PD(no)VC-events from H1Lumi Dedicated Runs 66552-66553 is good sample for relation coefficient set defining (data from this sample were collected with log rate near 10Hz - it means that calibration at Luminosity subbranch(FIC#2) was made sinchronously with any off-line calibration with this sample). Yesterday and today it was made first attempt to define relation coefficient set with tools which were used at May'93 for the same procedure. It was got only subsample from full sample of ET&PD&no(VC)-events - first 12400 events from both Runs. It was made only due to very slow passing of IBM-jobs at L class. High mentioned sample was processed only with K-jobs. It was made 3 steps of calibration and 4 steps of relation coefficient set testing. The next digits were produced with test jobs: 1) May'93 Relation Coefficient Set: sample entries LOOK's peak finder Mean/Sigma Min.value Max.value all events 12599 26.87 0.89 26.48 2.05 9.85 50.72 |Xet|<6.6 10822 26.90 0.86 26.76 1.65 9.85 50.72 vcdep<0.3 9049 26.89 0.86 26.75 1.61 9.85 50.72 2) after first step of calibration: sample entries LOOK's peak finder Mean/Sigma Min.value Max.value all events 12209 26.63 0.84 26.35 2.03 9.47 48.90 |Xet|<6.6 10368 26.58 0.83 26.54 1.63 9.47 48.90 vcdep<0.3 8679 26.51 0.82 26.51 1.58 9.47 48.90 3) after 2nd step of calibration: sample entries LOOK's peak finder Mean/Sigma Min.value Max.value all events 12199 26.66 0.83 26.36 2.03 9.45 49.18 |Xet|<6.6 10388 26.69 0.82 26.56 1.63 9.45 49.18 vcdep<0.3 8690 26.64 0.80 26.53 1.57 9.52 49.18 4) after 3rd step of calibration: sample entries LOOK's peak finder Mean/Sigma Min.value Max.value all events 12196 26.65 0.83 26.35 2.02 9.51 49.20 |Xet|<6.6 10406 26.70 0.82 26.56 1.64 9.51 49.20 vcdep<0.3 8704 26.67 0.80 26.53 1.57 9.56 49.20 As always there are not large difference between 2-nd and 3rd step of calibration - it means that 2 steps of recalibration is enough for getting of final result. High mentioned procedure put on 7% better resolution at ETrec+Pdrec spectrum as it was with start values of relation coefficient set. These are the characteristics of all standard histos after 3rd step of calibration: Mean/Sigma Min.value Max.value VCdep-spectrum(12196 entries) - 0.141 0.23 -1.09 1.69 VCdep spectrum has not any statistics after 1.5GeV. ETrec-spectrum(12196 entries) - 15.30 2.50 3.42 32.13 ETrec spectrum has maximum near 16.5 GeV and range of significant statistics from 4GeV up to 22 GeV; PDrec-spectrum(12196 entries) - 11.04 2.38 2.07 30.62 PDrec spectrum has maximum near 9.0 GeV and range of significant statistics from 6GeV up to 19.5 GeV. Xpd and Ypd distributions - mean/sigma values: histos Look's peak finder Xpd(cm) -0.27 / 1.43 -0.56 / 1.28 Ypd(cm) 0.20 / 0.53 0.25 / 0.343 These are values of relation coefficients after 3rd step of calibration with sigmas at its distributions and numbers of entries into its histos --> Relation Coeff.Set for ET 0.000 1.224 0.000 1.227 1.217 0.000 1.224 0.000 1.276 1.309 1.360 1.213 1.276 0.000 1.031 0.945 1.077 0.992 1.238 0.921 1.165 1.025 1.019 1.021 1.046 1.073 1.056 1.049 0.944 0.840 1.117 1.009 0.947 0.943 1.038 1.234 1.069 1.362 1.272 1.195 1.399 1.443 0.000 1.237 1.243 1.169 1.451 1.189 1.246 --> Relation Coeff.Set for PD 1.104 1.301 1.843 1.116 0.851 1.755 0.864 0.932 0.976 1.163 0.927 0.982 0.986 0.956 1.056 0.987 0.984 0.908 0.955 0.873 1.128 0.906 0.919 0.882 0.832 --> Sig. at Relation Coeff.Set for ET 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.001 0.000 0.000 0.000 0.000 0.009 0.005 0.001 0.006 0.003 0.000 0.004 0.009 0.007 0.010 0.008 0.009 0.009 0.007 0.006 0.011 0.013 0.007 0.004 0.006 0.007 0.001 0.003 0.007 0.003 0.001 0.011 0.028 0.000 0.008 0.000 0.000 0.000 0.001 0.000 --> Sig. at Relation Coeff.Set for PD 0.000 0.001 0.001 0.000 0.000 0.013 0.009 0.006 0.006 0.002 0.004 0.007 0.007 0.007 0.017 0.004 0.005 0.004 0.005 0.001 0.001 0.008 0.007 0.001 0.000 --> Ent. at Relation Coeff.Set for ET 0 1 0 1 16 0 1 0 1 2 2 1 2 0 57 29 47 15 30 3 5 2215 3480 2428 1906 1535 1277 605 672 388 790 585 563 600 407 9 3 7 5 6 10 7 0 3 2 1 1 2 1 --> Ent. at Relation Coeff.Set for PD 6 9 9 4 6 147 354 600 241 66 552 3550 5861 3038 355 109 953 2215 1155 108 12 5 17 11 1 High mentioned digits mean that really it were defined relation coef- ficients for 3 central layers at both detectors (zero value means full absence of statistics and relation coefficient was not defined). High mentioned procedure was made with the next jobs at H1KFOM.TSTVC- PD6-library: #C553PS1 002A06 SAVED BY H1KFOM ON 93/11/06 AT 05:11 #C553PS2 00350A SAVED BY H1KFOM ON 93/11/06 AT 13:50 #C553PS3 003A06 SAVED BY H1KFOM ON 93/11/06 AT 14:12 #T553PS1 00230C SAVED BY H1KFOM ON 93/11/06 AT 03:07 #T553PS2 002D0C SAVED BY H1KFOM ON 93/11/06 AT 05:47 #T553PS3 003A0E SAVED BY H1KFOM ON 93/11/06 AT 14:14 #T553PS4 004004 SAVED BY H1KFOM ON 93/11/06 AT 14:45 For high mentioned procedure it was used as input data set - H1KFOM.R66553LP.A00-cartridge with selected ET&PD&no(VC)-events from all 14820 events of Run66552 and first 71614 events of Run66553. There are 14004 ET&PD&no(VC) selected events at this cartridge. In principle all this procedure can be repeated for full sample of Lumi pure events from these two Runs. But it is not possible to make so quickly as with K-class jobs. It is possible to check so-called classics calibration (without using of LREP- and LRPP- banks contents). As seems resolution must be better. Final distributions and 100th tuple it is possible to find at H1KFOM. DLOOKT.R66552 - LOOK data set. 06.11.1993 ==================== H1KFOM === TLV1-bits at 100th tuples. It was produced 100th tuple for Run66533 data - almost the same as previous but at case of absence TLV1-bank at event (Random-trigger label) s86,s93,s94 and s95 are labeled with value 2 (not 0 and not 1). This tuple is kept at H1KFOM.CLOOK.R66553-data set as it was promised at my 05.11.93 message. Test of high mentioned 12 events shows that it were really events without TLV1-bank - so -no any problems with cross check of subtriggers and trigger elements. 06.11.1993 ==================== H1KFOM === Classical Calibration66553 It was tested classical calibration at the sample ET&PD&no(VC)-events on the same 12400 events which were used for Relation Coefficients Set producing. The next digits were produced with test job: after 3rd step of calibration: sample entries LOOK's peak finder Mean/Sigma Min.value Max.value all events 12190 26.58 0.83 26.39 2.03 9.55 48.95 |Xet|<6.6 10408 26.62 0.82 26.58 1.66 9.55 48.95 vcdep<0.3 8637 26.55 0.78 (!) 26.54 1.60 9.55 48.95 High mentioned procedure was made with the next job at H1KFOM.TSTVC- PD6-library: #TCL553K 004E0F SAVED BY H1KFOM ON 93/11/06 AT 21:31 For high mentioned procedure it was used as input data set - H1KFOM.R66553LP.A00-cartridge with selected ET&PD&no(VC)-events from all 14820 events of Run66552 and first 71614 events of Run66553. There are 14004 ET&PD&no(VC) selected events at this cartridge. The next values of CC were produced after 3rd step of calibration: --> ET Calibr.Coefficients Final Status 4.015 4.016 4.015 4.016 4.017 4.015 4.015 4.015 4.012 4.021 4.013 4.014 4.015 4.015 4.678 3.154 3.578 3.238 4.069 3.833 4.005 4.494 4.930 4.926 4.797 4.387 4.969 3.688 3.955 3.670 4.086 3.631 3.379 3.415 2.970 4.022 4.016 3.416 3.365 3.711 3.522 3.590 4.015 3.970 4.013 4.014 3.984 4.011 4.013 --> PD & VC Calibr.Coefficients Final Status 2.525 2.774 2.993 2.621 2.528 2.237 1.789 2.672 2.460 2.312 2.044 2.924 3.510 3.128 2.301 2.247 1.107 2.504 3.248 2.612 2.498 1.882 2.262 2.722 2.525 3.817 --> ET Channels Final Weights 0.650 0.663 0.650 0.659 0.680 0.650 0.652 0.650 0.656 0.667 0.658 0.653 0.652 0.650 1.247 1.259 1.001 1.397 0.898 0.669 0.740 798.341 860.476 571.132 372.656 278.321 196.654 80.299 3.847 2.403 4.663 3.574 4.072 4.739 5.739 0.703 0.753 0.742 0.828 0.729 0.971 1.004 0.650 0.773 0.654 0.651 0.735 0.663 0.654 --> PD & VC Channels Final Weights 0.665 0.694 0.674 0.655 0.657 1.610 5.436 8.873 4.153 0.906 23.112 331.181 815.241 344.490 18.708 2.611 14.859 28.335 16.033 1.384 0.680 0.972 1.554 0.950 0.651 0.650 Only 7 channels at ET and 8 channels at PD were calibrated with this pure statistics. 100th tuple for this data is kept at H1KFOM.ZLOOKT.R66552 LOOK-data set 06.11.1993 ==================== H1KFOM === Lumi RUN# 370 ............ Start lumi run370 date: 6/11/93 Collision time: 11:46:06 - 16:19:22 Currents (mkA): Ie= 1107.00, Ip= 1700.00 Peak luminosity: ???.??? mb-1 s-1 Peak spec.lumi : ???.?????? mb-1 s-1 mA-2 Number of collided b.: 10 Number of e-filled b.: 12 Number of p-filled b.: 10 Number of e-pilot b.: 2 Number of p-pilot b.: 0 HERA total lumi: ????.?????? mkb-1 H1 Run total lumi: ????.?????? mkb-1 H1 data taken lumi: ????.?????? mkb-1 H1 Runs: 66904 - 66907 (temporary logging) HERA - physics: 0.0 mkb-1 H1Run - physics: 0.0 mkb-1 H1data - physics: 0.0 mkb-1 H1 Runs: 66904 - 66907 (temporary logging) P.S.: "Mashine Shift"-announcement 07.11.1993 ==================== H1KFOM === Lumi RUN# 371 ............ Start lumi run371 date: 6/11/93 -7/11/93 Collision time: 21:30:00 - 04:48:30 Currents (mkA): Ie= 8500.00, Ip= 8500.00 <-- from picture Peak luminosity: 375.000 mb-1 s-1 <---/ Peak spec.lumi : 390.000000 mb-1 s-1 mA-2 <---/ Number of collided b.: 84 Number of e-filled b.: 94 Number of p-filled b.: 90 Number of e-pilot b.: 10 Number of p-pilot b.: 6 HERA total lumi: 4400.69 mkb-1 <-- for both #370 ų H1 Run total lumi: 4214.91 mkb-1 <--/ H1 data taken lumi: 2454.56 mkb-1 <--/ H1 Runs: 66907 - 66914 (after 66910 - permanent logging) HERA - physics: 0.0 mkb-1 H1Run - physics: 0.0 mkb-1 H1data - physics: 0.0 mkb-1 H1 Runs: 66907 - 66914 (after 66910 - permanent logging) P.S.1 "Mashine Shift"-announcement P.S.2 During this Run it was crossed boundary 1.0pb-1 (HERA produced Integrated Luminosity since 27.06.93). It was happened near 23:00 06 November 1993. P.S.3 At 23:30 it was started Run 66913 with similar trigger setting as at Run 66553 (s86 as PD with threshold 2.5 GeV). Log rate was choosen not very large specially for attempt to registrate some changing at CC during changing of beam currents. Run started when Ip=8.5 mA and Ie=6.6 mA. P.S.4 At 01:59 it was started Run 66914 with similar trigger setting but with tuned prescales because of low luminosity. 07.11.1993 ==================== H1KFOM === Relation Coefficient Set.. It was produced relation coefficients set after processing of all events from sample ET&PD&no(VC) at Runs 66552-66553. All these events were preliminary selected on cartridge data set: H1KFOM.G66553LP.A00. There are 21433 events at this cartridge. It were repeated all procedure which was made 06/11/93 for all sample. There were no significant improvements at resolution digits at ETrec+PDrec histos. The best resolution after 3 steps of calibra- tion and with apperture and VCdep cuts is 0.81 GeV. This is the result of relation coefficients set calculation: --> Relation Coeff.Set for ET 0.000 1.224 0.000 1.227 1.217 0.000 1.224 0.000 1.276 1.310 1.360 1.213 1.276 0.000 1.021 0.970 1.140 0.990 1.239 0.920 1.170 1.029 1.016 1.019 1.045 1.085 1.066 1.057 0.965 0.932 1.115 1.013 0.958 0.948 1.052 1.238 1.070 1.335 1.235 1.186 1.289 1.354 0.000 1.232 1.243 1.165 1.437 1.189 1.246 --> Relation Coeff.Set for PD 1.104 1.301 1.833 1.116 0.851 1.640 0.864 0.928 0.984 1.142 0.927 0.976 0.979 0.947 1.033 0.970 0.978 0.910 0.951 0.920 1.128 0.890 0.933 0.890 0.832 --> Sig. at Relation Coeff.Set for ET 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.001 0.000 0.000 0.000 0.000 0.012 0.004 0.010 0.005 0.003 0.001 0.003 0.010 0.009 0.008 0.008 0.011 0.009 0.006 0.006 0.016 0.011 0.008 0.003 0.005 0.005 0.001 0.003 0.005 0.004 0.001 0.021 0.022 0.000 0.007 0.001 0.002 0.016 0.001 0.000 --> Sig. at Relation Coeff.Set for PD 0.001 0.001 0.013 0.001 0.001 0.026 0.008 0.005 0.007 0.004 0.004 0.007 0.008 0.008 0.021 0.004 0.004 0.004 0.004 0.005 0.001 0.006 0.006 0.003 0.000 --> Ent. at Relation Coeff.Set for ET 0 1 0 1 25 0 1 0 2 3 2 1 3 0 93 71 90 22 43 7 12 3809 5969 4207 3282 2690 2191 1012 1231 957 1345 1007 994 1003 671 15 6 17 12 13 14 11 0 4 4 8 5 3 1 --> Ent. at Relation Coeff.Set for PD 13 15 22 9 8 232 627 1050 442 88 917 6044 10138 5351 643 179 1627 3852 2105 235 17 10 24 15 1 Final distributions and 100th tuple it is possible to find at H1KFOM. DLOLKT.R66552 - LOOK data set. 07.11.1993 ==================== H1KFOM === Relation Coeff. Test ..... Relation Coefficients Set was tested with sample of ET&PD&no(VC) events from last H1Lumi Dedicated Run - 66913. High mentioned sample was preliminary selected on H1KFOM.G66913LP.A00 cartridge data set. There are 22512 events at this cartridge. It was produced 100th tuple for high mentioned sample of events with using of last Relation Coefficients Set (see my message 07.11.93). It was produced good resolution digits on ETrec+PDrec spectra: sample entries LOOK's peak finder Mean/Sigma all events 20917 26.60 0.81 26.17 1.91 |Xet|<6.6 18143 26.61 0.80 26.46 1.51 vcdep<0.3 16882 26.59 0.80 26.43 1.49 100th tuple is kept at H1KFOM.ALOOKT.R66913 - LOOK data set. 07.11.1993 ==================== H1KFOM === Data sets with sel. events There are 4 data sets with selected H1Lumi monitoring events collected after begin of September 93 (when FADC-shifts problem was decided): H1KFOM.EP248EPV.ZZ K28180 78048081 00001 ACS-E-C 4166 events of ET&PD&VC -trigger (Runs 60989-61407) H1KFOM.EP248B29.Z K28968 78048081 00001 ACS-E-C 12657 events of VC=1 trigger (Runs 60990-61357) H1KFOM.EP248.Z K31974 78048081 00001 ACS-C 52605 events of ET&PD&no(VC)-trigger (Runs 60998-61303) H1KFOM.EP308XXX.A00 K37674 78048081 00001 ACS-E-C 30980 events of VC=1 trigger (Runs 64662-65559) First three data sets were prepared due to activity of N.Gogitidze, last data set - by V.Andreev and me. Nelly started her selection from H1Runs of epRun#248, Vova started from H1Runs of epRun#308. The next banks are presented at these data sets: RSELECT BOSOUTPUT BLOCK = "RUNEVENT" 'HEAD' 'HEAR' 'LREE' 'LRPE' & 'LREP' 'LRPP' 'TLV1' 'TEL1' 'LRTN' 'TSTC' 08.11.1993 ==================== H1KFOM === Relation Coeff. Test ..... Relation Coefficients Set was tested with sample of ET&PD&no(VC) events from last H1Lumi Dedicated Run - 66914. High mentioned sample was preliminary selected on H1KFOM.G66914LP.A00 cartridge data set. There are 83231 events at this cartridge. It was produced 100th tuple for high mentioned sample of events with using of last Relation Coefficients Set (see my message 07.11.93). It was produced good resolution digits on ETrec+PDrec spectra(again): sample entries LOOK's peak finder Mean/Sigma all events 77834 26.64 0.80 26.17 1.85 |Xet|<6.6 67115 26.64 0.79 26.47 1.38 vcdep<0.3 62793 26.63 0.78 (!) 26.43 1.49 |Xet|<6.0 57051 26.66(!)0.76 (!!) 100th tuple is kept at H1KFOM.ALOOKT.R66914 - LOOK data set. Xpd-distribution seems as unwaitable (a little). 09.11.1993 ==================== H1KFOM === Relation Coeff. Test ..... Relation Coefficients Set was tested with Nelly's sample of ET&PD&no(VC)-events which was collected on data set H1KFOM.EP248B30.Z. There are 80435 events at this sample. These events were selected from H1Runs 60998 up to 61570 - from epRun#248 up to epRun#260 - from 05.09.93 11:57 up to 10.09.93 05:54. It was produced 100th tuple for high mentioned sample of events with using of last Relation Coefficients Set (see my message 07.11.93). It was produced good resolution digits on ETrec+PDrec spectra(again): sample entries LOOK's peak finder all events 79588 26.59 0.83 |Xet|<6.0 65863 26.60 0.81 vcdep<0.3 42461 26.53 0.76 (!) 100th tuple is kept at H1KFOM.ALOOKT.R60998 - LOOK data set. 10.11.1993 ==================== H1KFOM === First attempts of VC-clbr. It were made first attempts to make VCsp-calibration at all possible data sample from last part of H1 Data Taking'93. 1) Nelly's sample (H1KFOM.EP248EPV.ZZ) - from Run60989 up to 61407: 4166 events: 4.287 <---- value od CC for VCsp; 131.610 <---- weight 2) H1Lumi's Dedicated Runs 66552-66553 - sample H1KFOM.G66553LR.A00 19080 events: 3.129 <---- value od CC for VCsp; 694.629 <---- weight 3) H1Lumi's Dedicated Run 66613 - sample H1KFOM.G66613LR.A00 13017 events: 2.400 <---- value od CC for VCsp; 455.418 <---- weight 4) H1Lumi's Dedicated Run 66614 - sample H1KFOM.G66614LR.A00 36000 events: 2.585 <---- value od CC for VCsp; 1211.558 <---- weight All values of CC for Veto are preliminary. It is needed to play with start values of CCvcsp. At all high mentioned jobs CCvcsp(start)=3.614 I asked Nelly to make only ET&PD&VC-selection at neareast days for getting more statistics at her sample. 11.11.1993 ==================== H1KSOL === Trig.eff. Runs 64662-65559 Trigger efficiency monitoring for H1 Runs#64662-65559 (ep308-336) was done offline. Monitoring trigger - s93 (VETO). Pictures are kept in file H1KSOL.MLOOK.R64662. Figs.310-312 - pd,et,pd+et trig.sum (Bit=1) Figs.410-412 - _______ " __________ (all) Figs.510-512 - corresp.trig.eff. Trigger summs from LRTN bank were taken for this monitoring. 12.11.1993 ==================== H1KFOM === HV on VCsp and VCtr OFF... Today during evening shift it was switched off HV on both photomulti- pliers of Water Cherenkov Counter. VC-rate was near 500 kHz during existing at HERA of high currents (BG conditions were very bad). 13.11.1993 ==================== H1KFOM === Data sets with sel. events There are 4 data sets with selected H1Lumi monitoring events collected after begin of September 93 (when FADC-shifts problem was decided): All events were selected with analyze of TEL1-bank contents (not TLV1) H1KFOM.EP248EPV.ZZZZ ( 5.994Mbyte)(16)(0) K28180 78048081 00001 ACS-E-C 5260 events of ET&PD&VC -trigger (Runs 60989-61567) H1KFOM.EP248B29.Z5 (20.269Mbyte) (5)(0) K28968 78048081 00001 ACS-E-C 15760 events of VC=1 trigger (Runs 60990-61800) H1KFOM.EP248B30.ZZZ (121.678Mbyte) (5)(1) K41468 78048081 00001 ACS-E-C 94475 events of ET&PD&no(VC)-trigger (Runs 60998-61747) H1KFOM.EP308XXX.A00 (38.584Mbyte+x) (0) K37674 78048081 00001 ACS-E-C 30980 events of VC=1 trigger (Runs 64662-65559) (x - due to selection without dedicated runs) First three data sets were prepared due to activity of N.Gogitidze, last data set - by V.Andreev and me. Nelly started her selection from H1Runs of epRun#248, Vova started from H1Runs of epRun#308. The next banks are presented at these data sets: RSELECT BOSOUTPUT BLOCK = "RUNEVENT" 'HEAD' 'HEAR' 'LREE' 'LRPE' & 'LREP' 'LRPP' 'TLV1' 'TEL1' 'LRTN' 'TSTC' There are two cartridges with selected VC=1 events and ET&PD&no(VC)- events from first part of H1 Data Taking'93 (these events were selected by Nelly Gogitidze with A.Semenov's program - with analyze of B29 or B30 at TLV1-banks). H1KFOM.B29.POT3.ZZ K31273 78048081 00001 ARCHIVE-C 26024 events of VC=1(s93)-trigger (Runs 51132-?????) H1KFOM.B30.POT3.ZZZZ K30944 78048081 00001 ARCHIVE-C 109406 events of ET&PD&no(VC)-trigger (Runs 50694-59961) Most part of events at Last two cartridges were got with non-right FADC-shifts. 13.11.1993 ==================== H1KFOM === Tuning of CC for VC ...... At the all possible samples of 2nd period H1 Data Taking'93 it were tuned calibration constants for VCsp-channel: 1) Nelly's sample: H1KFOM.EP248EPV.ZZZ ( 5.510Mbyte) (8)(0) K28180 78048081 00001 ACS-E-C 4579 events of ET&PD&VC -trigger (Runs 60989-61500) CCvcsp = 4.436 Mev/FADCcount ---------------------------- 2) Vova's sample: H1KFOM.EP308XXX.A00 (38.584Mbyte+x) (0) K37674 78048081 00001 ACS-E-C 30980 events of VC=1 trigger (Runs 64662-65559) CCvcsp = 3.262 Mev/FADCcount ---------------------------- 3) H1Lumi's Dedicated Runs during #367th epRun (Run66552-66553) H1KFOM.G66553LR.A00 K40087 78048081 00001 ACS-E-C 19080 events of ET&PD&VC-trigger (Runs 66552-66553) CCvcsp = 3.129 Mev/FADCcount ---------------------------- 4) H1Lumi's Dedicated Run during #371th epRun (Run66913) H1KFOM.G66913LR.A00 K41298 78048081 00001 ACS-E-C 13017 events of ET&PD&VC-trigger (Run 66913) CCvcsp = 2.392 Mev/FADCcount ---------------------------- It were produced 100th tuples for each from high mentioned sample with using of last Relation Coefficient Set (07.11.93-message at LLC) and with using of high mentioned different CCvcsp. These tuples are kept at the next LOOK-data sets: H1KFOM.WBOOKT.R60989 H1KFOM.WAOOKT.R64662 H1KFOM.W1OOKT.R66552 H1KFOM.WBOOKT.R66913 It is possible to make 07.06.93 Levonian's method for final tuning (accounting of the second photon). Without Levonian's tuning there are the next digits at ETrec+PDrec spectra (LOOK's peak finder): 1) Nelly's sample: Mean = 26.43 Sigma = 1.25 (all sample) Mean = 26.50 Sigma = 1.22 (|Xet|<6.0cm) 2) Vova's sample: Mean = 26.45 Sigma = 1.28 (all sample) Mean = 26.43 Sigma = 1.19 (|Xet|<6.0cm) 3) R66552-66553 : Mean = 26.55 Sigma = 1.23 (all sample) Mean = 26.40 Sigma = 1.22 (|Xet|<6.0cm) 3) R66913 : Mean = 26.23 Sigma = 1.22 (all sample) Mean = 26.38 Sigma = 1.18 (|Xet|<6.0cm) Before VC-calibration (with using of start value 3.616 for all samples the next digits were produced (with the same Relation Coefficients) 1) Nelly's sample: Mean = 26.02 Sigma = 1.33 (all sample) Mean = 26.05 Sigma = 1.25 (|Xet|<6.0cm) 2) Vova's sample: Mean = 26.19 Sigma = 1.27 (all sample) Mean = 26.18 Sigma = 1.26 (|Xet|<6.0cm) 3) R66552-66553 : Mean = 26.74 Sigma = 1.23 (all sample) Mean = 26.70 Sigma = 1.21 (|Xet|<6.0cm) 3) R66913 : Mean = 27.13 Sigma = 1.31 (all sample) Mean = 27.21 Sigma = 1.31 (|Xet|<6.0cm) It were made different number of calibration steps for different samples (criterium of last step was minimal difference between start value and final value of CCvcsp) due different statistics: 1) Nelly's sample: 3 steps 3.616 --> 4.287 --> 4.420 --> 4.436 Number of events passed through CALIBRA = 4042 2) Vova's sample: 3 steps 3.616 --> 3.298 --> 3.267 --> 3.262 Number of events passed through CALIBRA = 3757 3) R66552-R66553: 2 steps 3.616 --> 3.129 --> 3.129 Number of events passed through CALIBRA = 18801 4) R66913 : 2 steps 3.616 --> 2.400 --> 2.392 Number of events passed through CALIBRA = 12191 The next jobs were used for calibration of VCsp and CCvcsp-testing (producing of high mentioned 100th tuples) H1KFOM.TSTVCPD7-library: #C552PS2 001D05 SAVED BY H1KFOM ON 93/11/13 AT 13:20 #C662PS3 002B0C SAVED BY H1KFOM ON 93/11/13 AT 14:09 #C913PS2 001A07 SAVED BY H1KFOM ON 93/11/12 AT 13:38 #C989PS3 002901 SAVED BY H1KFOM ON 93/11/13 AT 13:34 #T552PS2 001201 SAVED BY H1KFOM ON 93/11/10 AT 16:32 #T662PS2 004301 SAVED BY H1KFOM ON 93/11/13 AT 14:51 #T913PS3 00310D SAVED BY H1KFOM ON 93/11/13 AT 14:14 #T989PS3 003C01 SAVED BY H1KFOM ON 93/11/13 AT 14:19 13.11.1993 ==================== H1KFOM === 100 tuples for PhotonShape It were produced 100th tuples from samples which are connected with high mentioned 4 periods of H1DataTaking'93. These tuples were produced for possibility Photon Shape Global Shift defining at different periods of H1DataTaking'93. All 100th tuples have the same structure as they had for H1Lumi's dedicated Runs (as at ZLOOK.R..-LOOK-data sets). The next samples were got for producing of 100th tuples: 1) Nelly's sample: H1KFOM.EP248B29.Z5 (20.269Mbyte) (5)(0) K28968 78048081 00001 ACS-E-C 15760 events of VC=1 trigger (Runs 60990-61800) 2) Vova's sample: H1KFOM.EP308XXX.A00 (38.584Mbyte+x) (0) K37674 78048081 00001 ACS-E-C 30980 events of VC=1 trigger (Runs 64662-65559) 3) Run 66553 - H1Lumi's Dedicated Run at the end of #367th epRun HERA04.H1RAWD.C9304034 -C HERA04.H1RAWD.C9304035 -C 115377 events with H1Lumi's standard trigger setting exept s86 = PD with low(2.5GeV) threshold 4) Run 66913 - H1Lumi's Dedicated Run during of #371th epRun HERA04.H1RAWD.C9304071 -C HERA04.H1RAWD.C9304072 -C 92785 events with H1Lumi's standard trigger setting exept s86 = PD with low(2.5GeV) threshold 100th tuples are kept at the next LOOK-data sets: Output file: H1KFOM.ALOOKQ.R60994 Output file: H1KFOM.ALOOKQ.R64662 Output file: H1KFOM.ALOOKQ.R66553 Output file: H1KFOM.ALOOKQ.R66913 It were used last Relation Coefficients Set (07.11.93) and last calibration coefficients for VC (13.11.93). CCvcsp were got different for each sample - according to message 13.11.93. Jobs for producing of high mentioned tuples are kept at library: H1KFOM.TSTVCPD8 at the next data sets: +R553 00170B SAVED BY H1KFOM ON 93/11/13 AT 17:45 +R662 001105 SAVED BY H1KFOM ON 93/11/13 AT 17:24 +R913 001B06 SAVED BY H1KFOM ON 93/11/13 AT 17:50 +R990 000F07 SAVED BY H1KFOM ON 93/11/13 AT 17:07 14.11.1993 ==================== H1KFOM === Photon Shape Fit ......... It were estimated preliminary values of global energy shifts at Photon Arm of H1Lumi detectors with S.Levonian's tool. Options: ep-fit (not e-gas), fit from 7 GeV up to 36 GeV, pdrec histos for fit - 70 bins from 0 up to 35 GeV. 1) Nelly's sample: 4579 events of ET&PD&VC -trigger (Runs 60989-61500) It was fitted 50th histo at H1KFOM.BLOOKQ.R60994 LOOK data set which was produced from high mentioned 100th tuple for this sample. (only with PDREC.lt.999 condition). Shift = 0.976 Resolution 21.6 2) Vova's sample: 30980 events of VC=1 trigger (Runs 64662-65559) It were fitted 50&51 histos at H1KFOM.BLOOKQ.R64662 LOOK data set which was produced from high mentioned 100th tuple for this sample. (50th with PDREC.lt.999 condition). (51st with PDREC.lt.999.and.PD.eq.1 conditions ). Shift = 0.982 Resolution 24.7 - for 50th histos Shift = 0.982 Resolution 24.8 - for 51st histos 3) H1Lumi's Dedicated Runs during #367th epRun (Run66552-66553) 19080 events of ET&PD&VC-trigger (Runs 66552-66553) It were fitted 50&60&61 histos at H1KFOM.BLOOKQ.R66552 LOOK data set which was produced from high mentioned 100th tuple for this sample. (50th with s86.eq.1.and.PDREC.lt.999 conditions). (60th with s93.eq.1.and.PDREC.lt.999 conditions ). (61st with s93.eq.1.and.PDREC.lt.999.and.pd.eq.1 conditions ). Shift = 1.010 Resolution 32.8 - for 50th histos Shift = 1.020 Resolution 27.8 - for 60th histos Shift = 1.021 Resolution 27.6 - for 61st histos 4) H1Lumi's Dedicated Run during #371th epRun (Run66913) 13017 events of ET&PD&VC-trigger (Run 66913) It were fitted 50&60&61 histos at H1KFOM.BLOOKQ.R66913 LOOK data set which was produced from high mentioned 100th tuple for this sample. (50th with s86.eq.1.and.PDREC.lt.999 conditions). (60th with s93.eq.1.and.PDREC.lt.999 conditions ). (61st with s93.eq.1.and.PDREC.lt.999.and.pd.eq.1 conditions ). Shift = 0.966 Resolution 31.5 - for 50th histos Shift = 0.976 Resolution 24.0 - for 60th histos Shift = 0.976 Resolution 24.0 - for 61st histos 14.11.1993 ==================== H01LNS === PD-SPECTRA-FITTING ....... PD-SPECTRA-FITTING, AFTER A.FOMENKO (not the last). In A.Fomenko's results for PD-spectra-fitting, LL6 27.10.93, there are no chisq-probability and behind this, some unsuccessful fits may be hidden, that might be important for understanding of PD-performance. In order to clarify the situaion and to test stability of fit-results, I have made some systematic fits, extended for different Emin,Emax,Np. To make comparison easier, the SL's code was updated in a way that fit results are printed in one line, some results are given in %%, etc. SUMMARY OF THE USED DATA. Five lumi-data-runs were used: One (A) run by V.Andreev, very old, as "a reper point", and Four cases exactly by A.Fomenko, to clarify the present situation. R 64662 was at the change of HV for VC, R 65575 was at 26.7 GeV e-beam FITTING VARIANTS. All the fits were done for (ep) and (eGas), as S.Levonian's (separated) For each run the fits were done twice, in dependance on Emin or Emax: - At fixed Emax==30.Gev, for different Emin, increas.by steps of 2.Gev, - At fixed Emin==...Gev, for different Emax, decreas.by steps of 1.Gev, Comment: the starting Emin were selected by the program, automatically. Comment: Ecorr == (S.L.'s_param_3 - 1.0)*100. (%%) == fitted_correction Comment: Res == (S.L.'s_param_4)*100. (%%) == as in Res/sqrt(E)(10%?) THE "BEST" RESULTS OF PD-SPECTRA-FITTING (four divisions) ----------------------------------------------------------------------- @ lrun ifg emin emax np ty e chfl chi2 prob% Ecorr +- err Res+-er kEV ----------------------------------------------------------------------- ----------------------------------------------------------------------- (ep) - fit emin from 2.Gev, up by 2.Gev, emax == 30.Gev ----------------------------------------------------------------------- ----------------------------------------------------------------------- R 33611 21 6.0 30.0 48 ep 0 fitOK 36. 79.7% ( 3.9+-0.3)% (21+-1)% 93 R 64662 21 14.0 30.0 32 ep 0 fitOK 33. 25.0% ( 3.8+-0.7)% (26+-4)% 10 R 65575 50 16.0 30.0 28 ep 0 fitOK 36. 5.0% ( 7.4+-0.3)% (21+-2)% 39 R 65575 51 16.0 30.0 28 ep 0 fitOK 36. 5.0% ( 7.4+-0.3)% (21+-2)% 19 R 65575 52 20.0 30.0 20 ep 0 fitOK 20. 20.0% ( 7.9+-0.1)% (20+-1)% 45 ----------------------------------------------------------------------- (eGas) - fit emin from 2.Gev, up by 2.Gev, emax == 30.Gev ----------------------------------------------------------------------- R 33611 21 10.0 30.0 40 eG 0 fitOK 38. 38.3% ( 2.1+-0.3)% (27+-1)% 93 R 64662 21 4.0 30.0 52 eG 0 fitOK 56. 19.5% ( 2.2+-0.7)% (30+-4)% 10 R 65575 50 18.0 30.0 24 eG 0 fitOK 27. 12.9% ( 6.4+-0.3)% (24+-2)% 39 R 65575 51 18.0 30.0 24 eG 0 fitOK 27. 12.9% ( 6.4+-0.3)% (24+-2)% 19 R 65575 52 24.0 30.0 12 eG 0 fitOK 12. 17.3% ( 7.4+-0.2)% (21+-1)% 45 ----------------------------------------------------------------------- (ep) - fit emin == 5.Gev, emax from 30.Gev, down by 1.Gev ----------------------------------------------------------------------- R 33611 21 5.0 30.0 50 ep 0 fitOK 45. 51.4% ( 3.8+-0.3)% (22+-1)% 93 R 64662 21 5.0 30.0 50 ep 0 fitOK 58. 11.3% ( 4.1+-0.7)% (25+-4)% 10 R 65575 50 6.5 30.0 47 ep 0 infER 185. 0.0% ( 7.0+-0.0)% (24+-0)% 39 R 65575 51 12.5 30.0 35 ep 0 fitOK 52. 1.1% ( 7.3+-0.3)% (22+-2)% 19 R 65575 52 18.0 30.0 24 ep 0 fitOK 41. 0.4% ( 8.0+-0.1)% (19+-1)% 45 ----------------------------------------------------------------------- (eGas) - fit emin == 5.Gev, emax from 30.Gev, down by 1.Gev ----------------------------------------------------------------------- R 33611 21 5.0 30.0 50 eG 0 fitOK 107. 0.0% ( 1.6+-0.3)% (30+-1)% 93 R 64662 21 5.0 30.0 50 eG 0 fitOK 55. 16.9% ( 2.2+-0.7)% (30+-4)% 10 R 65575 50 6.5 30.0 47 eG 0 fitOK 279. 0.0% ( 4.9+-0.3)% (35+-3)% 39 R 65575 51 12.5 30.0 35 eG 0 fitOK 82. 0.0% ( 5.8+-0.3)% (28+-2)% 19 R 65575 52 18.0 30.0 24 eG 0 fitOK 47. 0.1% ( 6.9+-0.1)% (24+-1)% 45 Comment:To see ALL numerical results of the fits put * to the next line LIST 'H01LNS.LUMI4(J1311FSS)' COMPARISON OF A.FOMENKO'S FITS WITH THE CORRESPONDING POINTS OF THESE ----------------------------------------------------------------------- Comment: I suppose, that A.F.'s Emax==36.GeV is a mistake, it is =30(?) 1) PDREC-histo(21) from VC=1 trigger for long time period: fit from 7 GeV up to 36 Gev, 46 points, ch= 54.3, e-p, eps=0.01 GlobalShift = 1.040+-0.007, Resolution(B)= 25.42+-4.37% R 64662 21 14.0 30.0 32 ep 0 fitOK 33. 25.0% ( 3.8+-0.7)% (26+-4)% 10 2) PDREC-histo(50) from PD=1 trigger for Run65575 fit from 13 GeV up to 36 Gev, 34 points, ch= 91.0, e-gas, eps=0.001 GlobalShift = 1.057+-0.003, Resolution(B)= 28.95+-2.23% R 65575 50 18.0 30.0 24 eG 0 fitOK 27. 12.9% ( 6.4+-0.3)% (24+-2)% 39 3) PDREC-histo(51) from PD1=1 trigger for Run65575 fit from 14 GeV up to 36 Gev, 32 points, ch= 69.5, e-gas, eps=0.001 GlobalShift = 1.059+-0.003, Resolution(B)= 27.84+-2.21%, R 65575 51 18.0 30.0 24 eG 0 fitOK 27. 12.9% ( 6.4+-0.3)% (24+-2)% 19 4) PDREC-histo(52) from PD2=1 trigger for Run65575 fit from 20 GeV up to 36 Gev, 20 points, ch=41.31, e-gas, eps=0.01 GlobalShift = 1.068+-0.003, Resolution(B)= 24.76+-1.16% R 65575 52 24.0 30.0 12 eG 0 fitOK 12. 17.3% ( 7.4+-0.2)% (21+-1)% 45 SOME CONCLUSIONS: 1. A.Fomenko's results are in a good agreement with the corresponding points of these results, even if they are not identical. 2. The dependance on Emin is rather trivial - starting from some Emin, up, all the fits have resulted in about the same (a good stability). 3. The dependance on Emax is more important, at least for me - all is OK for Emax = 30->29->28 Gev, but lower all fits have failed. I gess it is because of the sharp end of the bremsstrahlung spectrum but I have dissapointed: the well increasing left-hand-region of the spectrum does not work, as I wanted it to work. M.b., trivial. 4. The "Best"-fit results show: - The selected PD-spectra are well fitted (a good probability value) - The Ee-correction (for these runs) is not small: ((2 - 7) +/- .5)% - The PD-Resolution (in this conditions) is bad: ~25% instead of 10% 5. It seems reasonable to continue fitting at other runs and triggers. 6. I dont understand all the staff, I shall try to do, m.b. update this L.Shtarkov, Moscow, 14.11.93 PS: I have not seen the previous LLO by A.Fomenko, so .... 15.11.1993 ==================== H1KFOM === Histos for PhotonShapeFits It is needed to submit S.Levonian's tool for all possible PDREC-spectra from high mentioned 4 periods of data taking (Nelly's sample, Vova's sample, H1Lumi Dedicated Runs 66552-66553 and H1Lumi Dedicated Run 66913). All these spectra were produced from contents of 100th tuples which were produced with last release of Relation Coefficients set (07.11.93) and last values of CCvcsp(13.11.93). As You remember CCvcsp's are different for all high mentioned data samples. All PDREC histos have 70 bins with 0.0Gev as Min value and 35.0 GeV as max value. All PDREC histos were produced from 100th tuples which are kept at ALOOKQ...-LOOK-data set family (see message 13.09.93 "100th tuples for Shape Fits") and wrote into BLOOKQ...LOOK-data set family. All possible PDREC-spectra are kept at the next LOOK-data sets: 1) Nelly's sample: H1KFOM.BLOOKQ.R60994 data source: H1KFOM.EP248B29.Z5 (20.269Mbyte) (5)(0) K28968 78048081 00001 ACS-E-C 15760 events of VC=1 trigger (Runs 60990-61800) 50th histo - do 'hs(50)=nt(100)(pdrec)@(pdrec.lt.999) PDREC-spectrum for VC=1 trigger 6749 entries 2) Vova's sample: H1KFOM.BLOOKQ.R64662 data source: H1KFOM.EP308XXX.A00 (38.584Mbyte+x) (0) K37674 78048081 00001 ACS-E-C 30980 events of VC=1 trigger (Runs 64662-65559) H1KFOM.BLOOKQ.R64662 LOOK data set 50th histo - do 'hs(50)=nt(100)(pdrec)@(pdrec.lt.999) PDREC-spectrum for VC=1 trigger 8063 entries 51st histo - do 'hs(51)=nt(100)(pdrec)@(pdrec.lt.999.and.pd.eq.1) PDREC-spectrum for VC&PD trigger 7689 entries 52nd histo - do 'hs(52)=nt(100)(pdrec)@(pdrec.lt.999.and.pd1.eq.1) PDREC-spectrum for VC&PD1 trigger 4265 entries 53rd histo - do 'hs(53)=nt(100)(pdrec)@(pdrec.lt.999.and.pd2.eq.1) PDREC-spectrum for VC&PD2 trigger 2506 entries 3) H1Lumi's Dedicated Runs during #367th epRun (Run66552-66553) H1KFOM.BLOOKQ.R66553 data source: HERA04.H1RAWD.C9304034 -C HERA04.H1RAWD.C9304035 -C 115377 events with H1Lumi's standard trigger setting exept s86 = PD with low(2.5GeV) threshold 50th histo - do 'hs(50)=nt(100)(pdrec)@(s86.eq.1.and.pdrec.lt.999) PDREC-spectrum for PD=1 trigger with low (2.5GeV) threshold 36192 entries 51st histo = 52nd histo do 'hs(51)=nt(100)(pdrec)@(s86.eq.1.and.pdrec.lt.999.and.pd1.eq.1) PDREC-spectrum for PD(2.5GeV)&PD1 14370 entries 53th histo do 'hs(53)=nt(100)(pdrec)@(s86.eq.1.and.pdrec.lt.999.and.pd2.eq.1) PDREC-spectrum for PD(2.5GeV)&PD2 7703 entries 60st histo do 'hs(60)=nt(100)(pdrec)@(s93.eq.1.and.pdrec.lt.999) PDREC-spectrum for VC=1 trigger 21896 entries 61st histo do 'hs(61)=nt(100)(pdrec)@(s93.eq.1.and.pdrec.lt.999.and.pd.eq.1) PDREC-spectrum for VC&PD trigger 16789 entries 62nd histo do 'hs(62)=nt(100)(pdrec)@(s93.eq.1.and.pdrec.lt.999.and.pd1.eq.1) PDREC-spectrum for VC&PD1 trigger 6773 entries 63rd histo do 'hs(63)=nt(100)(pdrec)@(s93.eq.1.and.pdrec.lt.999.and.pd2.eq.1) PDREC-spectrum for VC&PD2 trigger 3678 entries 4) H1Lumi's Dedicated Run during #371th epRun (Run66913) H1KFOM.BLOOKQ.R66913 data source: HERA04.H1RAWD.C9304071 -C HERA04.H1RAWD.C9304072 -C 92785 events with H1Lumi's standard trigger setting exept s86 = PD with low(2.5GeV) threshold 50th histo - do 'hs(50)=nt(100)(pdrec)@(s86.eq.1.and.pdrec.lt.999) PDREC-spectrum for PD=1 trigger with low (2.5GeV) threshold 31512 entries 51st histo = 52nd histo do 'hs(51)=nt(100)(pdrec)@(s86.eq.1.and.pdrec.lt.999.and.pd1.eq.1) PDREC-spectrum for PD(2.5GeV)&PD1 13176 entries 53th histo do 'hs(53)=nt(100)(pdrec)@(s86.eq.1.and.pdrec.lt.999.and.pd2.eq.1) PDREC-spectrum for PD(2.5GeV)&PD2 7003 entries 60st histo do 'hs(60)=nt(100)(pdrec)@(s93.eq.1.and.pdrec.lt.999) PDREC-spectrum for VC=1 trigger 8748 entries 61st histo do 'hs(61)=nt(100)(pdrec)@(s93.eq.1.and.pdrec.lt.999.and.pd.eq.1) PDREC-spectrum for VC&PD trigger 6605 entries 62nd histo do 'hs(62)=nt(100)(pdrec)@(s93.eq.1.and.pdrec.lt.999.and.pd1.eq.1) PDREC-spectrum for VC&PD1 trigger 2834 entries 63rd histo do 'hs(63)=nt(100)(pdrec)@(s93.eq.1.and.pdrec.lt.999.and.pd2.eq.1) PDREC-spectrum for VC&PD2 trigger 1513 entries As seems it is needed to make different fits (ep,e-gas) with different Emin (due to PD1 and PD2 existing) and with may be only 28,29,30 Emax for cross check. After this it will be possible to decide - which correction factors it is needed to get for future reprocessing. May be preliminary estimations which were made earlier (see 14.11.93 message) it is needed to check else one with all histos (may be with high mentioned L.N.Shtarkov's tool? - for cross-check? As I know reprocessing must start after 3 days. 15.11.1993 ==================== H01LNS === FITS FOR 19 PD-SPECTRA ... PD-FITS FOR 19 PD-SPECTRA, AFTER A.FOMENKO. After A.Fomenko's request, I have done PD-fits for 4 groups of PD- spectra, according to A.F.'s LLO 15.11.93. The total number of fits for 4 groups is = 1 + 4 + 7 + 7 = 19 fits. I should stress that selection of LOOKs by A.Fomenko was quiet a job! At present, The results can be seen, if you put * to the next line: LIST 'H01LNS.JF1511MI' There is a possibility, that they will be updated or even improoved, but I show them at Zeitnote to recieve any comments and understanding For any look-file, fits were done for different Emin, at Emax==35GeV. Emin were chousen by the program, as E for Y=Ymax (then Y decreasing) N-of-points varied according to (Emax-Emin), up to minimal = 8 points The dependance of fits on Emax is not yet done ... The 5-th group of data is not yet ready for fitting ... I gess that the next step must be SOME SELECTION of the "BEST fit", searching in dependance on Emin, then the table of fits will be cut. It is evident, that this selection is not evident. E.g., any probabi- lity grater then, say, 1% are equally good, etc. Probably, looking at the "BAD fits" is also useful, that is why I show many variants - look and select ... Independantly of this selection, some COMPARISON of different cases can be made, something is evident - look and compare ... I shall compare and select, either, and shall write later ... The definitions of parameters are as following: Ecorr == (S.L.'s_param_3 - 1.0)*100. (%%) == fitted_correction. Resolution == (S.L.'s_param_4)*100. (%%) == as in Res/sqrt(E)(10%?). Note, especcially: Probability(%); (ep)/(eGas); infER=infiniteERros =>error=0.0; error=* is too big; N-of-events is in kEV. May be useful to look at my LLO 14.11.93. L.Shtarkov, Moscow, 15.11.93 18.11.1993 ==================== H1KFOM === Relation Coefficients Test It were tested last values of Relation Coefficients (07/11/93) with 1/8 of Nelly's sample of ET&PD&no(VC) events which are kept at H1KFOM.EP249B30.ZZZ- cartridge dataset (class K). It were processed 12865 selected monitoring events with contents of TEL1-bank: ET=1 and PD=1 and VC=0. At this sample it were included events (pure LUMI events) from 60998 (epRun#248) up to 61083 (epRun#250) - from 05/09/93 11:57 up to 06/09/93 07:43 - from 32 H1Runs. It were produced the next digits: 1) all sample: Mean = 26.47 Sigma = 0.76 12776 entries 2) |Xet|<6.0cm Mean = 26.50 Sigma = 0.73 9614 entries 3) VCdep<0.2GeV & Mean = 26.47 Sigma = 0.67(!!)6052 entries |Xet|<6.0cm It is really good result! Note that I did not make correction on Photon Shape Shift (0.965 as with last L.N.Shtarkov's fits - see message of H01LNS from 15/11/93). 100th tuple for high mentioned sample is kept at H1KFOM.ALOOKA.R60998 LOOK-data set. Job for producing of 100th tuple is kept at H1KFOM.TSTVCPD9-library: #T248S0 000E0B SAVED BY H1KFOM ON 93/11/18 AT 18:54 19.11.1993 ==================== H01LNS === A SUMMARY AFTER PD-FITS.. A SUMMARY AFTER PD-FITS. 00000101 It is a Summary for PD-fits that were introduced in LLO of 15.11.93. 00000210 1.) A-priory, Fitting in dependence on Emin makes unnecessary the 00000500 Fits at different thresholds. The Table of Fits has confirmed it.00000501 2.) There is a reasonable STABILITY of results in dependence on Emin.00000511 Its possible to reject the BAD Fits: Prob<1% or "INFinite_ERrors"00000512 Its possible to select the BEST Fit: The smallest Emin at Prob>5%00000513 3.) The BAD fits (among the 19 spectra): 00000520 R66913 fg 60,61,62,63 (not fg 50,51,53) - see the Table bellow. 00000530 4.) The BEST fit lines, if it is wanted: put * to the next line: 00000700 LIST 'H01LNS.JF1511BF' 00000800 5.) Some Summary for the Best Fits. 00000901 For any spectrum there is a number of the Best-fit-lines in Table00000902 and the E-shift% varies by ~0.2%, that is inside the error ~0.5%.00000903 Strictly speaking, this needs of averaging, that is not done, and00000904 the results that are following were taken, as the Best, "by eye".00000905 I think this not important, but it is possible to make averaging.00000906 00000908 Run fig prob% E_shift% +/-err Res_B% +/-e 00000909 ----------------------------------------------- 00000910 R60994 fg50 >10% (-3.5 +/- 0.6)% (25 +/- 2)% 00000911 00000912 R64662 fg50 >10% (-2.3 +/- 0.5)% (26 +/- 2)% 00000913 R64662 fg51 >10% (-2.3 +/- 0.5)% (26 +/- 2)% 00000914 R64662 fg52 ~ 5% (-2.4 +/- 0.5)% (26 +/- 2)% 00000915 R64662 fg53 ~ 4% (-2.5 +/- 0.5)% (26 +/- 2)% 00000916 00000920 R66553 fg50 > 5% (+2.6 +/- 0.4)% (26 +/- 1)% 00000930 R66553 fg51 > 5% (+2.6 +/- 0.4)% (26 +/- 1)% 00000931 R66553 fg53 > 5% (+2.8 +/- 0.4)% (26 +/- 1)% 00000932 00000940 R66553 fg60 >10% (+2.9 +/- 0.5)% (24 +/- 1)% 00000950 R66553 fg61 >10% (+2.9 +/- 0.5)% (24 +/- 1)% 00000951 R66553 fg62 >10% (+2.9 +/- 0.5)% (24 +/- 1)% 00000952 R66553 fg63 >10% (+2.9 +/- 0.5)% (24 +/- 1)% 00000953 00000960 R66913 fg50 >10% (-1.5 +/- 0.4)% (24 +/- 1)% 00000970 R66913 fg51 >10% (-1.5 +/- 0.4)% (24 +/- 1)% 00000971 R66913 fg53 >10% (-1.3 +/- 0.4)% (24 +/- 1)% 00000972 00000980 R66913 fg60 <0.0% (-2.0 +/- 0.6)% (20 +/- 3)% 00000981 R66913 fg61 <0.0% (-2.0 +/- 0.6)% (20 +/- 3)% 00000982 R66913 fg62 <0.0% (-2.0 +/- 0.6)% (20 +/- 3)% 00000983 R66913 fg63 <0.0% (-4. +/- 4. )% (30 +/-30)% 00000984 00000985 6.) Some first conclusions from the above Table: 00000989 - Well seen, that 5x=50,6x=60, that is, 1.) is confirmed. 00000991 - That is, for Applications, it's enough to select fg==50. 00000992 - For R66913 6x, prob% is <0.0%, but the results are reasonable. 00000994 - THE MOST INTERESTING - THERE ARE POSITIVE AND NEGATIVE SHIFTS%.00000995 - The Shift Value is from -4% to +2% - don't know, is big or low?00000996 - The Resol(B%) Value is (20 - 25)% - more than 10%, complicated.00000997 00001000 7.) Dependence on the Trigger ... to be added. 00001502 8.) Dependence on the Detector-Structure ... to be added. 00001702 9.) The choice of either (ep), or (eGas) ... to be added. 00001812 00001813 10.) I don't know, whether it's reasonable to make use of fit results.00001814 00001822 PS: This LLO was done a little in a hurry, there is a hope: 00001902 - To add discriptions of the used spectra. 00001903 - To add the conclusions at 7.)-9.) points. 00001904 - To make fits for some other cases of spectrum. 00001912 - Later, to come to more general conclusions. 00002002 L.Shtarkov, Moscow, 18.11.9300002102 19.11.1993 ==================== H1KFOM === new set of PDREC-spectra.. as continuation of 15.11.93 message: 5) H1Lumi's Dedicated Run during #371th epRun (Run66914) H1KFOM.BLOOKQ.R66914 data source: HERA04.H1RAWD.C9304073 (only 1/4 part of Run) -C 114205 events with H1Lumi's standard trigger setting exept s86 = PD with low(2.5GeV) threshold 50th histo - do 'hs(50)=nt(100)(pdrec)@(s86.eq.1.and.pdrec.lt.999) PDREC-spectrum for PD=1 trigger with low (2.5GeV) threshold 41798 entries 51st histo do 'hs(51)=nt(100)(pdrec)@(s86.eq.1.and.pdrec.lt.999.and.pd1.eq.1) PDREC-spectrum for PD(2.5GeV)&PD1 17732 entries 53th histo do 'hs(53)=nt(100)(pdrec)@(s86.eq.1.and.pdrec.lt.999.and.pd2.eq.1) PDREC-spectrum for PD(2.5GeV)&PD2 9685 entries 60st histo do 'hs(60)=nt(100)(pdrec)@(s93.eq.1.and.pdrec.lt.999) PDREC-spectrum for VC=1 trigger 11656 entries 61st histo do 'hs(61)=nt(100)(pdrec)@(s93.eq.1.and.pdrec.lt.999.and.pd.eq.1) PDREC-spectrum for VC&PD trigger 8593 entries 62nd histo do 'hs(62)=nt(100)(pdrec)@(s93.eq.1.and.pdrec.lt.999.and.pd1.eq.1) PDREC-spectrum for VC&PD1 trigger 3649 entries 63rd histo do 'hs(63)=nt(100)(pdrec)@(s93.eq.1.and.pdrec.lt.999.and.pd2.eq.1) PDREC-spectrum for VC&PD2 trigger 2032 entries 70th histo do 'hs(70)=nt(100)(pdrec)@(s86.eq.1.and.pdrec.lt.999.and.vcdep.lt.0.2) PDREC-spectrum for PD-trigger (without shower at VC - 0.2 GeV thresh. 4668 entries -------------------------------------- 71st histo do 'hs(71)=nt(100)(pdrec)@(s86.eq.1.and.pdrec.lt.999.and.vcdep.lt.0.5) PDREC-spectrum for PD-trigger (without shower at VC - 0.5 GeV thresh. 6893 entries -------------------------------------- 72nd histo do 'hs(72)=nt(100)(pdrec)@(s86.eq.1.and.pdrec.lt.999.and.vcdep.lt.1.0) PDREC-spectrum for PD-trigger (without shower at VC - 1.0 GeV thresh. 11220 entries -------------------------------------- 73rd histo do 'hs(72)=nt(100)(pdrec)@(s86.eq.1.and.pdrec.lt.999.and.vcdep.lt.1.5) PDREC-spectrum for PD-trigger (without shower at VC - 1.5 GeV thresh. 15934 entries -------------------------------------- 80th histo do 'hs(80)=nt(100)(pdrec)@(s93.eq.1.and.pdrec.lt.999.and.vcdep.lt.0.2) PDREC-spectrum for VC-trigger (without shower at VC - 0.2 GeV thresh. (only with unefficiencies effect of VC=1) 232 entries 81st histo do 'hs(81)=nt(100)(pdrec)@(s93.eq.1.and.pdrec.lt.999.and.vcdep.lt.0.5) PDREC-spectrum for VC-trigger (without shower at VC - 0.5 GeV thresh. (only with unefficiencies effect of VC=1) 1435 entries 82nd histo do 'hs(82)=nt(100)(pdrec)@(s93.eq.1.and.pdrec.lt.999.and.vcdep.lt.1.0) PDREC-spectrum for VC-trigger (without shower at VC - 1.0 GeV thresh. (only with unefficiencies effect of VC=1) 3609 entries 83rd histo do 'hs(82)=nt(100)(pdrec)@(s93.eq.1.and.pdrec.lt.999.and.vcdep.lt.1.5) PDREC-spectrum for VC-trigger (without shower at VC - 1.5 GeV thresh. (only with unefficiencies effect of VC=1) 5362 entries 20.11.1993 ==================== H1KFOM === Luminosity from VC,ET rate As You remember (see LL6 24.01.93 and 02.02.93 messages of H1KFOM) at first monthes of this year me and V.Andreev had attempts to estimate PLTEFF*ACCC*CrossSectionVisible for alternative Luminosity calculations: It was proclaimed that for VC=1 trigger this digit is 128.1 mb for ET&PD&no(VC) - 4.75mb. It was alived high mentioned tool and it were checked PLTEFF*ACCC* *CrossSectionVisible for VC and ET triggers at the data #248th epRun. The next preliminary digits were defined: 124.4 +- 4.8 - for VC-trigger 28.5 +- 1.0 - for ET-trigger As it was possible to see from behaivour of this digits with time(with supposition that Lumi with ET&PD rate method was measured right) - VC-trigger rate method could overestimate Luminosity value at the start of LumiRun (as seems it exists some VC-trigger rate dependence from high currents - my be effective threshold value was changed may be something else). Luminosity values calculated with ET-trigger rates and with 28.5 mb as PLTEFF*ACCC*CrossSectionVisible put more or less the same Luminosity values behaivour with time as ET&PD-method. 22.11.1993 ==================== H1KFOM === HERA Shut Down ........... Today morning at 06:00 last HERA e-beam'93 was dumped (Imax was near 30 mA). It was started HERA Shut Down (up to 8 April 94). 22.11.1993 ==================== H1KFOM === Lumi from VC,PD,ET,EP!V... It was tested posibility of Current Luminosity measurement with using of Rate/Current methodics with another H1Lumi trigger bits and trigger elements: VC,PD,ET,ET&PD&no(VC). It were used contents of LRTS-banks which are kept at BCKUPTCL-cartridges (100 cartridges at all H1DataTaking'93). It were selected only 2 epRuns (248th and 308th) from high mentioned data sets. These epRuns are kept at the next cartridges: F11JOL.H01.BCKUPTCL.R060989.R061082 (epRun#248 - near 240 min) F11JOL.H01.BCKUPTCL.R064646.R064785 (epRun#308 - nbear 310 min) epRun #248 - H1Runs from 60994 up to 61018 (from 05/09/93 11:32:40 up to 05/09/93 16:16:03) epRun #308 - H1Runs from 65658 up to 64680 (from 08/10/93 03:38:26 up to 08/10/93 08:47:43) It were used the same methodics as it was proclaimed 24.01.93 and 02.02.93 at LL6-messages of H1KFOM. 1) supposition that Luminosity current value which is kept at LRTL- bank is right; 2) calculation of the values for all K: A(K) = (TotalRate(K) - PilotsRate(K)*(IeTotal/IePilots))*10.**27 where: TotalRate - sum of rates from collided and e-pilot bunches PilotsRate - sum of rates from pilot bunches IeTotal - sum of e-bunch currents from collided and e-pilot bunches IePilots - sum of e-bunch currents from e-pilot bunches K = VC | PD | ET | ET&PD | ET&PD&no(VC) 3) calculation of values for all K: B(K) = 1./(LRTLBANK(232)*10.**25/A(K)) where: B(K) - estimated values of PLTEFF*ACCC*CrossSectionVisible (at mb units) K = VC | PD | ET | ET&PD | ET&PD&no(VC) 4) filling of histos (for estimation of Mean and Sigma) and vectors (for estimation of behaivour with time of these values) VEC(1) = time (at sec) VEC(2) = B(K) CALL SVEC (7X0000,1,VEC) IF(B(K).LT.AMX(K).AND.B(K).GT.AMN(K)) CALL SHS (7X0001,0,B(K)) where: AMN(K)=0. AMX(K)= 200 or 100 or 50 or 10 for different K X = 0 for VC-rates X = 1 for ET-rates X = 2 for ET&PD-rates X = 3 for ET&PD&no(VC)-rates X = 4 for PD-rates After processing of high mentioned epRuns (with using of LRTS,LRTL and TCUR -bank contents) it were produced the next LOOK-data set with high mentioned histos and vectors: 1) for epRun#248: Output file: H1KFOM.TLOOK1.R60994 VEC 500000 1 878 t(sec); f\_{WC}(Hz) VEC 510000 1 878 t(sec); f\_{ET}(Hz) VEC 520000 1 878 t(sec); f\_{LR}(Hz) VEC 530000 1 878 t(sec); f\_{LP}(Hz) VEC 540000 1 878 t(sec); f\_{PD}(Hz) VEC 700000 1 878 t(sec); (PLRTF*SIGMA*ACCC)mb VEC 710000 1 878 t(sec); (PLRTF*SIGMA*ACCC)mb VEC 720000 1 878 t(sec); (PLRTF*SIGMA*ACCC)mb VEC 730000 1 878 t(sec); (PLRTF*SIGMA*ACCC)mb VEC 740000 1 878 t(sec); (PLRTF*SIGMA*ACCC)mb HS 700001 0 844 PLTEFF*ACCC*CrossS(mb); PLTEFF*ACCC*Cr... HS 710001 0 844 PLTEFF*ACCC*CrossS(mb); PLTEFF*ACCC*Cr... HS 720001 0 844 PLTEFF*ACCC*CrossS(mb); PLTEFF*ACCC*Cr... HS 730001 0 844 PLTEFF*ACCC*CrossS(mb); PLTEFF*ACCC*Cr... HS 740001 0 846 PLTEFF*ACCC*CrossS(mb); PLTEFF*ACCC*Cr... 2) for epRun#308: Output file: H1KFOM.TLOOK1.R64658 VEC 500000 1 1373 t(sec); f\_{WC}(Hz) VEC 510000 1 1373 t(sec); f\_{ET}(Hz) VEC 520000 1 1373 t(sec); f\_{LR}(Hz) VEC 530000 1 1373 t(sec); f\_{LP}(Hz) VEC 540000 1 1373 t(sec); f\_{PD}(Hz) VEC 700000 1 1373 t(sec); (PLRTF*SIGMA*ACCC)mb VEC 710000 1 1373 t(sec); (PLRTF*SIGMA*ACCC)mb VEC 720000 1 1373 t(sec); (PLRTF*SIGMA*ACCC)mb VEC 730000 1 1373 t(sec); (PLRTF*SIGMA*ACCC)mb VEC 740000 1 1373 t(sec); (PLRTF*SIGMA*ACCC)mb HS 700001 0 1257 PLTEFF*ACCC*CrossS(mb); PLTEFF*ACCC*Cr... HS 710001 0 1255 PLTEFF*ACCC*CrossS(mb); PLTEFF*ACCC*Cr... HS 720001 0 1315 PLTEFF*ACCC*CrossS(mb); PLTEFF*ACCC*Cr... HS 730001 0 1262 PLTEFF*ACCC*CrossS(mb); PLTEFF*ACCC*Cr... HS 740001 0 1323 PLTEFF*ACCC*CrossS(mb); PLTEFF*ACCC*Cr... Some digits from high mentioned histos and vectors: #248th epRun: PLTEFF*ACCC*CrossSectionVisible (mb) VC PD ET ET&PD ET&PD&no(VC) Mean value 124.5 63.5 28.7 27.6 5.3 Sigma 4.8 0.5 0.5 0.5 0.4 #308th epRun: PLTEFF*ACCC*CrossSectionVisible (mb) VC PD ET ET&PD ET&PD&no(VC) Mean value 175.7 80.8 32.5 29.0 5.3 Sigma 4.2 0.2 0.5 0.3 0.4 It were fixed difference at time dependence of B(K)-values for different K and different epRuns.As seems B(K) for K=ET and fo K=ET&PD are more or less stable with time - all B(K) with VC or no(VC) trigger element are unstable with time (as seems due to changing of effective threshold with high rates). B(K) for PD-trigger bit is unstable with time too but there is less degree of dependance as for VCtrigger bits. #248th epRun: PLTEFF*ACCC*CrossSectionVisible (mb) VC PD ET ET&PD ET&PD&no(VC) Max value 156.0 70.0 30.5 28.5 5.5 Time of max t0 t0 t0 t0 tmax Min value 124.0 62.5 28.5 27.5 5.2 Time of min tmax t0+20min.t0+20min. t0+20min. t0 #308th epRun: PLTEFF*ACCC*CrossSectionVisible (mb) VC PD ET ET&PD ET&PD&no(VC) Max value 181.0 79.0 32.8 29.0 6.0 Time of max tmax tmax t0 t0 t0 Min value 145.0 76.0 32.2 29.0 5.0 Time of min t0 t0 tmax tmax tmax It is needed to process some set of ep-Runs for study of high mentioned dependences and for final choice of alternative Rate/Current Lumi Measurements (PD or ET or something else). Examples of jobs are kept at H1KFOM.TSTVCPD9-library at data sets: #C211193 00470B SAVED BY H1KFOM ON 93/11/21 AT 23:56 #D211193 004A01 SAVED BY H1KFOM ON 93/11/21 AT 23:56 22.11.1993 ==================== H1KFOM === Lumi from VC,PD,ET,EP!V... Some digits for comparison of two neigbour epRun(with B=0 and with B>0) #271st epRun (last epRun with B=0) 18/09/93 07:04-09:31 H1Runs 62549-62551 time - near 75 min. Ie = 15.2 mA --> 10.7 mA Ip = 10.95mA --> 10.58mA Lumi=(5.5--->3.4)*10**29 cm-2s-1 beam position - and near 0.0 cm PLTEFF*ACCC*CrossSectionVisible (mb) VC PD ET ET&PD ET&PD&no(VC) Mean value 137.1 82.5 40.9 34.4 7.6 Sigma 2.5 1.8 0.8 0.5 0.3 Max value 143.0 82.5 40.9 34.4 7.8 Time of max t0 t0 t0 t0 tmax Min value 135.0 82.5 40.9 34.4 7.0 Time of min tmax tmax tmax tmax t0 Rates (kHz) VC PD ET ET&PD ET&PD&no(VC) Rate (t0) 200.0 125.0 80.0 52.0 11.5 Rate (tmax) 115.0 75.0 46.0 30.0 7.0 #272nd epRun (first epRun with B>0 after pause on B=0) 18/09/93 18:21-02:43 H1Runs 62572-62609 time - near 455min. Ie = 11.7 mA --> 4.2 mA Ip = 11.1 mA --> 10.8 mA Lumi=(4.6--->1.8)*10**29 cm-2s-1 beam position - and near -0.5 cm PLTEFF*ACCC*CrossSectionVisible (mb) VC PD ET ET&PD ET&PD&no(VC) Mean value 122.3 81.0 31.0 28.9 6.9 Sigma 1.8 1.1 0.5 0.2 0.1 Max value 132.0 81.0 31.0 28.9 6.9 Time of max t0 tmax t0 t0 tmax Min value 122.0 79.0 31.0 28.9 6.5 Time of min tmax t0 tmax tmax t0 Rates (kHz) VC PD ET ET&PD ET&PD&no(VC) Rate (t0) 89.0 53.0 34.0 19.2 4.4 Rate (tmax) 29.0 20.0 11.5 6.5 1.6 23.11.1993 ==================== F11LEV === Minutes of the LPI meeting #################################################################### ## ## ## Minutes of the LPI group meetings at DESY, 17,19,21-Nov-1993 ## ## ## #################################################################### (S. Levonian, 22-Nov-1993) Participants: Baranov, Belousov, Eliseev, Fomenko, Gorbov, Kotelnikov, Lebedev, Levonian, Rusakov, Smirnov, Soloviev, Usik. Agenda: 1) Lumi system operation in 93 2) Status of the different methods of lumi measurements 3) Schedule for the winter shutdown 4) Organization, manpower and visits in 94 Note: Not all questions which need discussion were discussed, but only most important and urgent ones. Therefore we decided to meet again 27/28 December in Moscow Exact time and place to be fixed and announced later. Lumi system operation in 93 --------------------------- In general, the system worked reasonably well. However a lot of prob- lems were encountered, preventing us from the required performance. A calibration is still (during three years!) the most crucial and not completely solved item. It directly influences two major points: o systematic error in the absolute luminosity o eTAG photoproduction sample quality Improvement compared to 92: the use of the permanent online calibration to take into account CC variation in time (mainly because of the different loading rates). Still unsolved: a reliable determination of the relation coefficients (RC) between lumi and photoproduction branches and automatic proce- dure for the eventual global correction of ET and PD energies. Two methods have been proposed and tried for RC determination: a) the use of the real lumi events taken through the photoproduction branch (A.F.) and b) direct measurements using a light source or simply the generated electronic signals (I.Sh.) Both methods have their advantages and drawbacks; therefore both must be brought to the end. Method a) needs a lot of monitoring events, or a special dedicated lumi runs; it also is able to define only a small fraction of RC (where the statistics is available). On the other hand, it gave so far the correct results, which is most important. Method b) is technically easier, it does not need a lot of real events and thus can be done at any time. It also allows to obtain RC for all cells. The practical results however, was not stable (and sometimes simply wrong). It seems, in the present form a systematical error of the method is too big. There are obvious ways to improve this method and this should be tried during the shutdown. The only reliable method to cross check an absolute energy scale and to define a possible global shift between ET and PD arms which has been proposed and tried offline is a gamma-energy distribution fit. It was proven to work, however for the next year it must be o turned to the automatic procedure running in the L4, and o improved to take into account a multiple photon events The problems/errors in the luminosity measurements can be classified in the following way: a) short in time (20-30 sec) but large in value lumi spikes (ca. 2 nb-1 each) b) bad measurements because of the wrong or absent e/p currents c) the integrated lumi is not erased at the beginning of the H1 runs (additional systematic error from this is of the order of 0.5% in average, but can be up to 20-50% for very short runs) d) not zeroed integrated lumi in LRTL at the end of ep-collisions (this already was corrected in October) All these counting errors (exept c) were corrected by my tool running in the IBM logging job, so the values in the database are corrected for these "counting" errors, but not the values in the raw data (LRTL,LRTN banks) Another "feature" of the H1 lumi monitor, which the outside people particularly did not like, was that the measurements for H1 physics (Fomenko) were often different from those for HERA (Usik). This MUST be avoided in future! F. and U. should sit together and define as common procedure as possible, taking the best from the two. Finally, a real life has demonstated, that the integrated luminosity per H1 run is not sufficient for the analysis (the H1 detector is not yet stable enough for that). Therefore a different possibililites were re-discussed once again in the close expert's meeting two weeks ago. We decided to create another bank(s) in the database in addition to the existing LH1R, which should contain the luminosity information and the detector status bits for each 10 sec. All necessary input data are available (100 cartridges with KEEP information) and this has to be done as soon as possible. The systematic error from the background subtraction procedure was directly checked several times, on the occasions when the p-beam was lost and only electrons left in HERA. It was always better than 1.2%, and 0.8+-0.2% in average. There were other well known problems, like o 90 nb-1 of e-tag data loss due to the wrong LUMI branch timing in August o HV trips in ET and PD which in addition were discovered too late (still rather rare, but expected more frequently in 94, due to 5-10 times higher rates) o bad behaviour of Veto counter in October (degraded efficiency and wider noise distribution) o saturation of the ET and PD counting at very high rates, pointing to the preamplifiers problem (observed 2 times, R(ET*PD) satruration limit was about 400 kHz) o not perfect lumi data presentation for HERA, and also not stable enough line connection and software (this has improved at the end) Status of the different methods of lumi measurements ---------------------------------------------------- Summarizing shortly, the status is bad. The LUMM bank was proposed in May. There were no feedback on that, and no real attempt to use LRTS information to calculate luminosity by PD-rate method. Only after many additional requests some activity has started, but again, without clear result and definite conclusion. Here the tasks were defined and I can repeat it again: o to compare different methods of lumi measurements using LRTS data (ratio to the basic ET*PD method, short and long term stability etc.) o define what is still missing for the implementation of these methods: ET*PD for different |X(ET)| cuts, PD-rate, gamma E- flux o compare systematic errors of all methods o write the code (or at least an algorithm) for online, which should create finally LUMM bank; just the fitting of the PD threshold from time to time is not sufficient! The 93 data show, that our ET acceptance was changing several times. Some of these changes in the ET-energy shape was due to the bad calibration only, but there were also "true" variations. Therefore, it was (and perhaps will be) impossible to follow all these variations in Monte Carlo with the precision better than 5-10%. We can however to extract the ET-acceptance for the lumi events from the data. This was tried recently and gave a good result. Here again an absolute calibration of PD is extremely crucial! Having solved this problem we can reduce a systematic error due to ET-acceptance from the present 4.5% to 1-2% and thus to reach an overall error better than 3-4% (compared to 7% in 92). To conclude, the present experience suggests to use PD-method (either rate, or flux) as a basic one for most of the physics analyses, while our present ET*PD method seems to remain the best for e-tag photopro- duction physics (because ET-acceptance systematics to large extent cancells in the e-tag data). It is not excluded however, that the coincidence method will give the best of all result, provided the PD absolute calibration is done within 1-2% in the whole energy range. Schedule for the winter shutdown -------------------------------- A task list for the shutdown (which has started today) was discussed in details and will be reviewed in December in Moscow again. A) Detectors (Soloviev, Malinovski, Gorbov, Usik, Eliseev, ...) ------------ Here the main aim is to prepare the system to the stable comissioning at 5 times higher average rates (and 10-20 times higher peak rates). Everything in this part MUST be ready by 1st of April. o Veto Counter upgrade (Gorbov, Soloviev, Malinovski, Usik + ?) (see also written proposal from A.Usik - not available at IBM) - replace 2 PM in the old VC and prepare this VC to the running as a spare to the new VC - investigate used PM at Troitsk (Belousov + ?) - build and test new VC at DESY: final design to be choosen by L.Gorbov after consultations with other experts; main features: at lest 3 PM (2 veto + 1 spectrometric); PM's should be removed from the median plane - decide on the final position of VC w.r.t. PD and on the shielding (collimators, "anti-synchrotron" tube); proposals requiring significant work (e.g. modification in the VC-table) should be proven by the quantitative arguments! - decide on the "devider" scheme o Preamplifiers (Sheviakov, Eliseev, Smirnov?) - decide on the main scheme, but prepare prototypes for both proposals made by Sheviakov and Eliseev - mass production and testing at DESY - final adjustment at the beam B) Trigger (Sheviakov) ---------- Here a big progress was made compared to 92 status. Main complaints to Egor: missing documentation and missing spare electronics. o prepare the documentation o improve and make usable ET+PD energy sum o decide on the best definition of the VC-bit (my opinion: (V1.and.V2).or.VS) o prepare several "standard" settings for the 8 lumi trigger elements; the main setting should allow for: - lumi (several methods) - eTAG physics - calibration of ALL detectors (ET,PD,VC) o prepare the production of the spare trigger electronics (which might be eventually done by somebody else, if I.Sh. has no time for that) C) Online s/w (Fomenko) ------------- See also A.F. task list in the lumi logbook, part 7. o fix all known bugs/problems in the L estimate (for LRTL bank) o implementation of the alternative methods ==> LUMM bank o set up a general scheme for the detector calibration (ET,PD,VC) including all steps (+ Levonian) o optimise a 2nd method of RC determination (+ Soloviev, Gogotidze Sheviakov) o a special task preparing all necessary histograms for L4/L5 o improve a standard lumi system operating (see a proposal for the panel with 4 software buttons, lumi logbook 28/09/93) this should be fully automatic! o one screen/menu collecting all useful plots in one place (like beam profile histos, 2d coordinate distr., energy spectra, trigger efficiencies, histograms sending to L4/L5 etc.) o include the VC veto channel(s) into the readout in order to make possible further offline VC trigger verification and realistic simulation D) Slow Control, Data transfer and presentation (Usik) ----------------------------------------------- o improve the stability of the H1-HERA connection (finish already started work) o improve the presentation quality (scales, user friendliness,...) o add more options via the user menu o prepare the possibility to display different lumi scalers (lumi measurements by different methods) o H1 lumi for ZEUS and ZEUS lumi for H1 o better alarm/warning handling (e.g. HV trips) o check, that the polarization data arrive into the H1 data stream E) Offline tasks (Levonian, others are welcome) ---------------- o prepare 93 data reprocessing (already done by yesterday) - LREC changes (VC energy is added to LPDR bank as a second row, additional calibration bank is used to correct for the global energy shift between ET and PD, LRTE bank is no longer used, a bit corrected coordinate rec. for PD distinguishing the cases with no, low and high VC-dep) - new calibration banks into database; update old ones - a special FPACK class (24) is reserved for all interesting "lumi" events (all events having at least one LUMI trigger element set in TEL1 bank, or having non-zero reconstructed energy in ET or PD); this should allow a fast selection of "our" events from the common files/tapes o LH1T bank and tool (lumi measurements per each 10 sec) - to meet the requirements of the physics analysis o L4/L5 software for - luminosity'94 - calibration'94 o prepare a set of efficient sample jobs for common use o prepare a short documentation on the existing offline tool which is related to the LUMI system o try to improve the coordinate reconstruction in ET/PD (Shtarkov) Organization, manpower and visits in 94 --------------------------------------- Trying to understand what prevents our group from more efficient work, the following points of general concern were mentioned, but the discussion was carefully avoided: o still too few people contribute (do a REAL work) o the "key" experts have no (or not enough) motivation compared to others (in best traditions of a standard "soviet" approach) o insufficient responsibility (e.g. only 3 people reacted on the message about shutdown plans) o only few people are able to perform a well defined task,producing a clear result o even less people are able to formulate necessary tasks themselves o lack of the professionality, preventing some tools from the general use, and reducing the performance of those which must be used, because nothing else exists ==> Conclusion: clearly, a better organization is needed. What could be improved immediately? Proposals for further discussion: o better information exchange - clean lumi logbook (split lumi run summaries into a separate automatically filled file; make your messages short and clear) - do we need other files? (see proposal of L.N.S.) - a permanently updated list of tasks with names of responsibles o finish each task either with the short LPI note, or H1 internal note; this helps to formulate the results better and is useful for future o regular reports from ALL LPI members - on the H1 meetings (important contributions) - on LPI seminars (mandatory for ALL, not only always the same 4-5 people, but also for bosses, and some of our visitors coming often to DESY, but existing only virtually for us: nobody knows why they came, what they did, may be something very useful? then tell us, don't keep your achievements in secret) LPI quota in 1993 was 60 man*months, paid by DFG. On top of this, ca. 12 months were added from DESY budget. The information, how it was used can be obtained from S.Rusakov (or H.Kleinebenne). The same amount (60mm) was requested for 1994. No answer from DFG yet. Nevertheless, a preliminary estimate was made assuming this value. A.S.Belousov gave the following list of names and associated terms of visits for the next year. Note, that not all LPI members were mentioned. The list is not final and does not contain exact dates. This have to be discussed once again in Moscow, Dec.27/28. Be prepared. Andreev ....... 4 months Eliseev ....... 3 Fomenko ....... 6 Gogitidze ..... 1 Kotelnikov .... 4 (work in the muon group) Malinovski .... 3 Sheviakov ..... 6 Smirnov ....... 3 (backward upgrade) Soloviev ...... 8 Usik .......... 6 Vazdik ........ 4 ==> Total of 48 months 23.11.1993 ==================== H01LNS === FIT 5 GROUP OF PD-SPECTRA In continuation of the PD-Fitting that was reported in some above LLO, after A.Fomenko's request, I did 20.11.93 PD-fits for the 5)th group of PD-spectra, according to A.F.'s LLO xx.11.93, See AF's description! At present, to gain the time and LLO-space, the full Table of Results for the all 5)-fits can be seen, if you put * to the next line: LIST 'H01LNS.JF2011MI' There is a possibility, that it will be updated or even improoved. A draft conclusion from the new fits is that nothing news appiered, particualarly, selection by "no-shower in VC" does not change the results significantly, and does not decrease the Resolution to ~10%. I hope yet to make more general Conclusions from the reported fits. Sorry for a delay, L.Shtarkov, 23.11.93, Moscow. 23.11.1993 ==================== F11LEV === Remark on PD-fitting...... A small remark on the previous message of L.N.S. A selection of VC=0 events (or E(VC)<1 GeV in VC=1 sample) DO give a better resolution. It is seen "by eye". I tried this in presence of A.Fomenko, so he can confirm this observation. A fit may not be always stable, but I've got on the events from the run 66914 the PD resolution of 8%/sqrt(E)! 23.11.1993 ==================== H01LNS === ON USE OF PD-FIT-E-SHIFTS I AM STRONGLY WORRIED ABOUT POSSIBLE USES OF THE PD-FIT-ENERGY-SHIFTS In particular, I have issued some figures which were srongly based on 4 (5) sets of LOOKs that were SELECTED and reported by A.Fomenko. Unfortunately, the trigger programs are highly dangerous and my LLO questions on this subject were not at all answered. That is why I am sure, IT IS NECESSARY BEFORE ANY USES, TO CHECK SELECTION OF EVENTS FOR FITS AND TO TEST THE FIT-RESULTS VIA SOME MORE GENERAL PROCEDURE. As for Selection of events, now I am trying to test Stability of fits in dependence of triggers etc., it will take me a few days, not hours As for a General test, I PROPOSE the following, that must be done strongly for the Sets that have been selected and fitted: - Check the position of the sum of PD+ET Energy (? not sure), - Correct the CCs according to the reported E-Shifts, - Make Re-Reconsctruction of these very Sets, - Make the Fits again, now for the Re-Reconstructed Sets, - Check that the New Shifts have come to the Zeroes, - Check that the sums of PD+ET have come to the "right"(?) position, - ? This procedure can be different in details, I am not an expert. - ? My Estimate for the necessary time: till Friday, better Monday=29.11.93 L.Shtarkov, 23.11.93, Moscow. 24.11.1993 ==================== H1KSOL === VETO dismounting ......... Veto (water) counter was taken from its place in photon arm this morning. Here below there is usefull information about cabling and position of it: Cabling - HV - #26 VETOv (most distant from e beam) - #27 VETOs (closest to the beam) Signal - #39 VETOv - #38 VETOs The hight of the upper Veto level relative to the table plane - 184mm. Other geometrical points are on the table. 24.11.1993 ==================== H1KSOL === Dosimeters ............... Dosimeters were taken from the following places: #1 - in front of element #21 of ET, #2 - on the side surface of the el#21 near e beam on the table #3 - back surface of the VETO counter #4 - in the center of the PD detector on the table Sorry but dosimeter # 2 was lost during transportation from HN to DESY. 25.11.1993 ==================== H1KFOM === Data set with sel. events As addition to 13.11.93 message - for today it was appeared new data set (so-called 5th sample of VC=1 events). This data set must be ready up to start of reprocessing. But reprocessing was started (as I know 23.11.93) from Run 62575 (#272nd ep-collisions) when field at Main Solenoid of H1 was ON again after September'93 pause. Due to some problems with FPACK the selection of VC=1 events from MONITRIG-cartridges was delayed on almost one week. Pavel Binko did not decide problem but found some set of FPACK-options with using of which all is OK. Yu.Soloviev, Nelly and me put all his(her,my) day quotas on selection jobs and now we have first release of so-called 5th data sample. Due to absence of any LOG-information (up to Run63666) about at which MONITRIG-cartridge is kept which H1Run (now this information is preparing by J.Olsson due to our and somebody else asking) we started selection from Run62633 (not 62575 as S.Levonian asked) - but as seems these events more useful for GlobalShift defining for H1Runs range 62575-64642 then events from so-called Nelly's sample (60990-61800). It were produced (yesterday and today) 7 cartridges: H1KFOM.MTP2.B29.L9301881.A00 H1KFOM.MTP2.B29.L9301899.A00 H1KSOL.MTP2.B29.L9301924.A00 H1KSOL.MTP2.B29.L9301949.A00 H1KSOL.MTP2.B29.L9301974.A00 H01GOG.MTP2.B29.L9302075.A00 H01GOG.MTP2.B29.L9302100.A00 All events from these cartridges were pressed into single one: H1KFOM.EP272B29.A00 K13784 78048081 00001 ACS-E-C 12625 events of VC=1 trigger (Runs 62633-63789) Examples of jobs for selection and pressing are kept at H1KFOM.TST- VCPD8-library at the next data sets: $NL29F1 003311 SAVED BY H1KFOM ON 93/11/23 AT 18:05 - sel. $PRESS5B 004712 SAVED BY H1KFOM ON 93/11/25 AT 19:12 - press. 25.11.1993 ==================== H1KFOM === new set of PDREC-spectra.. as continuation of 19.11.93 message: (first attempt - no large statistics !!!!) 6) So-called 5th sample of VC=1 monitoring events from Runs62633-63789 H1KFOM.BLOOKQ.R62633 data source: H1KFOM.EP272B29.A00 K13784 78048081 00001 ACS-E-C 12625 events of VC=1 trigger (Runs 62633-63789) 5005 entries into 100th tuple for reconstruction it were used Relation Coefficients (07.11.93) and CCvcsp from 13.11.93 (for Nelly's sample). CCvcsp = 4.436 Mev/FADCcount ---------------------------- 60st histo do 'hs(60)=nt(100)(pdrec)@(s93.eq.1.and.pdrec.lt.999) PDREC-spectrum for VC=1 trigger 4584 entries 61st histo do 'hs(61)=nt(100)(pdrec)@(s93.eq.1.and.pdrec.lt.999.and.pd.eq.1) PDREC-spectrum for VC&PD trigger 4441 entries 80th histo do 'hs(80)=nt(100)(pdrec)@(s93.eq.1.and.pdrec.lt.999.and.vcdep.lt.0.2) PDREC-spectrum for VC-trigger (without shower at VC - 0.2 GeV thresh. (only with unefficiencies effect of VC=1) 4 entries 81st histo do 'hs(81)=nt(100)(pdrec)@(s93.eq.1.and.pdrec.lt.999.and.vcdep.lt.0.5) PDREC-spectrum for VC-trigger (without shower at VC - 0.5 GeV thresh. (only with unefficiencies effect of VC=1) 57 entries 82nd histo do 'hs(82)=nt(100)(pdrec)@(s93.eq.1.and.pdrec.lt.999.and.vcdep.lt.1.0) PDREC-spectrum for VC-trigger (without shower at VC - 1.0 GeV thresh. (only with unefficiencies effect of VC=1) 377 entries 83rd histo do 'hs(82)=nt(100)(pdrec)@(s93.eq.1.and.pdrec.lt.999.and.vcdep.lt.1.5) PDREC-spectrum for VC-trigger (without shower at VC - 1.5 GeV thresh. (only with unefficiencies effect of VC=1) 886 entries 26.11.1993 ==================== H01LNS === NEW FIT of the 5-th GROUP The new 5) group of PD-spectra was fitted, RUN 62633, according to A.Fomenko's LLO of 25.11.93. Description are in A.F.'s LLO At present, to gain the time and the LLO-space, The full Table of Fit-Results for this portion of 5)-fits can be seen, if you put * to the next line: LIST 'H01LNS.JF2611MI' There is a possibility, that A.F.'s data will be updated, then the fits will be re-done, the Table will be replaced. SEE 27.11.93 Comment: look at the header-lines "l-text1,4: ..." = Selection by A.F. Comment: look at the last columnes, Entr is shown for the Fit-interval. A draft Conclusion from the above cited Table of results: NONE of the fits were fully sucsessful, because they have resulted in 1) Probability(Chi2) < 0.1%, or 2) infinite-Errors, or 3) Errors>100%. Nevertheless the numerical results for E-Shifts & Resolut can be seen, if somebody would prefer to forget of 1-3) then see via the above LIST. I have not yet analyzed the causes for fits failure, should be trivial. L.Shtarkov, 26.11.93, Moscow. 26.11.1993 ==================== H1KFOM === Data set with sel. events It was created new data set with selected VC=1 events from Run62633 up to Run64102 or from epRun#273 up to epRun#300 or from 19/09/93 16:17:50 up to 01/10/93 19:57:10. It was possible due to last Nelly's activity with selection of VC=1 events. So we have data set H1KFOM.EP273B29.A00 K14483 78048081 00001 ACS-E-C 21998 events of VC=1 trigger (Runs 62633-64102) After Run 62773 (see message of H1KFOM at LL7 from 21/09/93) it were introduced PD1 and PD2 -triggers - it means that it will be possible to create "large" set of PDREC-spectra for fits and its cross-check (from epRun#274). 26.11.1993 ==================== H1KFOM === new set of PDREC-spectra.. as continuation of 25.11.93 message: (second attempt - a little more statistics !!!!) 6) So-called 5th sample of VC=1 monitoring events from Runs62633-64102 H1KFOM.DLOOKQ.R62633 data source: H1KFOM.EP273B29.A00 K14483 78048081 00001 ACS-E-C 21998 events of VC=1 trigger (Runs 62633-63789) 8820 entries into 100th tuple for reconstruction it were used Relation Coefficients (07.11.93) and CCvcsp from 13.11.93 (for Nelly's sample). CCvcsp = 4.436 Mev/FADCcount ---------------------------- 60st histo do 'hs(60)=nt(100)(pdrec)@(s93.eq.1.and.pdrec.lt.999) PDREC-spectrum for VC=1 trigger 7993 entries 61st histo do 'hs(61)=nt(100)(pdrec)@(s93.eq.1.and.pdrec.lt.999.and.pd.eq.1) PDREC-spectrum for VC&PD trigger 7778 entries 62nd histo - only for events with RUN > 62773 do 'hs(62)=nt(100)(pdrec)@(s93.eq.1.and.pdrec.lt.999.and.pd1.eq.1) PDREC-spectrum for VC&PD1 trigger 4108 entries 63rd histo - only for events with RUN > 62773 do 'hs(63)=nt(100)(pdrec)@(s93.eq.1.and.pdrec.lt.999.and.pd2.eq.1) PDREC-spectrum for VC&PD2 trigger 2399 entries 80th histo do 'hs(80)=nt(100)(pdrec)@(s93.eq.1.and.pdrec.lt.999.and.vcdep.lt.0.2) PDREC-spectrum for VC-trigger (without shower at VC - 0.2 GeV thresh. (only with unefficiencies effect of VC=1) 4 entries 81st histo do 'hs(81)=nt(100)(pdrec)@(s93.eq.1.and.pdrec.lt.999.and.vcdep.lt.0.5) PDREC-spectrum for VC-trigger (without shower at VC - 0.5 GeV thresh. (only with unefficiencies effect of VC=1) 77 entries 82nd histo do 'hs(82)=nt(100)(pdrec)@(s93.eq.1.and.pdrec.lt.999.and.vcdep.lt.1.0) PDREC-spectrum for VC-trigger (without shower at VC - 1.0 GeV thresh. (only with unefficiencies effect of VC=1) 602 entries 83rd histo do 'hs(82)=nt(100)(pdrec)@(s93.eq.1.and.pdrec.lt.999.and.vcdep.lt.1.5) PDREC-spectrum for VC-trigger (without shower at VC - 1.5 GeV thresh. (only with unefficiencies effect of VC=1) 1515 entries 27.11.1993 ==== H01LNS ===== FITS FOR NEW-NEW 5) GROUP OF PD-SPECTRA == The NEW-NEW 5) group of PD-spectra was fitted, RUN 62633, according to A.Fomenko's LLO of 26.11.93. Description are in A.F.'s LLO At present, to gain the time and the LLO-space, The full Table of Fit-Results for this portion of 5)-fits can be seen, if you put * to the next line: LIST 'H01LNS.JF2711MI' Watch that this Table must be used istead of the JF2611MI. There is a possibility, that A.F.'s data will be updated, then the fits will be re-done, the Table will be replaced again. Comment: look at the header-lines "l-text1,4: ..." = Selection by A.F. Comment: look at the last columnes, Entr is shown for the Fit-interval. A draft conclusion from both the cited Tables of results are the same: NONE of the fits were fully sucsessful, because they have resulted in 1) Probability(Chi2) < 0.1%, or 2) infinite-Errors, or 3) Errors>100%. Nevertheless the numerical results for E-Shifts & Resolut can be seen, if somebody would prefer to forget of 1-3) then see via the above LIST. I have not yet analyzed the causes for fits failure, should be trivial. L.Shtarkov, 27.11.93, Moscow. 27.11.1993 ==== H01LNS ===== SELECTION OF "LUMI-EVENTS", again ? ====== For God's sake, people, write, please, to LLO: THE MAIN CONDITIONS THAT ARE BEING USED FOR SELECTION OF "LUMI-EVENTS" (Events for a possible PD-fit; 1)-5) groups reported by A.Fomenko, etc) 3 months ago, at LLO7 22.08.93, I asked for the following Conditions: 1)Bank's-Name 2)Variable's-Name-in-Bank 3)Subtrigger's-Word-Number 4)Raw-or-Actual-Trigger 5)Trigger-Signals-of-H1 etc.,less significant. I have not found a definite answer - Sorry if I have missed an answer. Now I repeat these questions, because of difficulties in understanding of PD-spectra, Energy-Shifts, and CC-corrections(!) (at least for me). Just one example: It is usially written like "VC-trigger Selection...". But, for VC some questions arrised: Is it s93 or VC-signal from Lumi? (these are different in many cases); Was the VC-threshold used or No? Was the Flag of sucsessful reconstruction used or Not? At different answers the selected PD-spectra are different qualitatively, in shape. The other problem: it is highly difficult to find in LLO the TEMPORARY, NOT STANDARD ASSIGMENTS FOR THE LUMI_TRIGGER_BITS, E.G., s86 -> pd instead of vc, "super-low-threshold", etc. ?? I think, the assigments must be reflected in LUTR-bank, but not sure, Its dangerous for all H1, I PROPOSE to make some automatic memoryzation I seem to understand that LLO is a little overflown, but the above asked information is highly important for PD-fits and CCs L.Shtarkov, 27.11.93, Moscow. 27.11.1993 ==== H1KFOM ===re:SELECTION OF "LUMI-EVENTS", again ? ====== I can make attempt to answer on L.N.S.'s questions at last message at LLO (from 27.11.93 "...For God's sake..."). >For God's sake, people, write, please, to LLO: >THE MAIN CONDITIONS THAT ARE BEING USED FOR SELECTION OF "LUMI-EVENTS >(Events for a possible PD-fit; 1)-5) groups reported by A.Fomenko, etc > >3 months ago, at LLO7 22.08.93, I asked for the following Conditions: >1)Bank's-Name 2)Variable's-Name-in-Bank 3)Subtrigger's-Word-Number >4)Raw-or-Actual-Trigger 5)Trigger-Signals-of-H1 etc.,less significant >I have not found a definite answer - Sorry if I have missed an answer. >Now I repeat these questions, because of difficulties in understanding >of PD-spectra, Energy-Shifts, and CC-corrections(!) (at least for me). "LUMI-EVENTS" at all high mentioned samples mean that it was hitted one of the H1Lumi trigger elements from Photon Arm (VC,PD) or it were hitted more then one trigger elements (VC&PD,PD&PD1,PD&PD2 etc.). 1) at all high mentioned samples 1)-5) it were used different procedu res of "LUMI-EVENTS"-selection. a) for 1st(Nelly's sample),2nd(Vova's sample) and 5th(last sample) it was used the next procedure: - as input data it were got so-called MONITRIG-cartridges (HERA03.H01POT2.MONITRIG.C930XXXX -family; - at each H1 event which was read from high mentioned cartridge it was found TEL1-bank and it was analyzed trigger element from H1LumiTrigger - VC - if VC=1 at current H1 Event all needed for future analyze LUMI-banks and H1CTRIG-banks were written into output data set - if VC=0 - no output - go to the next event. b) for 3rd (66552-66553 dedicated Runs), 4th (66913 dedicated Run) and for some intermediate sample (at which S.L. and You had attempts to obtain B-value near 10%) 66914 dedicated Run (at my message from 19.11.93 this sample was named as 5th - as seems it is mistake - before this me agreed that 5th sample will be from selected events from 62575-64662 H1Runs range - similar to Nelly's and Vova's samples) it was used the next procedure: - FOR SELECTION of "LUMI-EVENTS" it were used contents of 100th tuples (only interactively - without jobs). - 100th tuple structure was described many times at LLO NAM(1)='ETREC' ET reconstructed energy (GeV) NAM(2)='PDREC' PD-arm reconstructed energy (GeV) NAM(3)='VCDEP' VetoCounter deposited energy (GeV) NAM(4)='XET' Xet - reconstructed coordinate (cm) NAM(5)='YET' Yet - reconstructed coordinate (cm) NAM(6)='XPD' Xpd - reconstructed coordinate (cm) NAM(7)='YPD' Ypd - reconstructed coordinate (cm) NAM(8)='TIME' time at minutes from start of first H1Run at sample NAM(9)='BUNCH' - bunch number NAM(10)='PD1' - value of trigger element PD1 NAM(11)='PD2' - value of trigger element PD2 NAM(12)='PD' - value of trigger element PD NAM(13)='VC' - value of trigger element VC NAM(14)='ET' - value of trigger element ET NAM(15)='ETAG' - value of trigger element eTAG NAM(16)='RUN' - H1 Run number NAM(17)='S86' - value of trigger bit s86 NAM(18)='S93' - value of trigger bit s93 NAM(19)='S94' - value of trigger bit s94 NAM(20)='S95' - value of trigger bit s95 CALL BNT (100,0,NVEC,NAM) where PD1,PD2,PD,VC,ET,ETAG - values of trigger elements from TEL1-bank and S86,S93,S94,S95 - values of trigger bits (subtrig- gers) (Actual subtriggers - not Raw) - from TLV1-bank (as exeption 2-value for all S86-S95 if event with random trigger 2) at all high mentioned samples 1)-5) it were used similar procedu- res of "LUMI-EVENTS"-reconstruction and making of PDREC-spectra: - H1Lumi reconstruction was made with March93-release of LREC (only Relation Coefficient set and CCvcsp were got not from H1DataBase - from LRINIT-subroutine) and it were installed non-standard PDdepmin and ETdepmin values as 0.5GeV (at two places - at LREC1 and at LUREAD) - for all high mentioned samples it were produced 100th tuples with high mentioned structure. Filling of high mentioned 100th tuples was made if both arms were succesfully reconstructed or one of the two arms is succesfully reconstructed (at last case values 999. were written into non-reconrusted arm values (Erec X,Y).Tuples were written into LOOK-data sets(ALOOKQ,CLOOKQ); - for all 5 samples it were used the same RelationCoefficients values (07.11.93) and different CCvcsp (13.11.93) exept 5th sample (it was used CCvcsp for Nelly's sample - nearest to 5th sample); - it were made (interactively) all possible PDREC-histos for each samples and were written into new LOOK-data set (BLOOKQ-for 4 samples and DLOOKQ-for 5th). > >Just one example: It is usially written like "VC-trigger Selection..." >But, for VC some questions arrised: Is it s93 or VC-signal from Lumi >(these are different in many cases); Was the VC-threshold used or No >Was the Flag of sucsessful reconstruction used or Not? At different >answers the selected PD-spectra are different qualitatively, in shape. As seems all these questions are answered earlier but for any case: a) s93 (or VC=1) - signal from H1LumiTrigger Logic was used for SELECTION - no any attention on VCdep; b) VCdep used (as You remember) for making of 70th an 80th families of histos (only during making of PDREC-spectra) - and for posibility extract events without any showers at VC; c) VC-threshold(-s) was used with HW of H1LumiTriggerLogics for making of VC-trigger element value (0 or 1) VC-thresholds at SW area (0.2,1.0,1.5) were used only for VCdep during filling of 70th and 80th histos families; d) Flag(-s) of succesful reconstruction was(were) used during making of 100th tuples which are the source of all PDREC- histos. >The other problem: it is highly difficult to find in LLO the >TEMPORARY, NOT STANDARD ASSIGMENTS FOR THE LUMI_TRIGGER_BITS, >E.G., s86 -> pd instead of vc, "super-low-threshold", etc. ?? NOT STANDARD ASSIGNEMENT(-s) or non-standard H1LumiTrigger Set were used no so often: 1) during first H1Lumi-dedicated Runs with standalone e-beam 65570-65584 family of H1 Runs; see LL7: 20.10.1993 -------------------- H1KFOM --- Standalone e-beam H1Lumi's 2) during 66552-66553 dedicated Runs with ep-collisions; 01.11.1993 -------------------- H1KFOM --- H1Lumi's Dedicated Runs.. 3) during 66913-66914 dedicated Runs with ep-collisions during 'Machine Studies'-announcement. see LLC: 07.11.1993 -------------------- H1KFOM --- Lumi RUN# 371 ............ (that's all). >I think, the assigments must be reflected in LUTR-bank, but not sure You are right - LUTR-bank contents keep this information about non-standard setting of H1Lumi-trigger elements and subtriggers (as cosequence); >Its dangerous for all H1, I PROPOSE to make some automatic memoryzatio All needed remarks were written into Large H1 LogBook which is filled at the H1 Shift Crew responsibility area and into Electronic LogBook. Really it was written into LL7 or LLC. LUTR-banks keep all needed information about H1LumiTriggerElemen setting at any current H1Runs. Changing(-s) of H1LumiTrigger Setting is made by hand (by expert hand only for today) and LUTR (at RunStartRecord) reflects these changing "automatically" - due to filling of LUTR-bank is only reading from fixed cells of some memory and writting of these values into some variable of LUTR-bank. Changing is made "by hand" - filling is automatical. 27.11.1993 ==================== H1KFOM === Data set with sel. events It was created new data set with selected VC=1 events from Run62633 up to Run64594 or from epRun#273 up to epRun#307 or from 19/09/93 16:17:50 up to 07/10/93 11:38:52. It was possible due to last Nelly's activity with selection of VC=1 events. So we have data set H1KFOM.EP273C29.A00 K14483 78048081 00001 ACS-E-C 27745 events of VC=1 trigger (Runs 62633-64594) Run 64662 is seen. Else one Nelly's selection and 5th sample will be finished. The next H1Run range VC=1 events exist at Vova's sample. I proposed to Nelly making of selection from Run65586 and up to End Of H1DataTaking (VC=1 events). When HERA03.H01POT3.MONITRIG.C93XXXXX-family of cartridges will appeare it will be possible to use CLASS 24 for selection of LUMI_EVENTS and CPU-usage for selection will be not so huge. 03.12.1993 ==================== F11LEV === Visits to DESY in 1994 ... For information of the LPI group. Please find below a list of names and associated terms for "DFG paied" visits to DESY in 1994. Left column - what was presented at LPI group meeting 21/11/93. Right column - what was actually given to the H1 secretariat by bosses as an official request. All questions - to S.V.Rusakov and/or A.S.Belousov. Name months main task (as discussed in our seminar) Andreev .......... 4 4 Eliseev .......... 3 - detector preamplifiers Fomenko .......... 6 6 online s/w Gogitidze ........ 1 1 Kotelnikov ....... 4 4 work in the muon group Malinovski ....... 3 3 Sheviakov ........ 6 6 trigger Smirnov .......... 3 3 work in the SPACAL group Soloviev ......... 8 8 lumi detectors Usik ............. 6 6 lumi data transter/presentation/SC Vazdik ........... 4 4 Baranov .......... - - Belousov.......... - - Lebedev .......... - 6 Rusakov .......... - 6 Shtarkov ......... - 3 Sum: 48 60 18.01.1994 ==================== F11LEV === LUMI monitoring events ... The reprocessing of the '93 fall data (after H1 field back, 320 nb-1) is finished. Presently an official DST is being created from the POTs. Below a corresponding list of LUMI monitoring events is given. Note, that only trigger and lumi banks were selected, and only class 24 was considered (class 24 means: either any lumi related trigger bit was set, or/and ET(PD) rec. energy was above min. value 3(2) GeV): SELECT BOSINPUT RECNAME="RUNEVENT" CLASS=24 SELECT BOSINPUT BLOCK = HEAD HEAR TEL1 TLV1 TSTC ZRCT & LRTN LREE LRPE LREP LRPP LETR LPDR LRXY LIST OF LUMI MONITORING FILES FOR 93 FALL DATA ---------------------------------------------- File names are: F11LEV.LMONI93.Axx Selected from : HERA03.H1POT4.MONITRIG.C9305000-C9305511 actual trigger of the event xx events 1st run last run size[Mb] eTAG VC LUMI Other Random ---------------------------------------------------------------------- 00 114506 62575 63507 194.3 29965 12193 63363 8643 341 01 119813 63662 64576 198.0 33644 11981 69116 4719 343 02 121957 64576 65389 197.9 39303 23387 51564 7293 410 03 113841 65421 66219 182.7 35617 25886 37005 14795 538 04 69597 66219 66609 113.0 15736 16293 31644 5686 238 ---------------------------------------------------------------------- Sum: 539714 154K 90K 252K 41K 2K ---------------------------------------------------------------------- There exist also LOOK datasets for these files, each containing one n-tuple (ca. 8 Mb per file) with the format: NAM(1) = 'RUN' NAM(2) = 'RQ' ! run quality: 0-unknown, 1-good, 2-medium NAM(3) = 'BUNCH' NAM(4) = 'BTYPE' ! bunch type, see H1EP92 NAM(5) = 'ATRIG' ! actual trigger: 1=eTAG,2=VC,4=ET*PD*!VC NAM(6) = 'ESTRIG' ! ET+PD+VC trigger energy (from LRTN) NAM(7) = 'EETRIG' ! ET trigger energy ( -"-) NAM(8) = 'EPTRIG' ! PD+VC trigger energy ( -"-) NAM(9) = 'EE' ! electron rec. energy (LETR) NAM(10)= 'EP' ! photon rec. energy (LPDR - 1st row) NAM(11)= 'EV' ! VC-dep. energy (LPDR - 2nd row) NAM(12)= 'XE' ! rec. coordinates NAM(13)= 'YE' NAM(14)= 'XP' NAM(15)= 'YP' NAM(16)= 'ET' ! lumi system trigger bits NAM(17)= 'PD' NAM(18)= 'VC' NAM(19)= 'ETAG' F11LEV.YLOOK.LMONI93.A00 FAST01 3010200E 00000 DASD F11LEV.YLOOK.LMONI93.A01 FAST05 3010200E 00000 DASD F11LEV.YLOOK.LMONI93.A02 FAST05 3010200E 00000 DASD F11LEV.YLOOK.LMONI93.A03 FAST02 3010200E 00000 DASD F11LEV.YLOOK.LMONI93.A04 FAST19 3010200E 00000 DASD 27.01.1994 ==================== H1KFOM === Data set with sel. events Due to last Nelly's activity (which started before posibility of monitoring events selection with option CLASS=24) it is possible to use new data sets with ET&PD&n(VC)=1 trigger: All events from these Nelly's cartridges were pressed into next ones: H1KFOM.EP261D30.A00 K34384 78048081 00001 ACS-E-C 122823 events of ET&PD&nVC=1 trigger (Runs 61890-63421) H1KFOM.EP261E30.A00 K23918 78048081 00001 ACS-E-C 97544 events of ET&PD&nVC=1 trigger (Runs 63426-64917) H1KFOM.EP261F30.A00 K24397 78048081 00001 ACS-E-C 125175 events of ET&PD&nVC=1 trigger (Runs 64997-65588) These data sets do not include BOS-banks which were born with H1REC - - only banks from H1RawData Flow (H1Lumi's and H1CentralTrigger): SELECT BOSINPUT BLOCK=HEAD HEAR LREE LRPE LREP LRPP TLV1 TEL1 LRTN TSTC Selected from : HERA03.H1POT2.MONITRIG.C9301741-C9302542 It is clear that S.Levonyan's data sets with Lumi-events (see his message at LLC from 18.01.1994) are better. But any work must be finished with ready product. I think that high mentioned data sets will be useful for some work (recalibration etc.). 08.02.1994 ==================== H1KFOM === Data set with sel. events Due to last Nelly's activity (which started before posibility of monitoring events selection with option CLASS=24) it is possible to use new data sets with ET&PD&n(VC)=1 trigger: All events from these Nelly's cartridges were pressed into next ones: H1KFOM.EP261D30.A00 K34384 78048081 00001 ACS-E-C 122823 events of ET&PD&nVC=1 trigger (Runs 61890-63421) H1KFOM.EP261E30.A00 K23918 78048081 00001 ACS-E-C 97544 events of ET&PD&nVC=1 trigger (Runs 63426-64917) H1KFOM.EP261F30.A00 [65561-65584] K24397 78048081 00001 ACS-E-C 125175 events of ET&PD&nVC=1 trigger (Runs 64997-65588) H1KFOM.EP261G30.A00 [66201-66205] K31467 78048081 00001 ACS-E-C 65175 events of ET&PD&nVC=1 trigger (Runs 65589-66531) SELECT BOSINPUT BLOCK=HEAD HEAR LREE LRPE LREP LRPP TLV1 TEL1 LRTN TSTC Selected from : HERA03.H1POT2.MONITRIG.C9301716-C9302767 [XXXXX-XXXXX] - means presence of H1Lumi dedicated Runs at monitoring events. As You remember we had 4 H1Runs periods with H1Lumi dedicated Runs 1)standalone 26.7 GeV e-beam & standalone 12. GeV e-beam H1 Runs from 65561 up to 65584 .R65570 - standard H1Lumi's trigger setting .R65572 - s86=PD, s93=PD2 .R65574 - s86=PD, s93=PD2 .R65576 - s86=ET&PD, S93=PD1 2)standalone 26.7 GeV e-beam (after p-lost at epRun#353) H1 Runs from 66201 up to 66205 .R66202 - standard H1Lumi's trigger setting .R66203 - standard .R66204 - standard .R66205 - standard 3)ep-collisions (last 0.5 hour of epRun#367) H1 Runs from 66551 up to 66553 .R66552 - s86=PD with new low threshold .R66553 - s86=PD with new low threshold. 4)ep-collisions (Mashine Studies announcements) H1 Runs from 66911 up to 66914 .R66913 - s86=PD with new low threshold .R66914 - s86=PD with new low threshold. It was created new data set with selected VC=1 events from Run62633 up to Run64662 or from epRun#273 up to epRun#308 or from 19/09/93 16:17:50 up to 08/10/93 03:45:51. It was possible due to last Nelly's activity with selection of VC=1 events. So we have data set H1KFOM.EP273D29.A00 K34592 78048081 00001 ACS-E-C 30315 events of VC=1 trigger (Runs 62633-64662) Selected from : HERA03.H1POT2.MONITRIG.C9301857-C9302221 So we have the next data sets with VC=1 and ET&PD&nVC=1 events for H1Run after 62575: it is possible to use F11LEV's data sets which were proclaimed at his message from 18.01.1994 and some data sets which were produced earlier by Nelly, Vova and me: a) VC=1 events H1KFOM.EP273D29.A00 ( Runs 62633 - 64662) 30315 events H1KFOM.EP308XXX.A00 ( Runs 64662 - 65559) 30980 events b) ERT&PD&nVC=1 events H1KFOM.EP261D30.A00 ( Runs 61890 - 63421) 122823 events H1KFOM.EP261E30.A00 ( Runs 63426 - 64917) 97544 events H1KFOM.EP261F30.A00 ( Runs 64997 - 65588) 125175 events H1KFOM.EP261G30.A00 ( Runs 65589 - 66531) 65175 events 11.02.1994 ===== dice2 ======== H1KFOM === H1Lumi Calibration at dice 1.Input Data Preparing: ----------------------- Today it were made some attempts to move some useful for calibration procedure data sets from IBM-cartridges on dice2-storage. Due to high mentioned problems with stagin procedure at dice2-environment (see message from 09.02.94 - Answer from Bernet) ".... Some of your problems, I have myself. 1st: stageing: I always used the command stagein -f HERA03.H1POT4.MONITRIG.C9305006 TESTFIL1 but since about one month I can't stage any cartradges also. I didn't have to stage, so I haven't looked for the problem. The person you shuold ask about is Ralf Gerhards. ...." and no any answer from Ralf Gerhards it was decided to try moving of data sets through FTP-protocol (binary) First step - making of STAGEI-procedure at IBM with all data sets mentioned at LLC message from 08.02.94 "Data set with sel.events" a) VC=1 events H1KFOM.EP273D29.A00 ( Runs 62633 - 64662) 30315 events H1KFOM.EP308XXX.A00 ( Runs 64662 - 65559) 30980 events b) ERT&PD&nVC=1 events H1KFOM.EP261D30.A00 ( Runs 61890 - 63421) 122823 events H1KFOM.EP261E30.A00 ( Runs 63426 - 64917) 97544 events H1KFOM.EP261F30.A00 ( Runs 64997 - 65588) 125175 events H1KFOM.EP261G30.A00 ( Runs 65589 - 66531) 65175 events All high mentioned data sets were successfully staged into DESY IBM disc storage. Second step: FTP-protocol can get data sets with names H1KFOM.XXXXXXXX ( you must define only XXXXXXXX-name at FTP-protocol). After staging it were produced the next names of staged data sets on disk: $.H1KFOM.EP273D29.A00 $.H1KFOM.EP308XXX.A00 $.H1KFOM.EP261D30.A00 $.H1KFOM.EP261E30.A00 $.H1KFOM.EP261F30.A00 $.H1KFOM.EP261G30.A00 For posibility of FTP-protocol using it was decided to make copy of high mentioned data set from stage-area of DESY IBM into privat catalog with commands: X COPY "$.H1KFOM.EP273D29.A00" TESTFIL1 X COPY "$.H1KFOM.EP308XXX.A00" EP308XXX.A00 X COPY "$.H1KFOM.EP261D30.A00" EP261D30.A00 X COPY "$.H1KFOM.EP261E30.A00" EP261E30.A00 X COPY "$.H1KFOM.EP261F30.A00" EP261F30.A00 X COPY "$.H1KFOM.EP261G30.A00" EP261G30.A00 First two copies were finished succesfully (36 Mbytes and 39.7 Mbytes). 3rd,4th and 5th were finished abnormally due to unposibility of allocation very large disk spaces. 6th copy was made with B37-error ( but 54 Mbytes of data were copied). Third step: Through FTP-protocol it were copied TESTFIL1, EP261D30.A00 and EP261G30.A00 into dice2 stage-area: Now I have at directory /stage/dice2/data20/h1/fomenko the next data sets with H1Lumi monitoring events (from Nelly's and Vova's samples): -rw-r--r-- 1 fomenko h1 36313200 Feb 11 18:05 testfil1 -rw-r--r-- 1 fomenko h1 39750480 Feb 11 20:52 ep308xxx.b00 -rw-r--r-- 1 fomenko h1 54054000 Feb 11 21:06 ep261g30.b00 2.Preparing and Submission of Calibration Job: ---------------------------------------------- As it was mentioned earlier (see message at mylog3 from 07.02.94 "First experience on dice2") it was prepared one of jobs from H1Lumi Calibration Package at directory: /home/local/fomenko/calibr94 -rwxr-xr-x 1 fomenko h1 1785872 Feb 11 21:17 Calibr94B -rw-r--r-- 1 fomenko h1 265 Feb 11 21:16 MakefileB -rw-r--r-- 1 fomenko h1 259751 Feb 11 21:15 mainB.f drwxr-xr-x 2 fomenko h1 512 Feb 11 21:20 R drwxr-xr-x 2 fomenko h1 512 Feb 6 18:19 Qcards drwxr-xr-x 2 fomenko h1 512 Feb 11 21:24 Out drwxr-xr-x 2 fomenko h1 512 Feb 11 21:21 Hbk At this directory: mainB.f (more then 7000 FORTRAN-lines) - FORTRAN source of the first step calibartion procedure (for Relation Coeffici- ent Set defining); MakefileB - makefile for creation of Calibr94B- application (compilation and linking); Calibr94B - executable program unit (ready for submission); Qcards - directory (folder) for keeping of BOS text banks which are input for high mentioned application (not H1 Data Base); Out - directory (folder) for collection of some output data sets from submitted Calibr94B-application; Hbk - directory for HBOOK-environment (as the rest from used example); R - directory with scripts for submission of Calibr94-family applications. /home/local/fomenko/calibr94/R -rw-r--r-- 1 fomenko h1 2467 Feb 11 21:20 s_calibr94B Application Calibr94B was submitted into diceS-queue with issuing at directory /home/local/fomanko/calibr94/R/ of command: qsub s_calibr94B After this it is possible to look for status of submitted job at dice2 -environment with command: qstat -a command You can see waiting at queue (only 4 jobs of S-class are running simultaneously) and running status and disappearing from all status (Q and R) - it means that apllication finished. You can go to /home/local/fomenko/calibr94/Out/ directory and find at its contents: -rw-r--r-- 1 fomenko h1 349 Feb 11 21:24 calibr94B -rw-r--r-- 1 fomenko h1 44956 Feb 11 21:24 calibr94_11049.out You can go to /home/local/fomenko/ directory and find at its contents (directory with LOOK-binary file): drwxr-xr-x 2 fomenko h1 512 Feb 11 21:33 A_calibr94_11049 You can go to /home/local/fomenko/ directory and find at its contents (directory with LOOK-binary file): -rw-r--r-- 1 fomenko h1 1614600 Feb 11 21:32 H1KFOM.ALOOK1.R65589 As seems high mentioned data sets at different directories are the full set of needed listings and data sets. Results of calibration procedure You can find at calibr94_11049.out data set. LOOK-binary file must be copied into h1poll-area for posibility of getting it with /h1/look/look20302-application. I make it with command: cp H1KFOM.ALOOK1.R65589 /shift/dice2/data20/h1/fomenko First view on output listing shows that high mentioned input data set ep261g30.b00 (single data set with ET&PD&nVC-events) is without normal end - due to high mentioned B37 at the 2nd step of preparing data for dice2. There are 42095 ET&PD&nVC-events at this data set. First view on LOOK-pictures which were pproduced by this job shows reasonable results. CPU time which was used for Calibr94B-application at dice2-environment is 183.74 sec.(as seems a little less then DESY IBM CPU usage with this number of events). At IBM calibration procedure at class L got near 38-40K events (300 sec,). There are some little remarks to Calibr94b-application running but in principle H1LumiCalibration Package at dice2 environment is working. 18.02.1994 ===== dice2 ======== H1KFOM === H1Lumi Calibration at dice I prepared enough long description of H1Lumi Calibration Package installation at dice2-environment. For anybody who interested put * into first column of the next line: list 'h1kfom.dice2exp' 22.02.1994 ===== dice2 ======== H1KFOM === Test of SGI Challenge .... Today it were tested the possibility of SGI Challenge working with large quantity of SGI-CPU-consumers: before this only 2 jobs with L, 4 jobs with M-class and 4 jobs with S-class were working: Now You can see: 4 + 16(!!!) + all needed S-consumers(!!!) Fantastic!!! this is time slice of SGI Challenge CPU consumers near 15:10 today: [dice2] /shift/dice2/data18/h1/fomenko/testclbr/A_testclbr22_13784 $ qs Request I.D. Owner Queue Start Time Time Limit Total ------------- ------ -------- -------- ----------- ---------- ---------- -- vtxshift 15051 kruener D_L 2/19 09:36 4 1:13 2 23: PlayWithSusy 15577 schuhman D_L 2/21 16:42 4 1:13 0 18: testphan.run 15603 walkowia D_L 2/22 01:42 4 1:13 0 9: mkidx3.qsub 15885 bernet D_L 2/22 13:50 4 1:13 0 0: job... 15894 arpagaus D_L selection1 15587 meyerpo C_M 2/22 06:37 0 23:53 0 6: G. Adamski 15610 flamm C_M 2/22 12:35 0 20:00 0 1: nqs.24712 15611 joachim C_M 2/22 13:49 0 5:33 0 0: batch21_19.15 15625 krueger C_M 2/22 13:50 1 0:00 0 0: h1r9321_19.26 15634 krueger C_M 2/22 13:50 1 0:00 0 0: sub93 15681 kermiche C_M 2/22 13:51 0 23:53 0 0: sub93 15688 kermiche C_M 2/22 13:51 0 23:53 0 0: phan.92 15787 siewert C_M 2/22 14:13 1 0:00 0 0: tamen02b 15801 reinshag C_M 2/22 14:14 0 13:53 0 0: bouge de la ! 15810 neyret C_M 2/22 14:15 0 20:00 0 0: tout arrive a qui sait attendre 15829 neyret C_M 2/22 14:16 0 20:00 0 0: batch_job 15866 hufnagel C_M 2/22 14:31 1 0:00 0 0: eglsel.nqs 15875 h1kmco C_M 2/22 14:32 1 0:00 0 0: so Miisli 15898 beck C_M 2/22 14:33 1 0:00 0 0: Blue men can"t sing the whites 15905 newmanpr C_M 2/22 14:55 0 2:46 0 0: phan.noFtrks 15906 sell C_M 2/22 14:58 1 0:00 0 0: Le temps est l'ami du sage 15913 delbuono C_M dk0_only.qsub 15918 grab C_M dd0_only.qsub 15919 grab C_M Jokerman 15914 wobisch B_S 2/22 14:51 0 1:00 0 0: teff2.qsub 15916 bernet B_S 2/22 14:56 0 1:00 0 0: main.s 15917 gellrich B_S 2/22 14:59 0 1:00 0 0: 22.02.1994 ==================== F11LEV === Few remarks on LUMI-93 ... Due to the lack of the time, I did not describe in the logbook some important and interesting results concerning LUMI system in 93. Here I try to recollect briefly what has been done in January-94. 1) After reprocessing of the fall-93 data (H1 runs 62575-66609) all lumi monitoring events were selected on separate cassetes (see LLO from 18.01.93) 2) The analysis of these events gave the following results: - the global absolute calibration is ok within 1.5-2% (remaining shift in the energy scale was estimated to be 1.5%: ET must be multiplied by 1.015 and PD - by 0.985) - there were 3 periods with different VC trigger efficiency: H1 runs VC-efficiency 62575-64400 0.755 +- 0.015 64401-65710 0.844 +- 0.011 65711-66609 0.869 +- 0.010 these periods exactly correspond to the changes of HV at VC. Note, that the "theoretical" eff. should be 0.900+-0.010. The only explanation I see is as follows. Change of HV changes the low threshold in E(VC) dependence. Due to the large fluctuations between E(VC) and E(PD) for the given photon energy, this results in the change of the plateau value for VC-eff versus real photon energy. This means, that we cannot use ET*PD*noVC trigger for the absolute lumi measurements, since its cross section is strongly dependent on the HV. Therefore, we were right, using ET*PD for lumi. - on the other hand, VC events were used to cross check our online lumi estimate (so called gamma rate method). This is ok as soon as we measure VC-eff. and take it into account. For the sample of G+M runs having this monitor trigger, the compa- rison is OK: L(ET*PD online) = 303.4 nb-1 L(VC offline) = 300.3 nb-1 thus, the method works and must be implemented online for 94 data - the shape of the photon energy for the trigger VC is not good: it shows bad energy resolution after averaging over all statistics with different calibration sets, and also some deficit of events at low energies (might be VC-eff. effect) 3) The amount of different triggers suggests for the future: o reduce eTAG rate (increase prescale) by a factor of 3-4 o reduce ET*PD*noVC rate (increase prescale) by a factor of 2-3 o increase VC rate (decrease prescaling) by a factor of 3-4 o remove the trigger noET*PD*VC (was not used at all) o add new triggers: ET*PD*VC with the rate of the order of 0.2 * ET*PD*noVC PD with low threshold (E > 2-3 GeV) PD with high threshold (E > 10 GeV) 4) It was found, that the energy resolution of our system does not follow the relation: sigma(E)/E = a + b/sqrt(E) (1) It turned out, that the resolution is worse for the low E_PD (high E_ET), and becomes better for the higher E_PD (lower E_ET). When using an expression (1) with a=1%, the value of b changes from 12.5% to 8-11% (depending on the run period) 5) It was found, that the ET acceptance for 1993 period can only be described by my Monte Carlo if one assumes an orbit offset in the horizontal direction of the order of -0.22 cm. R.Brinkmann said, that it well could be (as an effective description of different real effects). Thus B-H normalization cross section used online, only by pure luck happens to be more or less correct! This is a lesson for future: we must to monitor our ET-acceptance at L4 level permanently! 6) Further investigations by Fomenko showed that the Rel.Coeff. during the given period has also changed around 2-5 Oct. After he finishes the new calibration for the second part of the data, I will reprocess LREC once again for all LUMI and ETAG data (which should be relatively easy and fast). 22.02.1994 ===== dice2 ======== H1KFOM === Relation Coefficients Sets It is proposed to use 2 different sets of Relation Coefficients for H1 Run range 62575-66609. Up to now (during last reprocessing) it was used only one Relation Coefficients Set for all event from high mentioned region. If You remember - this Relation Coefficients Set was defined on data which were written during H1Lumi Dedicated Runs 66552-66552 (near 21K events with ET&PD&nVC=1 trigger, see message 07.11.93 at LLC). Now it is almost obviously that during mentioned at LL7 troubles at H1,HERA and H1Lumi at 2-5 October - something changed with Relation Coefficients Set too. Firstly for Relation Coefficients sets defining it were used not so-called 10Hz stream of ET&PD&nVC-events at H1RAWD-cartridges but selected from 62575-66609 H1 Runs ET&PD&nVC-events which were got from F11LEV.LMONI93.A00-A04 cartridges and pressed into 2 cartridges with only ET&PD&nVC=1 trigger and with districted set of banks (without BOS-banks produced by H1REC-package). H1KFOM.ETPDNV.LMONI93.DICE2.A00 K40982 78048081 00001 ACS-E-C H1KFOM.ETPDNV.LMONI93.DICE2.A01 K44871 78048081 00001 ACS-E-C These cartridges keep more then 250K ET&PD&nVC=1 triggered events. Selection and pressing of high mentioned data was made at dice2- environment (as exercise) and shifted into IBM-cartridges with standard dice2-environment (without using of ftp-protocol). First attempt "to improve" existing Relation Coefficient Set put unwaitable result - there are 2 different time periods at Run range 62575-66609 (up to 2-5 October 93 troubles and after) which are differed significantly with Relation Coefficient Set. (there are 2 good peaks at histos 8400-8424 and 8500-8548 which are used for Relation Coefficients defining). Note that during H1Lumi Dedicated Runs we never looked for this type contents of high mentioned histos. Looking for behaivour of individual Relation Coefficients with time (for more statistical channels - 21,22,23,24,25,26,27 at ET and 10,11,12,13,14 at PD) put the next conclusion: it is needed to share high mentioned data on two parts: 1) 62575-64471 2) 64472-66609. These parts of data are differed significantly. There are differencies inside high mentioned regions (but - as seems - 2nd order significancy ). As seems last differencies were connected with OnLineProcedure of put into start values of all calibration coefficients.The reason of intervantion into OnLine Calibration was (at most cases) evidence of bad energy distributions at OnLine collected histos (as example - trips of HV or another troubles). After three iteration calibration for each time period events samples it were obtained the next values of Relation Coefficients: for H1 Runs 62575 - 64471 ------------------------- ==> Relation Coeff.Set for ET 0.000 1.230 1.239 1.227 1.231 0.000 1.223 1.213 1.238 1.238 1.472 1.212 1.252 1.267 1.224 0.937 0.965 0.949 1.001 0.977 1.097 0.958 0.957 0.962 0.990 1.005 1.003 0.982 1.029 0.980 1.093 1.034 0.951 0.904 0.895 1.177 1.066 0.990 0.915 0.907 0.911 0.985 1.211 1.130 1.056 0.977 1.181 1.149 1.220 ==> Relation Coeff.Set for PD 1.436 1.354 1.540 1.226 0.969 1.016 1.018 1.003 1.054 1.174 1.017 1.066 1.043 1.038 1.048 1.031 1.013 1.001 0.999 1.003 1.564 0.916 0.900 0.950 0.839 ==> Entries into Relation Coeff.Set histos for ET 0 8 1 4 32 0 1 1 19 10 22 4 5 8 1357 372 162 108 125 81 86 14002 21541 15610 12420 10270 8488 3860 4941 5653 5989 4229 3758 3116 2194 22 12 37 30 21 29 15 1 8 10 18 15 5 2 ==> Entries into Relation Coeff.Set histos for PD 108 81 150 137 37 597 6374 13335 6740 909 2699 19567 37464 22444 3122 405 3592 8313 5000 739 78 31 87 47 20 for H1 Runs 64472 - 66609 ------------------------- ==> Relation Coeff.Set for ET 1.223 1.209 1.216 1.246 1.233 1.233 1.220 1.211 1.247 1.201 1.521 1.212 1.275 1.337 1.018 1.002 1.083 1.098 1.084 1.097 1.066 1.039 1.026 1.023 1.066 1.112 1.108 1.107 1.016 1.006 1.197 1.046 1.057 0.952 1.057 1.298 1.105 1.087 1.039 1.111 1.171 1.130 1.221 1.143 1.125 0.987 1.129 1.068 1.214 ==> Relation Coeff.Set for PD 1.394 1.308 1.358 1.207 0.906 0.984 0.884 0.923 0.926 1.156 0.897 0.950 0.948 0.934 0.961 0.944 0.966 0.918 0.888 0.935 1.485 0.839 0.814 0.915 0.827 ==> Ent. at Relation Coeff.Set for ET 3 24 2 8 56 3 5 4 28 11 33 10 12 14 628 617 347 180 222 116 69 24050 37223 25992 20859 17420 14112 6152 8139 9208 10513 7928 8149 6841 4268 75 53 116 123 115 195 97 2 20 44 57 53 31 23 ==> Ent. at Relation Coeff.Set for PD 119 111 166 142 32 727 6692 13596 5596 1145 4416 32983 64776 39114 5361 603 6335 15981 8794 1221 83 52 100 73 11 Some digits for comparison: Sample 1:(Run62575-64471): with H1DataBase values of RC after 3rd iteration mean sigma cuts entries mean sigma entries 26.82 0.9 all sample 124533 26.65 0.86 124390+179 26.78 0.87 |xet|<6.0cm 90713 26.69 0.84 90807+113 26.70 0.84 vcdep<0.2GeV 63443 26.61 0.80 63370+ 90 26.70 0.82 both cuts 46769 26.63 0.78 46812+ 54 Sample 2(Run64474-66609): with H1DataBase values of RC after 3rd iteration mean sigma cuts entries mean sigma entries 26.65 0.80 all sample 125164+164 26.69 0.80 125170+168 26.76 0.78 |xet|<6.0cm 92128+118 26.70 0.77 92240+117 26.64 0.79 vcdep<0.2GeV 101386+124 26.66 0.79 101407+127 26.64 0.77 both cuts 74789+ 86 26.68 0.76 74904+ 86 P.S.1) At producing of high mentioned digits I did not use so called Global Photon Shape Shift constants (5 different values for high mentioned Run range (from -9.6% up to +3.4%)-see LESC-bank at H1DB). 2) as start values for 1st iteration it were got H1DataBase Relation Coefficients; 3) it were used 2 different CCvcsp values at high mentioned Run Range which are kept at LRFC-bank at H1DB (4.3 for first sample and 3.1 for second one). As seems it is needed (for any case) recalibrate CCvcsp too. 4) all work for recalibration Relation Coefficients Sets was made at dice2-environment (as exercise). All sources, BOS-text banks and scripts for submitting of calibration jobs at dice2 are kept at the next directories at dice2: /home/local/fomenko/Calibr94 /home/local/fomenko/Testclbr /home/local/fomenko/Testclbr1. 5) LOOK-binary data sets produced at dice2 environment are kept at the next data sets at DESY IBM storage: with H1DB values of RC for both samples: H1KFOM.TLOOK01.R62575 FAST17 3010200E 00000 DASD H1KFOM.TLOOK01.R64472 FAST03 3010200E 00000 DASD after 3rd iteration: H1KFOM.TLOOK31.R62575 FAST20 3010200E 00000 DASD H1KFOM.TLOOK32.R64472 FAST04 3010200E 00000 DASD 23.02.1994 ===== dice2 ======== H1KFOM === Attempts of VC-calibration It were made first attempts of VCsp-calibration with new Relation Coefficients sets. Preliminary results are the next: 1) First sample: 4.3 (start value) ---> 4.398(1st iteration) ---> 4.407(2nd iteration ---> 4.768(tuning- optimal factor 1.0819) final calibration weight per iteration = 192 2) Second sample: 3.1 (start value) ---> 3.079(1st iteration) ---> 3.079(2nd iteration ---> 3.541(tuning - optimal factor=1.150) final calibration weight per iteration = 320 Attempts of tuning were made at the next LOOK-data sets: H1KFOM.VLOOKT2A.R62575 FAST01 3010200E 00000 DASD H1KFOM.VLOOKT2B.R64472 FAST20 3010200E 00000 DASD high mentioned LOOK-data sets were made with CCvcsp after 2nd iteration (4.407 and 3.079). Some digits: Sample 1: CCvcsp=4.3 CCvcsp=4.407 CCvcsp=4.768 26.44 1.31 26.45 1.32 all events 26.47 1.27 26.49 1.29 26.68 1.282 |xet|<6.0 Sample 2: CCvcsp=3.1 CCvcsp=3.079 CCvcsp=3.541 26.47 1.31 26.45 1.32 all events 26.44 1.26 26.41 1.27 26.68 1.250 |xet|<6.0 Statistics for calibration is very pure (4K events for first sample and near 9K events for 2nd sample). It is difficult to make some final conclusions. 24.02.1994 ===== dice2 ======== H1KFOM === Attempts of VC-calibration It were repeated attempts of calibration VCsp at the both samples of events (4370 events of ET&PD&VC at Sample 62575-64471 and 9081 ones at Sample 64472-66609) with different start values. Each calibration was proclaimed as finished if after next iteration CCvcsp was not changed (weights after each iteration were kept and incremented at the next iteration - it is new at calibration procedure of VCsp). The next results were obtained: First sample (62575-64471): 4.300(start value)-->4.250(1st step)--->4.268(2nd step)---> 0.65 134.9 272.8 <---weights -->4.272(3rd step)--->4.273(4th step)---> 411.1 549.4 -->4.273(5th step)--->...the same value at the next steps 4.768(start value)-->4.285(1st step)--->4.273(2nd step)---> 0.65 141.4 280.1 <---weights -->4.273(3rd step)--->...the same value at the next steps Second sample (64472-66609): 3.100(start value)-->3.007(1st step)--->2.987(2nd step)---> 0.65 357.4 700.0 <---weights -->2.987(3rd step)--->...the same value at the next steps 3.541(start value)-->3.015(1st step)--->2.988(2nd step)---> 0.65 373.7 716.6 <---weights -->2.987(3rd step)--->...the same value at 1057.9 the next steps As seems at both samples were achived (as seems) final values of calibration coefficient for VCsp. CCvcsp = 4.273 - for Runs 62575-64471 CCvcsp = 2.987 - for Runs 64472-66609 I don't know is needed tuning procedure at these cases - may be due to low statistics it will be not good. Attempts of tuning (in principal) it is possible to make with using of contents 100th ntuple at the next LOOK-binary data sets which were made with 4.273 and 2.987 values: H1KFOM.VLOOKT3A.R62575 FAST13 3010200E 00000 DASD H1KFOM.VLOOKT3B.R64472 FAST03 3010200E 00000 DASD "technolgy" of tuning: bins 200 0 40 do '/lis1=(xet.lt.6.0.and.xet.gt.-6.0)' do 'hs(200)=nt(100)(etrec+pdrec-vcdep+A*vcdep)@(lis1)' peaks 200 (for a=1.01,1.02,1.03,1.04 etc.) a - correction factor for 4.273 or for 2.987 You must find "optimal" correction factor (Mean value and Sigma of 26.69 peak must be best). Some digits: Sample 1: CCvcsp=4.3 CCvcsp=4.273 26.45 1.31 26.40 1.28 all events 26.51 1.22 26.41 1.24 |xet|<6.0 Sample 2: CCvcsp=3.1 CCvcsp=2.987 26.61 1.32 26.46 1.32 all events 26.40 1.26 26.44 1.28 |xet|<6.0 Note: some differencies with yesterday digits can be explain with today using of pure VC-calibration events (only ET&PD&VC-events). which were preliminary selected into new data set. Tools for Veto Calibration and for Test of defined CCvcsp are kept at home/local/fomenko/vetoclbr and home/local/fomenko/tstvclbr directories at dice2-environment. 25.02.1994 ===== dice2 ======== H1KFOM === new set of PDREC-spectra.. It were produced 11 possible PDREC-spectra for each from high mentioned samples (62575-64471 and 64472-66609) for defining of possible Global Photon Spectra Shifts (after last recalibration): All histos were made with bins 70 0 35 It were used new Relation Coefficients sets for each sample and new CCvcsp-values (see messages from 23.02 and 24.02.94): a) First sample (H1Runs 62575-64471): H1KFOM.BLOOKQ.R62575 data source: H1KFOM.VCEQ1.LMONI93.DICE2.A00 K22714 78048081 00001 ACS-E-C this data set was produced from F11LEV.LMONI93.A00-..A04 with selection-C of events with VC=1 trigger (s93=1). 24391 events with VC=1 trigger 50th histo - do 'hs(50)=nt(100)(pdrec)@(pdrec.lt.999) PDREC-spectrum for VC=1 trigger 13454 entries 51st histo do 'hs(51)=nt(100)(pdrec)@(pdrec.lt.999.and.pd.eq.1) PDREC-spectrum for VC&PD-trigger 9113 entries 52nd histo do 'hs(52)=nt(100)(pdrec)@(pdrec.lt.999.and.pd1.eq.1) PDREC-spectrum for VC&PD1-trigger 5499 entries 53th histo do 'hs(53)=nt(100)(pdrec)@(pdrec.lt.999.and.pd2.eq.1) PDREC-spectrum for VC&PD2-trigger 1860 entries 60st histo do 'hs(60)=nt(100)(pdrec)@(s93.eq.1.and.pdrec.lt.999) PDREC-spectrum for VC=1 trigger 12252 entries 61st histo do 'hs(61)=nt(100)(pdrec)@(s93.eq.1.and.pdrec.lt.999.and.pd.eq.1) PDREC-spectrum for VC&PD trigger 8314 entries 62nd histo do 'hs(62)=nt(100)(pdrec)@(s93.eq.1.and.pdrec.lt.999.and.pd1.eq.1) PDREC-spectrum for VC&PD1 trigger 5063 entries 63rd histo do 'hs(63)=nt(100)(pdrec)@(s93.eq.1.and.pdrec.lt.999.and.pd2.eq.1) PDREC-spectrum for VC&PD2 trigger 1677 entries 80th histo do 'hs(80)=nt(100)(pdrec)@(s93.eq.1.and.pdrec.lt.999.and.vcdep.lt.0.2) PDREC-spectrum for VC-trigger (without shower at VC - 0.2 GeV thresh. (only with unefficiencies effect of VC=1) 17 entries 81st histo do 'hs(81)=nt(100)(pdrec)@(s93.eq.1.and.pdrec.lt.999.and.vcdep.lt.0.5) PDREC-spectrum for VC-trigger (without shower at VC - 0.5 GeV thresh. (only with unefficiencies effect of VC=1) 284 entries 82rd histo do 'hs(82)=nt(100)(pdrec)@(s93.eq.1.and.pdrec.lt.999.and.vcdep.lt.1.5) PDREC-spectrum for VC-trigger (without shower at VC - 1.5 GeV thresh. (only with unefficiencies effect of VC=1) 3848 entries b) Second sample (H1Runs 64472-66609): H1KFOM.BLOOKQ.R64472 data source: H1KFOM.VCEQ1.LMONI93.DICE2.A00 K22714 78048081 00001 ACS-E-C this data set was produced from F11LEV.LMONI93.A00-..A04 with selection-C of events with VC=1 trigger (s93=1). 83151 events with VC=1 trigger 50th histo - do 'hs(50)=nt(100)(pdrec)@(pdrec.lt.999) PDREC-spectrum for VC=1 trigger 31807 entries 51st histo do 'hs(51)=nt(100)(pdrec)@(pdrec.lt.999.and.pd.eq.1) PDREC-spectrum for VC&PD-trigger 19832 entries 52nd histo do 'hs(52)=nt(100)(pdrec)@(pdrec.lt.999.and.pd1.eq.1) PDREC-spectrum for VC&PD1-trigger 9277 entries 53th histo do 'hs(53)=nt(100)(pdrec)@(pdrec.lt.999.and.pd2.eq.1) PDREC-spectrum for VC&PD2-trigger 4866 entries 60st histo do 'hs(60)=nt(100)(pdrec)@(s93.eq.1.and.pdrec.lt.999) PDREC-spectrum for VC=1 trigger 26984 entries 61st histo do 'hs(61)=nt(100)(pdrec)@(s93.eq.1.and.pdrec.lt.999.and.pd.eq.1) PDREC-spectrum for VC&PD trigger 17132 entries 62nd histo do 'hs(62)=nt(100)(pdrec)@(s93.eq.1.and.pdrec.lt.999.and.pd1.eq.1) PDREC-spectrum for VC&PD1 trigger 8090 entries 63rd histo do 'hs(63)=nt(100)(pdrec)@(s93.eq.1.and.pdrec.lt.999.and.pd2.eq.1) PDREC-spectrum for VC&PD2 trigger 4205 entries 80th histo do 'hs(80)=nt(100)(pdrec)@(s93.eq.1.and.pdrec.lt.999.and.vcdep.lt.0.2) PDREC-spectrum for VC-trigger (without shower at VC - 0.2 GeV thresh. (only with unefficiencies effect of VC=1) 462 entries 81st histo do 'hs(81)=nt(100)(pdrec)@(s93.eq.1.and.pdrec.lt.999.and.vcdep.lt.0.5) PDREC-spectrum for VC-trigger (without shower at VC - 0.5 GeV thresh. (only with unefficiencies effect of VC=1) 3039 entries 82rd histo do 'hs(82)=nt(100)(pdrec)@(s93.eq.1.and.pdrec.lt.999.and.vcdep.lt.1.5) PDREC-spectrum for VC-trigger (without shower at VC - 1.5 GeV thresh. (only with unefficiencies effect of VC=1) 12618 entries. Tools for preparing of 100th ntuples which were used for producing of high mentioned spectra is kept at /home/local/fomenko/NtupleMaker- directory at dice2-environment. Produced at dice2-environment LOOK-binary data sets were transmitted into DESY IBM storage through ftp-protocol and are kept at the next data sets: H1KFOM.ALOOKQ.R62575 FAST01 3010200E 00000 DASD H1KFOM.ALOOKQ.R64472 FAST07 3010200E 00000 DASD 26.02.1994 ==================== H1KFOM === PDspectra Fits............ Very preliminary estimation shows that Global Energy Shift Factors after recalibration have the next values: a) for sample 1 (62575-64471) -0.8% b) for sample 2 (64472-66609) -4.5% 27.02.1994 ===== dice2 ======== H1KFOM === Alternative Lumi Measurem. It was installed tool at dice2-environment for estimation (measure- ment) of BSCrossSectionVisible-values for different H1LUMI-trigger elements and trigger bits (VC,ET,ET&PD,ET&PD&nVC,PD,PD1 and PD2). This tool was defined at 22.11.93 message at LLC (future LL8). Tool was tested with 3 different epRuns: 307,365 and 366. As You can see:307th epRun was got after 1st VC-HV-changing, 365 and 366 - after 2nd VC-HV-changing. #307th epRun(Peak Lumi 8*10**29 cm-2s-1, Ie=13.5 mA, Ip=13.2 mA) PLTEFF*ACCC*CrossSectionVisible (mb) VC PD ET ET&PD ET&PD&no(VC) PD1 PD2 Mean value 196.3 76.5 30.1 28.6 4.6 34.9 9.5(?) Sigma 2.1 1.3 0.5 0.2 0.1 0.8 0.3(?) #365th epRun(Peak Lumi 8*10**29 cm-2s-1, Ie=13.0 mA, Ip=14.0 mA) PLTEFF*ACCC*CrossSectionVisible (mb) VC PD ET ET&PD ET&PD&no(VC) PD1 PD2 Mean value 244.9 74.8 31.0 28.3 3.9 33.4 26.9(?) Sigma 6.9 1.5 0.6 0.4 0.1 0.7 0.7(?) #366th epRun(Peak Lumi 3.5*10**29 cm-2s-1,Ie= 7.0 mA, Ip=13.0 mA) PLTEFF*ACCC*CrossSectionVisible (mb) VC PD ET ET&PD ET&PD&no(VC) PD1 PD2 Mean value 270.4 79.4 33.6 29.4 3.9 35.1 18.5(?) Sigma 6.6 1.8 0.7 0.5 0.1 0.9 0.7(?) Rates (kHz) #307th epRun(Peak Lumi 8*10**29 cm-2s-1, Ie=13.5 mA, Ip=13.2 mA) VC PD ET ET&PD ET&PD&no(VC) PD1 PD2 Rate (t0) 235.0 92.0 55.0 30.0 4.5 43.0 25.0 Rate (tmax) 72.0 27.0 18.0 9.2 1.5 12.0 8.0 #365th epRun(Peak Lumi 8*10**29 cm-2s-1, Ie=13.0 mA, Ip=14.0 mA) VC PD ET ET&PD ET&PD&no(VC) PD1 PD2 Rate (t0) 410.0 135.0 80.0 50.0 7.0 63.0 36.0 Rate (tmax) 115.0 38.0 20.0 12.5 1.8 16.0 9.0 #366th epRun(Peak Lumi 3.5*10**29 cm-2s-1,Ie= 7.0 mA, Ip=13.0 mA) VC PD ET ET&PD ET&PD&no(VC) PD1 PD2 Rate (t0) 185.0 55.0 32.0 20.0 2.7 25.0 14.1 Rate (tmax) 72.0 20.0 13.0 8.0 1.1 9.2 5.3 All high mentioned digits it is possible to look for from LOOK- binary data files which were produced at dice2-environment and were transfered through ftp-protocol into DESY IBM-storage: H1KFOM.TLOOK1.R64595 FAST12 3010200E 00000 DASD H1KFOM.TLOOK1.R66426 FAST03 3010200E 00000 DASD H1KFOM.TLOOK1.R66480 FAST13 3010200E 00000 DASD Rates: VEC 500 1 3398 t(sec); f\_{WC}(Hz) VEC 510 1 3398 t(sec); f\_{ET}(Hz) VEC 520 1 3398 t(sec); f\_{LR}(Hz) VEC 530 1 3398 t(sec); f\_{LP}(Hz) VEC 540 1 3398 t(sec); f\_{PD}(Hz) VEC 550 1 3398 t(sec); f\_{PD1}(Hz) VEC 560 1 3398 t(sec); f\_{PD2}(Hz) Estimations of BSCrossSectionVisible: HS 701 0 3027 (mb); VC_PLTEFF*ACCC*CrossSection HS 711 0 3070 (mb); ET_PLTEFF*ACCC*CrossSection HS 721 0 3109 (mb); ET*PD_PLTEFF*ACCC*CrossSection HS 731 0 3082 (mb); ET*PD*nV_PLTEFF*ACCC*CrossSection HS 741 0 3090 (mb); PD_PLTEFF*ACCC*CrossSection HS 751 0 3076 (mb); PD1_PLTEFF*ACCC*CrossSection HS 761 0 3027 (mb); PD2_PLTEFF*ACCC*CrossSection Behaivours of estimated values with time: VEC 700 1 3398 t(sec); (VC_PLRTF*SIGMA*ACCC)mb VEC 710 1 3398 t(sec); (ET_PLRTF*SIGMA*ACCC)mb VEC 720 1 3398 t(sec); (ET*PD_PLRTF*SIGMA*ACCC)mb VEC 730 1 3398 t(sec); (ET*PD*nV_PLRTF*SIGMA*ACCC)mb VEC 740 1 3398 t(sec); (PD_PLRTF*SIGMA*ACCC)mb VEC 750 1 3398 t(sec); (PD1_PLRTF*SIGMA*ACCC)mb VEC 760 1 3398 t(sec); (PD2_PLRTF*SIGMA*ACCC)mb I cannot understand estimated values of BSCrossSection for PD2-trigge element for #307,#365,#366th epRuns. May be I mistaked at calculations (I'll test later). Tool is kept at /home/local/fomenko/LumiAlternative-directory. 28.02.1994 ===== dice2 ======== H1KFOM === Alternative Lumi Measurem. After talking with I.Sheviakov it was became clear that 5th group of bunch trigger element rates (which was connected with PD2) have 9-bunches shift. After correction it were produced new high mentioned LOOK-data sets and it were corrected values for BSCrossSectionVisible for PD2-trigger element: #307th epRun(Peak Lumi 8*10**29 cm-2s-1, Ie=13.5 mA, Ip=13.2 mA) PLTEFF*ACCC*CrossSectionVisible (mb) VC PD ET ET&PD ET&PD&no(VC) PD1 PD2 Mean value 196.3 76.5 30.1 28.6 4.6 34.9 20.2 Sigma 2.1 1.3 0.5 0.2 0.1 0.8 0.6 #365th epRun(Peak Lumi 8*10**29 cm-2s-1, Ie=13.0 mA, Ip=14.0 mA) PLTEFF*ACCC*CrossSectionVisible (mb) VC PD ET ET&PD ET&PD&no(VC) PD1 PD2 Mean value 244.9 74.8 31.0 28.3 3.9 33.4 17.9 Sigma 6.9 1.5 0.6 0.4 0.1 0.7 0.5 #366th epRun(Peak Lumi 3.5*10**29 cm-2s-1,Ie= 7.0 mA, Ip=13.0 mA) PLTEFF*ACCC*CrossSectionVisible (mb) VC PD ET ET&PD ET&PD&no(VC) PD1 PD2 Mean value 270.4 79.4 33.6 29.4 3.9 35.1 18.9 Sigma 6.6 1.8 0.7 0.5 0.1 0.9 0.6 Bunches shift is the next: 211 --> 0 212 --> 1 ....... Don't forget about Rates(0) <---> Rates(1) before shift 218 --> 7 OR 219 --> 8 Counts(0)<---> Counts(1) 0 --> 9 1 --> 10 ........ 210 --> 219 03.03.1994 ===== dice2 ======== H1KFOM === Pedestal Distributions.... At 1st March 1994 it was written Run68906 with H1CentralTrigger and Luminosity branches with Random 10kHz trigger. HV at all H1Lumi channels was off. It was possible to make some investigations of pedestal behaivour at H1Lumi channels and its sums (ETdep,PDdep and ETdep+PDdep) without noice from photomultipliers (only FADC pedestal behaivour). It were collected near 190K events with Random trigger. It were produced the next digits which are characteristics of pedestal distributions at each channels (as Relation Coefficient Set and CCvcsp were got last digits which were proclaimed at LLC 22.02.94 and 24.02.94). Mean Values(GeV) at Pedestal Distr. for ET cells -0.003 0.000 -0.004 0.002 -0.016 0.001 -0.009 0.008 -0.006 0.003 -0.002 -0.002 0.000 0.000 0.001 -0.001 0.002 -0.010 -0.001 -0.011 -0.008 -0.012 -0.001 -0.009 -0.006 -0.006 -0.007 -0.009 -0.002 -0.005 -0.002 -0.003 -0.004 0.002 0.002 0.002 0.002 -0.004 -0.001 -0.001 0.002 -0.005 -0.005 -0.005 0.004 -0.008 -0.001 -0.007 0.007 Mean Values(GeV) at Pedestal Distr. for PD cells 0.003 0.001 0.002 -0.010 -0.003 -0.005 -0.001 -0.004 -0.010 -0.005 -0.002 -0.005 -0.005 -0.002 -0.008 0.002 0.007 -0.005 0.016 0.009 0.026 0.003 0.008 0.013 0.008 0.045 Sigma Values(GeV) at Pedestal Distr. for ET cells 0.029 0.022 0.030 0.035 0.060 0.020 0.033 0.026 0.024 0.024 0.033 0.030 0.028 0.049 0.033 0.019 0.016 0.018 0.018 0.029 0.026 0.025 0.017 0.021 0.016 0.019 0.020 0.021 0.018 0.016 0.016 0.018 0.015 0.012 0.015 0.024 0.018 0.025 0.018 0.016 0.016 0.019 0.023 0.015 0.019 0.028 0.015 0.015 0.022 Sigma Values(GeV) at Pedestal Distr. for PD cells 0.027 0.023 0.022 0.026 0.027 0.017 0.016 0.014 0.017 0.022 0.022 0.018 0.023 0.016 0.018 0.015 0.027 0.022 0.021 0.026 0.061 0.018 0.022 0.018 0.027 0.057 Entries into Histos for Pedestal Distr. (ET) 181251 170727 183863 183975 187309 181646 185028 178449 181912 181181 177522 182735 170361 187682 185777 183008 173722 176100 179694 186837 183816 185227 177204 186151 183897 183258 180595 181587 171464 186334 169953 184655 181383 166286 176067 185732 176948 182869 180741 179038 181515 180818 185448 178335 166333 172386 181616 180619 181723 Entries into Histos for Pedestal Distr. (PD) 181664 181870 179778 183543 185260 173284 181067 180493 183734 172629 183769 179543 184117 181407 183606 180881 185173 183405 186866 185315 187616 183679 184501 185958 184992 185128 Averaged values Means and Sigma(GeV) for ET cells -0.003 0.023 Averaged values Means and Sigma(GeV) for PD cells 0.001 0.023 Mean,Sigma(GeV) and Entries Values for ETdep -0.433 0.851 190885 Mean,Sigma(GeV) and Entries Values for PDdep 0.090 0.532 191494 Mean,Sigma(GeV) and Entries Values for ETdep+PDdep -0.344 1.013 190657 Tool is kept at /home/local/fomenko/tstvclbr-directory at Makefile68906,main68906.f,lureadt68906.f,lrinit68906.f- members. For processing of high mentioned Run68906 it were staged two temporary data sets into h1stage area of dice2: dice2Ù fomenko/tstvclbr $ sfdir -p h1stage -rwxr-xr-x 1 fomenko h1 304947720 Mar 2 11:48 HERA04.H1TEMP. .G6831V00 -rwxr-xr-x 1 fomenko h1 305086320 Mar 2 12:13 HERA04.H1TEMP. .G6832V00 Processing of 190K events got near 10min CPU time at SGI Challenge. It is possible to see high mentioned digits at LOOK-binary data set H1KFOM.LOOK1.R68906 which was produced at dice2-environment and was transferred through ftp-protocol into DESY IBM-storage: H1KFOM.LOOK1.R68906 FAST09 3010200E 00000 DASD HS 5 1 191500 ET dep (incl.low) HS 7 1 191500 PD dep (incl.low) HS 9 1 191500 PD+ET dep (low) HS 5001 1 181664 E_pd / cell(GeV) ........... HS 5025 1 184992 E_pd / cell(GeV) HS 5026 1 185128 E_vc / cell(GeV) HS 6001 1 181257 E_et / cell(GeV) ........... HS 6049 1 181729 E_et / cell(GeV) HS 5200 1 49 MeanValues at Pedestal Distr. ET HS 5201 1 25 MeanValues at Pedestal Distr. PD HS 5202 1 49 SigmaValues at Pedestal Distr. ET HS 5203 1 25 SigmaValues at Pedestal Distr. PD 04.03.1994 =====H01LNS ===== PD-spectra-fits R62575,R64472 ============ I have done PD-spectra fitting that was recommended by A.Fomenko at 25.02.1994, see up. To see the resulting tables you can "star" the following commands: LIST 'H01LNS.JPF.R62575' LIST 'H01LNS.JPF.R64472' Sorry for delay with the jobs and even the report to LLO. -------- Unfortunately, I dont yet have any comments on the results ... Anybody is free to understand in his own way, and to compare with his own fitting. Also, it is possible to compare with some earlier fittings reported in this LLO (see H01LNS up). Lollii Shtarkov, Moscow. 04.03.1994 ===== dice2 ======== H1KFOM === Pedestal Distributions.... Today during H1Lumi test with H1CDAQ it were written two Runs on temporary cartridges with the same condition as at 68096 (see message from 03.03.94). It were Runs 69091 and 69108 (47055+62947 ev.). Today morning Yu.Soloviev switched on HV on all channels of H1Lumi. As seems situation with sigmas became better then with HV off on ETdep,PDdep and ETdep+PDdep distributions: Mean,Sigma(GeV) and Entries Values for ETdep -0.426 0.865 108850 Mean,Sigma(GeV) and Entries Values for PDdep -0.013 0.491 109377 Mean,Sigma(GeV) and Entries Values for ETdep+PDdep -0.436 0.995 108704 Mean Values(GeV) at Pedestal Distr. for ET cells -0.003 -0.001 -0.009 -0.001 -0.010 -0.003 -0.008 0.002 0.000 0.003 -0.001 0.003 0.003 0.000 0.005 0.000 0.000 -0.009 -0.003 -0.013 -0.008 -0.009 0.000 -0.007 -0.005 -0.003 -0.005 -0.007 -0.003 -0.005 -0.003 -0.003 -0.003 0.000 0.001 -0.003 0.002 -0.008 -0.002 -0.004 0.000 -0.005 -0.003 -0.003 0.002 -0.008 -0.001 -0.004 0.006 Mean Values(GeV) at Pedestal Distr. for PD cells 0.002 0.004 0.002 -0.008 -0.005 -0.003 -0.001 -0.002 -0.009 -0.004 -0.003 -0.004 -0.005 -0.003 -0.007 0.000 0.006 -0.007 0.009 0.014 0.009 0.002 0.005 -0.001 0.004 0.003 Sigma Values(GeV) at Pedestal Distr. for ET cells 0.034 0.021 0.032 0.034 0.055 0.029 0.033 0.026 0.024 0.025 0.035 0.028 0.029 0.048 0.036 0.018 0.017 0.018 0.019 0.031 0.026 0.025 0.017 0.020 0.017 0.019 0.021 0.020 0.018 0.016 0.016 0.017 0.015 0.013 0.016 0.025 0.018 0.023 0.017 0.015 0.016 0.018 0.022 0.014 0.019 0.028 0.014 0.014 0.022 Sigma Values(GeV) at Pedestal Distr. for PD cells 0.027 0.025 0.026 0.030 0.025 0.018 0.015 0.014 0.021 0.025 0.020 0.018 0.025 0.016 0.018 0.015 0.024 0.021 0.023 0.028 0.057 0.015 0.019 0.016 0.026 0.044 Entries into Histos for Pedestal Distr. (ET) 104118 97152 103788 105764 106808 104443 105341 100836 102016 103493 104660 103872 99624 107063 106615 99518 100699 102893 102561 107235 105272 104528 103661 106349 104582 104150 102994 103421 102049 105335 96746 103663 103126 84253 97768 106552 98261 102353 103460 103394 104522 101091 103269 99014 93833 101223 100343 95991 103695 Entries into Histos for Pedestal Distr. (PD) 101372 103865 104225 105004 105946 104282 101435 103123 105552 105831 104268 102642 104949 103155 104527 104162 105203 103905 105739 106642 106511 103535 104820 104702 107487 103232 Averaged values Means and Sigma(GeV) for ET cells -0.003 0.023 Averaged values Means and Sigma(GeV) for PD cells 0.000 0.023 It is possible to see high mentioned digits at LOOK-binary data set H1KFOM.LOOK1.R69091 which was produced at dice2-environment and was transferred through ftp-protocol into DESY IBM-storage: H1KFOM.LOOK1.R69091 FAST20 3010200E 00000 DASD At both cases (HV ON and HV OFF) e04,e13 and p20 have another sigmas characteristics then others channels(more then another on factor 2 or more). VCsp-channel has sigma=0.044 with HV ON and sigma=0.057 GeV with HV OFF (more then average sigma for both arms channels (0.023 GeV). 05.03.1994 ===== FIC#1 ======== H1KFOM === New CMOS-installation..... Due to preparing of new release of H1LumiOnLineFIC1-program (previous release name was H1LumiOnLineFIFOC) with additional readout routine ReadNC (12 FADC-channels) and LRNA-bank producing at H1LUMI-data flow I got some additional memory at Absolute Addresses Page at FIC#1 local DPM memory mapping (from $2002CD5C up to $2002D2A8). It was the reason for changing of CMOS-option installation at FIC#1 environment. Before yesterday: a) for installation CMOS-option: b) for removing of CMOS option: CMOS CMOS 1 1 10FE 10FF . . After 04.03.94: a) for installation CMOS-option: b) for removing of CMOS option: CMOS CMOS 1 1 E5FE E5FF . . Naturally it were changed Reboot_both_FICs and Reboot_FIC1 data set for rebooting of FIC#1 program (new Load addrees is 2002FA00, before yesterday this address was 2002CE00). Old Release of OnLineProgram for FIC#1 which was used at most part of Autumn93 H1DataTaking now renamed into H1LumiOnlineFIC1Old and moved into another directory HD_LPI:MPW:FomenkoAfter02.02.94:Applica- tions. At this directory Reboot_FIC1, Reboot_FIC2, Reboot_both_FICs data sets and old release of OnLineProgram for FIC#2 are kept. Old release of OnLineProgram for FIC2 which was used most part of Autumn93 H1DataTaking now renamed into H1LumiOnLineFIC2Old. If anybody wants to use old releases - change names at high mentioned directory: H1LumiOnLineFIC1 ---> H1LumiOnLineFIC1New H1LumiOnLineFIC1Old ---> H1LumiOnLineFIC1 H1LumiOnLineFIC2 ---> H1LumiOnLineFIC2New H1LumiOnLineFIC2Old ---> H1LumiOnLineFIC2 Conventions are next: 1)H1LumiOnLineFIC1 and H1LumiOnLineFIC2 - are the new (fresh, current) releases if at directory there are not H1LumiOnLineFIC1New or H1LumiOnLineFIC2New programs. 2)H1LumiOnLineFIC1Old and H1LumiOnLineFIC2Old are the old (Autumn93) releases if at directory there are not H1LumiOnLineFIC1New or H1LumiOnlineFIC2New programs. Creation dates for old (Autumn93) releases: H1LumiOnLineFIC1 (26.02.94, 15:44). H1LumiOnLineFIC2 (16.02.94, 18:51). They are full doublers of releases which are kept at old working directory (HD_LPI:MPW:FomenkoAfter08.01.93:NewFICReconstruction.09.03. 93:) with names H1LumiOnLineFIFOC and H1LumiOnlineRec. Old release of H1LumiOnLineFIC1 is working with new CMOS-option. Now CMOS option is ON. 08.03.1994 ===== FIC#1 ======== H1KFOM === H1LumiOnLineFIC1 releases. It was prepared the new release of H1LumiOnLineFIC1.The next changings with previous (Autumn93) release were made: 1) it was realized the I.Sheviakov's request: a) it was added ReadOut of 12 first FADC-channels of 11th FADC1001 (B040A000-B040AFFF as its are seen from MacII); b) it was added new bank LRNA into H1LumiDataFlow (as temporary bank - may be for only for H1DatTaking94) for supporting of posibility to get some amplitude information from neutron counters c) ReadoutError (4C000000) which was used for labeling bad LRTC-banks at previous releases now will label so called "bad" LRNA-banks if its will appear during readout operation; d) average incrementing of ReadoutTime (time up to FrontEndReady issuing from Luminosity-branch) is not large - near 100-150 mksec; 2) it was upgraded ReadOut chain for VETE-bank - bufffers for VETE- bank ReadOut now is disposed at Local FIC1 DPM (at previous releases its were disposed at ordinary DPM). As it is waited - it must decrement ReadOutTime for Luminosity branch (up to 10 %). 3) after this two changings ReadOutTime with standard bank set a little less then at old release (5-7%). Creation dates for new releases: H1LumiOnLineFIC1 (07.03.94, 17:28). First Run with new LRNA-bank was written 08.03.94 - Run 69234. It was prepared but not tested yet else one release of H1LumiOnLine- FIC1-program at which: 1) it was removed creation LRTE-bank and all ReadOut chains connected with this creation (due to message of S.Levonyan that he removed this bank presence analyze at H1LumiRec-package); 2) it was involved so-called prescale factor for LRTN and LRTF-banks - it means that LRTN and LRTF banks will be presented not at all H1Events but only at each 10th or more rare.Prescale factor for appearence LRTN,LRTF-banks at H1DataFlow is one of H1LumiDAQ options - You can change it from 1 (no prescale) up to very large number. LRTN-bank will be present at first event of each H1Run. Default prescale factor value for LRTN/LRTF-banks is 10. 09.03.1994 ==================== F11LEV === Remark on new online ..... 1) Removing LRTE is ok with me, but suppressing (prescaling) LRTN is bad due to the following reasons: a) it is used in the Event Display b) in the analysis this is the only source of the trigger energies and if not present in each event, makes some analyses impossible 2) I have an impression, that the changes in the online program does not follow the priority. Very often non-important options/news are added, while some urgent items are still missing (see e.g. a task list we discussed in Nov-Dec '93). 10.03.1994 ==================== F11LEV === Payment for LPI in 1994... As you might know, we have asked DFG to pay 60 men-months for H1 work in 1994, our usual amount during last years. As I was told today, DFG gave a negative answer, refusing to pay stay expences for LPI members in 1994. Sofar DESY takes care, however there is no guarantee that '94 LPI request will be satisfied on 100%. 11.03.1994 ===== FIC#1 ======== H1KFOM === H1LumiOnLineFIC1-release.. It was prepared (but not tested with H1CDAQ yet) the new release of H1LumiOnLineFIC1 with posibility of putting into LRPE-bank (only into this bank - no any additional information at LRPF-bank) of response from WaterCounter (veto channel). This channel is labeled as 26th input (if it exists at event). There are no any changing at LRPP- bank structure and length (there is no calibration for WCvc). Previous release which was proclaimed earlier (LRTE-bank absence and prescale factor 10 for LRTN/LRTF-banks) was tested with H1CDAQ. As example You can see J.Olsson's RunSummary for Run 69357 which was recorded yesterday into IBM cartridge. >>> Start RUN Summary --------------------- <<<<< NRUN NrofEvts Ev.Fst Ev.Lst DATE TIME <---> DATE TIME 69357 25183 0 4FFFFFFF 940310 113335 940310 113742 SYSMOD DAQMDE XI FEB CTLVRS DETSTA I.LUMI R-Devs 00005801 80940045 0202 003C 00000000 BEFF01BC 0 35 CTCNF1 CTCNF2 TRIG1 TRIG2 NMASKST0 NMASKST1 NMASKST2 NMASKST3 FFDE0204 07EE2F2D 00000040 00000000 00000000 00000000 00000000 00000000 Eelec Eprot Polari Ecurr Pcurr Bfield Bcurr Patmo e-p-bunches 0 0 0 0 0 0 0 1968 7 19 BBL3:BEFF01BC HVST:005F7FFF BANKs:Name Freq Av.Length Mn.Length Mx.Length 48454144:HEAD 25150 26 26 26 54535443:TSTC 50298 10 10 10 4C524545:LREE 25150 47 31 50 4C525045:LRPE 25150 25 20 27 54475050:TGPP 25148 53 53 53 4C525446:LRTF 2515 19 19 19 4C52544E:LRTN 2515 14 14 14 48454152:HEAR 25150 17 17 17 5A584354:ZXCT 1 1274 1274 1274 5A524354:ZRCT 25149 22 22 22 4B454550:KEEP 33 1 1 1 56455445:VETE 25150 29 29 29 5A544255:ZTBU 1 226 226 226 4C525453:LRTS 26 6853 6853 6853 54435443:TCTC 1 72 72 72 54454C31:TEL1 50296 35 35 35 5A545052:ZTPR 1 86 86 86 4C525050:LRPP 25150 14 14 14 4C524550:LREP 25150 26 26 26 54435444:TCTD 1 35 35 35 54435446:TCTF 1 10 10 10 54435449:TCTI 1 21 21 21 4C555452:LUTR 1 40 40 40 5433494F:T3IO 25148 52 52 52 4C524E41:LRNA 25150 37 37 37 58434344:XCCD 8 36 15 80 Average Ev. Length: 1.749 kBytes; Min & Max 1.102 29.590 TSTC checks & statistics (NLOG(0:3),TSTCLG(1:8,9:16) 0 3 25139 8 25148 0 0 0 0 0 0 0 25145 0 0 0 0 0 0 0 RunStart/KEEP/FARM/RunEnd records on file F11JOL.H01.TESTCAL.G6341V00 Output Data Set Name HERA04.H1RAWD.C9400018 Permanent >>> End RUN Summary -------------------- <<<<< Really You cannot find LRTE-banks at this data set and You can see that only at 10% of events there exists LRTN/LRTF-banks (H1Lumi dead-time decremented at average on 200 mksec(!!!) from 1600-1700. H1 Event Display at H1CDAQ-area worked normally with absence of LRTN- bank - there is absence of Trig.energies and Current Value of Lumi at pictures. We must find some decisions of decrementing of H1Lumi Dead Time. H1Lumi one of the branches at H1CDAQ which have largest dead time. I want to put Your attention on the last lines of RunSummary: TSTC checks & statistics (NLOG(0:3),TSTCLG(1:8,9:16) 0 3 25139 8 25148 0 0 0 0 0 0 0 25145 0 0 0 0 0 0 0 This information means that at 3 events from 25K Luminosity branch it was not sent any data about some L2Keep-events. It is very old problem - I spent some days on fixing of this bug and without success. There are no visible bugs. 11.03.1994 ===== H1CDAQ ======= H1KFOM === SGI-Logging Mode at H1CDAQ From yesterday Manfred Zimmer recommended to log all data sets on SiliconGraphics. It is new option at H1DAQ-supervisor. Today there is categorical request: write all data sets only(!) on SiliconGraphic. As seems there are no large changing at finding of data sets at IBM-storage. All SGI-data sets are dumped into DESY IBM and Jan Olsson's summary fixed first data sets which were recorded through SGI-logging option. This is the information from Jan Olsson: "...... SENDMAIL message Date: THU, 10 MAR 94 10:48:06 +0100 From: Alexander Fomenko Subject: SGI Logging To: Jan-Erik Olsson A.M. Fomenko, DESY -FH1K- Notkestrasse 85, D 2000 Hamburg 52 Tel. (040) 8998-2568(121 1b),-4356(Hera Nord), -2433(125 1c) (permanent address: Lebedev Phys.Inst., Moscow, Russia) NODE: DHHDESY3 USERID: H1KFOM Hello, Dear Jan, I am sorry for my unwaitable mail. I have some questions which are connected with new mode of data logging at H1 CDAQ. I mean SGI Logging. Today Manfred Zimmer wrote at H1 CDAQ Pinboard -"please log to SGI" Some questions due to this logging mode: - will be possible to use Your covinent tools "CARTL" and H1Data service for getting some information about RawData cartridges which will be logged at this mode? - where SGI logged data sets are disposed - or - is it possible to find its at dice2-environment (may be special pool exists at dice2- disk storage for keeping of high mentioned data sets) - or - may be SGI-disk is intermediate area and it is possible to find SGI-logged data sets at IBM cartridges? - as example - at H1CDAQ Supervisor Screen it is possible to see that data are written into C9400001-cartridge. Which is full name of this data set at dice2-environment [XXXXX.YYYYY]ZZZZZZ.C9300001 and which full name will be at IBM-environment - ordinary name HERA04.H1RAWD.C9400001 is busy now (first 17 data set names are used with first permanent datat sets at 94) C9400001-C9400017? Answer me please. Sincerely, Alexander Fomenko ------------------------------------------------------------------------------- READMAIL message stored : THU, 10 MAR 94 11:56:45 MEZ Date: THU, 10 MAR 94 11:36:57 +0100 From: Jan-Erik Olsson Subject: MAIL from F11JOL at DHHDESY3 To: Alexander Fomenko Hi! The idea is to use CARTL still, since all data will be on the IBM cartridges anyhow. CARTL is not yet available on SGI, but will be soon. No access to data on SGI cartridges is foreseen, only use the IBM ones, which will be available just as if there had been a dump on IBM. The CARTL information on IBM will be available also just as soon. The same numbering as on IBM is followed. Please ignore the information on the Manfred screen, till it has been properly set, which will happen this afternoon. The cartridge numbering will join up with the present IBM cartridges. More details will be given by Ralf Gerhards in the H1 meeting this afternoon... Cheers, Jan 14.03.94 ====================== H1KSOL ==== New ET geometry .......... Today the platform with ET was moved to the e-beam pipe as much as was possible. From now on we have a new ET geometry: /| e-beam pipe //| !axis ! ( 149.3 +/- 0.3) mm (it was 159.8 mm) //| ! |<----------------------------------------->| //| ! | ! | ET-center //| ! | ! ( 117.5 +/- 0.2) mm ---+--- //| ! | !<-------------------------------------->| //| ! | ! | //| | //| d(beam pipe) = 63.5 mm (from drowing) | //| | //| | /|<-------- Side plane of magnet element 34QS10 | 14.03.1994 ===== FIC#2 ======== H1KFOM === H1LumiOnLineFIC2-release.. It was prepared the new release of H1LumiOnLineFIC2-program. At this release it was realized some requests to OnLine Luminosity calculation with ET*PD-method (LRTL-bank). a)it was made more right (as seems) calculation of Integral Luminosity values (RB1 and RB11) at first period of H1Run operation (from start H1Run up to first Luminosity measurement at this Run). As start Run it was choosen moment of getting at Luminosity branch of the first L2Keep event at this H1Run (not RunStart-signal). b) it was made additional protection of Luminosity calculation from so called "spikes". All bunch current values will be tested on reasonable values before using its at all calculations. 1000 mkA and 7 mkA values which were used at FindPilotFIC.f routine were choosen for checking of e-bunch current values. Before this high mentioned e-bunch current values were checked only on positive values. 15.03.94 ====================== H1KSOL ==== New ET geometry again..... Real diameter of e-beam pipe as it turned out differs from value on drowing.I measured it by different tools and got the same result: 63.8 mm.So the new values are: /| e-beam pipe //| !axis ! ( 149.4 +/- 0.3) mm (it was 159.8 mm) //| ! |<----------------------------------------->| //| ! | ! | ET-center //| ! | ! ( 117.5 +/- 0.2) mm ---+--- //| ! | !<-------------------------------------->| //| ! | ! | //| | //| d(beam pipe) = 63.8 mm (direct measurment) | //| | //| | /|<-------- Side plane of magnet element 34QS10 | 16.03.1994 ===== FIC#1 ======== H1KFOM === First Run with VCv-channel Today it was possible (due to H1CDAQ was free some time) to test new release of H1LumiOnlineFIC1 with VCv-channel at H1Lumi-readout. From the next RunSummary You can see that maximal length of LRPE-bank now is 28 (before this this digit was 27). >>> Start RUN Summary --------------------- <<<<< NRUN NrofEvts Ev.Fst Ev.Lst DATE TIME <---> DATE TIME 69734 37260 0 37227 940316 105139 940316 105705 SYSMOD DAQMDE XI FEB CTLVRS DETSTA I.LUMI R-Devs 5801 80940145 202 1BB 0 0 0 42 CTCNF1 CTCNF2 TRIG1 TRIG2 NMASKST0 NMASKST1 NMASKST2 NMASKST3 FFDE0200 7E80000 40 102 0 0 0 0 Eelec Eprot Polari Ecurr Pcurr Bfield Bcurr Patmo e-p-bunches 0 0 0 0 0 0 0 1002 182 75 BBL3:F6EF21B8 HVST:605B78FF BANKs:Name Freq Av.Length Mn.Length Mx.Length 48454144:HEAD 37219 13 13 13 54535443:TSTC 37224 10 10 10 4C524545:LREE 37215 46 34 50 4C525045:LRPE 37215 26 20 28 54475050:TGPP 37218 53 53 53 4C525446:LRTF 3722 19 19 19 4C52544E:LRTN 3722 14 14 14 48454152:HEAR 37219 17 17 17 5A584354:ZXCT 1 1360 1360 1360 5A524354:ZRCT 37219 22 22 22 4B454550:KEEP 41 1 1 1 56455445:VETE 37215 29 29 29 4C525453:LRTS 33 6853 6853 6853 59345454:Y4TT 37218 12 12 12 54435443:TCTC 1 72 72 72 54454C31:TEL1 74436 35 35 35 5A545052:ZTPR 1 1 1 1 4C525050:LRPP 37215 14 14 14 4C524550:LREP 37215 26 26 26 54435444:TCTD 1 35 35 35 54435446:TCTF 1 10 10 10 54435449:TCTI 1 21 21 21 4C555452:LUTR 1 40 40 40 4C524E41:LRNA 37215 37 37 37 4D4F4449:MODI 37224 2 2 2 4D4F444F:MODO 37224 2 2 2 4D4F4453:MODS 37224 2 2 2 58434344:XCCD 8 276 20 500 59544C34:YTL4 37218 4 4 4 59564152:YVAR 37218 6 6 6 Average Ev. Length: 1.563 kBytes; Min & Max 1.035 28.805 TSTC checks & statistics (NLOG(0:3),TSTCLG(1:8,9:16) 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 IGGDRU(IRU) 0 RunStart/KEEP/FARM/RunEnd records on file TESTCAL.R0069734 Output Data Set Name HERA04.H1RAWD.C9400023 Permanent Cartridge:: Average Event Length: 3.627 kBytes Cartridge:: Total nr of events and banks 686181199050 ==> 261.496 MB >>> End RUN Summary -------------------- <<<<< It were not fixed any problems with new length of LRPE-bank at all H1CDAQ-chain. H1LumiOnLineFIC1 was tested with Random trigger (10kHz and 100 Hz) from H1CTrig. It were fixed logging rate near 115Hz during 10kHz random trigger rate and near 85 Hz during 100 Hz trigger rate. Data Logging was made through SGI Logging H1CDAQ option. Problems with "Luminosity Timeout during running on Event XXXXX" are existing now too (as at all previous releases). There are some details: when SGI-logging was overflowed (at 12:17 today it was fixed message from FARM - "FARM:No SGI buffer available") I switched off any logging of data and started Run 69739 with 100 Hz random trigger rate (near 30 Hz of L2keep-rate was fixed). Run 69739 was started at 12:18:11 and was stopped at 12:49:44 and it was not fixed any warning messages as "Luminosity Timeout during running on Event XXXXX". It is pity but as seems it is impossible to make L2Keep-rate more then 30 Hz without any Logging Task involved into H1CDAQ-operation. 16.03.1994 ===== dice2 ======== H1KFOM === H1LumiOnline-sources...... From "H01LNS's proposal and F11LEV's re:proposal" (see e-mail of S.Levonyan from 09.03.94 to all H1Lumi-people): ".... # Some info-proposals: "To copy from MAC's to IBM(LPI.PER) any readable source of ON-line" A: I agree. But for .... ...." It is possible to find all sources of program units which were used for H1LumiOnline-programs creation (for FIC#1 and FIC#2). Most of this sources are FORTRAN sources (with RTF-conventions and with Assembler Motorola 68020 insertions). There are some additional files with memory mapping of H1Lumi's DPM, with scripts for program and libraries creation. There are H1LumiRec-package which was got at March93 from current CMZ-file of H1REC-package and (a little) changed due to RTF-conventions. These sources are from last Autumn'93 releases. Moving into IBM-storage is not very good due to problems with tab-symbols, line length etc. Anybody can read high mentioned sources at dice2-environment at the next directories: drwxr-x--- 2 fomenko h1 2560 Feb 26 16:27 H1LumiOnlineFIC1 drwxr-x--- 2 fomenko h1 2048 Feb 18 17:21 H1LumiOnlineFIC2 These directories are disposed at /home/local/fomenko -directory. As You can see access rights are installed for group on read. All sources for monitoring tools (SuperCard technology) are written as Pascal-procedures and (as seems) will be not understandable without SuperCard scripts where its are using. 16.03.1994 ==================== F11LEV === B-H cross section for 1994 For the new ET position, announced by Yu.Soloviev, and new electron beam energy expected in 1994 data taking (27.53 GeV) the value of the accepted B-H cross section (coinsidence method: trigger ET*PD) is given in a table below (syst. error due to the possible 1mm ET-position shift is 1%): ----------------------------------------------------------------------- XCUT[cm] = 7.70 7.60 7.00 6.60 6.00 5.50 ACCEPTANCE= 0.52695 0.51463 0.47275 0.44251 0.39858 0.36278 SIGMA[mb] = 32.38 31.62 29.05 27.20 24.47 22.30 ----------------------------------------------------------------------- Note, that in calculation a standard e-beam optics were assumed (i.e. as in 1992). 17.03.94 ====================== H1KSOL === Dosimeters ............... 20 dosimeters were instaled in the Lumi detectors areas yesterday. Their positions are described below: ET : ( see sketch for details) #1 - on the side surface of element 21 near beam plane (will be moved with table). #2 - on the thin plastic support in front of #1 (will be not moved with the table) #3 - on the side surface of element 27 near beam plane (will be moved with table). #4 - on the side surface of ET (el.27) inside lead case near beam plane close to the front face of ET (will be moved) #5 - on the side surface of lead case in front of #4 (will be moved) #6 - on the side surface of ET close to beam pipe inside lead case close to case for HV dividers near beam plane. #7 - on the side surface of lead case in front of #6. Front view of ET Top view of ET ------- e-beam || ------- || | ! ^ || | | || e-beam| ! | #7*||*|#6 | ||<---- lead case + #1* *#3 | || | | || | ! | || | | || | ! | || | | || ------- | || | | || | || | | || | || | #4|*||*#5 | || |#1 | || | || *-------* || + * #3 #2 * - indicates position of dosimeter VC: #8 - near e-beam plane on the edge of the front side of VC near to e-beam pipe #9 - the same as #8 but on the edge far from e-beam pipe #10- near e-beam plane on the edge of the back side of VC far from e-beam pipe #12- the same as #10 but near e-beam pipe #11- in the center of back surface of VC near e-beam plane #13- in the center of side sufface near e-beam pipe behind the lead shielding #15- the same as #13 but in front of lead shielding #14- the same as #13 but on side surface far from e-beam pipe #16- the same as #14 but in front of lead shielding Top view of VC * #16 ---------------- lead shielding ---------------- ____ * #9 __*_ | #10 * ____ #14 | | | | |*#11 | | *|____ | #8 ____ | * ____|*#12 #13 ---------------- ---------------- lead shielding * #15 +---------------> e-beam PD: #17- on the front side near el.#10 (will be moved) #18- in the center of the front side (el.#12)(will be moved) #19- on the front side near el.#14 (will be moved) #20- on the thin plastic support in front of #18 (will be not moved) Front view of PD ------------- | | | | | | | | #17 *| * |* #19 +e-beam | #18 | | | | | | | ------------- 18.03.1994 ===== FIC#2 ======== H1KFOM === LUMM-bank contents ....... Yesterday it was prepared the new release of H1LumiOnLineFIC2- program with posibility of new LUMM-bank filling and preparing it for retranslation into H1DataFlow - as KEEP-bank (once per 10 sec. together with LRTL-bank). LRTS-bank will be removed from H1DataFlow forever (as seems) and will be replaced with LUMM-bank. Release was tested with test pulses and as seems is working reasonable. Contents of LUMM-bank is oriented on Luminosity measurement with PD1-trigger element (220 bunch rates of PD1-trigger element and 1 total rate of PD1-trigger element will be presented at LUMM-bank). Total rate od PD1-trigger element will be got from 32bit HW counter (the same type which was used for getting ET&PD and VC total rate). Bunch rates will be got from Histo Memory. As You remember - at last part of H1DataTaking93 - 3rd input of Histo Memory was connected to PD1-trigger element (0th-VC,1th-PD,2nd-ET,3rd-PD1,4th-PD2). Time stamp (as H1Lumi Local Timer contents) of 31x220 counters readout moment will be as 222th B32-word at this bank. It is proposed to put into contents of this bank 4 histos with photon arm energy spectra: 1) PDrec for sample with last 5000 (or 10000) events - with 75 bins from 0. up to 35.Gev with underflow,overflow and bad reconstruction additional bins and with time stamps of first and last events at spectrum; 2) PDdep for sample with last 5000 (or 10000) events - with the same characteristics; 3) PDrec for sample with last 5000 (or 10000) events - from pilot-bunches only; 4) PDdep for sample with last 5000 (or 10000) events - from pilot-bunches only. Deposited energy spectra are attractive due to posibilty of getting of large statistics at short time (absence of reconstruction CPU needs). Spectra with PDrec/PDdep from pilot bunches will be useful for investigation of PhotonEnergyFlux-method of Lumi-calculation. And the last - each 10 sec. it will be calculated 27 different rates which will be useful for monitoring and investigation at off-line. It is proposed to use 9 trigger elements and trigger bits for high mentioned investigation abd monitoring: 1. VC; 2. PD; 3. ET; 4. PD1; 5. PD2; 6. ET&PD; 7. ET&PD&nVC; 8. PD&nVC; 9. ET&nPD&nVC. For each from high mentioned trigger elements (bits) it will be calculated 3 Rates: 1. total trigger rate which is calculated as sum of all 220 bunch rates for this trigger element/bit; 2. trigger rate which is calculated as sum of collided and e-pilot bunch rates for this trigger element/bit; 3. trigger rate which is calculated as sum of e-pilot bunch rates for this trigger element/bit. The length of LUMM-bank will be 220+1+1+(1+75+1+1+2)*4+(3*9)+2(mini header) = 571 longwords = 2284 bytes. As You remember the length of LRTS-bank was 6853 longwords = 27412 bytes. Yesterday it was prepared and tested the filling of 220+1+1+(3*9)+ 2(mini header) values. Histos preparing and filling will be made later. Format of future LUMM-bank will be B32. Full description of LUMM-bank will be prepared later. It was made accounting of bunch rate shifts (rate(0)<--->rate(1)) and mentioned at message from 28.02.1994 9-bunch shift for 4th input into Histo Memory (PD2 was connected last time to 4th input). 19.03.1994 ==================== H1KFOM === LUMM-bank description.... (project) ! ! BANKname BANKtype !Comments ! TABLE LUMM B32 ! LUmi_Measurements_with_alternative ! -Methods ! (non-event bank at KEEP-environment) ! ATTributes: ! ----------- ! PD1 - is one of H1Lumi Trigger Elements and one from H1Lumi Photon ! Arm Energy Triggers (PD,PD1,PD2); ! PD_LTH1- low threshold value for PD1 it is possible to find at LUTR- ! bank contents (HW-setting). ! !COL ATT-name FMT Min Max !Comments ! ! ! negative value at the next 220 variables means technical problems ! with rate calculations 1 PD1RT000 I -INF +INF ! Rate of Photon Arm Energy Trigger(PD1) ! in bunch 0, in 0.1 Hz units ! (averaged over the last 10 sec.) 2 PD1RT001 I -INF +INF ! Rate of Photon Arm Energy Trigger(PD1) ! in bunch 1, in 0.1 Hz units ! (averaged over the last 10 sec.) ! ................................................................... 220 PD1RT219 I -INF +INF ! Rate of Photon Arm Energy Trigger(PD1) ! in bunch 219, in 0.1 Hz units ! (averaged over the last 10 sec.) 221 PD1RTTOT I 0 +INF ! Total Photon Arm Energy Trigger Rate ! (PD1), in 0.1 Hz units ! (averaged over the last 10 sec.) 222 LLTLUMM I -INF +INF ! H1Lumi Local Timer value (2950Hz tick) ! at Luminosity measurement moment ! ! negative value at the next 27 variables means technical problems with ! rate calculations !-------------- VC - trigger element ---------------------------------- 223 VCRTOT | -INF +INF ! Total Water Cherenkov Counter Trigger ! Rate (VC), in 0.1 Hz units, ! as sum of all 220 VC-trigger rates, ! (averaged over the last 10 sec.) 224 VCRTOT1 | -INF +INF ! Total' Water Cherenkov Counter Trigger ! Rate (VC), in 0.1 Hz units, ! as sum of VC-trigger rates at collided ! and e-pilot bunches ! (averaged over the last 10 sec.) 225 VCRTOT2 | -INF +INF ! Total" Water Cherenkov Counter Trigger ! Rate (VC), in 0.1 Hz units, ! as sum of VC-trigger rates at all ! e-pilot bunches ! (averaged over the last 10 sec.) !-------------- PD - trigger element ---------------------------------- 226 PDRTOT | -INF +INF ! Total Photon Arm Energy Trigger ! Rate (PD), in 0.1 Hz units, ! as sum of all 220 PD-trigger rates, ! (averaged over the last 10 sec.) 227 PDRTOT1 | -INF +INF ! Total' Photon Arm Energy Trigger ! Rate (PD), in 0.1 Hz units, ! as sum of PD-trigger rates at collided ! and e-pilot bunches ! (averaged over the last 10 sec.) 228 PDRTOT2 | -INF +INF ! Total" Photon Arm Energy Trigger ! Rate (PD), in 0.1 Hz units, ! as sum of PD-trigger rates at all ! e-pilot bunches ! (averaged over the last 10 sec.) !-------------- ET - trigger element ---------------------------------- 229 ETRTOT | -INF +INF ! Total Electron Arm Energy Trigger ! Rate (ET), in 0.1 Hz units, ! as sum of all 220 ET-trigger rates, ! (averaged over the last 10 sec.) 230 ETRTOT1 | -INF +INF ! Total' Electron Arm Energy Trigger ! Rate (ET), in 0.1 Hz units, ! as sum of ET-trigger rates at collided ! and e-pilot bunches ! (averaged over the last 10 sec.) 231 ETRTOT2 | -INF +INF ! Total" Electron Arm Energy Trigger ! Rate (ET), in 0.1 Hz units, ! as sum of ET-trigger rates at all ! e-pilot bunches ! (averaged over the last 10 sec.) !------------- PD1- trigger element ----------------------------------- 232 PD1RTOT | -INF +INF ! Total Photon Arm Energy Trigger ! Rate (PD1), in 0.1 Hz units, ! as sum of all 220 PD1-trigger rates, ! (averaged over the last 10 sec.) 233 PD1RTOT1 | -INF +INF ! Total' Photon Arm Energy Trigger ! Rate (PD1), in 0.1 Hz units, ! as sum of PD1-trigger rates at ! collided and e-pilot bunches ! (averaged over the last 10 sec.) 234 PD1RTOT2 | -INF +INF ! Total" Photon Arm Energy Trigger ! Rate (PD1), in 0.1 Hz units, ! as sum of PD1-trigger rates at all ! e-pilot bunches ! (averaged over the last 10 sec.) !------------- PD2- trigger element (as default setup) ---------------- !------------- may be switched on another trigger element ------------- 235 PD2RTOT | -INF +INF ! Total Photon Arm Energy Trigger ! Rate (PD2), in 0.1 Hz units, ! as sum of all 220 PD2-trigger rates, ! (averaged over the last 10 sec.) 236 PD2RTOT1 | -INF +INF ! Total' Photon Arm Energy Trigger ! Rate (PD2), in 0.1 Hz units, ! as sum of PD2-trigger rates at ! collided and e-pilot bunches ! (averaged over the last 10 sec.) 237 PD2RTOT2 | -INF +INF ! Total" Photon Arm Energy Trigger ! Rate (PD2), in 0.1 Hz units, ! as sum of PD2-trigger rates at all ! e-pilot bunches ! (averaged over the last 10 sec.) !------------- ET&PD - trigger bit ------------------------------------ 238 RLUMTOT | -INF +INF ! Total Rough Luminosity Trigger Bit ! Rate (ET&PD), in 0.1 Hz units, ! as sum of all 220 ET&PD-trigger rates, ! (averaged over the last 10 sec.) 239 RLUMTOT1 | -INF +INF ! Total' Rough Luminosity Trigger Bit ! Rate (ET&PD), in 0.1 Hz units, ! as sum of ET&PD-trigger rates at ! collided and e-pilot bunches ! (averaged over the last 10 sec.) 240 RLUMTOT2 | -INF +INF ! Total" Rough Luminosity Trigger Bit ! Rate (ET&PD), in 0.1 Hz units, ! as sum of ET&PD-trigger rates at all ! e-pilot bunches ! (averaged over the last 10 sec.) !------------- ET&PD&nVC - trigger bit -------------------------------- 241 PLUMTOT | -INF +INF ! Total Pure Luminosity Trigger Bit ! Rate (ET&PD&nVC), in 0.1 Hz units, ! as sum of all 220 ET&PD&nVC-trigger ! rates (averaged over the last 10 sec.) 242 PLUMTOT1 | -INF +INF ! Total' Pure Luminosity Trigger Bit ! Rate (ET&PD&nVC), in 0.1 Hz units, ! as sum of ET&PD&nVC-trigger rates at ! collided and e-pilot bunches ! (averaged over the last 10 sec.) 243 PLUMTOT2 | -INF +INF ! Total" Pure Luminosity Trigger Bit ! Rate (ET&PD&nVC), in 0.1 Hz units, ! as sum of ET&PD&nVC-trigger rates at ! all e-pilot bunches ! (averaged over the last 10 sec.) !------------- PD&nVC - trigger bit ----------------------------------- 244 PHDTOT | -INF +INF ! Total Photon Detector Trigger Rate ! (PD&nVC), in 0.1 Hz units, ! as sum of all 220 PD&nVC-trigger ! rates (averaged over the last 10 sec.) 245 PHDTOT1 | -INF +INF ! Total' Photon Detector Trigger Rate ! (PD&nVC), in 0.1 Hz units, ! as sum of PD&nVC-trigger rates at ! collided and e-pilot bunches ! (averaged over the last 10 sec.) 246 PHDTOT2 | -INF +INF ! Total" Photon Detector Trigger Rate ! (PD&nVC), in 0.1 Hz units, ! as sum of PD&nVC-trigger rates at ! all e-pilot bunches ! (averaged over the last 10 sec.) !------------- ET&nPD&nVC - trigger bit (eTAG) ------------------------ 247 ETAGTOT | -INF +INF ! Total eTAG Trigger Bit Rate ! (ET&nPD&nVC), in 0.1 Hz units, ! as sum of all 220 ET&nPD&nVC-trigger ! rates (averaged over the last 10 sec.) 248 ETAGTOT1 | -INF +INF ! Total' eTAG Trigger Bit Rate ! (ET&nPD&nVC), in 0.1 Hz units, ! as sum of ET&NPD&nVC-trigger rates at ! collided and e-pilot bunches ! (averaged over the last 10 sec.) 249 ETAGTOT2 | -INF +INF ! Total" eTAG Trigger Bit Rate ! (ET&nPD&nVC), in 0.1 Hz units, ! as sum of ET&nPD&nVC-trigger rates at ! all e-pilot bunches ! (averaged over the last 10 sec.) ! next 320 values are 4 Photon Arm Energy spectra under PD1-trigger; ! these spectra consist off last 10000 events (time stamps of first and ! last events at spectra are presented at this data) ! each energy spectrum consists of 75 bins from 0. up to 37.5 GeV !----- PDrec - reconstructed energy spectrum for all bunches events --- 250 PDRECUN | 0 +INF ! underflow bin (less then 0.0 GeV) 251 PDREC01 | 0 +INF ! 1st bin contents [ 0.0 - 0.5 GeV) 252 PDREC02 | 0 +INF ! 2nd bin contents [ 0.5 - 1.0 GeV) ! ........... 325 PDREC75 | 0 +INF ! 75th bin contents [37.0 - 37.5 GeV) 326 PDRECOV | 0 +INF ! overflow bin (more then 37.5 GeV) 327 PDRECBD | 0 +INF ! "bad" reconstruction counter 328 LLTFRSR | -INF +INF ! H1Lumi Local Timer value for first ! event at PDrec spectrum 329 LLTLSTR | -INF +INF ! H1Lumi Local Timer value for last ! event at PDrec spectrum !----- PDdep - deposited energy spectrum for all bunches events ------- 330 PDDEPUN | 0 +INF ! underflow bin (less then 0.0 GeV) 331 PDDEP01 | 0 +INF ! 1st bin contents [ 0.0 - 0.5 GeV) 332 PDDEP02 | 0 +INF ! 2nd bin contents [ 0.5 - 1.0 GeV) ! ........... 405 PDDEP75 | 0 +INF ! 75th bin contents [37.0 - 37.5 GeV) 406 PDDEPOV | 0 +INF ! overflow bin (more then 37.5 GeV) 407 PDDEPBD | 0 +INF ! "bad" SW-cluster finder result counter 408 LLTFRSP | -INF +INF ! H1Lumi Local Timer value for first ! event at PDdep spectrum 409 LLTLSTP | -INF +INF ! H1Lumi Local Timer value for last ! event at PDdep spectrum !----- PDrec - reconstructed energy spectrum for e-pilot bunches events 410 PDPRECUN | 0 +INF ! underflow bin (less then 0.0 GeV) 411 PDPREC01 | 0 +INF ! 1st bin contents [ 0.0 - 0.5 GeV) 412 PDPREC02 | 0 +INF ! 2nd bin contents [ 0.5 - 1.0 GeV) ! ........... 485 PDPREC75 | 0 +INF ! 75th bin contents [37.0 - 37.5 GeV) 486 PDPRECOV | 0 +INF ! overflow bin (more then 37.5 GeV) 487 PDRECBD1 | 0 +INF ! "bad" reconstruction counter 488 LLTFRSR1 | -INF +INF ! H1Lumi Local Timer value for first ! event at PDrec spectrum 489 LLTLSTR1 | -INF +INF ! H1Lumi Local Timer value for last ! event at PDrec spectrum !------ PDdep - deposited energy spectrum for e-pilot bunches events --- 490 PDPDEPUN | 0 +INF ! underflow bin (less then 0.0 GeV) 491 PDPDEP01 | 0 +INF ! 1st bin contents [ 0.0 - 0.5 GeV) 492 PDPDEP02 | 0 +INF ! 2nd bin contents [ 0.5 - 1.0 GeV) ! ........... 565 PDPDEP75 | 0 +INF ! 75th bin contents [37.0 - 37.5 GeV) 566 PDPDEPOV | 0 +INF ! overflow bin (more then 37.5 GeV) 567 PDDEPBD1 | 0 +INF ! "bad" SW-cluster finder result counter 568 LLTFRSP1 | -INF +INF ! H1Lumi Local Timer value for first ! event at PDdep spectrum 569 LLTLSTP1 | -INF +INF ! H1Lumi Local Timer value for last ! event at PDdep spectrum END TABLE 20.03.94 ====================== H1KSOL === Max.Shower C gas counter.. The prototype of Max Shower Cherenkov gas counter was installed yesterday on the movable platform of ET. The distance between axe of counter and e-beam axe is about 290 mm. Signal of this counter is connected to one of the el #48 of ET, HV is connected to channel #26 of crate #2,HV value of 1700 V was set,that corresponds to bias circuit current 0.62 ma. 21.03.1994 ===== FIC#2 ======== H1KFOM === H1LumiOnLineFIC2-release.. It was issued the new release of H1LumiOnLineFIC2-program with supporting of PDrec/PDdep-spectra making for LUMM-bank. Preparing of data for histos filling will be made later. It were changed important constants at FIC2-SW-environment as it was recommended by S.Levonian at his 16.03.94's message at LLC. Energy of e-beam is installed on 27.53 GeV (this value was 26.7 GeV). SigmaVisible for ET&PD-triggered Bremstrahlung events is installed on 32.38 mb (this value was 28.3 mb at H1DataTaking'93). Beam Energy is important constant for calibration procedure and PLTEFF-determination. SigmaVisible - for Current Luminosity Value estimation. 21.03.1994 ==================== H1KFOM === LRNA-bank description..... This bank (as seems) will be presented at H1DataFlow temporary (may be only at H1DataTaking94-period). The structure of this bank is the same as our LREF,LRPF-banks. Format is B16 too. As seems it is better to make pedestal substraction, linearization and ADC-counts making at OffLine step. Lenght of this bank is fixed. Mini-header exists. ! BANKname BANKtype !Comments ! TABLE LRNA B16 ! Lumi_Response_Neutron_counters_fAdc ! (presents temporary at H1LumiDataFlow) ! ATTributes: ! ----------- !COL ATT-name FMT Min Max !Comments ! 1 NCELL I 0 11 ! cell (channel) number 2 PULSE12 I 0 65535 ! 1st and 2nd bins of NCELL FADC window ! packed into B16; high byte - 1st, low ! byte - 2nd; 3 PULSE34 I 0 65535 ! 3rd and 4th bins of NCELL FADC window 4 PULSE56 I 0 65535 ! 5th and 6th bins of NCELL FADC window 5 PULSE78 I 0 65535 ! 7rd and 8th bins of NCELL FADC window 6 PULSE910 I 0 65535 ! 9th and 10th bins of NCELL FADC window END TABLE 21.03.94 ====================== H1KSOL === Max.Shower C counter ..... On the request of L.N.Shtarkov - a sketch of "new" counter: ===-------------------------------- | | \ | | \<-- mirror | | \ | | \ | | \ ===------------------------------ - ^ ^ | | | | | | 21 mm of PB | | | | | | | | | square tube with |_______| cross section 5*5 cm**2 |! !| and length 60 cm |! PM !| ( Aluminium) |! !| |!_____!| | | | | ------- We had not too much possibilities for choice of place to install it so there is an iron post in front of this counter which has the following cross section: ______ incoming ||----|| --------> || ||<-- the thickness of walls 2mm electrons|| || ||____|| ------ 20mm 22.03.1994 ==================== F11LEV === More about new C counter To make things clear. The aim of the installation of the gas Cerenkov counter *NOW* was: 1) To verify that this proposal will work at the conditions similar to the expected (high rates, E(e)=20-25 GeV) 2) If it works, to get a rough idea on the expected rates, amplitudes at PM and may be even an energy resolution. There are two bad things in present realization (see sketch by Yu.Sol.) 1) PM is oriented in upper direction (above beam-pipe) and thus is not sufficiently shielded against possible high background. 2) The counter is installed too far from the beam axis (290 mm), and according to MC, only very small part of scattered electrons with an average energy of 8 GeV can reach it. From the real data it follows, that in addition one may expect ca. 20-22 GeV off-momentum electrons, but neither rate, nor acceptance of those could be reliably estimated. Therefore, we may get two scenarios: a) off-momentum electrons are not in the acceptance of the gas C. In this case almost nothing can be verified (very low rate of 8 GeV electrons) b) part of the off-momentum electrons can reach the gas C. Then we may see significant amplitudes, but we cannot even guess how to extrapolate observed rates. Also, if amplitudes will be very small, then an ambiguity appears: is it a responce on 8-10 GeV electrons and no off- momentum e- are accepted, or 8 GeV electrons are below threshold (give small signal in PM) and what we see are high energy off-momentum electrons? Nevertheless, the fact, that a new device was prepared, installed and connected to the readout in such a short term is indeed a success. Some very preliminary results still can be expected. Perhaps, we could try to re-install it to a better position at the eventual access to the tunnel during the data taking. P.S. Q. to Egor. Which readout channel is used for C, 48 or 6? I need it to declare correctly a "dead" channel of ET for H1REC. 23.03.1994 ==================== H1KFOM === LUMM-bank update ........ It was decided to add some data into LUMM-bank after discussion with I.Sheviakov. It will be added else 160 longwords or B32-words into "bottom" of LUMM-bank contents. These words will keep data about trigger efficiencies behaivour with energy for PD1-trigger bit. These data as seems will help at determination of real thresholds values for PD1rec and PD1dep - which will be needed for "visible cross-section" corrections due to possible shift of thresholds values with high rates presence. PD1dep-spectra is used only for posibility to get large energy-spectra statistics at more ore less short time periods (for sensitivity to rates changing). As seems there is simple connection between PD1rec and PDdep spectra for calculation of PD1rec-threshold value calculation from PD1dep-spectra. It is proposed to use VC-trigger element as trigger (at Luminosity subbranch) for collection of all LUMM-bank spectra: a)---- PDrec - reconstructed energy spectrum for all bunches events --- b)---- PDdep - deposited energy spectrum for all bunches events ------- c)---- PDrec - reconstructed energy spectrum for e-pilot bunches events d)---- PDdep - deposited energy spectrum for e-pilot bunches events --- All high mentioned spectra will be built from VC=1 triggered events but with PD1-trigger element presence (really VC&PD1-subset of events). It will be built else two Photon Arm Energy spectra (inside FIC#2 environment) - for preparing of of high mentioned two additioanl regions at "bottom" of LUMM-bank - 5th and 6th histos with PDrec energy spectra for VC=1 and PDdep-energy spectra for VC=1. These spectra will be used only for calculation of PD1rec- and PD1dep- thresholds behaivour with energy. So, additional part of LUMM-bank: ! next 160 values are 2 arrays with PD1-trigger element threshold ! behaivour (for PDrec and PDdep spectra) ! each array consists of 78 bins (unfl,1-75,ovfl,"bad") and ! two time stamp words ! each array element was calculated as int(1000.*ratio(i)) !----- PDrec(VC&PD1)/PDrec(VC) ratio array ----------------------- 570 RATIOUNR | 0 +INF ! underflow ratio (less then 0.0 GeV) 571 RATIO01R | 0 +INF ! 1st bin ratio ( 0.0 - 0.5 GeV] 572 RATIO02R | 0 +INF ! 2nd bin ratio ( 0.5 - 1.0 GeV] ! ........... 645 RATIO75R | 0 +INF ! 75th bin ratio (37.0 - 37.5 GeV] 646 RATIOOVR | 0 +INF ! overflow ratio (more then 37.5 GeV) 647 RATIOBDR | 0 +INF ! "bad" reconstruction ratio 648 LLTFRSRR | -INF +INF ! H1Lumi Local Timer value for first ! event at PDrec spectrum 649 LLTLSTRR | -INF +INF ! H1Lumi Local Timer value for last ! event at PDrec spectrum !----- PDdep(VC&PD1)/PDdep(VC) ratio array ----------------------- 650 RATIOUND | 0 +INF ! underflow ratio (less then 0.0 GeV) 651 RATIO01D | 0 +INF ! 1st bin ratio ( 0.0 - 0.5 GeV] 652 RATIO02D | 0 +INF ! 2nd bin ratio ( 0.5 - 1.0 GeV] ! ........... 725 RATIO75D | 0 +INF ! 75th bin ratio (37.0 - 37.5 GeV] 726 RATIOOVD | 0 +INF ! overflow ratio (more then 37.5 GeV) 727 RATIOBDD | 0 +INF ! "bad" SW-cluster finder result ratio 728 LLTFRSPD | -INF +INF ! H1Lumi Local Timer value for first ! event at PDdep spectrum 729 LLTLSTPD | -INF +INF ! H1Lumi Local Timer value for last ! event at PDdep spectrum 23.03.94 ====================== H1KSOL === HERA schedule .... On the Operational meeting today F.Brasse report about changes in HERA schedule. Now it is the following: 28.03-30.03 Start of moving H1 into beam. No cooling LAr and no power for detector. 7.04 H1 Detector completely closed. 8.04 Start of e-beam (night) ~16.04 p-beam only studies ~16.05 ep collisions studies ~22.05 Lumi for experiments inbetween 16.04 and 16.05 there will be e beam probably also. 94 filling scheme 156 colliding bunches, 12 e- and 15 p pilot bunches. The filling scheme will be based on 9 bunches packets. Length of dump kicker gap and exact positions of e-pilot bunches may change. 23.03.1994 ==================== H1KFOM === First Runs with LUMM-banks Yesterday somebody had attempts to include Luminosity branch into H1CDAQ-operation. New release of H1LumiOnLineFIC1-program operated at FIC#1 board. We have first H1Runs with LUMM-bank presence at H1Data Flow (last 160 B32-words was not included yet - release is not prepared So - Runs 70154, 70155 and 70156 are first Runs with LUMM-banks at H1DataFlow. As I remember at this period some playing with test pulses was at Luminosity subbranch - and I put (by hand) 10 first e-beam currents (as e-pilot bunches). As seems it must be found some reasonable rates at LUMM-banks. As You can see LRTS-bank is absent LRPE-bank has maximal length as 28 (presence of WCvc-channel), LRTN and LRTF are prescaled on factor 10, LRNA-bank is presented too. All new features of H1LumiOnlineFIC1 are presented at low mentioned Jan Olsson's RunSummary: >>> Start RUN Summary --------------------- <<<<< NRUN NrofEvts Ev.Fst Ev.Lst DATE TIME <---> DATE TIME 70154 13770 0 4FFFFFFF 940322 164529 940322 164736 SYSMOD DAQMDE XI FEB CTLVRS DETSTA I.LUMI R-Devs 5801 80940145 212 39 0 9FEF29BC 0 50 CTCNF1 CTCNF2 TRIG1 TRIG2 NMASKST0 NMASKST1 NMASKST2 NMASKST3 FFDE0210 7E80000 40 102 0 0 2000 0 Eelec Eprot Polari Ecurr Pcurr Bfield Bcurr Patmo e-p-bunches 0 0 0 0 0 0 0 1968 1 192 BBL3:9FEF29BC HVST: 1FF8FF BANKs:Name Freq Av.Length Mn.Length Mx.Length 48454144:HEAD 13722 13 13 13 54535443:TSTC 41162 10 10 10 4C524545:LREE 13722 46 34 50 4C525045:LRPE 13722 26 20 28 43524A44:CRJD 9100 34 28 236 43525A44:CRZD 10 29 28 44 4C524543:LREC 13 51 51 51 4C525043:LRPC 13 28 28 28 44535434:DST4 13720 28 28 28 54475050:TGPP 13720 53 53 53 44434634:DCF4 1 1 1 1 444D5034:DMP4 1 54 54 54 444D5334:DMS4 1 512 512 512 44545334:DTS4 1 512 512 512 44454E44:DEND 1 1 1 1 4C525446:LRTF 1372 19 19 19 4C52544E:LRTN 1372 14 14 14 544C5631:TLV1 1 5 5 5 48454152:HEAR 13722 17 17 17 5A584354:ZXCT 1 1274 1274 1274 5A524354:ZRCT 13721 22 22 22 4C52544C:LRTL 13 236 236 236 4B454550:KEEP 48 1 1 3 56455445:VETE 13722 29 29 29 5A544255:ZTBU 1 226 226 226 4C525453:LRTS 1 6853 6853 6853 54435443:TCTC 1 72 72 72 54454C31:TEL1 27440 30 8 35 5A545052:ZTPR 1 1 1 1 54435355:TCSU 4 4 4 4 44525334:DRS4 25 512 512 512 43525044:CRPD 13720 90 76 284 44504144:DPAD 13720 8 8 8 4C525050:LRPP 13722 14 14 14 4C524550:LREP 13722 26 26 26 44524634:DRF4 1 120 120 120 54435444:TCTD 1 35 35 35 54435446:TCTF 1 10 10 10 54435449:TCTI 1 21 21 21 4C555452:LUTR 1 40 40 40 4C524E41:LRNA 13722 37 37 37 4C554D4D:LUMM 14 571 571 571 58434344:XCCD 7 190 25 490 Average Ev. Length: 2.053 kBytes; Min & Max 2.102 61.188 TSTC checks & statistics (NLOG(0:3),TSTCLG(1:8,9:16) 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 IGGDRU(IRU) 0 RunStart/KEEP/FARM/RunEnd records on file TESTCAL.R0070154 Output Data Set Name HERA04.H1RAWD.C9400029 Permanent >>> End RUN Summary -------------------- <<<<< 24.03.1994 ==================== H1KFOM === First Runs with LUMM-banks Yesterday evening it was tested new release of H1LumiOnlineFIC1 program with posibility of retranslation of updated LUMM-bank into H1DataFlow. Run 70322 is the first H1Run with updated LUMM- bank. NRUN NrofEvts Ev.Fst Ev.Lst DATE TIME <---> DATE TIME 70322 69398 0 4FFFFFFF 940323 194215 940323 195209 SYSMOD DAQMDE XI FEB CTLVRS DETSTA I.LUMI R-Devs 5801 80940145 202 1BB 0 9EEFA2B8 0 189 BANKs:Name Freq Av.Length Mn.Length Mx.Length 48454144:HEAD 69212 13 13 13 54535443:TSTC 69212 10 10 10 4C524545:LREE 69203 46 34 50 4C525045:LRPE 69203 26 19 28 4C554D4D:LUMM 60 731 731 731 59345454:Y4TT 69210 12 12 12 Output Data Set Name HERA04.H1RAWD.C9400031 Permanent Old problem of "Luminosity branch timeout during running of event XXX" is not resolved up to now. Errors of this type were fixed yesterday during running with 10000 Hz Random trigger from H1CTRIG. Under this conditions (almost 100% dead time and a little more then 100 Hz L2Keep's rate) it was fixed 7 errors of high mentioned type - it means - once per 10K events and once per 1 min. There are 2 types of high mentioned errors -1) event XXX never appears at H1DataFlow after high mentioned timeout message; 2) event XXX appears at H1DataFlow (when EventBuilder canceled its number from waited event numbers list) and after 20 sec. - all involved into H1CDAQ-operation branches put errors of high mentioned type. Today it will be held technical DAQ tests: "...... Sender: Eckhard Elsen To: "A.M. Fomenko" Dear Colleagues, as you may have heard we plan to do technical DAQ tests tomorrow and the Thursday next week, i.e. 24.3.94 and 31.3.94 in both cases starting around 9 h. For the tests tomorrow could you make sure that your component is in state that we can excercise - L3 Rejects and - High rate data taking? Best regards, Eckhard Elsen ....." 24.03.1994 ==================== H1KFOM === Technical DAQ tests ...... Today morning it were held technical DAQ test with our Luminosity branch. It were recorded 3 Runs with H1CTRIG and Luminosity branches. It were tested different L2keep rates (up to 400 Hz) with L3Reject presence. Preliminary results are reasonable. Dead time with 200 Hz L2keep-rate and with L3Keep-rate near 95-96 Hz was fixed near 26-28%. Averaged readout time - near 1440 mksec. It was used last release of H1LumiOnLineFIC1-program (see high mentioned messages - LRNA-bank,prescale LRTN/LRTF,WCvc at ReadOut, LUMM-bank etc.). It's creation date: 23 March 1994, 19:38. High mentioned problem of timeout is not understandable yet. It were not fixed timeout situations with some changings at H1CDAQ- environment. There is chance that not Luminosity branch is the reason of timeout situation. It would be nice but it must be investigated and understood. 24.03.1994===================== H1KBEL === Checking Photomultipliers Standard conditions for checking the PM-tubes FEU-85 and 147. 1. Using LED type HLMP-150 or K-101 (Hewlett Pakkard) Distance from LED to photocathode: for FEU-85 - 240mm for FEU-147 - 140mm LED excited by positive impulse: pulse hight =1 volt, length= 20ns. Output pulses from normal PM tubes schould have the amplitudes: from FEU-85 - 300mv +-100mv from FEU-147 - 150mv +-50mv The length of the pulses in both cases should be near 30ns. 2. Using plastic scintillator. Standard plastic scintillator with dimensions: diameter 20mm hight 50mm (cylinder) was connected to photocathode by silicon optical grease. Under the gamma radiation Co-60 or cosmic rays the output pulses of both types of PMs were the same as in the case of LED (see 1.) 24.03.1994 ==================== H1KFOM === Technical DAQ tests result As You know today from 09:00 up to 12:30 it were held technical DAQ tests. Luminosity branch was tested twice: 1) from 09:25:53 up to 09:43:28 with H1CTRIG branch only - Runs 70331, 70332,70333; 2) from 11:55:10 up to 12:26:38 together with 1,3,4,5,8 branches (CentralTrigger,CaloADC,CentralTracker,ForwardTracker,Muons) - Runs 70398,70399. All high mentioned Runs were written through SGI Logging option of H1CDAQ on temporary cartridges (HERA04.H1TEMP.G7106V00 and HERA04.H1TEMP.G7108V00 - ...G110V00). Short characteristics of each Run: | 70331 70332 70333 70398 70399 --------------|------------------------------------------------------- L2KeepRate,Hz | 117.1 118.4 360.0 82.8 230.0 L3KeepRate,Hz | 117.1 118.4 89.2 82.8 53.5 L3Reject | - - + - + Nevents | 34297 48065 27581 8114 94599 Duration,sec | 293 406 309 98 1767 SystemMode,hex| 5801 5001 5001 5001 5001-->4005001 ----------------------------------------------------------------------- SystemMode 5801 means status ON of H1CDAQ-option "Branch Timeout implememted". At SystemMode 5001 and 4005001 this option was Off. Only with status ON of high mentioned H1CDAQ-option it is possible to observe at H1CDAQ Supervisor screen red messages with " timeout during running with event XXXX". For Run 70331 it were fixed the next messages: 09:25:53 - Start Run 09:26:17 - CentralTrigger timeout during running with event 333 09:26:39 - CentralTrigger timeout during running with event 369 09:27:00 - CentralTrigger timeout during running with event 371 09:27:22 - CentralTrigger timeout during running with event 373 09:27:43 - Luminosity timeout during running with event 6258 09:29:09 - Luminosity timeout during running with event 22609 09:29:31 - CentralTrigger timeout during running with event 22609 09:30:14 - Luminosity timeout during running with event 30175 09:30:36 - Luminosity timeout during running with event 31377 09:30:46 - Stop Run And really at Run70331 it is fixed missing of some data (banks): At "right" Run (without timeout situations) there are some "rules" (from simple relations - connected with number of TSTC-banks at Run) - for normally stopped Run (not aborted) and FarmON option: N(HEAD) = N(TSTC) + 3 (RunStartRecord + EndRunRecord + L4Histos) N(HEAR) = N(TSTC) + 2 (RunStartRecord + EndRunRecord) N(LREE) = N(TSTC) (the same for LRPE,LREP,LRPP,VETE,LRNA banks) N(TGPP) = N(TSTC) N(ZRCT) = N(TSTC) + 1 (RunStartRecord) N(Y4TT) = N(TSTC) + 1 (EndRunRecord) N(MODI) = N(TSTC) + 1 (EndRunRecord) This "rules" are absolutely right for Runs 70332,70333,70398,70399. But for high mentioned Run 70331: N(TSTC) = 34207 OK N(HEAD) = 34205 (must be 34210) 5 banks are missing N(HEAR) = 34204 (must be 34209) 5 banks are missing N(LREE) = 34203 (must be 34207) 4 banks are missing N(TGPP) = 34202 (must be 34207) 5 banks are missing N(ZRCT) = 34203 (must be 34208) 5 banks are missing N(Y4TT) = 34203 (must be 34208) 5 banks are missing N(MODI) = 34208 (must be 34208) OK These digits are corellated with LogMessage contents: a) really at 5 events of Run 70331 (333,369,371,373 and 22609) it were fixed problems with CentralTrigger timeout; b) really at 4 events of Run 70331 (6258,22609,30175 and 31377) it were fixed problems with Luminosity timeout. As seems switching ON of high mentioned option have some influence on situations with timeout appearence (when it is switching ON - there are some problems with data missing at H1DataFlow). From another hand - may be switching off this option is blocking passing into tapes of events with banks missing and timeout problems really exist at branches? 25.03.1994 ==================== H1KFOM === Timeout Problem .......... Fri, 25 Mar 94 09:17:59 MET From: Eckhard Elsen Subject: Analysis of Runs To: Alexander Fomenko cc: h1+Haynes@desy.de ..... I agree with you that switching the timeout option changes the system behaviour dramatically. ..... My understanding of the timeout option is that when the bit is off, the partial events are never sent to CDAQ. Hence the buffers should be blocked and be lost for the rest of the run.I would then expect a series of messages at the end of the run indicating the partial events that were never matched. This we did not observe; not even in the run with almost 100000 accepted events. .... 28.03.1994 ==================== H1KFOM === Timeout Problem .......... From: Bill Haynes Subject: Analysis of Runs To: H1KFOM@dsyibm.desy.de >I am most worried by the fact that in run 70331 event 22609 first a >timeout is observed in the Luminosity Branch and then in the Central igger >TrBranch 22 secs later. I imagine the sequence would be > >:CTrig delivers event 22609. >:other events are delivered >:22 secs later Lumi delivers event 22609. > BILL : That must be exactly what is happening. Correct. Screw-up in Lumi STC software logic somewhere. Remember we have seen similar pecularities in the STC/event number fromthe muon system occassionally... sometimes a system delivers NO number... sometimes the wrong number... sometimes the same number twice !!! >My understanding of the timeout option is that when the bit is off, >the partial events are never sent to CDAQ. Hence the buffers should be >blocked and be lost for the rest of the run. BILL : Correct >I would then expect a series >of messages at the end of the run indicating the partial events that re >we never matched. This we did not observe; not even in the run with >almost 100000 accepted events. BILL : INCORRECT .. when timeout bit is off you only get the LAST EVENT NUMBER (ie the run-end record) timed out ... all other events are d ... zilche NOTE: This has NOT always been the case .. in the distant past all s event pending would be timed out.. Then too many people complained that it added to the overhead of run-ends ... so this feature was eliminated since over a year now. > >How does the timeout option work and why do we not have a problem when >the bit is off? (or do we not see it?). BILL : see above > >On the other hand: What is the maximum time that may elapse in Lumi >branch before a record is shippe dto CDAQ? Do you have a guess for ? >that BILL ; At end of 1993 run timeout was extended from several hundred isecs mill.It is now of the order of 1-2 seconds !!! Should be more than enough, even for the Calo OS-9 !!! =---------------------------------------------------------------------------- 07.04.1994 ==================== H1KFOM === "Zero"-corrupted BOS-bank. SENDMAIL message Date: THU, 7 APR 94 16:31:44 +0200 From: Alexander Fomenko Subject: re: tstc checks lumi To: Eckhard Elsen A.M. Fomenko, DESY -FH1K- Notkestrasse 85, D 2000 Hamburg 52 Tel. (040) 8998-2568(121 1b),-4356(Hera Nord), -2433(125 1c) (permanent address: Lebedev Phys.Inst., Moscow, Russia) NODE: DHHDESY3 USERID: H1KFOM > Received: from VXDSYA.DESY.DE by DSYIBM.DESY.DE (IBM MVS SMTP V2R2.1)with T > Thu, 07 Apr 94 12:32:51 MET > Thu, 7 Apr 1994 10:32:17 GMT > Date: Thu, 07 Apr 1994 12:32:17 +0200 (MDT) > From: Eckhard Elsen > Subject: TSTC Checks Lumi > To: "A.M. Fomenko" > Message-id: > MIME-version: 1.0 > Content-type: TEXT/PLAIN; charset=US-ASCII > Content-transfer-encoding: 7BIT > > Dear Colleagues, > occasionally the TSTC banks read by all branches do not agree. Here > is one example: > > Run Event 71361 4646 > TSTC Banks: > Sy Stat Bnch Revolution All_BuXings BuXings_L1Atve L1Keep > 0 C71E 38 6174702 0000000050F80FD3 00020000416255CD 4646 > 0 0760 38 6174702 0000000050F80FD3 00100000416255CD 4645 > 0 0760 38 6174702 0000000050F80FD3 00200000416255CD 4645 > 0 0760 38 6174702 0000000050F80FD3 00400000416255CD 4645 > 0 0760 38 6174702 0000000050F80FD3 00800000416255CD 4645 > 0 0620 38 6174702 0000000050F80FD3 01000000416255CD 4645 > 0 0760 38 6174702 0000000000000000 02000000416255CD 4645 > 0 0760 38 6174702 0000000050F80FD3 04000000416255CD 4645 > 0 0760 38 6174702 0000000050F80FD3 08000000416255CD 4645 > > The Lumi provides a zero (000000000..) for All_BuXings and for the > L2Keep number. This is very strange because that part of the event > was given to Bill with the correct event number, namely 4646 > (otherwise it would not have been built).All other numbers are > correct and agree. > > To me it smells like a software problem... > > The event can be found on F22ELS.TSTC.R713610. > > Does this help you to locate the problem? > > Cheers, Eckhard Hello, Dear Eckhard, Thanks for Your mail. As it is possible to suppose it is error inside of Luminosity branch readout. This type of error (zeros at unwaitable places) was fixed firstly some years ago (some from our BOS-banks had zero length or "zero"-name of bank). I inserted checking of BOS-banks names and reasonable length of BOS-banks into procedure before putting of H1Lumi-BOS-banks into Bill's DPM (before CALL MEBREADY (...)). At all cases of fixing of this type errors Luminosity branch is reporting Readout Error into H1CDAQ. Program scalers of high mentioned situations shows that this situations are enough rare (but sometimes are fixed). At Your mentioned case "zero" was fixed at place which is not name of bank and which is not length of bank - readout error was not fixed - but really this data were corrupted inside Lumi's DPM (after reading and before or during of moving to Bill's DPM). L2Keep-event number is reported to H1EventBuilder not at TSTC -bank but as parameter of CALL MEBREADY(... - and it was not fixed problems with building of event. But H1Lumi's SW got this number from the same place which this number was got for TSTC-bank. It means that after getting of L2Keepevent-number something happened inside Luminosity branch SW (or HW(?)) and corrupted data inside Bill's memory which was filled with this event BOS- banks. I am sorry for this error - this is our old problem but up to now we cannot resolve it and fix possible reason of this type error. As seems this type of unwaitable "zeros" it is possible to wait at another H1Lumi-banks - but up to now nobody reported about 'unwaitable' zeros at H1Lumi-banks - I shall try to make some tools for off-line checking of unwaitable "zeros" at H1Lumi BOS-banks (LREE,LRPE,LRTN,LRPP,LREP,LRTL,LRTS,LUMM,LRNA,VETE,LREC, LRPC). Else one thanks for reporting of this (as seems very rare) type of error. Sincerely, Alexander Fomenko 08.04.1994 ===== FIC#2 ======== H1KFOM === H1LumiOnLineFIC2-release.. Today was made and tested (only with test pulses) new release of H1LumiOnLineFIC2-program. At this release it was inserted posibility of making photon arm reconstructed energy spectra for LUMM-bank filling. Additionaly it was introduced readout and processing of ET&PD&VC- events at Luminosity subbranch. These events will be used for VC-channel calibration. It was made assignement that e48 is dead channel at Luminosity subbranch reconstruction package. 08.04.1994 ==================== H1KFOM === Banks Definition Updates.. Today it was received e-mail from Eckhard Elsen with the next contents "..... READMAIL message stored : FRI, 8 APR 94 17:35:25 MESZ Fri, 8 Apr 1994 15:15:56 GMT Date: Fri, 08 Apr 1994 17:15:55 +0200 (MDT) From: Eckhard Elsen Subject: LRTL Limits Errors Sender: Eckhard Elsen To: "A.M. Fomenko" Reply-to: Eckhard Elsen Message-id: MIME-version: 1.0 Content-type: TEXT/PLAIN; CHARSET=US-ASCII Content-transfer-encoding: 7BIT Dear friends, U.Martyn is checking the contents of our banks against the limits given in the bank descriptions. He finds some errors in the LUTR/LRTL banks. Could you check this and if necessary ask him to update the description accordingly. Best regards, Eckhard Elsen LUTR: Number of incorrect attributes 3 Attribute Name Min Max Nrows PD_LTH2 0 30000 1 Irow Content 1 35000 Attribute Name Min Max Nrows ET_T_SUM 0 65535 1 Irow Content 1 952967163 Attribute Name Min Max Nrows ET_CLUST 0 65535 1 Irow Content 1 -468881824 ***************** *ENTITY 410 LRTL* ***************** Number of incorrect attributes 3 Attribute Name Min Max Nrows RATE1 0 +INF 1 Irow Content 1 -17655 Attribute Name Min Max Nrows RATE2 0 +INF 1 Irow Content 1 -17261 Attribute Name Min Max Nrows RATE220 0 +INF 1 Irow Content 1 -17511 ....." After checking of situation at on-line me and Egor propose to make updating of atrributes definition at existing LUTR and LRTL-banks. It is proposed to change LRTL-bank attributes definition as: ................... ! negative values of the next 220 values mean technical problems ! of Rate calculations 1 RATE0 I -INF +INF ! Rate of rough luminosity trigger in ! Bunch 0, in 0.1 Hz units ! (averaged over the last 10 sec.) 2 RATE1 I -INF +INF ! RATE of rough luminosity trigger at ! Bunch 1, in 0.1 Hz units ! (averaged over the last 10 sec.) ! ................................................................... 220 RATE219 I -INF +INF ! Rate of rough luminosity trigger in ! Bunch 219, in 0.1 Hz units ! (averaged over the last 10 sec.) 221 RTLUMI I 0 +INF ! Total Rough Luminosity Trigger Rate ! in 0.1 Hz units ! (averaged over the last 10 sec.) 222 PLTEFF I 0 1000 ! Precise Lumi Trigger Eff (in 0.1%) ! (on the basis of the last 10000events) 223 RLTEFF I 0 1000 ! Rough Lumi Trigger Efficiency(in 0.1%) ! (on the basis of the last 100 events) 224 XPROF0 I 0 10000 ! Current Mean Value of X-coordinate ! of the beam profile at PD-face ! X(cm) = (XPROF0-5000)*0.001 ! (on the basis of the last 900 events) 225 XPROF1 I 0 10000 ! Current Mean Value of X-coordinate ! of the beam profile at PD-face ! X(cm) = (XPROF1-5000)*0.001 ! (on the basis of the last 50 events) 226 YPROF0 I 0 10000 ! Current Mean Value of Y-coordinate ! of the beam profile at PD-face ! Y(cm) = (YPROF0-5000)*0.001 ! (on the basis of the last 900 events) 227 YPROF1 I 0 10000 ! Current Mean Value of Y-coordinate ! of the beam profile at PD-face ! Y(cm) = (YPROF1-5000)*0.001 ! (on the basis of the last 50 events) 228 ACCC I 0 1100 ! Acceptance correction coeff (in 0.001) ! (after analisis of last 50 events) 229 FADCEFF I 0 1000 ! Current FADC Efficiency (in 0.1%) 230 RC I 0 +INF ! Revolution Counter value 231 LLT I -INF +INF ! h1Lumi Local Timer value 232 CTLV I 0 +INF ! Current Total Luminosity Value ! (averaged over the last 10 sec., in ! units of 10**25 cm-2s-1) 233 RUNLUMI I 0 +INF ! Total Integrated Luminosity from the ! beginning of the run ! (in units 10**25 cm-2 = 0.01 mb-1) 234 RH1LUMI I 0 +INF ! H1 Integrated Luminosity from the ! beginning of the run (during H1Live) ! (in units 10**25 cm-2 = 0.01 mb-1) ....................... It is proposed to change LUTR-bank atributes definition as: ! ! BANKname BANKtype ! BANK LUTR B32 ! ! ---------------------------------------------------------------------- ! L1 trigger setting description for the Luminosity System. ! Created online in LUMI DAQ, transmitted to IBM in RunStart record, ! stored at the H1 Database, if differs from the current DB-bank. ! LUMI trigger produces 8 trigger bits available at CTL: ! ---------------------------------------------------------------------- ! TB0: Monitor_1 (defined by the TB0_BITS and TB0_MASK) ! TB1: Monitor_2 (defined by the TB1_BITS and TB1_MASK) ! TB2: SUM = LUMI bit 7 (defined by the TB2_BITS and TB2_MASK) ! TB3: ET = LUMI bit 3 (defined by the TB3_BITS and TB3_MASK) ! TB4: PD = LUMI bit 4 (defined by the TB4_BITS and TB4_MASK) ! TB5: VC = LUMI bit 5 (defined by the TB5_BITS and TB5_MASK) ! TB6: ETAG = LUMI bit 6 (defined by the TB6_BITS and TB6_MASK) ! TB7: Monitor_3 (defined by the TB7_BITS and TB7_MASK) ! ---------------------------------------------------------------------- ! Definition of the 16 internal LUMI bits used for the TB0-TB7: ! ---------------------------------------------------------------------- ! bit 0 CALP = Calibration pulses ! bit 1 ETTOP = Topological enegry sums in ET ! bit 2 CFS = Cluster Finder Solution ! bit 3 ET = ET_LTH < E(ET) < ET_HTH ! bit 4 PD = PD_LTH < E(PD) < PD_HTH ! bit 5 VC = (VC(v) > VC_VTH) .OR. (VC(s) > VC_STH) ! bit 6 ETAG = ET * noPD * noVC ! bit 7 SUM = ETPD_LTH < E(ET)+E(PD) < ETPD_HTH ! bit 8 MN2 = MN2_TH < E(MN2) ! bit 9 MN1 = MN1_TH < E(MN1) ! bit 10 VCV1 = VC_VTH1 < VC(v) ! bit 11 VCS1 = VC_STH1 < VC(s) ! bit 12 PD2 = PD_LTH2 < E(PD) ! bit 13 PD1 = PD_LTH1 < E(PD) ! bit 14 ET2 = ET_LTH2 < E(ET) ! bit 15 ET1 = ET_LTH1 < E(ET) ! ---------------------------------------------------------------------- ! Trigger conditions for the LUMI trigger bits TB0, TB1, ... TB7 can be ! obtained in a following way (1 = true, 0 = false): ! TBx = IAND(TBx_MASK,TBx_BITS) .AND. .NOT.IAND(TBx_MASK,NOT(TBx_BITS)) ! ---------------------------------------------------------------------- ! ! COL Att-name FMT MIN MAX !Comments 1 ETPD_LTH I 0 60000 ! basic (ET+PD) low threshold [MeV] 2 ETPD_HTH I 0 60000 ! basic (ET+PD) high threshold [MeV] 3 ET_LTH I 0 60000 ! basic ET low threshold [MeV] 4 ET_HTH I 0 60000 ! basic ET high threshold [MeV] 5 PD_LTH I 0 60000 ! basic PD low threshold [MeV] 6 PD_HTH I 0 60000 ! basic PD high threshold [MeV] 7 VC_VTH I 0 60000 ! basic VC-veto threshold [MeV] 8 VC_STH I 0 60000 ! basic VC-spectr. threshold [MeV] ! 9 ET_LTH1 I 0 60000 ! ET low threshold 1 [MeV] 10 ET_LTH2 I 0 60000 ! ET low threshold 2 [MeV] 11 PD_LTH1 I 0 60000 ! PD low threshold 1 [MeV] 12 PD_LTH2 I 0 60000 ! PD low threshold 2 [MeV] 13 VC_STH1 I 0 60000 ! VC-spectrometric threshold 1 [MeV] 14 VC_VTH1 I 0 60000 ! VC-veto threshold 1 [MeV] 15 MN1_TH I 0 60000 ! MN1-threshold [MeV] 16 MN2_TH I 0 60000 ! MN2-threshold [MeV] ! 17 PD_Y_TH I 0 60000 ! PD y-layers low threshold [MeV] 18 PD_X_TH I 0 60000 ! PD x-layers low threshold [MeV] 19 ET_Y_TH I 0 60000 ! ET y-layers low threshold [MeV] 20 ET_X_TH I 0 60000 ! ET x-layers low threshold [MeV] 21 ET_Y_SUM I 0 60000 ! ET y-topol.sums threshold [MeV] 22 ET_X_SUM I 0 60000 ! ET x-topol.sums threshold [MeV] ! 23 TB0_BITS I 0 65535 ! definition of the trigger bit 0 24 TB0_MASK I 0 65535 ! mask for the trigger bit 0 25 TB1_BITS I 0 65535 ! definition of the trigger bit 1 26 TB1_MASK I 0 65535 ! mask for the trigger bit 1 27 TB2_BITS I 0 65535 ! definition of the trigger bit 2 28 TB2_MASK I 0 65535 ! mask for the trigger bit 2 29 TB3_BITS I 0 65535 ! definition of the trigger bit 3 30 TB3_MASK I 0 65535 ! mask for the trigger bit 3 31 TB4_BITS I 0 65535 ! definition of the trigger bit 4 32 TB4_MASK I 0 65535 ! mask for the trigger bit 4 33 TB5_BITS I 0 65535 ! definition of the trigger bit 5 34 TB5_MASK I 0 65535 ! mask for the trigger bit 5 35 TB6_BITS I 0 65535 ! definition of the trigger bit 6 36 TB6_MASK I 0 65535 ! mask for the trigger bit 6 37 TB7_BITS I 0 65535 ! definition of the trigger bit 7 38 TB7_MASK I 0 65535 ! mask for the trigger bit 7 ! 39 ET_T_SUM I -INF +INF ! definition of the ET-topol. Sum 40 ET_CLUST I -INF +INF ! definition of the ET-cluster type ! END BANK Welcome any critisizm and remarks.Asking to S.Levonian to retranslate these changings to H.U.Martyn or make this changings himself at official H1-banks definitions. If no any remarks with LUMM/LRNA-contents - as seems it is needed to move its definitions into official H1-banks definitions too. These banks are presented at H1DataFlow and it would be good if some definitions exist at lbanks-service (see proposals at LLC - messages of H1KFOM - 19.03.94,21.03.94 and 23.04.94). 09.04.1994 ==================== H1KFOM === 0.5 Gbyte Cartridges .... At nearest H1DataTaking H1RawData will be written into 0.5 Gbyte capacity cartridges. This is examples of H1RAWD.C94XXXXX-filling at current cosmic Run (Runs71677-71686,08/04/94 from 18.51 up to 19.42) files and directories in POOL h1stage of OWNER fomenko -rwxr-xr-x 1 fomenko h1 500041080 Apr 9 17:16 HERA04.H1RAWD.C9400344 -rwxr-xr-x 1 fomenko h1 500041080 Apr 9 17:25 HERA04.H1RAWD.C9400345 11.04.1994 ==================== H1KFOM === First HERA e-beam 94 .... From yesterday, Sunday, 10 April 1994 afternoon it was fixed first HERA e-beams at this year. It was filled only one bunch (mainly 0th bunch). Sometimes 2 bunches existed at HERA e-ring (0th and 200th, 0th and 3th). Ietotal max was fixed near 230 mkA last night near 23:00. HERA e-beam activity continues at Monday all day. Beam energy is 12 GeV. 12.04.1994 ==================== H1KFOM === HERA e-beam ............. Tonight it was continued HERA e-beam activity. From 02:00 up to 08:40 it was long e-beam filling with Ie total near 5 mA at start of filling (near 02:00). Ee =12 GeV. Number of bunches was near 50. At 08:40 e-beam was dumped from Ie total near 800 mkA. It were not recorded any H1Runs this night - it is impossible to look for situation with number of bunches and bunch currents at offline. 12.04.1994 ==================== H1KSOL === HV trip on VETOs .... Today morning HV trip on VETOs was found. There were no message on slow control messages panel about it. I report it to Guenter Sigurd and we have to investigate this with Samvel Kazarian. I did try simply to switch on this channel (#38 crate 2) but in a few minutes it went down by trip again. Then I lowered HV preset on this channel to 450 V and switch on it again - no trip. Then I subsequently increased HV with step of 100 V up to previous value - 750 V - no trip. Now it is OK but the reason is not clear. In one hour the problem came again (trip),it is under investigation, probably we will need an access in the tonnel. 13.04.1994 ==================== H1KFOM === HERA e-beam ............. Tonight it was continued HERA e-beam activity from 18:00 yesterday up to 06:00 today. Longest e-filling was from 03:00 up to 06:00 Ie total (max) = 4.2 mA at 03:00. At 05:00 (as seems) it was injected additinal portion of electrons into HERA e-ring and Ie total was incremented from 3 mA up to 4 mA. At 06:00 e-beam was dumped. Today all day and (may be) all night - up to tomorrow morning HERA will stop all activity due to H1 request on access for H1 Main Solenoid cooling system repairing. 13.04.1994 ==================== H1KFOM === Alan's requests ......... Alan Campbell together with Felix Sefkov proposed to make some changings at H1LumiDataFlow. Due to request of S.Levonian about TSTC-bank presence at BCKUPTCL-files (now these files are named HERA03.H1TCL.C94XXXXX where XXXXX=00001,00002 etc.) and possible rejection of nearest to LRTL-bank appearence H1Events - it was proposed to move TSTC bank from Luminosity branch to KEEP-environment if LRTL-bank will be presented at current event. Today Alan Campbell reported then he looked for LRPE-bank with zero-length. As seems it is situation from high mentioned "zero- corrupted" bank problem. As seems I forgot to make check of LRPE and LRPP-banks length on reasonable values before moving into Bill's memory. 13.04.1994 ==================== H1KFOM === e-beam Bunch Structure... As it is possible to see from TCUR-banks contents (Runs 72053-72072, from 12.04.94 17:41:50 up to 12.04.94 22:09:23) yesterday it was tested new e-beam bunch sructure for 170 bunches filling: 9 filled(01) + 2 empty(01) + 9 filled(02) + 2 empty(02) + ... ...+ 9 filled(12) + 2 empty(12) + 14 filled(13) + 2 empty(13) + + 9 filled(14) + 2 empty(14) + .... ...+ 9 filled(17) + 2 empty(17) + 9 filled(18) + 8 empty(18) + + 3 filled(19) + 8 empty(19). There were 50 empty and 170 filled e-bunches. 14.04.1994 ==================== H1KFOM === HERA e-beam ............. Today early morning it was began HERA e-beam activity again. Injection of last long filling was started at 05:00. At 06:15 it was made additional injection and this filling was finished only at 08:10 with ebeam dump. Ie total max was fixed at 05:45 with value near 6.3 mA. From 08:20 up to 09:10 it was fixed one bunch filling with Ie total near 100 mkA. After 09:35 one bunch (0th) filling exists at HERA e-ring with Ie=180 mkA. Machine Studies announcement. 15.04.1994 ==================== H1KFOM === HERA e-beam ............. Yesterday all day and tonight it was fixed permanent (almost without large pauses) HERA e-beam activity. Yestearday near 17:18 it was fixed Ietotal = 10.5 mA with 126 bunches. At 18:47 it was first attempt of e-ring magnets ramping. Ebeam was reached 19 GeV and e-beam was lost. Tonight it were fixed near 10 attempts of e-ring filling with Ie total not more then 4 mA. It is waited that today afternoon HERA people will begin HERA p-beam study(up to the next weekend). 15.04.94 ===================== H1KSOL === Trips on ET HV channels.. 7 ET HV channel were found under trip status this morning. This are: ## 9,16,21,23,24,26,34. After switching HV on - no problem (exept of #38 in crate 2-VETOs) during 2 hours. No alarms were registered !!! 15.04.1994 ==================== H1KFOM === HERA e-beam activity...... Today it was fixed some activity of HERA e-beam (from 10:00 up to 10:15, from 11:00 up to 11:50, from 12:00 up to 12:20, from 12:25 up to 13:33). During last filling (Ie total = 195 mkA, 0th bunch only) it was made attempt of ramping. We looked for at HERA-screen energy 27.55 GeV during 2-3 min. At 13:33 27.55 Gev e-beam was lost (as seems something happened with syncronisation). 15.04.1994 ==================== H1KFOM === Possible Trips Scenario... During night HERA-e-beam activity it was written very long Run72304 on temporary cartridges (from 02:18:17 up to 09:00:12) and HV-status must be reflected at Olsson's RunSummary: If You will use Jan Olsson's CARTL-tool with RUNSUM 72304 option You can see on Your screen the next info: a) information about Run: Nrun, number of wriiten events, number of L2Keep-events, RunStart Data/Time, RunStop Data/Time: ".... 72304 170563 0 164586 940415 21817 940415 90012 ....." b) information about BBL3-fixed "Bad Status of HV" at H1Lumi: "..... ............................ Lumi 0: OFF evs.143069-143121 Lumi 0: OFF evs.150905-155836 Lumi 1: OFF evs.150905-155836 ..." it means that from L2KeepEvent 143069 up to 143121 and from 150905 up to 155836 it were fixed situations with "Bad Status of HV" at H1Lumi detector (at first events region - only Lumi0 (as seems - mentioned at H1KSOL-message trips), at second events region - both Lumi0 and Lumi1 were off (as seems - Soloviev's morning intervention - first of all he switched off both crates after this - switched on). c) You can find some problems with HV (at the same time - between 143069 and 143121 events) on another subdetectors of H1: ............................. Liq.Ar: OFF evs.143069-143121 TOF: OFF evs.143069-143121 ............................... Veto2: OFF evs.143069-143121 ............................... d) You can identify time for high mentioned 'common' problem with HV status: "... 72304 2609 141437 144010 940415 80313 940415 80929 ...... Output Data Set Name HERA04.H1TEMP.G7539V00 Events: 141437- 144010 ..." it means that between 08:03:13 and 08:09:29 it was happened something what was the reason of trips. e) if You see at the history of HERA e-beam at this time - You can find that at this period e-beam was lost. So - possible scenario of last trips at ET: e-beam was lost at H1 interaction point region ---> trips at LAr, ToF, VetoWall and H1Lumi --> BBL3 alarm at H1CDAQ ---> desactivation of alarms from H1 shift crew (but trips at H1Lumi were kept) after 144010th event. It means that only 20 min. trips were existed at ET (from alarm near 08:00 up to Soloviev's intervention between 08:21 and 08:27). 16.04.1994 ==================== H1KFOM === HERA p-beam ............. Today morning at HERA-TV-screen it was appeared information which means that HERA-people started HERA p-beam study. HERA p-ring magnets are tuned on 40 GeV energy (injection energy). No any p-beams were fixed up to now."Machine Studies" announcement. 17.04.1994 ===== dice2 ======== H1KFOM === Else One About ET-Trips.. It were looked for different H1Lumi trigger element rates during part of Run72304 (from 07:08:34 up to 07:57:05 15.04.94). At 07:56 it were fixed the next rates at VC,PD and ET-trigger elements (from LUMM-bank contents): RateVC = 180 kHz RatePD = 120 kHz RateET = 300 kHz (!!!) RateET&PD = 9.5 kHz RateEtag = 300 kHz (!!!) If You remember ordinary rate at ET-trigger element at the last H1DataTaking was fixed as value not more then 80kHz. As seems namely this overrate was the reason of high mentioned ET trips. It is possible to look for behaivours with time of different H1Lumi trigger elements and bits rates at LOOK-data set H1KFOM.LOOK.R72304 at vectors 20000(VC),30000(PD),40000(ET),50000(PD1),70000(ET&PD), 80000(ET&PD&nVC),90000(PD&nVC),100000(eTAG). t0=07:08:34 15/04/94. LOOK-data set was produced at dice2-environment - see directory home/local/fomenko/LUMMbank - and moved into IBM storage with FTP- protocol. Shortly about e-beam at this period: - at 07:08 Ietotal=3.5 mA, before e-beam lost = 2.85 mA. - max bunch current - near 140 mkA - number of bunches 48 - bunch structure: 3 filled(01) + 2 empty(01) + 3 filled(02) + 2 empty(02) + ... ...+ 3 filled(12) +150 empty(12) - at 07:33 it was fixed ramping of e-ring - after ramping it were fixed some activity of HERA-people with changing of position of e-beam and during changing of Y at 07:56 e-beam was lost. All these characteristics of e-beam it is possible to see from LOOK-binary file H1KFOM.LOOKA.R72304 (1000-1219 - family of vectors with e-bunch currents, 802-805 family of vectors with beam position, 1501 histo with Ie total value behaivour with time). 18.04.1994 ==================== F11LEV === Is gas C counter working? Has anybody tried to analyse lumi raw data in order to see whether our new gas Cherenkov counter is working and feeling the beam? What is needed is very simple: 1) find some runs on the permanent data, taken in presence of e-beam in the ring; may be even at different conditions (current, energy) 2) look into LREE raw bank and to plot response for different bunches (filled and empty) from a) channel #48 (gas C) b) channel #21 (closest to the pipe) c) total sum from channels 0-47 (excluding gas C) You don't need reconstruction and even correct calibration - just use raw amplitudes from LREE. The question is: do we see any correlation between C and ET behaviour? 19.04.1994 ==================== H1KFOM === About Ch-Gas Response..... As seems high mentioned S.Levonyan's proposal is unrealyzed at offline due to: 1) it was not switched on any H1Lumi-trigger into H1 Trigger Coctail during H1Runs which were recorded during HERA e-beam presence. As seems at most part of events must be fixed only pedestal information (as at cosmic Run); 2) there were mostly temporary modes of recording of raw data during e-beam presence at HERA (at this moment 13:00 19/04/94 there are only H1Runs from 12/04/94 23:38 up to 12:24 19/04/94 at 200 cartridges with temporary data); 3) only at little peace of data it was ON e-beam synchronisation (main part of data were recorded with ON p-beam synchronization) - it means that it would be impossible to check filled/empty bunches. 4) e-beam was 'non-standard' - it was looked for 20 ns shift at timing (2 FADC-bins). It was not made any tuning of Photoproduction branch FADCs shifts. 5) it were fixed only one or two attempts of e-ring magnets ramping (mostly - unsuccesful). It was attempt to look for on FADC-counts (at Luminosity subbranch) for e48-channel (Ch-Gas Responce) at Friday 15/04/94 during last e-filling and last attempt of e-beam ramping making - (see message of H1KFOM from 15.04.94 "HERA e-beam activity") - with Ie=195mka by Yu.Soloviev with Egor's Histos tools. Trigger for monitoring was installed on ET&PD&nVC-events. As I remember - with 12 GeV e-beam it was seen some response at e48's histo (at Egor's made FADC-counts - up to 66 value). As seems this response was a little (almost negligable) more then responses at e47,e46,e45 etc. When e-beam ramping started Yura zeroed e48'th histos and we looked for that e48-response was became more and more and went to out of high boundary of e48's histo. After e-beam was 27.55 (only 2-3 minutes) we cannot find the boundaries of e48's response - due to absence of time. So - no any significant results of e48- responce up to now - it will be possible to make at the next week when "good" e-beam must be filled into HERA. 20.04.1994 ==================== H1KFOM === First HERA p-beam 94 .... Tonight and today morning it was fixed first HERA p-beam 94. First 'non-zero' value of Ip total was fixed at 03:15 20.04.1994. There was little Ip total current at HERA p-ring during almost all period from 03:15 up to now (09:57). Ip total was not more then 100mkA Energy of p-beam is 40 GeV (injection energy). 21.04.1994 ==================== H1KFOM === HERA p-beam .............. Yesterday during all day and evening and tonight it was continued HERA p-beam study (protons energy - 40 GeV, number of bunches - 1, bunch current - not more then 140 mkA). 22.04.1994 ==================== H1KFOM === HERA p-beam .............. Yesterday during all day and evening and tonight it was continued HERA p-beam study. From 23:00 yesterday it was started injection of more then one p-bunch into HERA p-ring. Ip total(max) during this night was fixed near 2.1 mA. 22.04.1994 ===== FIC#1 ======== H1KFOM === H1LumiOnLineFIC1-release.. It was made new release of H1LumiOnLineFIC1-program due to Egor's request on involving of additional GPTP-module into H1Lumi Readout. Additional portion of information are placed into bottom of VETE- bank with additional 27 B32-words (it means that after -4,-3,-2,-1,0, 1,2,3,4 bunch x-ing with 64 trigger elements from H1VetoWall follows the new portion of information with -4,-3,-2,-1,0,1,2,3,4 bunch x-ing but for 32 trigger elements from H1 Roman Pots subsystem + 32 zeros bits). Readout time incremented on 200 mksec and became 1500 mksec. This release will be used only episodically when request from W.Bartel's team will be issued. Its creation date is 22.04.94 12:14. This release was tested with H1CDAQ. It was written short H1Run on permanent cartridge - Run 72818 with extended VETE-bank contents. 23.04.1994 ==================== H1KFOM === HERA p-beam .............. Tonight it was continued Mashine Study of HERA p-ring. During all night it were fixed the set of injections with Ip total not more then 200 mkA. Yesterday at day time it was fixed injections with 10 p-bunches. Today evening it is possible to wait of starting e-beam program at HERA. 23.04.1994 ===== dice2 ======== H1KFOM === Tool for LRNA-bank tests It was prepared tool for LRNA-bank contents testing at dice2-environm. Egor wrote 3 H1Runs on temporary cartridge at 22.04.94 from 01:15:36 up to 01:25:50 (Runs 72742,72743 and 72744) with different FADCs shifts for LRNA-bank contents readout (143 for Run 72742, 133 - for Run 72743 and 153 for Run 72744). High mentioned Runs were recorded during stable 40GeV p-beam existing at HERA p-ring with Ip=1.2 mA. NCsum is analog trigger sum of all responses from 10 Neutron Counters and will be additional Trigger Element at H1 Trigger Level1. NCsum was used as trigger with low threshold (0.5 GeV at Egor's H1Lumi trigger elements scale). This trigger bit was connected to eTAG (s86) and s86 - trigger bit was used as source of triggers for high mentioned Runs. NCsum was connected with MN2(or MN1?)-trigger element too and was retranslated through one of 8 H1Lumi trigger elements wires to H1CentralTriggerLogic. Analog value of NCsum is written into 10th input of LRNA-bank. 11th input of LRNA-bank is empty (spare) as seems. Tool is kept at /home/local/fomenko/LUMMbank-directory: MakeLUMM72742c - example of make file for creation of executable module LUMM72742c.f - example of main-program s_LUMMbank11 (at R-subdirectory) - example of script for job starting. Tool is useful for looking for on FADC-pulses shape, making of responses histos (with linearization and FADC-counts making). Raw data were got from temporary cartridge HERA04.H1TEMP.G7662V00 and it were selected only useful banks into Egor's data set at cartridge on DESY IBM (H1KSHE.H1LUMI.R072742.A00). LOOK-file which was produced at dice2-environment was retranslated into DESY IBM storage with name H1KFOM.LOOK.R72742. Histos 500001-500012 keep FADC pulses shapes for all 12 inputs. Histos 400001-400012 keep Neutron Counters responces at FADC counts (after linearization, pedestals substraction and making of single FADC counts per channel). 23.04.1994 ===== dice2 ======== H1KFOM === Tool for Ch.GasCnt tests It was prepared tool for Cherenkov Gas Counter Prototype response behaivour tests (as S.Levonyan proposed at his 18.04.94 message): This tool produces ntuple 200000 with the next contents: NB - bunch number from TSTC-bank of current event CS - bunch status (0,1,2,3-values) -thresh.=7mkA as at on line monitoring =0 - empty =1 - collided =2 - e-pilot =3 - p-pilot E48FADC - FADC-counts at e48-channel ETGEV - ET deposited energy (with current calibration and last values of relation coefficients) at GeV ETFADC - ET deposited energy at FADCcounts without accounting of any calibration Tool was tested with set of Runs which were recorded during last e-beam study. As example - H1KFOM.LOOK.R72228 - LOOK-data set keeps high mentioned ntuple. Tool is kept at /home/local/fomenko/LUMMbank - directory: MakeLUMM72228a - example of make file for creation of executable module LUMM72228a.f - example of main-program s_LUMMbank10(at R-subdirectory) - example of script for job starting. At high mentioned LOOK-data set it is possible to see vectors 20000,30000,40000,50000,60000,70000,80000,90000 and 100000 with different H1Lumi Trigger Elements/Bits behaivours with time during high mentioned e-fillings(test of LUMM-bank contents). It was collected some data sets with H1Lumi and H1Ctrig-BOS banks from last period of e-filling: 1) from HERA04.H1TEMP.G7460V00- ...G7466V00 cartridges (exept G7461V00) Runs 72089 - 72095 ( 13/04/94 from 04:13 up to 06:05) (Ie total up to 3.3 mA) H1KFOM.H1LUMI.R072089.A00 K11325 78048081 00001 ACS-E-C 2) from HERA04.H1RAWD.C9400400- ...C9400401 cartridges Runs 72228 - 72256 ( 14/04/94 from 18:35 up to 21:54) (Ie total up to 4.0 mA) H1KFOM.H1LUMI.R072228.A00 K32038 78048081 00001 ACS-E-C H1KFOM.H1LUMI.R072228.A01 K18007 78048081 00001 ACS-E-C 3) from HERA04.H1TEMP.G7529V00- ...G7536V00 cartridges (exept G7529V00) Part of long Run 72304 (15/04/94 from 07:08:34 up to 07:57:05) (Ie total up to 3.6 mA) H1KFOM.H1LUMI.R072304.A00 K44331 78048081 00001 ACS-E-C Selection of needed banks was made at dice2-environment. Tools are kept at /home/local/fomenko/stage1 and /home/local/fomenko/Selection directories. As examples: at stage1-directory: H1TEMP.G7530 - script for staging of data set from temporary data set from IBM-environment to dice2 h1stage pool at Selection-directory: main3.f - example of main program for selection Makefile3 - example of make file for creating of executable module H1LumiSelection8.qsub - script for data selection job starting FPACK8.options - FPACK-options for selection of needed banks (get data from data set at h1stage pool--> -> temporary put into h1pool pool ---> -> stageout into DESY IBM cartridge) 23.04.1994 ===== FIC#1 ======== H1KFOM === Remark on H1RunPB message Today at H1RUN pinboard appeared message of H1 Run Coordinator which connected with operation of Luminosity branch at H1CDAQ-environment: ".... H1-Status 22/23.4.94 ... Lumi Can not run together with other branches. Experts are working on that. They will use the access to HERA on Monday to inspect their hardware problem. ... Kerstin Borras ..." Firstly this problem appeared at 21.04.1994 when one of H1 Shift crew persons asked me about this. First reaction was - reboot FIC#1 and start again - but result the same - Luminosity branch was monitored at H1CDAQ Supervisor as branch without RunPrepareAcknowlege (RedLine with no any event at Run). H1Lumi monitoring tool shows that during request from Bill (through MultiEventBufferRequest) of needed buffer for putting into it of prepared RunStartRecord (with LUTR-bank) - Bill's routine puts strange response (Run Start Record is not required). Checking of SystemMode status at H1CDAQ Supervisor shows that RunStartRecord is needed. Before this date always was reaction - OK. It is first problem of this type. Together with H1 Shift Person we made some investigations of behaivour Luminosity branch with different set of H1 - branches involved into H1CDAQ. We found that only(!) if ForwardMuonTrigger branch is involved into H1CDAQ operation - high mentioned problem exists. Before this date Luminosity branch worked more or less properly with all H1 branches. No any changings it were made at last time at H1LumiOnLinrFIC1-program. Old release operated (from begin of April 94). As seems it is problem not from our side. 24.04.1994 ==================== H1KFOM === HERA e-beam again ...... Yesterday evening HERA-crew started HERA e-beam study again. Almost permanently e-beam is existing at HERA e-ring. Ie total (max) was fixed tonight near 01:30 with value near 400 mkA. Ee=12 GeV. Now (at 10:00) Ie=100 mkA, number of bunches 1, Ee=12 GeV, Existing filling was started at 04:30 tonight and is continue up to now with two additiional injections (at 06:40 and at 07:30). There is "Injection"-announcement at HERA-TV-screen. 24.04.1994 ===== FIC#1 ======== H1KFOM === Remark on H1RunPB message As Jan Olsson summaries show: Last H1Runs with 'no problem' at compatibility Luminosity branch with ForwardMuonTrigger Branch was Run 72631 with 1,5,6,8,9,10 branches involved into H1CDAQ operation. This Run was written 20/04/94 from 11:01:47 up to 11:06:15 on permanent cartridge. Problem of high mentioned uncompatibility was fixed at 21.04.94 near 11:30 (Runs 72702,72707,72708 and 72710 were mentioned at my previous message Runs with investigation of compatibility Luminosity branch with another branches) 72702 (1,9) - OK 72703-72706 - not OK all atempts with 10 branch (Forward Muon Trigger) 72707 (1,7,8,9) - OK 72708 (1,5,7,8,9) - OK High mentioned Runs were recorded into permanent cartridge. At Friday morning operation meeting at North Hall Jura Soloviev said about uncompatibility Luminosity branch with only(!) Forward Muon Trigger brunch. It was 22.04.94 near 09:25. At Friday 22.04.94 at 09:36 was fixed Run 72780 which was recorded on permanent cartridge with 1,9 and 10 branches involved into H1CDAQ- operation. From this Run it was started new stage of compatibility Luminosity branch with Forward Muon Trigger branch. After this it were recorded the set of H1Runs without any problems of high mentioned uncompatibility Luminosity branch. As seems somebody from Forward Muon Trigger team made some changing at their OnLine program. 25.04.1994 ==================== H1KFOM === HERA e-beam ............. Yesterday during all day e-beam was existed at HERA e-ring. Up to 13:30 it was studied only one bunch filling with Ie total near 100 mkA and Ee=12 GeV. Near 13:30 it was injected 40-bunches e-beam with Ie total at start time near 5.2 mA. This filling was continued up to 19:00 yesterday. Before midnight it was started new powerful injection and near 01:00 tonight Ie total was fixed more then 10 mA. This filling existed up to morning. During day e-fillings we wrote the set of H1Runs with our triggers: ET&PD&n(VC), VC and PD1(as eTAG). Details will be reported later. Now 2 hour interruption of HERA e-beam study is planning due to access at H1 area. Y.Soloviev, P.Smirnov are planning to go into HERA tunnel and check VetoCounter (sp) channel - divider or cabel or something else. During last e-filling it were not fixed any trips at VCs - Yura tested VCs with large HV (more then 1 kV). 25.04.1994 ==================== H1KSOL === VETOs HV... Since last Friday HV on VETOs was at nominal value i.e. 700 V, during a few hours on Sunday it was set on 1300 V - no trip. Today we used access to HERA tunnel to check situation with VETOs, no visible failures were found during this short access. Probably some short sircuit was burned out itself. Now both VETO channel are on nominal value of HV: VETOs - 700 V VETOv - 900 V 25.04.1994 ==================== H1KFOM === H1Lumi's Runs at 24.04.94 Yesterday it were recorded 3 sets of H1Lumi's Runs during e-fillings with Ee=12 GeV. 1)2 H1Runs (73040,73041) during of first e-filling with Ie total near 100 mkA and 1 bunch only; from 12:18 up to 13:12. 2)2 H1Runs (73042,73043) during injection of the second e-filling. Ie total was changed from 0.2 mA up tp 5.2 mA; from 13:25 up to 13:48. 3)8 H1Runs (73044-73049,73052,73053) during of the second e-filling. Ie total was changed from 5.1 mA up to 2.7 mA; 40 bunches; from 13:48 up to 16:56. All high mentioned H1Runs were recorded on 8 permanent cartridges from HERA04.H1RAWD.C9400438 up to HERA04.H1RAWD.C9400445 - almost 4 Gbytes of data. It were selected only 3 trigger bits at H1Trigger Coctail (s86,s93 and s94). s86 was as ordinary eTAG first time (with large prescale). s93=VC and s94=ET&PD&nVC were main event sources for high mentioned H1Runs data sets. At some time it was changed setting of s86 trigger bit on PD1- trigger (see LUTR-bank) and prescale factors were tuned on equal portions of data from all 3 trigger bits. It were installed so-called 'thresholds for 12 GeV e-beam' for H1Lumi Trigger elements (see LUTR-bank contents). It is possible to make some investigation of e48-element behaivour with 12 GeV e-beam (Egor's OnLine Monitoring shows that behaivour of e48-element response histos is another as e47,e46 etc. - something presents at Ch.Gas.Counter). It is possible to try making of first calibration with 12 GeV. It is possible to investigate on-line calibration of VC-counter. Get this data and enjoy (N-joy) its. 25.04.1994 ==================== H1KFOM === 27.551 GeV HERA e-beam... Today afternoon after high mentioned short break HERA e-beam study started again. As seems main task of afternoon HERA shift crew - to make ramping. After appearence of first e-beam at HERA e-ring it were made 3 attempts of e-ring ramping. First two attempts were unsuccesful (e-beam with 6 bunches and Ie=480 mkA was lost on 16 Gev first time and with 3 bunches and Ie=280 mkA was lost on 25 GeV). Third attempt of ramping of 3 bunches with Ie=330 mkA was succesful. At 15:45 Ee was fixed as 27.551 GeV and this beam existed at HERA e-ring almost 0.5 hour (up to 16:12). e48-element response is presented at each ET&PD-triggered event. 25.04.1994 ==================== H1KFOM === H1Lumi's Runs with 27.551 Fourth attempt of ramping today was succesful too and at 16:45 it was appeared 27.551 GeV e-beam with Ie=311 mkA and with three bunches. It were recorded H1Lumi's Runs during existing of this beam. Runs 73271,73272 - on temporary cartridge Runs 73273,73274 - on permanent cartridge. As seems Run 73274 is the best (it was decremented Random Trigger Rate as more as possible, it were tuned threshold of H1Lumi Trigger Elements on '26.7 GeV e-beam' values, prescale factors for s93 and s94 were tuned on equal portion of s93 and s94 trigger bits, s86 was switched on too with large prescale as PD1). Number of events at Runs: 23410,5985,33179 and 22048. At first 3 Runs large quantity of Random-trigger events. 26.04.1994 ==================== H1KFOM === HERA e-beam ............. From yesterday evening almost 8 e-filling were registered. Ie total max was fixed near 01:20 tonight as 9 mA. 26.04.1994 ==================== H1KFOM === 27.550 GeV HERA e-beam... Today at 12:45 it was started more or less powerfull e-filling with 42 bunches and Ie total near 5.2 mA. This e-beam was ramped up to energy Ee=27.550 GeV and now this e-beam is living at HERA (13:22) with Ie total=4.5mA. At 14:40 e-beam was lost. H1CDAQ was busy at this time so there are no any new H1Lumi Runs. 26.04.1994 ===== dice2 ======== H1KFOM === Attempts of processing ... It were made first attempts of recorded H1Lumi's Runs processing. It were processed two samples: 1) 12 GeV e-beam sample (HERA04.H1RAWD.C9400441, Runs 73044-73049, near 177000 events); 2) 27.551 GeV e-beam sample (HERA04.H1RAWD.C9400470, Run 73274, near 23000 events). It were produced LOOK-binary data sets at dice2-environment and moved into DESY IBM storage: H1KFOM.LOOK.R73044 FAST16 3010200E 00000 DASD H1KFOM.LOOK.R73274 FAST12 3010200E 00000 DASD These LOOK-data sets include early mentioned 200000th ntuple with e48FADC-counts etc. - for posssible investigations of e48-response. Histos 200001-200032 at last LOOK-data set are examples of Photon Arm Energy spectras which are retranslated through LUMM-bank contents. As seems all is OK with these spectra (they have statistics near 7000 events per spectrum) and it is possible to try Photon Energy Shift procedure with these spectra. It were tested contents of 26-element of LRPC-bank. This element retranslates value of current VCsp calibration coefficient - as it is possible to see from vectors 6026th at the next LOOK-binary data sets: H1KFOM.LOOKA.R73044 FAST09 3010200E 00000 DASD H1KFOM.LOOKA.R73274 FAST05 3010200E 00000 DASD CCvcsp is changed during Runs (but always only incremented). It must be checked else one and understood - may be there are some bugs Tools for first processing are kept at /home/local/fomenko/LUMMbank and /home/local/fomenko/LumiAlternative - directories: LUMMbank-directory: LUMM73044b.f - examples of main program LUMM73274a.f MakeLUMM73044a - examples of make files MakeLUMM73274a s_LUMMbank73470 - examples of scripts for job starting (at R-subdir.) s_LUMMbank73044 LumiAlternative-directory: main73044.f - examples of main program main73274.f Makefile73044c - examples of make files Makefile73274c s_AlternativeLumi73044b - examples of scripts for job starting s_AlternativeLumi73274b 27.04.1994 ==================== H1KFOM === HERA e-beam ............. Tonight it was continued HERA e-beam study. Night HERA e-filling had started near 01:00 and Ie max was fixed as 1.8 mA near 01:30. This e-beam existed up to morning. At 08:45 it was started new powerful e-beam injection. At 09:20 it was fixed largest Ie total at this year (15.7 mA) with 96 bunches. At 09:25 most of e-beam was lost and it was injected additional electrons. At 09:45 injection finished with 10.1 mA and 78 bunches. Now more or less stable e-beam conditions are existing. But e-beam is bad - its orbit is far from standard - as our BeamProfile tool shows. 27.04.1994 ==================== H1KFOM === 28.66 mA HERA e-beam..... Today afternoon it was started the second powerful e-beam injection. At 12:45 Ie total was 16.2 mA, at 12:46 - 19.7 mA, at 13:18 - 23.9 mA at 13:24 - 27.5 mA, at 13:26 - 28.66 mA with 180 bunches. Rates of our H1Lumi trigger elements were fixed as the next: VC - 2.5 MHz PD - 1.8 MHz (thresholds for 12 Gev e-beam) ET - 1.5 MHz ET1 -0.8 MHz PD1 -1.7 MHz VS1 -0.2 MHz NtrC-0.0013 MHz At this period something happened with e-beam and almost 90% of beam was lost (at lost period high mentioned rates was incremented on 2.7-3.3 MHz. It were fixed 2 channels with trips (21th ET-element and one channel which is connected to 37th output of CAEN crate 2). Trips were fixed with H1SlowControl System. Trips were canceled with standard manner. After this ocasion e-beam kept its bunch structure (180 bunches) but all bunch currents were decremented on 90%. Ie total became 3.3 mA. At 14:22 was finished additional injection (up to 25.0 mA) and now (15:34) the same beam is living at HERA e-ring with Ie total near 13.1 mA. High mentioned beam was with the same high mentioned bad orbit - most of photons from H1 IP was unvisible by H1LumiPhotonArm - - it was impossible to make some calibration etc. 27.04.1994 ==================== H1KFOM === VCv-Response Test........ Today I checked VCv-response at Run73274 data (27.55 GeV) and found that there is bug at LRPE-bank structure (not right value of hitted cells at miniheader - I am sorry - I forgot to update it - only lenght of LRPE-bank length was updated if VCv response existed). I shall try to fix this bug at H1LumiOnLineFIC1-release and report about later. Nevetheless- it is possible to decode the VCv-response at current bank structure too. I made it and You can see it at histo 9000 at H1KFOM.LOOK1.R73274 - LOOK-binary data set. This response is measured at FADC-counts. There are some cases of overflow - 20000 values. 28.04.1994===================== H1KBEL === Gas C counter estimate.. Simulation of the e.m. schower in 5Xo lead absorber was made by GEANT (V.Zverev). For electron energyes 10 GeV and 30 GeV and thres- hold 22 MeV the multiplisity and the angular distributions of the secondary electrons and positrons has been obtained. The multipli- cityes are: for 10 GeV - Mean = 20.6 R.M.S. = 5.9 for 30 GeV - Mean = 53.6 R.M.S. = 12.7 For our gas c counter acceptance we should have the numbers of secon- dary electrons per 1 initial: for 10 GeV - 7.6 for 30 GeV - 22.1 In the air each electron will emit 0.26 photons of the cerenkov radi- ation per 1 cm of his trak. If the efficiensy of light collection multiplied by quantum efficiensy of the photocathod = 0.1, then the number of the photoelectrons in the PM of gas c counter will be for 10 GeV - 12, for 30 GeV - 34. That means: we should signal from electrons, if they will be, but the pulce hight will be smaller, then from KRS. May be we should rise the HV in chanel 48. 28.04.1994 ==================== H1KFOM === HERA e-beam ............ From yesterday evening up to 06:50 this morning it was fixed permanent e-filling of HERA e-ring. More or less powerful injection was started at 0:30 and at 01:30 Ie total was fixed as 6.0 mA. At 04:30 it was injected 12.0 mA e-beam which existed almost 2 hours. Yesterday evening I put Movable Platforms into Parking Positions as Y.Soloviev recommended (near 18:30). Now all e-ring magnets are off and it is possible to wait a little interruption of HERA e-beam study and (may be) proton beam study will started with possible ramping up to 820 GeV. 29.04.1994 ==================== H1KFOM === HERA beams absence...... As seems - HERA people again returned to HERA e-beam study after yesterday attempts to inject p-beam. Yesterday it was fixed one bunch little current at p-ring with 40 GeV and nothing was fixed at night. Now it is possible to see status of e-ring magnets at 12 GeV. 29.04.1994 ===== dice2 ======== H1KFOM === 100th-ntuple for R73274 It was produced 100th ntuple for Run 73274 (best Run with 27.55 GeV e-beam) with using of Relation Coefficients set from 22.02.94 and Calibration Coefficient for VCsp from 24.02.1994 (nearest to H1DataTaking93's finish). Into contents of this ntuple was added 4 variables to standard set of variables which was used at Autumn93 (see my messsage at LL7 from 25.10.93) The structure of the 100th tuples is the next: NAM(1)='ETREC' ET reconstructed energy (GeV) NAM(2)='PDREC' PD-arm reconstructed energy (GeV) NAM(3)='VCDEP' VetoCounter deposited energy (GeV) NAM(4)='XET' Xet - reconstructed coordinate (cm) NAM(5)='YET' Yet - reconstructed coordinate (cm) NAM(6)='XPD' Xpd - reconstructed coordinate (cm) NAM(7)='YPD' Ypd - reconstructed coordinate (cm) NAM(8)='TIME' time at minutes from start of H1Run73264 NAM(9)='BUNCH' - bunch number NAM(10)='PD1' - value of trigger element PD1 NAM(11)='PD2' - value of trigger element PD2 NAM(12)='PD' - value of trigger element PD NAM(13)='VC' - value of trigger element VC NAM(14)='ET' - value of trigger element ET NAM(15)='ETAG' - value of trigger element eTAG NAM(17)='S86' - value of trigger bit s86 NAM(18)='S93' - value of trigger bit s93 NAM(19)='S94' - value of trigger bit s94 NAM(20)='S95' - value of trigger bit s95 new variables: NAM(21)='E48' - Ch.GasCounter Prototype Response (FADC units) NAM(22)='VCS' - VetoCounter Sp.Ch. Response (FADC units) NAM(23)='VCV' - VetoCounter Veto Ch. Response (FADC units) NAM(24)='CS' - bunch status (0,1,2,3-values) =0 - empty =1 - collided =2 - e-pilot =3 - p-pilot First release of high mentioned ntuple is kept at H1KFOM.LOOK2.R73274 LOOK-binary set. Anybody can play with its contents. H1KFOM.LOOK2.R73274 FAST08 3010200E 00000 DASD Ntuple was produced at dice2-environment and was moved into DESY IBM storage with ftp-protocol. Tool is kept at /home/local/fomenko/LUMMbank -directory: LUMM73274e.f - example of main program MakeLUMM73274e - example of make file s_LUMMbank73274 - example of script for job starting (at R-subdir.) 29.04.1994 ==================== H1KFOM === HERA e-beam ............ Today at 11:45 it was started e-beam filling (3 bunches, 6 bunches) with Ie total not more then 1.0 mA and with the same so called 'bad' conditions of e-beam when it is impossible to make something with our detectors. It is pity. 29.04.1994 ==================== H1KFOM === HERA e-beam ............ Today at 15:40 started more or less powerful e-filling into HERA e- ring. At 16:18 Ie total was 18.4 mA. At 16:38 e-beam was lost. At this period our Movable Platforms were at Parking positions. As Yura said it was fixed trip at 37 channel of 2nd CAEN crate (new max shower detector). Trip was unrecoverable - as seems this channel is dead after last overcurrents (firstly 2 days ago). At 16:45 it was began new powerful e-filling up to 15 mA. It was made ramping up to energy 27.5167 GeV. At 18:02 it was fixed Ie=11.1 mA with Ee=27.5167 GeV. 30.04.1994 ==================== H1KFOM === HERA e-beam & H1LumiRuns Tonight it was prolongated HERA e-beam activity. At 01:00 it was fixed Ie total as 12.0 mA. At 02:00 during the next e-filling it was fixed Ie total near 7.0 mA. This e-filling existed up to 06:30. At 10:40 today it was started new e-filling with Ie total max near 12.0 mA at 11:10. It was made ramping of e-ring magnets on energy Ee=27.516 GeV. There are 84 e-bunches with bunch structure 6 filled + 5 empty + 6 filled + 5 empty + etc. (14 trains). Beam conditions of last e-filling were better then at most of previous ones and it is possible to make calibration, collection of Photon Arm energy spectra, Beam position monitoring, Data Taking etc. There are the same shifted on 20 nsec FADCshifts (as ordinary when we have alone e-beam).I put shifted on 2 units FADCshifts on both branches It was decided to get some data with this beam. There are 3 subtriggers at H1LumiData - s86(as PD1), s93 and s94. Thresholds values (see LUTR-contents): VCs = VCv = 1.75 GeV PD1 = 3.25 GeV PD = 3.50 GeV ET = 5.00 GeV For Luminosity subbranch it were installed 4 triggers: ET&PD&n(VC) - for ET,PD channels calibration, beam position monitoring etc. ET&PD&VC - for VCs-calibartion VCv - for Photon Arm Energy Spectra collection - for LUMM bank filling PD1 - for Photon Arm Energy Spectra collection - for LUMM bank filling Yura Soloviev installed a little less HV on Ch.GasCounter Prototype PHEU - before this it was 2000 V, now 1900V. It was made for investigation of behaivour Ch.GasCounterPrototype response. First view on behaivour Ch.GasCounter Prototype response is not optimistic. As it was possible to see at FADC window there is no any synchroni- sation of pulses fronts from Ch.Gas with pulses from PD-cells. It were written four H1Lumi Runs with high mentioned conditions: Run 73833 (with first release of prescale gaps) 12:27:30 Ie=5.08mA-->4.85mA 33171 events Run 73834 (with second release of prescale gaps) 12:36:36 Ie=4.85mA-->3.00mA 258759 events Run 73835 (with attempt to write LREF and LRPF - banks - unsuccesf.) 13:29:50 Ie=3.00mA-->2.94mA 2127 events (90% ReadOut Errors) Run 73836 (with third release of prescale gaps) 13:32:40 Ie=2.94mA-->2.80mA 167279 events At 14:00 e-beam was dumped. At 14:15 new e-filling was started. At 14:30 there are 39 bunches with Ie total 7.65 mA (13 trains with 3 filled + 8 empty). Ie bunch max is 262 mkA. At 14:35 - 9.46 mA with 48 bunches (6+5+6+5+6+5+3+8+.. At 14:45 during attempt of ramping e-beam was lost (with 81 bunches). 30.04.1994 ==================== H1KFOM === SHOWLUMI is not operated Today I had attempt to see at LOOK-environment SHOWLUMI pictures. HERA03.H1LUMI-data set was at MIGRAT status and after remigration SHOWLUMI shows status 18/02/94 (may be during last operation of IBM Logging). May be due to SGI Logging implementation the SHOWLUMI option at LOOK and H1ED environments must be upgraded too? 30.04.1994 ==================== F11LEV === SHOWLUMI status .......... SHOWLUMI works, just file is no longer transmitted to IBM. The experts sofar did not make their mind on the final arrangement of all files which previously was located at IBM. As soon as a decision on H1LUMI file will be taken, the corresponding modifications will be implemented 01.05.1994 ==================== H1KFOM === HERA e-beam & H1LumiRuns Night from 30.04.94 on 01.05.94 (Russian Easter Night) was night of permanent HERA e-beam activity. This activity is not interrupted up to now (12:50). From midnight it were fixed 4 e-filings of HERA e-ring: 01:00 -04:00 (Ie max = 12 mA at 02:00) 04:15 -07:10 (Ie max = 18 mA at 05:00) 07:20 -10:15 (Ie max = 18 mA at 08:00) - ramping at 08:15 10:20 -??:?? (Ie max = 24 mA at 10:50) - ramping at 10:50 to Ee=27.515 There are 160 e-bunches at last e-filling. Bunch structure is the next: 97 filled + (2 empty + 9 filled)*5 + 10 empty +3 filled + + (8 empty + 3 filled )*5. I put our detectors at Working Positions at 12:00. Thresholds at H1Lumi Trigger Elements are installed the same as yesterday. At 12:50 with Ie=6.26 RateVC = 10.8 kHz, RateET&PD = 1.45kHz Beam conditions are not standard. Coordinate distribution at ET face is not ordinary (most of electrons hits are at right part of plot - far from beam). From 13:30 up to 13:50 it was recorded H1Run 73892 with the same trigger conditions as yesterday (near 65K events). At 13:45 e-beam was dumped. At 13:55 after e-ring magnets "massage" it was started 5th today e- filling. First injection - 42 bunches with Ie total = 6.8 mA,(3 filled + 8 empty) * 14. After second injection (14:03) - 84 bunches with Ie total = 9.0 mA (6 filled + 5 empty) * 14. After 3rd injection - (14:04) - 84 bunches with Ie total near 13.0 mA (6 filled + 5 empty)*14 After 4th injection (14:08) - 126 bunches with Ie total 19.2 mA (9 filled + 2 empty)*14. At 14:12 ramping was started on Ee = 27.515 GeV. At 14:16 ramping was finished with Ie total 18.8 mA (Rate VC = 81 kHz, Rate ET&PD = 10 kHz). At 14:25 it was started Run 73895 with the same trigger conditions as at Run 73892. This Run was recorded up to 15:50 when e-beam was dumped (or lost). Ie total before e-beam was lost was fixed as 9.0mA. At 16:04 started 6th today e-filling: 1st injection - 16:04 - Ie = 7.2 mA - 42 bunches (3+8)*14 2nd injection - 16:09 - Ie =10.4 mA - 57 bunches (3+8)*19 3rd injection - 16:13 - Ie =13.8 mA - 84 bunches (6+5)*8+(3+8)*12 4th injection - 16:15 - Ie =21.0 mA -126 bunches (9+2)*2+(6+5)*18 5th injection - 16:24 - Ie =24.1 mA -147 bunches (9+2)*9+(6+5)*11 Ramping Start - 16:26 Ramping Finish- 16:27 - Ie = 23.1 mA (RateVC=380kHz,RateET&PD=60kHz). Now (at 17:20) Ie=11.5mA. I put Movable Platforms into Parking Positions at 17:30. 02.05.1994 ==================== H1KFOM === PDtrigger Elements Problem As seems there are some problems with making of all PD-trigger elements set at H1Lumi Trigger Logic. It would be nice if this problem was connected with so-called 'bad e-beam condition' or something else. But as seems it is needed to think about low mentioned evidences and to make some conclusions or suppositions. At H1Lumi data collection there are 3 sets of H1Runs which were recorded with Ee=27.5 GeV: a) Runs 73271-73275 which were recorded at Monday 25.04.94 b) Runs 73833-73836 which were recorded at Saturday 30.04.94 c) Runs 73892-73895 which were recorded ar Sunday 01.05.94 For preliminary estimation of H1Lumi System behaivour with this beam it were processed some data from these Runs Sets and produced 100th ntuples which are kept at the next LOOK-binary data sets: H1KFOM.LOOK2.R73274 FAST08 3010200E 00000 DASD H1KFOM.LOOK.R73830 FAST16 3010200E 00000 DASD H1KFOM.LOOK.R73890 FAST12 3010200E 00000 DASD 100th ntuples structure is described at my message from 29.04.94 at LLC "100th-ntuple for R73274". It is possible to make the next manipulations at LOOK-interactive mode with any from high mentioned LOOK-binary set: do 'hs(201)=nt(100)(pdrec)@(s93.eq.1)' bins 200 0 35 do 'hs(201,1)=nt(100)(pdrec)@(s93.eq.1.and.pd.eq.1)' do 'hs(201,2)=nt(100)(pdrec)@(s93.eq.1.and.pd1.eq.1)' do 'hs(201,3)=nt(100)(pdrec)@(s93.eq.1.and.pd2.eq.1)' do 'hs(201,4)=nt(100)(pdrec)@(s93.eq.1.and.et.eq.1)' f 201 grid s93=1 means selection VC=1 triggered events. At first two Runs sets (25-30.04) it is possible to see reasonable behaviour of different PD-trigger-elements action on full VC=1 triggered events spectrum: PD1 PD PD2 HW 3.25 3.5 10 <--- as it were installed at Egor's card SW 4.50 5.1 13 <--- as it seen from 201 histos. At last Runs set (01.05.94): PD1 PD PD2 HW 3.25 3.5 10 <--- as it were installed at Egor's card SW 11.50 12.5 ~25.0 <--- as it seen from 201 histos. (PD2-marked events almost unvisible on 201 histos - its threshold it is possible to see from s86=1 events (PD1)). do 'hs(301)=nt(100)(pdrec)@(s86.eq.1)' bins 200 0 35 do 'hs(301,1)=nt(100)(pdrec)@(s86.eq.1.and.pd.eq.1)' do 'hs(301,3)=nt(100)(pdrec)@(s86.eq.1.and.pd2.eq.1)' do 'hs(301,4)=nt(100)(pdrec)@(s86.eq.1.and.et.eq.1)' f 301 grid One of the possible reasons - p11-element is not included into PD-trigger sum making. Response from p11 is at ReadOut but (as seems) this response is not involved into PD-analog sum making - it would be possible reason of high mentioned shift of SW-visible thresholds. Another influence of "p11-not involving into PD Trigger Sum" is seen at PD-coordinate plot (both OnLine at OffLine) - You can see pure statistics of hits at p11-region of plots. Else one possible influence - strange behaivour of ET-coordinate plot for pure Luminosity events - most of electron hits are at e27,e26,e25,e24 etc. Most statistical is e27-element area of plot. We never seen this picture before. It is looking as vice versa of ordinary ET coordinate plot. It is possible to explain with really high values of threshold at PD-trigger element (due to absence of p11 which is disposed at central layer and must put large quota into PD-Arm Trigger sum). Pure Luminosity Trigger with high threshold at PD-trigger element selects electrons with low energy - and ET-coordinate plot shows that it is really so - most of e-hits at regions of plot with low Ee. Welcome any another explanations. Asking to Egor - put please Your diagnose on this behaivour of PD-trigger elements. Main - p11 is at ReadOut - we see good ordinary response from p11 at H1Lumi events. Is it possible that p11 is not involving into making of PD Analog Sum at this case? 02.05.1994 ==================== H1KFOM === HERA e-beam ............ This night it was fixed ordinary HERA e-beam activity. It were fixed 3 e-fillings. Ie total max was fixed at last night e-filling with near 24 mA at 04:30. Today morning it was fixed 2 short e-fillings (from 08:30 up to 09:00 and from 09:45 up to 10:40). At last both cases it were made rampings and e-beam were lost after some minutes of 27.515 GeV e-beam living. At 13:56 it was started 3th today e-filing but only first injection was made (Ie=4 mA, 42 bunches, (3+8)*14 bunch structure) and after some minutes e-beam was dumped or lost. Our Movable Platforms are at Working Positions from 10:30 today. At 14:07 it was started 4th today e-filling - after first injection Ie = 4.3mA, 42 bunches, (3+8)*14 bunch structure. Ie =11.7mA, 84 bunches, (6+5)*8 + (3+8)*12 bunch structure (14:22) At 14:24 e-beam was lost. At 14:30 it was started 5th today e-filling. It was made ramping at 15:05. Ie total was 13.8 mA after ramping. Number of bunches 84 with (6+5)*14 bunch structure. At 15:45 e-beam was lost. On the first view the same problems with high effective value of PD- threshold exists today too. Online estimation of PD threshold value with Egor's monitoring tool (VC-trigger for monitoring) shows value near 9.0 GeV ( it must be near 3-3.5 GeV). As seems really there are some problems with p11's quote at PD Arm AnalogTrigger Sum. Rates after ramping finishing were fixed with strange ratio too: Rate VC - 96 kHz, Rate ET&PD - 5.8 kHz (as seems very low value for Ie total 13.4 mA) - only due to PD threshold high effective value. 02.05.1994 ==================== H1KFOM === PDtrigger Elements Problem At 16:10 it was started 6th today e-filling. At 16:30 it was collected 126 bunches with Ie = 21.8 mA with bunch structure (9+2)*14. At 16:46 ramping started from Ie = 20.6 mA and after Ee=19.712 GeV ramping was stopped and during almost 45 min. at HERA e-ring existed electrons with Ee=19.712 GeV. At 17:26 ramping was prolongated up to "standard" value of Ee=27.515 GeV and was finished at 17:28 with Ie= = 17.68 GeV. It was decided to play with PD threshold value for understanding of situation. So - with LowPDThr = 3.5 GeV at Egor's card and at LUTR- bank we had effective PD low threshold value 9.0 GeV (both - at OnLine and Offline). Yura put new values of LowPDThr = 1.0 GeV for PD and 0.75 for PD1. On-line Monitoring put 3.5-3.7GeV effective value for Low PD threshold.Coordinate distributions at PD,ET became reasonable without any strange situation with electron hitting at ET and p11 area missing at PD. As seems with this new values of PD threshold situation is similar as we looked for at 25.04.94 Runs with 3.5 GeV PD-threshold. Bunch statistics is reasonable - no any noises from empty bunches. Attempt to return on ordinary value of LowPDthr (3.5 GeV) put the same effect with effective value of PD-threshold - it became 9.GeV again and strange coordinate plots appeared again. It is clear that namely PD-trigger element is responsible for high mentioned situation but why - it is not understandable. After high mentioned tuning - RateET&PD = 26 kHz with Ie=15 mA. 03.05.1994 ==================== H1KFOM === H1Lumi Run with Ee=27.515 Yesterday after high mentioned tuning of PD-trigger elements low thresholds it was decided to write H1Lumi Run for off-line checking of H1Lumi system behaivour. It was written Run 73979 with the next trigger elements thresholds (see LUTR-bank contents): Vetov = 1.5GeV; Vetos = 1.25GeV; PD = 1 GeV; PD1 = 0.75 GeV; PD2 = 10 GeV (old value); ET = 5.25 GeV. HV at VCs = 730 V, at VCv=900 (at 01.05.94 Runs = 700 V and 850 V, at 30.04.94 Runs - 700V and 900 V) Run was recorded from 18:52 up to 19:14 (Ie=12.4 mA at 18:52). At Run 73979 it were collected 68K events with 3 H1Lumi's subtriggers s86(as PD1), s93 and s94. Prescale Gaps were tuned on more or less equal quotas of each subtriggers at data sample. 03.05.1994 ==================== H1KFOM === HV tuning at PD-channels.. Yesterday it was found (from OnLine values of Calibration Coeffici- ents) that response from p11-element is less on factor almost 2 from neighbour p12-element and on factor 5 from neighbour p16-element. It was decided to change HV at p11 and p16 elements: p11 - 1740V --> 1790V p16 - 1990V --> 1900V With high mentioned HV-changing it was written two H1Lumi's Runs with the same thresholds and subtrigger set as at Run 73979. Run 73981 144838 events from 19:54 up to 20:39 (Ie from 9.2 up to 7.3mA Run 73984 47418 events from 20:49 up to 21:07 (Ie from 7.0 mA to ???). As seems it is possible to try the making of first release of Relation Coefficients set with data from Run 73981. 03.05.1994 ==================== H1KFOM === HERA e-beam ............ Tonight it were fixed 4 e-fillings (from 22:00 up to 22:30, from 23:10 up to 01:45, from 02:00 up to 03:15 and from 03:20 up to 08:45). Ie total max was fixed during first night filling - near 20 mA. Today morning it were fixed 2 e-fillings (from 09:10 up to 09:40 and from 09:45 which is at progress now (10:51)). First today e-filling - Ie=20.33 mA, 126 bunches with complicated bunch structure (9+2)*4 + (6+5)*9 + (3+2+3+3)*5 + (3+8+3+7)*1. After ramping at 09:36 up to energy 27.573 GeV e-beam was lost). Second today e-filling - Ie=11.8 mA, 84 bunches with bunch structure (6+5)*8 + (3+8)*12. Ramping was made at 10:00 up to energy 27.573 GeV. Movable Platforms were all night at Working Positions. 03.05.1994 ==================== H1KFOM === PD Thresholds Diagnose.. After yesterday investigations it is possible to make possible diagnose of strange behaivour pf PD-trigger elements low thresholds. As seems something happened with digital-analog converter (DAC) which converts selected PD-trigger elements low thresholds at Egor's trigger monitoring card into real amplitudes thresholds fot trigger sums. As seems it was changed "calibration" of this conversion - now for all PD-trigger elements (PD,PD1 and PD2) low thresholds value 1 GeV (4 DAC units) corresponds to old value 3.5GeV(14 DAC units) As seems - all is OK with ET-trigger elements low thrsholds - Egor's monitoring tools show ordinary threshold value for ET-trigger elements with 5.25 GeV installed - 3.5-4.0 effective. As seems - "not involving p11" was looked for due to high value (9GeV) of PD low threshold and low level of responce from p11. Problem of 'not involving p11 into PDTriggerAnalogSum' is cancelling - it was not right supposition. 03.05.1994 ===== dice2 ======== H1KFOM === 100th-ntuple for R73979 It was produced high mentioned 100th ntuple for Run 73979 (only for part of Run due to problems with LOOK-space at dice2-environment). Relation Coefficients Set and CC for VCs were got as at previous 100th ntuple producing (22.02.94 release). LOOK-data set with 100th ntuple was moved into DESY IBM storage: H1KFOM.LOOK.R73979 FAST13 3010200E 00000 DASD Only 1/2 events of Run 73979 are at 100th ntuple. First view on contents of this tuple shows that all is OK with H1Lumi system after "tuning" of low thresholds for PD and PD1. SW-thresholds which are visible at pdrec-histos: PD1 - 3.5 GeV PD - 4.5 GeV acceptance threshold for PDrec spectrum (ET&PD events) - near 7 GeV. As You remember Run 73979 was written before last tuning of HV at p11 and p16 elements. Tool for creating of high mentioned binary LOOK-data set is kept at /home/local/fomenko/LUMMbank-directory: LUMM73979e.f - example of main program MakeLUMM73979e - example of make file s_LUMMbank73979 - example of script for job starting (at R-subdir.) 03.05.1994 ==================== H1KFOM === H1Lumi Run with Ee=27.573 Today during 2nd e-filling (which was finished at 14:30 with beam dump) it was recorded little Run 74078 (due to problems at SGI-logging) Run 74078 was started at 13:35 and finished at 13:43 (due to SGI-prob- lems). Ie total was at this period near 6 mA. Thresholds values were the next (see LUTR-bank contents too): VCs = 1.0 GeV; VCv = 1.5 GeV; PD = 1 GeV; PD1 = 0.5 GeV; PD2 = 3 GeV; ET = 5.25 GeV. Yura changed HV on VCs=700V and VCv=1000V. At high mentioned Run it was used the same 3 H1Lumi subtriggers - s86(asPD1),s93 and s94 with prescales on equal quotas at data sample. Number of events - 20995. Before start of this Run as seems it was made good calibration of ET,PD channels at Luminosity subbranch - energy resolution at OnLine etrec+pdrec spectrum was near 1.03-1.04 GeV (sigma). Note: H1 Main Solenoid is Off. 03.05.1994 ===== dice2 ======== H1KFOM === 100th-ntuple for R73981 It was produced high mentioned 100th ntuple for Run 73981 (only for part of Run due to problems with LOOK-space at dice2-environment). Relation Coefficients Set and CC for VCs were got as at previous 100th ntuple producing (22.02.94 release). LOOK-data set with 100th ntuple was moved into DESY IBM storage: H1KFOM.LOOK1.R73981 FAST01 3010200E 00000 DASD Only 30K events of Run 73981 are at 100th ntuple. As You remember Run 73981 was written after last tuning of HV at p11 and p16 elements. Tool for creating of high mentioned binary LOOK-data set is kept at /home/local/fomenko/LUMMbank-directory: LUMM73981e.f - example of main program MakeLUMM73981e - example of make file s_LUMMbank73981 - example of script for job starting (at R-subdir.) 03.05.1994 ==================== H1KFOM === HERA e-beam ...... Today at 15:00 it was started 3rd today e-filling. At 15:11 with Ie=13.83mA, with 75 bunches and with bunch structure (6+3)*5 + + (3+8)*15 it was started ramping of e-ring. Ramping was finished at 15:13 with Ee=27.573 GeV e-energy and with Ie=12.2 mA. Rates at H1Lumi Trigger Elements at end of ramping moment: VC - 500 kHz, ET*PD - 130 kHz. Now (17:35) Ie = 5.9 mA VCrate=130 kHz, ET&PDrate = 32 kHz. 03.05.1994 ==================== H1KFOM === H1Lumi Runs with Ee=27.535 At the final stage of 3rd today e-filling it were recorded the set of H1Lumi Runs with different HV at VCv. Run 74115 - HVvcv=1000V HVvcs=700V; Thresholds vcs=1.25 vcv=1.5 Run 74116 - HVvcv=1050V HVvcs=700V; Thresholds vcs=1.25 vcv=1.5 Run 74117 - HVvcv=1100V HVvcs=700V; Thresholds vcs=1.25 vcv=1.75 Run 74118 - HVvcv=1000V HVvcs=700V; Thresholds vcs=1.25 vcv=1.5 At each Run it were recorded near 55K events (at last Run - 100K). Luminosity Subbranch was good calibrated before data taking (as seems) Sigma=1.04 GeV at online etrec+pdrec online histo for ET&PD&nVC-events View on online calibration coefficients shows that CCp11 became better and CCp16 became better too(almost the same values as neighbour Energy of e-beam at HERA TV screen was changed during data taking from 27.573 up to 27.535 GeV. 04.05.1994 ==================== H1KFOM === HERA e-beam .............. Tonight it was fixed 3 e-fillings of HERA e-ring (from 23:40 up to 04:45, from 05:00 up to 05:45 and from 06:35 up to 09:50). Ie max during first filling was fixed near 16 mA, during second - 4 mA, during third - 7 mA. Now HERA e-magnets is off - HERA TV screen shows zero at Ee. Movable platforms are at Parking Positions from yesterday evening (after end of last H1Lumi Run 74118 with almost 100K events). Today H1CDAQ will be busy all day and all evening - at the morning Run Coordinator's meeting was adopted time schedule of today H1CDAQ using. 04.05.1994 ===== dice2 ======== H1KFOM === 100th-ntuples for R7411X It were produced high mentioned 100th ntuples for Runs 74115,74116 and 74117 (yesterday recorded for HV vcv investigations). This ntuples are kept at the next LOOK-binary data sets: H1KFOM.LOOK.R74115 FAST03 3010200E 00000 DASD H1KFOM.LOOK.R74116 FAST08 3010200E 00000 DASD H1KFOM.LOOK.R74117 FAST09 3010200E 00000 DASD All high mentioned LOOK-binary data sets were produced at dice2- environment and moved into IBM storage through ftp-protocol. Examples of tools are kept at /home/local/fomenko/LUMMbank-directory LUMM7411Xe.f - example of main program MakeLUMM7411Xe - example of make file s_LUMMbank7411X - example of script for job starting (at R-subdir.) where X=5,4,6. First view on contents of 100th ntuples shows that best energy resolution (Sigma at etrec+pdrec spectrum for ET*PD*nVC-events) is 0.90 GeV. May be this digit so large due to using of "old" relation coefficients. As seems calibration became again task number 0. There are a lot of data for attempts of calibration (with H1 Magnet ON and with H1 Magnet Off). It is pity but this data were collected with different HV-status at Veto Counter (now it is not decided which HV must be installed as default). May be after analysis of high mentioned 100th tuples it would be possible to select reasonable values. PD-trigger elements (with 'tuned' low thresholds 3.5GeV-->1GeV for PD 3.25GeV-->0.75 for PD1 and 10GeV-->3 GeV) are working with the same manner as at Saturday 30.04.94 - the same effective values. Ratio of vcv(FADC counts)/vcs(FADC counts) = 1/7 at all last Runs (74115-74117). Before this we looked for 1/5. OnLine calibration of VCS-channels is made with the same strange manner - we looked for values 27(!),20, and another large values at OnLine. And we looked for sensitivity of CCvcs to yesterday VChv changing. CCvcs was changed into reasonable side but with large values. Due to this large CCvcsp-values at H1CDAQ H1EventDisplay we looked for 'hugo' hits at VCS. This behaivour of VCcalibration will be checked - may be some bugs exist at this branch of calibration. May be selecti- on of ET*PD*VC-triggered events must be checked with additional analyze of VCs FADC-response without accounting of CCvcs-current value etc. This effect naturally corrupts PhArm Energy Spectra at LUMMbanks. These spectra are looked as far from reasonable. Rate of Photon Arm spectra collection is not more then 2Hz and I never looked for full statistics 10000 events at spectra and as seems it will be difficult to use its for thresholds correction or for Photon Arm Shift. Yesterday we looked for interesting behaivour of bunch statistics at 6th and 7th groups from Histo Memory. It were not fixed any statistics at first bunches of each train. All statistics for first bunches at each train was collected at 22 and 23 groups (PD1 and PD coincidence). It is not understandable why first bunch at each train does not put any statistics into 6th and 7th group of HistoMemory. May be some effects of pulse overlapping on neigbours bunches? Or something else? 04.05.1994 ==================== H1KFOM === HERA e-beam ............ First today e-filling started at 12:45 and finished at 14:28 with beam dump. Ie total max was fixed as 8.7 mA. Number of bunches - 39. Bunch structure - (3+8)*13. Ee after ramping - 27.521 GeV. Movable platforms were put into Working Position at 12:55. Rates were fixed after end of ramping with Ie=8.23 mA - Ratevc=128kHz RateET*PD = 35 kHz. Second today e-filling was started at 14:50 and is in progress now 15:20). Ie total = 2.64 mA, Ee = 11.971 GeV, injection status. 04.05.1994 ==================== H1KFOM === HERA e-beam ............ Second today e-filling was finished without ramping at 16:10. Third today e-fillng was started at 16:15. It were injected 42 bunches with Ie total near 10 mA and at 16:25 it was made ramping up to energy 27.521 GeV. Now third e-ffiling in progress - Ie=7.9 mA VCrate=60 kHz, ET&PDrate= 13.6 kHz (16:45). At 18:47 e-beam was dumped from Ie=4 mA. Beam conditions were non-standard - beam position was 1 cm low as ordinary and -0.5cm far from HERA center. H1CDAQ was busy and it was no any posibility to write some H1Lumi data. Movable platforms were put into Parking Positions at 19:21. 05.05.1994 ==================== H1KFOM === HERA e-beam ............ Last night it were fixed 7 e-fillings of HERA e-ring. Ie total max was fixed as value near 20 mA at three e-fillings. Ee =27.521 GeV and Ie = 13.8 mA at 07:15 (during last 7th e-filling). Now (08:22) there is pause at HERA e-beam activity. 05.05.1994 ==================== H1KFOM === HERA e-beam ............ Today (as seems) HERA-e-team returned to large e-beam currents study. Up to now (12:06) it were fixed 4 today e-fillings of HERA e-ring: 1) from 08:15 up to 09:30, Ie = 19 mA (Working Positions 09:05-09:30 2) from 09:55 up to 10:15, Ie = 25 mA 3) from 10:40 up to 11:25, Ie = 24 mA (Working Positions 11:20-11:25 4) from 11:35 and in progress now, Ie = 18.5 mA with 84 bunches and with (6+5)*14 bunch structure. First 3 e-fillings were finished during unsuccesful rampings which were started immediately after large current e-beam injection. Now it is made long pause after 1st injection (at 12:10 Ee=11.971 Ie = 16.6 mA). At 12:13 it was made additional injection (Ie=24.65mA, number of bunches - 120, bunch structure (9+2)*12 + (6+5)*2) and immediately ramping was started. At the Ee=20.366 GeV ramping was stopped at 12:19. At this stage of ramping RateVC is minimal (50 kHz after 540 kHz at the previous ramping step). Ramping was finished at 13:11 with Ee=27.521 GeV and Ie= 17.8 mA. VCrate=316 kHz and ET*PDrate=79 kHz at this moment. Movable Platforms were put into Working Positions at 13:16. At 13:18 e-beam was lost. 05.05.1994 ==================== H1KFOM === HERA e-beam ............ Today afternoon it were fixed 4 e-fillings (5th-8th): 5) from 13:25 up to 13:41, Ie = 25.2 mA 113 bunches 6) from 13:52 up to 14:49, Ie = 23.1 mA (Working Pos. 14:20 - 14:50) 107 bunches 7) from 15:17 up to 15:30, Ie = 24.8 mA 114 bunches 8) from 15:39 and in progress now (17:22) Ie = 23.1 mA with 105 bun- ches with (9+2)*10 + (6+5)*4 bunch structure. Last filling was succesful - ramping was made in 2 min. (from 15:52 up to 15:54 on 27.520 GeV with Ie=22.85 after end of ramping After ramping Movable Platforms were put into Working Positions: RateVC = 620 kHz, Rate ET*PD = 150 kHz (after end of ramping). At 17:00 Luminosity Subbranch is good calibrated but H1CDAQ is busy - we requested 2 Runs with 15 min. each with new HV sets at VC (HVvcs=800V, HVvcv=1000V now). At 17:35 H1CDAQ was free and H1ShiftCrew prposed to get it for H1Lumi Runs but at this time beam was lost. 05.05.1994 ==================== H1KFOM === HERA e-beam & H1Lumi Runs HERA e-beam activity was prolongated with 9th and 10th today e-fillings: 9) from 18:06 up to 18:15, Ie = 23.8 mA (Working Pos. 18:10 - 18:15) 120 bunches (6+5)*20 Rates after end of ramping: RateVC=360 kHz, RateET*PD=80 kHz 10) from 18:36 up to ??:??, Ie = 22.1 mA (Working Pos. 18:55 - ??:??) 99 bunches (6+5)*13 + (3+8)*7 Rates after end of ramping: RateVC=280 kHz, RateET*PD=70 kHz (before Moving of Platforms - immediately after ramping we looked for rate at VC=1.3 MHz). H1CDAQ was free and we recorded three H1LumiRuns: 74402 with HVvcv=1000V and NVvcs=800V - near 60K events (~ 15min) Ie was fixed as value near 18.3 mA at start Run moment 74403 with HVvcv=1050V and HVvcs=800V - near 4K events (~ 1min) Ie was fixed as value near 15.6 mA at start Run moment 74404 with HVvcv=1050V and HVvcs=800V - near 60K (new Prescales) Ie was fixed as value near 15.2 mA at start Run moment (~15min) Remark: H1 Main Magnet is ON. Thresholds values: VCv=1.5GeV VCs=0.75 GeV PD=1.0GeV PD1=0.75GeV PD2=3.0GeV ET=5.5GeV (see LUTR-bank). At the stop run 74404 moment it was fixed resolution 0.96 GeV (Sigma) at ETrec+PDrec-spectrum of ET&PD&nVC-events at Luminosity subbranch. After H1Lumi it was created queue on H1CDAQ - Ferencei and Vallee - we stopped last Run after 60K events. 06.05.1994 ==================== H1KFOM === HERA e-beam ............. Tonignt it were fixed 5 e-fillings and 6th e-filling which was started today early morning is in progress now (08:20). At all high mentioned e-fillings Ie total max was near 20-21 mA. At last (6th) e-filling Ie total after end of injection was fixed near 26 mA. Now Ee=27.521GeV. At 08:26 e-beam was dumped or lost. As seems there is problem of short life time of HERA e-fillings with Ie total more then 20 mA. 06.05.1994 ===== dice2 ======== H1KFOM === 100th-ntuple for R7440X It were produced high mentioned 100th ntuples for Runs74402,74403,74404 (yesterday recorded for HV vcs,vcv investigations). This ntuples are kept at the next LOOK-binary data sets: H1KFOM.LOOK.R74402 FAST16 3010200E 00000 DASD H1KFOM.LOOK.R74403 FAST07 3010200E 00000 DASD Data from 74403,74404 were combined into one LOOK-data set due to the same HV-conditions. The best resolution which it is possible to see at high mentioned data sets (Sigma at ETrec+PDrec spectrum at pure lumi-events) is 0.99 GeV. (with 22.02.94 release of Relation Coefficients Set). All is OK with thresholds PD,PD1,PD2,ET - its are looking for as reasonable. 06.05.1994 ==================== H1KFOM === HERA e-beam ............. Today it was fixed 6 e-fillings: 1) from 09:45 up to 11:50, Ie = 24 mA 2) from 12:00 up to 12:27, Ie = 24.5 mA 3) from 14:05 up to 15:20, Ie = 20 mA (Working Positions 14:50-15:20 4) from 15:27 up to 16:25. Ie = 20 mA (Working Positions 15:50-16:30 5) from 17:55 up to 18:24. Ie = 24 mA (Working Positions 18:15-18:30 with 126 bunches with bunch structure (9+2)*4 + (6+5)*9 + + (3+2+3+3)*5 + (3+8)*2 Today morning Yura asked some time at H1CDAQ for H1Lumi Runs but up to now it was not any posibility to collect data - e-beam with stable conditions is rare guest today at HERA e-ring. 6) from 18:35 up to ??:??. Ie = 20 mA (Working Positions 18:55-??:?? with 126 bunches with bunch structure (9+2)*4 + (6+5)*9 + + (3+2+3+3)*5 + (3+8)*2 Beam conditions were far from standard - YPD= -1.8 -1.9 cm We did not write any data due to bad beam conditions. H1CDAQ was got for CosmicRun together with stable e-beam. Ee at HERA TV-screen was changed on 27.492 GeV value (after some minutes after end of ramping when it was fixed as ordinary value 27.521 GeV). 07.05.1994 ==================== H1KFOM === HERA p-beam ............. Today morning it was started HERA p-beam study. First HERA p-beam was fixed near 10:30 today. Ip total was fixed near 12:30 with value near 1.65 mA. 10.05.1994 ==================== H1KFOM === HERA p-beam ............. Yesterday evening and tonight it was fixed almost permanent activity of HERA p-beam. Ip total max was fixed yesterday evening near 20:30 as Ip=1.6mA. At this moment (08:20) Ep =39.73GeV (injection energy) and Ip = 0.53 mA. After 10:00 today and up to 14:30 it were fixed four p-injections with Ip total more then 2.0 mA. 11.05.1994 ==================== H1KSOL === VETOv efficiency.... VetoV efficiency was measured from the data of LUMI stand alone RUNs taken during e-beam studies. Monitoring trigger - PD1 (bit S86,nonstandard setting) The results are cited below. HVv(kV) HVs(kV) Thrv(GeV) Thrs(GeV) Efficiency(%) 0.85 0.7 1.75 1.75 44.5+-1. 0.9 0.7 1.75 1.75 71 +-1. 1.0 0.7 1.5 1.25 80 +-1.5 1.05 0.7 1.5 1.25 86 +-1. 1.1 0.7 1.75 1.25 90 +-1.5 1.0 0.8 1.5 0.75 82 +-1.5 1.05 0.8 1.5 0.75 85 +-1.5 12.05.1994 ===== dice2 ======== H1KFOM === Relation Coefficients Sets It was made first attempt to produce Relation Coefficients Sets for two data samples which were recorded during last HERA e-beam activity. 1) First sample - data from Runs 73981 and 73984 which were recorded 02 May 1994 from 19.54 up to 21:07 with Ee= 27.515Gev and Ie from 9.2 mA (at 19:54) up to 7.0 (20:49). H1 Main Magnet was ON. Number of bunches 126 with bunch structure (9+2)*14. From high mentioned Runs it were selected 68835 events with ET&PD&nVC=1 trigger. For calibration procedures it were got 33731 events and used for calibration (after cuts) only 27060 events. 2) Second sample - data from Runs 74115 and 74118 which were recorded 03 May 1994 from 18.12 up to 19:39 with Ee= 27.573Gev and Ie from 5.0 mA (at 18:12) up to 4.3 (19:04). H1 Main Magnet was OFF. Number of bunches 75 with bunch structure (6+5)*5 +(3+8)*15. Ee at HERA TV-screen was changed during data taking from 27.573 GeV up to 27.535 GeV. From high mentioned Runs it were selected 41282 events with ET&PD&nVC=1 trigger. For calibration procedures it were got 30538 events and used bfor calibration (after cuts) only 25063 events. Both samples were collected after last PD11,PD16 HV changings. At both samples the same HV was installed at VCs and VCv: 700V, 1000V. For each from high mentioned samples it were made 3 calibration steps and 4 test of calibration steps. First test of calibration was madefor Relation Coefficients Set from 22.02.94 (last values of H1DataTaking93) It were got the same (22.02.94) relation coefficients set for first steps of calibration with each sample. At both cases Ee was got as 27.515 Gev (as calibration reper). At both cases Luminosity subbranch was good calibrated before data collection. At both cases CCvc was got as 3.1 value. e48 channel at ET was proclaimed as dead channel. For calibration quality it were used the next cuts: xetcut = (xet.gt.-6.0.and.xet.lt.6.0) vcdepcut = (vcdep.lt.0.2) These are the results of high mentioned 4 tests from ETrec+PDrec histo LOOK peak finder: Runs 73981 and 73984 all events xetcut vcdepcut xetcut&vcdepcut Step Mean Sigma Mean Sigma Mean Sigma Mean Sigma 1 27.37 1.22 27.49 1.18 27.54 1.16 27.64 1.14 2 27.34 1.04 27.38 0.99 27.53 0.95 27.55 0.92 3 27.34 1.01 27.34 0.98 27.47 0.92 27.48 0.90 4 27.36 1.00 27.38 0.97 27.49 0.91 27.52 0.89 Runs 74115 and 74118 all events xetcut vcdepcut xetcut&vcdepcut Step Mean Sigma Mean Sigma Mean Sigma Mean Sigma 1 26.85 1.06 26.90 1.01 26.83 1.04 26.85 1.00 2 27.42 1.00 27.50 0.95 27.46 0.97 27.52 0.93 3 27.43 0.99 27.50 0.94 27.50 0.97 27.52 0.92 4 27.45 0.99 27.52 0.95 27.49 0.96 27.53 0.92 So - the best energy resolution (sigma at ETrec+PDrec histo of ET&PD&nVC-triggered events) is 0.89 GeV at first sample and 0.92 GeV for second sample. As always it was fixed large difference at PDrec and ETrec energy distributions for sample with H1 Magnet ON and sample with H1 Magnet Off. Really the two dimensional (PDrec,ETrec) plots very differ for high mentioned samples (H1 Magnet influence). All possible energy and coordinate distributions for both samples it is possible to make and see from 100th tuples (with short items list - ETrec,PDrec,VCdep,Xet,Yet,Xpd,Ypd,time and Run). LOOK-binary data sets with high mentioned 100th tuples for 4th steps of calibration tests are kept at DESY IBM storage: H1KFOM.TLOOK3.R73981 FAST14 3010200E 00000 DASD H1KFOM.TLOOK3.R74115 FAST16 3010200E 00000 DASD These data set were produced at dice2-environment and were moved into IBM-storage with ftp-protocol. These are the results of first attempt of Relation Coefficients Sets producing: Runs 73981 and 73984: -------------------- ==> Relation Coeff.Set for ET 1.101 1.111 1.100 1.118 1.079 1.130 1.101 1.092 1.100 1.020 1.267 1.031 1.003 1.073 1.118 0.914 1.013 0.963 1.018 0.969 1.000 1.005 0.996 0.977 0.969 0.979 0.953 0.951 1.060 0.971 1.040 0.980 1.050 1.015 1.092 1.225 1.214 0.944 0.891 1.025 0.934 0.879 1.077 1.086 1.065 1.175 1.049 1.116 0.000 ==> Relation Coeff.Set for PD 0.977 0.927 0.957 0.896 0.923 0.611 1.029 0.978 1.052 0.997 0.991 1.020 1.030 1.042 1.012 0.965 1.718 0.957 0.983 0.953 0.951 0.658 0.753 0.854 0.926 ==> Sig. at Relation Coeff.Set for ET 0.000 0.001 0.000 0.000 0.000 0.002 0.002 0.000 0.005 0.018 0.027 0.002 0.020 0.024 0.012 0.012 0.027 0.021 0.012 0.022 0.013 0.016 0.015 0.011 0.013 0.016 0.017 0.012 0.004 0.010 0.008 0.017 0.010 0.008 0.009 0.000 0.005 0.008 0.004 0.004 0.000 0.000 0.000 0.002 0.001 0.010 0.005 0.000 0.000 ==> Sig. at Relation Coeff.Set for PD 0.001 0.012 0.010 0.002 0.000 0.009 0.010 0.015 0.016 0.011 0.012 0.034 0.024 0.016 0.019 0.009 0.024 0.011 0.013 0.022 0.002 0.013 0.009 0.002 0.000 ==> Ent. at Relation Coeff.Set for ET 3 8 3 2 13 6 7 7 18 32 56 22 27 29 869 538 918 1046 1141 1662 883 7031 11017 9197 7498 6380 5160 2435 2415 2137 870 834 632 309 302 4 8 15 19 10 9 3 1 3 3 8 4 1 0 ==> Ent. at Relation Coeff.Set for PD 39 37 81 84 6 136 1215 5009 5026 1010 876 7561 19726 17244 3988 243 2319 6375 5407 654 55 21 52 32 3 Runs 74115 and 74118: -------------------- ==> Relation Coeff.Set for ET 1.100 1.059 1.101 1.121 1.129 1.105 1.108 1.089 1.151 1.096 0.648 1.224 1.085 1.146 1.249 0.949 1.135 1.140 1.150 1.106 1.105 1.058 1.069 1.039 1.044 1.078 1.080 1.086 1.124 1.037 1.123 1.060 1.073 1.091 0.921 1.211 1.244 0.865 0.845 1.123 0.948 0.000 1.101 1.091 1.103 1.055 0.948 1.120 0.000 ==> Relation Coeff.Set for PD 0.937 0.866 0.800 1.022 0.937 0.582 0.906 1.020 1.047 0.752 0.938 1.004 0.965 0.976 0.983 0.934 0.835 0.941 0.967 0.932 0.881 0.714 0.809 0.992 0.932 ==> Sig. at Relation Coeff.Set for ET 0.000 0.010 0.001 0.004 0.007 0.009 0.000 0.005 0.011 0.010 0.040 0.019 0.010 0.008 0.005 0.009 0.002 0.007 0.003 0.008 0.008 0.011 0.014 0.012 0.009 0.016 0.010 0.005 0.004 0.005 0.006 0.010 0.009 0.008 0.008 0.003 0.007 0.001 0.002 0.001 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 ==> Sig. at Relation Coeff.Set for PD 0.002 0.002 0.012 0.002 0.002 0.004 0.027 0.005 0.013 0.007 0.006 0.017 0.010 0.012 0.006 0.017 0.014 0.015 0.007 0.004 0.006 0.008 0.006 0.003 0.000 ==> Ent. at Relation Coeff.Set for ET 11 32 16 23 43 17 16 53 137 130 142 118 93 51 1000 1408 1844 1809 1239 1148 429 7924 12813 10065 6558 3977 2170 817 3054 3724 1529 784 298 71 59 19 7 4 7 5 3 0 3 3 2 4 1 1 0 ==> Ent. at Relation Coeff.Set for PD 38 53 602 203 17 87 1421 8378 10112 2307 295 4344 17192 18392 4378 46 498 2344 2426 249 21 10 24 19 4 It is not understandable yet but interesting - as You can see from high mentioned results - relation coefficient for P16 is very differs for both samples (1.718 for first sample and 0.835 for second one). HV - same, start values the same - only difference at Ie-values (?), and another beam position (on X-axis shifted to HERA center)(?). All high mentioned procedure (14 jobs - (3+4)*2) was made at dice2 environment. Each job got near 13-15 min of SGI-CPU. Tools are kept at /home/local/fomenko/Calibr94 and .../Testclbr1 - directories: .../Calibr94 - directory: MakeLUMM73981A - make file for first step of Rel.Coeff.Defining MakeLUMM73981B - ------------- 2nd ----------------------------- MakeLUMM73981C - ------------- 3rd ----------------------------- MakeLUMM74115A - make file for first step of Rel.Coeff.Defining MakeLUMM74115B - ------------- 2nd ----------------------------- MakeLUMM74115C - ------------- 3rd ----------------------------- lrinit.f - subroutine with start values of Calibration Coeff. (the same for both samples) lrinit73981a.f - --------------- CC after 1st step lrinit73981b.f - --------------- CC after 2nd step lrinit74115a.f - --------------- CC after 1st step lrinit74115b.f - --------------- CC after 2nd step main73981x.f - example of main program for 3 calibration jobs(x=A,B,C main74115x.f - example of main program for 3 calibration jobs(x=A,B,C s_18H1LumiCalibration73981A - examples of scripts for job submitting s_18H1LumiCalibration74115A (at R-subdirectory) .../Testclbr1- directory: MakeLUMM73981A - make file for first step of Rel.Coeff.Test (22.02.94) MakeLUMM73981B - ------------- 2nd ------------------------ MakeLUMM73981C - ------------- 3rd ------------------------ MakeLUMM73981D - ------------- 4th ------------------------ MakeLUMM74115A - make file for first step of Rel.Coeff.Test (22.04.94) MakeLUMM74115B - ------------- 2nd ------------------------ MakeLUMM74115C - ------------- 3rd ------------------------ MakeLUMM74115D - ------------- 4th ------------------------ lrinit.f - subroutine with values of Relation Coeff. (22.04.94) (the same for both samples) lrinit73981a.f - --------------- RC after 1st step lrinit73981b.f - --------------- RC after 2nd step lrinit73981c.f - --------------- RC after 3rd step lrinit74115a.f - --------------- RC after 1st step lrinit74115b.f - --------------- RC after 2nd step lrinit74115c.f - --------------- RC after 3rd step main73981x.f - example of main program for 4 test RC jobs(x=A,B,C,D main74115x.f - example of main program for 4 test RC jobs(x=A,B,C,D s_TestCalibrationR73981BA - examples of scripts for job submitting s_TestCalibrationR74115BA (at R-subdirectory) Really these Relation Coefficients sets will be (can be) another for ep -collisions data (due to another values of FADC-shifts (on 20 nsec from ordinary values). We must be ready to collect these type of data for Relation Coefficients defining for ep-collisions data. 13.05.1994 ==================== H1KFOM === HERA p-beam ............. Today early morning (at 05:30) it was injected powerful HERA p-beam with Ip total near 28 mA. After ramping up to 820 GeV Ip total is near 24.7 mA and exists up to now (09:40) with good life time (not less then 99 hours). 13.05.1994 ===== FIC#1 ======== H1KFOM === RunStartRecord problems.. Today morning it was fixed the same situation with RunStartRecord prepare acknolegements as it was fixed at 23.04.94(see message at LLC) After Run Start - we have reaction of MEBRequest as "Run Start Record not required". All was OK before today morning. No any changings it was made at H1LumiOnLineFIC1-program. Old release is working. Yesterday evening, tonight and today morning (Runs 75033-75062 at temporary cartridges) Luminosity branch was included into H1CDAQ and it was not fixed any problems. All branches were ON at high mentioned Runs. Luminosity branch was Off from H1CDAQ-environment from 09:20 today. 13.05.1994 ==================== H1KFOM === VCs Trip Again .......... Yesterday evening it was happened VCs trip.Yura switched off HV at VCs today morning. H1 Shift Crew made attempts to switch on HV at VCs after trip (as Instruction requires) but trip was appeared again. At Monday access it will be made attempt to find something again. May be HV cabel will be changed on another one. HV at VCs was 800V before fixing of high mentioned trip. 13.05.1994 ===== dice2 ======== H1KFOM === VCs Calibration ......... It was made attempt to find Calibration Coefficients for VCs for high mentioned data samples. It was made preliminary selection of ET&PD&VC triggered events from high mentioned data samples and it were produced 2 data sets: -rwxr-xr-x 1 fomenko h1 68551560 May 12 13:54 ETPDVC.R073981.R073984.A00 -rwxr-xr-x 1 fomenko h1 81801720 May 12 13:57 ETPDVC.R074115.R074118.A00 This procedure was made at dice2 environment. Tool is kept at /home/local/fomenko/Selection - directory: H1LumiSelectionETPDVC.qsub - example of script for selection job subh H1LumiSelectionETPDVC1.qsub ---------------------------------------- MakefileETPDVC - makefile for application creatiotion mainETPDVC.f - example of FORTRAN-source for selection FPACKETPDVC.options - example of FPACK-options for selection FPACKETPDVC1.options - -------------------------------------- It was made 3steps VC-calibration procedure for each sample (3 steps at one job). It were made test steps for each sample (with produced values of CCvcs and early produced Relation Coefficients Sets. Tools for calibration and test are kept at /home/local/fomenko/vetoclbr and ..../tstvclbr - directories. ..../vetoclbr -directory: Makefile73981 - examples of makefile for VCs calibration application Makefile74115 s_vetoclbr73981 - examples of scripts for submitting of 3 steps jobs s_vetoclbr74115 ..../tstvclbr -directory: Makefile73981 - examples of makefiles for testing of CCvcs Makefile74115 s_ChVetoTest73981 - examples of scripts for submitting of test jobs s_ChVetoTest74115 These are the results of high mentioned tests from ETrec+PDrec histo LOOK peak finder: Runs 73981 and 73984 all events xetcut Step Mean Sigma Mean Sigma 1-3 27.41 1.28 27.40 1.23 Runs 74115 and 74118 all events xetcut Step Mean Sigma Mean Sigma 1-3 27.33 1.41 27.16 1.33 Energy spectrum for VCs-channel is looking as reasonable - as always visible part of spectrum up to 10 GeV. For calibrations and tests it were used 32931 ET&PD&VC-events from first sample and 29852 events from second one. It were produced 100th ntuples at test jobs which are kept at the next LOOK-binary data sets:. H1KFOM.VLOOKT.R73981 FAST08 3010200E 00000 DASD H1KFOM.VLOOKT.R74115 FAST05 3010200E 00000 DASD The main result - with the same HV status at VCs and VCv we have different Calibration Coefficients for VCs at these two data samples: Runs 73981,73984 --> CCvcs = 7.979 Mev/FADCcount Runs 74115,74118 --> CCvcs = 6.932 Mev/FADCcount HVvcs=700V and HVvcv=1000V during collection of Runs 74115 and 74118. 13.05.1994 ==================== H1KFOM === Which Must be BS SigmaVis? Today at H1 weekly meeting it was proclaimed that it is possible to wait ep-collisions with 12 GeV HERA e-beam at first days of nearest ep- collisions. As I remember - last attempt to make this type collisions (unsuccesful) was made at November 1992 at the finish of H1DataTaking92 (see messages 08.11.92 and 09.11.92 at LL5). At that time we put new Bremstrahlung Visible Cross Section into our Lumimetr. This value was calculated by S.Levonyan (with old HERA geometry and old ET-position etc.). This digit was proclaimed at H1 LPI Report 02-92 and was 22.0 mb (with Fast MC calculation only). Which Sigma Visible we can use at nearest Luminosity measurements with Ee=12 GeV? Really we can change and thresholds values on 12 GeV e-beam conditions. Due to mentioned at 01-03/05/94 problem with PD-thresholds these thresholds can be more rough then at previous thresholds installations on 12GeV e-beam conditions and may be it is needed to ask some time for short H1Lumi data taking with 12 GeV stable e-beam for possible thresholds investigations and calibration. 13.05.1994 ===== dice2 ======== H1KFOM === VCs Calibration ......... It was made attempts to produce calibration coefficients for VCs- channel at Runs 74116 (HVvcs=700, HCvcv=1050) and 74117 (HVvcs=700, HVvcv=1100). Next Calibration Coefficients Values were produced: Runs 74116 --> CCvcs = 4.314 Mev/FADCcount Runs 74117 --> CCvcs = 2.287 Mev/FADCcount These are the results of high mentioned tests from ETrec+PDrec histo LOOK peak finder: Run 74116 (10K events) all events xetcut Step Mean Sigma Mean Sigma 1-3 27.25 1.38 27.23 1.31 Run 74117 ( 9K events) all events xetcut Step Mean Sigma Mean Sigma 1-3 26.79(?) 1.47 26.84(?) 1.40 It were produced 100th ntuples at test jobs which are kept at the next LOOK-binary data sets:. H1KFOM.VLOOKT.R74116 FAST14 3010200E 00000 DASD H1KFOM.VLOOKT.R74117 FAST16 3010200E 00000 DASD Note that HVvcv now have large influence on CCvcs due to Egor connec- ted vcv and vcs channels. VCV has 1/5 response at vcs (through resistor scheme). It was discussed at March94 H1LPI meeting. 13.05.1994 ==================== F11LEV === B-H x-section for 12 GeV e For the beam configuration 12.0 GeV e- x 820.0 GeV p the value of the accepted B-H cross section (coinsidence method: trigger ET*PD) is given in a table below (syst. error due to the possible 1mm ET-position shift is 1%). Note however, that the acceptance at ET is 2-9 GeV (and so 3-10 GeV at PD). This requires rather low E-thresholds which perhaps is not so easy to achieve, if at all possible. Thus, the real trigger thresholds may cut the acceptance and make a visible cross section somewhat smaller. One has to investigate it on the real data. ----------------------------------------------------------------------- XCUT[cm] = 7.70 7.60 7.00 6.60 6.00 5.50 ACCEPTANCE= 0.45086 0.44116 0.40535 0.37913 0.34155 0.31072 SIGMA[mb] = 30.05 29.40 27.01 25.26 22.75 20.72 ----------------------------------------------------------------------- Note, that in calculation a standard e-beam optics were assumed (i.e. as in 1992). 13.05.1994 ===== dice2 ======== H1KFOM === VCs Calibration ......... It was made preliminary calibration of VCs-channel for H1Lumi Runs 74397-74402 and 74403-74404. At first data sample was installed HVvcv=1000V and HVvcs=800V. At the second one - HVvcv=1050V and HVvcs=800V. Data from these Runs were collected with H1 Magnet ON and with Ee=27.520 GeV. Number of bunches - 99. Bunch structure was (6+5)*13+(3+8)*7. Ie total was fixed as 20.0 mA at start of Run74397 and as 14.1mA at 19:43. High mentioned Runs were recorded from 19:04:59 up to 19:52:20 05/05/94. It was supposed that Relation Coefficients Set which was defined for Runs 73981,73984 (when H1 magnet was ON too) will work for high mentioned data samples. For any case these Relation Coefficients set were tested with ET&PD&nV triggered events from both samples. Results are not so good - due to this factor I put "preliminary" word into this VCs calibration. These are the results of high mentioned test from ETrec+PDrec histo LOOK peak finder: Runs 74397 - 74404 all events xetcut vcdepcut xetcut&vcdepcut Step Mean Sigma Mean Sigma Mean Sigma Mean Sigma 0 27.19 1.04 27.21 1.01 27.23 1.03 27.27 0.99 It is possible to see high mentioned results from 100th ntuple which is kept at the next LOOK binary data set: H1KFOM.TLOOK3.R74397 FAST16 3010200E 00000 DASD After ordinary 3 step calibration of VCs channel it were produced the next (preliminary) values of CCvcs for high mentioned data sets: Runs 74397-74402 --> CCvcs = 5.009 Mev/FADCcount Runs 74403-74404 --> CCvcs = 3.574 Mev/FADCcount I shall try to define Relation Coefficients Set for high mentioned data sample and after this to make recalibration of VCs again. 13.05.1994 ===== FIC#1 ======== H1KFOM === Luminosity is ON again... Today evening between 19:00 and 20:00 Luminosity branch was returned into H1CDAQ-enevironment and now (20:13) is operated at Run 75126 together with all another branches. I made nothing. As seems it is not our problem. 13.05.1994 ===== dice2 ======== H1KFOM === Relation Coefficients Set It was produced Relation Coefficients Set with ET&PD&nvVC-events from data sample Run74397-74404 (40129 events): ==> Relation Coeff.Set for ET 1.105 1.121 1.100 1.094 1.180 1.187 1.086 1.071 1.126 1.058 0.798 1.138 0.948 1.022 1.275 1.073 1.113 1.028 1.036 1.008 1.009 1.051 1.046 1.048 1.049 1.077 1.105 1.072 1.136 1.095 0.990 1.089 1.019 1.136 1.012 1.224 1.214 0.944 0.846 1.095 0.953 0.000 1.079 1.080 1.102 0.957 0.961 1.126 0.000 ==> Relation Coeff.Set for PD 0.968 0.830 0.870 1.070 0.942 0.818 1.077 0.972 0.998 0.921 1.050 0.944 0.967 0.965 0.987 0.942 0.978 0.998 0.925 1.008 0.892 0.698 0.833 1.039 0.930 ==> Sig. at Relation Coeff.Set for ET 0.001 0.003 0.001 0.012 0.001 0.002 0.002 0.000 0.012 0.006 0.018 0.014 0.006 0.011 0.011 0.006 0.006 0.007 0.005 0.007 0.011 0.014 0.010 0.011 0.012 0.011 0.068 0.010 0.005 0.005 0.005 0.012 0.007 0.014 0.007 0.003 0.000 0.006 0.002 0.002 0.000 0.000 0.000 0.000 0.001 0.006 0.006 0.001 0.000 ==> Sig. at Relation Coeff.Set for PD 0.002 0.005 0.024 0.001 0.001 0.003 0.008 0.005 0.008 0.003 0.006 0.016 0.012 0.017 0.009 0.003 0.006 0.005 0.004 0.004 0.000 0.001 0.010 0.015 0.000 ==> Ent. at Relation Coeff.Set for ET 7 27 13 22 89 33 25 13 62 93 109 123 106 75 812 427 822 900 913 1481 750 5850 8716 6901 5404 4355 3365 1614 2224 2747 741 785 337 202 155 24 9 19 17 17 7 0 1 3 4 24 4 5 0 ==> Ent. at Relation Coeff.Set for PD 25 29 71 122 8 166 1688 7521 8930 1718 453 3966 13239 14414 3625 54 500 1501 1448 264 12 7 23 18 1 These are the results of high mentioned test from ETrec+PDrec histo LOOK peak finder: Runs 74397 - 74404 all events xetcut vcdepcut xetcut&vcdepcut Step Mean Sigma Mean Sigma Mean Sigma Mean Sigma 3 27.56 0.99 27.61 0.96 27.51 0.97 27.57 0.94 These digits it is possible to see from histos which are producing from 100th ntuple which is kept at the next LOOK binary data set: H1KFOM.TLOOK4.R74397 FAST02 3010200E 00000 DASD It was fixed 164 events from 40129 (0.4%) at sample ET&PD&nVC events with overflow at p11-element (may be due to large Ie total values at this time period). 13.05.1994 ===== dice2 ======== H1KFOM === VCs recalibration...... After producing of new Relation Coefficient Set for data sample from Runs 74397-74404 it were recalibrated calibration coefficients for VCs: Runs 74397-74402 --> CCvcs = 4.457 Mev/FADCcount Runs 74403-74404 --> CCvcs = 3.197 Mev/FADCcount These are the results of tests from ETrec+PDrec histo LOOK peak finder: Runs 74397-74402 (17K events) all events xetcut Step Mean Sigma Mean Sigma 1-3 27.19 1.45 27.24 1.40 Runs 74403-74404 ( 9K events) all events xetcut Step Mean Sigma Mean Sigma 1-3 27.31 1.27 27.40 1.23 It were produced 100th ntuples at test jobs which are kept at the next LOOK-binary data sets:. H1KFOM.VLOOKT.R74397 FAST12 3010200E 00000 DASD H1KFOM.VLOOKT.R74403 FAST14 3010200E 00000 DASD 14.05.1994 ===== dice2 ======== H1KFOM === VCs Calibration Results These are final results of calibration VCs with different HV- conditions and different beam conditions (for all long data sample which were recorded 02/05/94, 03/05/95 and 05/05/94): CCvcs value HVvcs HVvcv Runs Ie total Date MeV/FADCcount V V mA 7.979 700 1000 73981,73984 9.2--> 7.0 02/05/94 6.932 700 1000 74115,74118 5.0--> 4.3 03/05/94 4.457 800 1000 74397-74402 20.0-->15.6 05/05/94 4.314 700 1050 74116 4.8--> 4.4 03/05/94 3.197 800 1050 74403-74404 15.6-->14.1 05/05/94 2.287 700 1100 74117 4.4--> 4.3 03/05/94 14.05.1994 ===== dice2 ======== H1KFOM === H1LumiRuns 100th ntuples.. It were prepared at dice2-environment 6 LOOK-binary data sets with 100th ntuple at each. All events were processed and put into these 100th tuples (s86-as PD1,s93 and s94). These tuples will be used for Photon Arm Energy Spectra preparing for possible definition of Global Energy Shift with S.Levonyan's tool. The structure of 100th ntuple is the same as it was proclaimed at message 25.10.93 at LL7: NAM(1)='ETREC' ET reconstructed energy (GeV) NAM(2)='PDREC' PD-arm reconstructed energy (GeV) NAM(3)='VCDEP' VetoCounter deposited energy (GeV) NAM(4)='XET' Xet - reconstructed coordinate (cm) NAM(5)='YET' Yet - reconstructed coordinate (cm) NAM(6)='XPD' Xpd - reconstructed coordinate (cm) NAM(7)='YPD' Ypd - reconstructed coordinate (cm) NAM(8)='TIME' time at seconds from start of first H1Run at data sample NAM(9)='BUNCH' - bunch number NAM(10)='PD1' - value of trigger element PD1 NAM(11)='PD2' - value of trigger element PD2 NAM(12)='PD' - value of trigger element PD NAM(13)='VC' - value of trigger element VC NAM(14)='ET' - value of trigger element ET NAM(15)='ETAG' - value of trigger element eTAG NAM(17)='S86' - value of trigger bit s86 NAM(18)='S93' - value of trigger bit s93 NAM(19)='S94' - value of trigger bit s94 NAM(20)='S95' - value of trigger bit s95 2-values at s86,s93,s94,s95 elements mean Random Trigger 999.-values at ETrec,PDrec,Xpd,Ypd,Xet,Yet mean bad reconstruction All binary LOOK-data sets were moved into DESY IBM storage and are kept at the next data sets: H1KFOM.ALOOKQ.R73981 FAST13 3010200E 00000 DASD H1KFOM.ALOOKQ.R74115 FAST04 3010200E 00000 DASD H1KFOM.ALOOKQ.R74116 FAST13 3010200E 00000 DASD H1KFOM.ALOOKQ.R74117 FAST07 3010200E 00000 DASD H1KFOM.ALOOKQ.R74397 FAST16 3010200E 00000 DASD H1KFOM.ALOOKQ.R74403 FAST01 3010200E 00000 DASD Three different Relation Coefficients Set and 6 different CCvcs values were used for producing of high mentioned tuples. 14.05.1994 ===== dice2 ======== H1KFOM === new set of PDREC-spectra.. It were produced at LOOK-interactive mode all Photon Arm Spectra early used for Global Energy Shift defining (50-53,60-63,70-73,80-82) bins 70 0 35 do 'hs(50)=nt(100)(pdrec)@(s86.eq.1.and.pdrec.lt.999)' do 'hs(51)=nt(100)(pdrec)@(s86.eq.1.and.pdrec.lt.999.and.pd.eq.1)' do 'hs(52)=nt(100)(pdrec)@(s86.eq.1.and.pdrec.lt.999.and.pd1.eq.1)' do 'hs(53)=nt(100)(pdrec)@(s86.eq.1.and.pdrec.lt.999.and.pd2.eq.1)' do 'hs(60)=nt(100)(pdrec)@(s93.eq.1.and.pdrec.lt.999)' do 'hs(61)=nt(100)(pdrec)@(s93.eq.1.and.pdrec.lt.999.and.pd.eq.1)' do 'hs(62)=nt(100)(pdrec)@(s93.eq.1.and.pdrec.lt.999.and.pd1.eq.1)' do 'hs(63)=nt(100)(pdrec)@(s93.eq.1.and.pdrec.lt.999.and.pd2.eq.1)' do 'hs(70)=nt(100)(pdrec)@(s86.eq.1.and.pdrec.lt.999.and.vcdep.lt.0.2)' do 'hs(71)=nt(100)(pdrec)@(s86.eq.1.and.pdrec.lt.999.and.vcdep.lt.0.5)' do 'hs(72)=nt(100)(pdrec)@(s86.eq.1.and.pdrec.lt.999.and.vcdep.lt.1.0)' do 'hs(73)=nt(100)(pdrec)@(s86.eq.1.and.pdrec.lt.999.and.vcdep.lt.1.5)' do 'hs(80)=nt(100)(pdrec)@(s93.eq.1.and.pdrec.lt.999.and.vcdep.lt.0.2)' do 'hs(81)=nt(100)(pdrec)@(s93.eq.1.and.pdrec.lt.999.and.vcdep.lt.0.5)' do 'hs(82)=nt(100)(pdrec)@(s93.eq.1.and.pdrec.lt.999.and.vcdep.lt.1.5)' All these spectra are kept at next LOOK-binary data sets at DESY IBM: H1KFOM.BLOOKQ.R73981 FAST08 3010200E 00000 DASD H1KFOM.BLOOKQ.R74115 FAST14 3010200E 00000 DASD H1KFOM.BLOOKQ.R74116 FAST08 3010200E 00000 DASD H1KFOM.BLOOKQ.R74117 FAST01 3010200E 00000 DASD H1KFOM.BLOOKQ.R74397 FAST09 3010200E 00000 DASD H1KFOM.BLOOKQ.R74403 FAST16 3010200E 00000 DASD Asking to L.N.Shtarkov to make first estimations of Global Energy Shifts for high mentioned 6 data samples. Note that I used for calibration Ee=27.515 GeV value as e-beam energy 15.05.1994 ==================== H1KFOM === HERA p-beam .............. Tonignt stable HERA p-beam existed at p-ring all night with Ip total near 21mA. Near 08:00 today morning this beam was dumped. Both p-ring and e-ring characteristics had appeared at HERA TV-screen. Now (11:30) 150 mkA 40 GeV HERA p-beam is existing at HERA p-ring. 15.05.1994 16.05H01LNS ===== PD-spectra-fits R73981 - R74403 (6 runs) According to A.M.F. of 14.05.94, PDFIT's have been done for 6 Runs. The starting E0 == 27.515 (instead of 26.7). Nothing more was changed in the program, comparing to the past, but: the printed line was extended for more then 80 positions. To see these extendings - use PF0 at FALCO, ect. Some figures in this extension are "duplicate" of the other figures, just for the sake of visual convenience, e.g. (1.xy-1.) -> xy%, etc. B.T.W., the lines "l_text ..." are taken exactly from A.F.'s(!). I think, e.g., (d(Emax) +/- error) is more convenient than others. The last figure in a line is a duplicate of probability in %%. The probability%% is the starting figure at any analysis, and it is naturally that many lines from the start display prob.< 1%. These lines should be IGNORED, they are printed for understanding. To look at the Tables of Fit-Results, as usually, STAR(*) the line: LIST 'H01LNS.JPDFIT.R73981' LIST 'H01LNS.JPDFIT.R74115' LIST 'H01LNS.JPDFIT.R74116' LIST 'H01LNS.JPDFIT.R74117' LIST 'H01LNS.JPDFIT.R74397' LIST 'H01LNS.JPDFIT.R74403' Some selection was done 16.05.94: The starting parts of fits, having probability < 1%, were deleted, and some blank lines were inserted for better reading. STAR(*) any of the next lines if you prefer such a selection: LIST 'H01LNS.JPDFIT.R73981.SEL1' LIST 'H01LNS.JPDFIT.R74115.SEL1' LIST 'H01LNS.JPDFIT.R74116.SEL1' LIST 'H01LNS.JPDFIT.R74117.SEL1' LIST 'H01LNS.JPDFIT.R74397.SEL1' LIST 'H01LNS.JPDFIT.R74403.SEL1' Some more important selection was done later 16.05.94: The "BEST" fits were chousen by some not well defined Ruls of Game, the corresponding lines were marked "BEST" at the place of "fitOK". All the other lines were not deleted for comparison ( == in .SEL1). STAR(*) any of the next lines if you prefer such a selection, then select all the "BEST" via a command "find BEST", or elswhere: LIST 'H01LNS.JPDFIT.R73981.SEL2' LIST 'H01LNS.JPDFIT.R74115.SEL2' LIST 'H01LNS.JPDFIT.R74116.SEL2' LIST 'H01LNS.JPDFIT.R74117.SEL2' LIST 'H01LNS.JPDFIT.R74397.SEL2' LIST 'H01LNS.JPDFIT.R74403.SEL2' Use "HREC" to retrieve the file if it has gone migrated. Use "find" to select the most interesting, e.g. select eG/ep, etc. I'd like to show the results as quick as possible, without conclusions, but the only one thing, the most important: MOST OF THE FIT-VARIANTS SHOW Emax < 27.515 , USUALLY BY THE VALUE ~ ((1.-3.) +/- 0.1) GEV. IT IS NOT A LARGE, BUT A SIGNIFICANT(!) VALUE. As an amateur detectiv, I suspect as possible criminals - the value "pdrec", and especcially, the contribution of VCdep energy. I suspect, even, that the SHAPEs of the used PD-spectra are not well. In the results there are some strange things that are looking as of a minor importance, but WHO knows (?). I'd like to go on ... L.Shtarkov, Moscow, LPI. 16.05.1994 ===== dice2 ======== H1KFOM === CCvcs tuning.............. Today it was made attempt of CCvcsp tuning with S.Levonian's proc- laimed procedure at LL7 (07.06.93) due to there are many ET*PD*VC- events at most of samples. The next results were obtained: CC(VC) = CC(Fom.14/05)*1.0683 for the run 73981 CC(VC) = CC(Fom.14/05)*1.0905 for the run 74115 CC(VC) = CC(Fom.14/05)*1.1108 for the run 74116 CC(VC) = CC(Fom.14/05)*1.4113 for the run 74117 (!!!!) CC(VC) = CC(Fom.14/05)*1.1180 for the run 74397 CC(VC) = CC(Fom.14/05)*1.0500 for the run 74403 CCvcs value HVvcs HVvcv Runs Ie total Date MeV/FADCcount V V mA 8.535 700 1000 73981,73984 9.2--> 7.0 02/05/94 7.559 700 1000 74115,74118 5.0--> 4.3 03/05/94 6.088 700 1050 74116 4.8--> 4.4 03/05/94 4.951 800 1000 74397-74402 20.0-->15.6 05/05/94 3.574 800 1050 74403-74404 15.6-->14.1 05/05/94 2.401 700 1100 74117 4.4--> 4.3 03/05/94 Tuning was made with ET*PD*VC=1 events from 100th ntuples of high mentioned ALOOKQ-data set for high mentioned 6 data samples. May be we can repeat procedure of Global Energy Shifts defining with new CCvcs values? 17.05.1994 ==================== H1KFOM === VCs PHEU÷r Replace.. Yesterday Petr and Yura replaced PHEU and divider at VCs-channel. It was made during yesterday plan access to HERA-tunnel. Attempt to change HV-cabel only did not put any success. As seems problem of mentioned from 12.04.94 trips at VCs-channel was connected with PHEU. HV status at VC for today: HVvcs = 900 V HVvcv = 930 V As seems we are at start point again with finding of optimal HV at Veto Counter PHEUs. 17.05.1994 ==================== H1KFOM === Waiting for ep-collisions. Yesterday H1 Run Coordinator asked H1Lumi-team to be ready for possible attempts of ep-collisions 17.05.94. 17.05.1994 ==================== H1KFOM === HERA p-beam .............. Today near 11:15 it was injected powerful HERA p-beam with Ip total near 25mA. After ramping (which was finished near 11:40) Ip=20 mA Ep=820 GeV. At 13:45 p-beam was lost. 17.05.1994 ==================== H1KFOM === 100th ntuples Reprocessing It were reprocessed 100th ntuples for H1Lumi Runs (02-05/05/94) with new values of CCvcs which were proclaimed at yesterday message. Its are kept at the next LOOK-binary data sets: H1KFOM.ALOOKQ1.R73981 FAST12 3010200E 00000 DASD H1KFOM.ALOOKQ1.R74115 FAST10 3010200E 00000 DASD H1KFOM.ALOOKQ1.R74116 FAST07 3010200E 00000 DASD H1KFOM.ALOOKQ1.R74117 FAST10 3010200E 00000 DASD H1KFOM.ALOOKQ1.R74397 FAST10 3010200E 00000 DASD H1KFOM.ALOOKQ1.R74403 FAST01 3010200E 00000 DASD New Photon Arm Energy spectra (50-53,60-63,70-73 and 80-83) are kept at the next LOOK binary data sets: H1KFOM.BLOOKQ1.R73981 FAST05 3010200E 00000 DASD H1KFOM.BLOOKQ1.R74115 FAST13 3010200E 00000 DASD H1KFOM.BLOOKQ1.R74116 FAST10 3010200E 00000 DASD H1KFOM.BLOOKQ1.R74117 FAST02 3010200E 00000 DASD H1KFOM.BLOOKQ1.R74397 FAST01 3010200E 00000 DASD H1KFOM.BLOOKQ1.R74403 FAST18 3010200E 00000 DASD 18.05.1994 ==================== H1KFOM === HERA p-beam & e-beams..... Yesterday at 18:40 it was started HERA p-beam injection. Injection was finished at 20:30. Ip total was fixed at this time near 22.7 mA. Number of bunches - 170. Bunch structure of p-beam - (10+1)*6 + (0+11) + (10+1)*6 + (0+5) + (10+1)*5. Ramping of p-ring was started at 20:30 and at 21:00 was finished with Ep=819.902 GeV and Ip=21.56 mA. HERA e-beam activity started near 22:00 yesterday. During night it was fixed 5 e-beam injections: 00:25 - 01:00 Ie total max near 5 mA 01:15 - 01:40 Ie total max near 5 mA 01:50 - 02:40 Ie total max near 5 mA 03:20 - 03:40 Ie total max near 5 mA 04:40 - 05:20 Ie total max near 6 mA P-beam was lost at 05:15 from Ip value near 18 mA. At our OnLine Rates History pictures only two last e-fillings were fixed. It were not fixed any Luminosity at last two e-fillngs - Total ET&PD-rate and BG ET&PD-rate were equal. At both last e-fillings rampings were made. The values of rates (after rampings): VC - 200 kHz; ET&PD - 7.5 kHz VC - 220 kHz; ET&PD - 9 kHz Thresholds at Egor's card were installed as ordinary for 27.53 GeV - PD - 1GeV, ET - 5.5 GeV, VCv - 1.75 GeV, VCs - 1.0 GeV 18.05.1994 ==================== H1KFOM === Possible LRTL-bank Upgrade Today it was talking with F.Sefkow about his and A.Campbell's propo- sal to include TSTC-bank from LUMI-branch into KEEP-banks when LRTL-bank is at current H1-events. Analyze of H1LumiOnLineFIC1-program shows that it is not so simply and more simpliest way is to add into LRTL-bank the same values which are kept at 3-6 words of current TSTC-bank: 3 nXing2 I 0 255 ! all beam crossings (bits 32-39) 4 nXing0 I -INF +INF ! all beam crossings (bits 0-31) 5 nXAct2 | 0 255 ! beam Xings during L1Atv (bits 32-39) 6 nXAct0 I -INF +INF ! beam Xings during L1Atv (bits 0-31) As seems it is needed to make long time ago promised LRTL-upgrade with adding of word with HERA Produced Luminosity value (from start of Current ep-collisions). F.Sefkow promised to discuss this question with E.Elsen and let me know if this decision will be accepted. 18.05.1994 ==== H01LNS == NEW PD-spectra-fits R73981 - R74403 (6 runs) According to A.M.F. of 17.05.94, NEW PD-spectra were fitted, 6 Runs, the same runs as were done before, LLO 15.05.94. The starting E0 == 27.515 (instead of 26.7). Some other comments and advise see in that LLO. ALL the fits having probability < 1%, were not printed. The BEST fits should be in the FIRST line for the file ("50" etc.) ALL the 6 runs are saved in ONE file, If you would like to see this file, STAR(*) the next line: LIST 'H01LNS.JPDFIT.R73981.R74403.BQ1 Unfortunately, the N of RUNs were not printed, you should know the sequence of runs: R73981, R74116(!), R74115(!), R74117, R74397, R74403 Use "find" to select the most interesting variants, etc. some comments see upper at LLO 15.05.94. L.Shtarkov, Moscow, LPI. 18.05.1994 ===== FIC#1,2 ====== H1KFOM === H1LumiOnLine Releases.. It were prepared new releases of H1LumiOnLinePrograms for FIC#1 and FIC#2. H1LumiOnLineFIC#2 was changed only for preparing of HERA Produced Luminosity Value (from last zeroing of Integral Luminosity Values) for LRTL-bank. The next changings were made at H1LumiOnLineFIC1: 1) it was changed FADC Shift for Neutron Counters FADCs from 0 to 143 (as it was recommended by Egor before his leaving of HH); 2) it was changed shift for readout of RomanPot's GPTP (added 7 BC to old value 9) 9-->16 - as it was recommended by Egor; 3) it was removed bug for posibility of writing LREF,LRPF-banks (mode 1); 4) it was changed LRTL-bank format - it were added 5 additional words 235 HERALUMI I 0 +INF ! HERA Produced Luminosity from the ! beginning of the current ep-collisions ! (in units 10**25 cm-2 = 0.01 mb-1) 236 nXing2 I 0 255 ! all beam crossings (bits 32-39) 237 nXing0 I -INF +INF ! all beam crossings (bits 0-31) 238 nXAct2 | 0 255 ! beam Xings during L1Atv (bits 32-39) 239 nXAct0 I -INF +INF ! beam Xings during L1Atv (bits 0-31) Releases were booted into both FICs after Run75760. Date and time of creation: H1LumiOnLineFIC1 - 18/05/94 15:23 H1LumiOnLineFIC2 - 18/05/94 15:00. 18.05.1994 ==================== H1KFOM === HERA p-beam .............. Today afternoon it was fixed powerful HERA p-beam with 180 bunches. Injection was started near 13:30 and at 14:21 Ip total was 30.23 mA. During ramping at 14:30 it was lost the half of Ip current and after attempt to ramping continuation p-beam was lost. Bunch structure: (10+1)*6 + (0+14) + (10+1)*6 + (0+5) + (10+1)*6. At 18:08 it was started injection of HERA p-beam again. At 18:26 injection was finished with Ip=27.68 mA and 180 bunches. Ramping was started at 18:27. Ramping was finished at 19:00 with Ip = 25.61 mA and Ep = 819.920 GeV. Bunch structure: (10+1)*6 + (0+11) + (10+1)*6 + (0+5) + (10+1)*6. 19.05.1994 ==================== H1KFOM === HERA p-beam & e-beam...... Yesterday after making of good p-beam at 19:00 HERA e-beam injection was started at 19:31 and was finished at 20:08 with Ie=4.953 mA and with 30 bunches. Bunch structure of e-beam - (9+2)*3 + 3 e-pilot bun- ches (67,68,69th bunches). At 20:10 e-ring ramping was started and was finished at 20:12 with Ie=4.350 mA and Ee=27.520 GeV. After ramping VCrate=112 kHz, ET&PD rate 6-7 kHz. After some manipulation of HERA-people (may be Synchrotron Radiation Masks were moved) at 20:24 VCrate =36 kHz and ET&PD rate 6 kHz with Ie=3.150mA. After end of e-ring ramping Movable Platforms were moved into Working Positions. Luminosity subbranch was good calibrated (I looked for 0.88 GeV energy resolution at OnLine ETrec+PDrec histo for ET*PD*nVC events). I asked H1CDAQ for calibration Run. But at 20:47 p-beam was lost. Nevetheless it were written some Runs with e-beam only (up to 21:06 when e-beam was dumped). But it was seen at on-line that something happened with synchronisation of e-beam at this period (bunch statistics was "rotated" and energy resolution became bad). So - for any case - these Runs must be labeled as not-good for calibration (but really it is possible to make some analysis of these data). First 3 Runs were written with temporary mode logging due to prescale factors tuning. Run 75819 was written on permanent cartridge and is labeled as H1Lumi Calibration Run with Ee=27.52 GeV. RUN 75816 EVS 62 ( 0-4FFFFFFF) XI 202 B= 0 H1RAWD.C9400786 RUN 75817 EVS 11974 ( 0- 11906) XI 202 B= 0 H1RAWD.C9400786 RUN 75818 EVS 1581 ( 0- 1538) XI 202 B= 0 H1RAWD.C9400786 RUN 75819 EVS 39613 ( 0-4FFFFFFF) XI 202 B= 0 H1RAWD.C9400786 Runs 75816-75819 were written from 20:49 up to 21:06. As at all previous H1Lumi Calibration Runs it were involved 3 H1Lumi subtriggers (s86 as PD1,s93 and s94). Thresholds values (see LUTR-banks contents): VCv=2 GeV, VCs=1 GeV, PD=1GeV, PD1=0.75 GeV, ET = 5.5 GeV. HVvcv =960V HVvcs = 650V. 19.05.1994 ==================== H1KFOM === HERA ep-beams&H1LumiRuns Today morning from 08:00 up to 09:15 it was prepared new HERA p-beam with Ip=20mA after injection and with Ip=17 mA after end of p-ring ramping. Number of bunches - 170. Bunch structure is the next: (10+1)*6 + (0+11) + (10+1)*6 + (0+5) + (10+1)*5. At 10:45 it was started HERA e-beam activity. First e-beam injection was finished near 11:00 with Ie = 3 mA and bunch structure (3+8)*12. It means that 3 e-pilot bunches and 137 p-pilot bunches existed. The same as yesterday evening behaiviour of H1Lumi system was fixed. Ratio of RateVC/RateET&PD was near 7. Luminosity subbranch was good calibrated (best energy resolution was near 0.81 GeV at OnLine energy spectra ETrec+PDrec foor ET&PD&nVC-events. We asked access to H1CDAQ but at this period (near 13:00) Calo investigation was made with local H1CTRIG timing-this local timing was the reason of changing at our monitoring facility - another behaivour of TotalET*PD and BkgEstimatedET*PD rates, bad calibration --> bad energy resolution etc. HERA-people phoned to H1 Run Coordinatior and syncronisation was returned on HERA p-beam but at this period e-beam was lost. At 15:10 it was started 2nd e-filling with the same p-beam. It were injected 42 bunches with Ie=5.4 mA. Bunch structure is (3+8)*14. There are 36 collided bunches, 6 e-pilot bunches and 134 p-pilot bunches. At 15:21 e-ring ramping was finished with Ee=27.52 GeV and Ie = 2.3 mA. At 15:23 it were fixed RateVC=26 kHz and RateET&PD=4 kHz. Luminosity subbranch was good calibrated again (Sigma = 0.85 GeV at OnLine Energy Spectra) and we asked H1CDAQ for H1LumiCalibration Run recording. It was written Run 75934 with near 150K events with the same subtriigers set as at all previous H1Lumi Calibration Runs (s86 as PD1, s93 and s94). H1 Magnet was Off at this time. Run 75934 was written from 16:09 up to 16:36 today. Thresholds were the same as at yesterday evening Run (see LUTR-bank contents). HVvcv=950V, HVvcs=670V (a little another as yesterday). At 18:05 e-beam was dumped. Movable Platforms were moved into Parking Positions. Ip = 16.68 and it is possible to wait for the next e-filling. H1 Main Magnet is Off. During last e-filling it were collected next Integral Luminosity values (due to some statistical and may be systematical errors): HERA Produced - 1.24*10**31 cm-2 H1 Run Gated - 0.61*10**31 cm-2 H1 Data Taken - 0.17*10**31 cm-2 These digits were collected from 15:20 up to 18:05 today. 19.05.1994 ==================== H1KFOM === Neutron Counters Run...... Between 1st and 2nd today e-filling on the request of R.Felst it was made attempt to look for behaivour of neutron counters with Ep=820 GeV and Ip=16.86mA (170 bunches). Yura made OnLine histos of Neutron Counters Responces and after this (after switching of 11th FADC from local synchronization on H1CTRIG syncronisation) it was written Run 75918 with only s86 as NC - trigger Run was written from 15:07 up to 15:15 today. It were recorded near 50K events. 11th FADC shift was installed on 143 value. It is needed to check - is it right or not. As OnLine Monitoring shows - there are no large changings at FADCshifts. As seems all channels (exept 2) have overflow situations. It is needed to check at off-line - how many overflows. Rate of NC-trigger element was fixed near 3.6-4.0kHz with thresh.1.25 19.05.1994 ===== FIC#2 ======== H1KFOM === H1LumiOnLine Release... Today it was prepared H1LumiOnLineFIC2-program new release. It was changed only scale of HERA Produced Luminosity Value at LRTL-bank. If we use 10**25 scale - maximal value of HERA produced luminosity will be near 20 nb-1 - it was good for H1DataTaking'93. Now we must be ready to fix more and more digits. S.Levonyan recommended to use 10**27 scale for high mentioned value. So .. 235 HERALUMI I 0 +INF ! HERA Produced Luminosity from the ! beginning of the current ep-collisions ! (in units 10**27 cm-2 = 1.0 mb-1) Date and time of release creation: 19 May 1994 12:57. It was booted into FIC#2 after Run 75961. 20.05.1994 ==================== H1KFOM === HERA p-beam & e-beam...... Today early morning injection of HERA p-beam was started(from 05:45) and at 06:00 Ip total was near 9.5 mA. After ramping Ip=5 mA with 60 bunches and bunch structure (10+1)*6. HERA e-beam activity startred at 11:45. HERA e-beam injection finished near 12:00 with Ie=3.7 mA and with 30 bunches. Bunch structure of e-beam is the next: (9+2)* 3 and 66,67,68th as e-pilots bunches. There are 27 - collided, 3 e-pilot and 33 p-pilot bunches. Movable platforms were moved into Working positions near 13:00. It were fixed the next rates at 13:00: RateVC=10kHz, RateETPD=1kHz. Up to now(17:00) it was not fixed any Luminosity. We had attempt to write H1Lumi Calibration Run but e-beam conditions after start of this Run were changed (RateVC became 120-140 kHz, sometimes 200 kHz, PD-rate was fixed near 20-30 kHz and ET rate - near 3 kHz, RateETPD was near 700-800 kHz but almost disappeared ET&PD&nVC-triggers (may be due to exetremly high rate of VC for these values of Ie total). High mentioned Run is 76024 which was started at 15:45 and was finished at 16:22 (permanent logging). 20.05.1994 ==================== H1KFOM === First Luminosity'94...... 20 May 1994 at 17:01 it was fixed first Luminosity'94 at H1 IP. Ie=2.15 mA, Ip=5.04 mA, Rate ET*PD total = 3.1 kHz, Rate ET*PD(bg)= 1.3 kHz. Current Luminosity value 6.5*10**28 cm-2s-1. Specific Lumi - 3.3*10**29 cm-2s-1mA-2. After start of Luminosity at H1 IP it was started H1Lumi Calibration Run 76027 (start of writting at 17:03) with the same subtriggers as early (s86 as PD1,s93 and s94). Run ended at 17:43 with 183K events. Thresholds at this Run were changed: PD=0.75, ET=5.75. Thresholds were changed after 16:22. H1 Main Magnet is ON. ep-collisions was finished at 19:15 due to p-beam was lost. At 19:20 e-beam was dumped. During last ep-collisions it were collected next Integral Luminosity values: HERA Produced - 0.416 nb-1 H1 Run Gated - 0.291 nb-1 H1 Data Taken - 0.104 nb-1 These digits were collected from 17:01 up to 19:15 today. 21.05.1994 ==================== H1KFOM === First Luminosity Run'94... HERA p-beam activity this night had started near 02:00. It was injected p-beam with Ip near 20 mA (near 04:00) but at 04:30 p-beam was lost. The next p-beam injection was started at 05:00 and near 06:00 injection was finished with Ip more then 20 mA. Significant part of p-beam was lost during p-ring ramping but at 06:20 it was finished p-ring ramping with Ep=819.920 GeV and Ip=7.4 mA. Bunch structure of p-beam is the next: (10+1)*6 + (0+11) + (10+1)*6 + (0+5) + (10+1)*5. HERA e-beam activity was started at 09:30 and at 10:35 e-beam injection was finished with Ie = 13 mA. It was made e-beam ramping very quickly with Ee=27.520 GeV and Ie =12 GeV. Bunch structure of e-beam is the next: (9+2)*13+(3+2)+(9+2)*4+(3+3)+(3+8)+(3+8). So we have - 150 collided bunches, 15 e-pilot bunches and 20 p-pilot bunches at this e-filling. At 10:37 Movable pltaforms were moved into Working Positions. At 10:40 it were fixed Second Luminosity'94 at H1 Interaction Point. At 11:00 it was finished Luminosity tunning at H1 IP. The max value of Current Luminosity was fixed at this period as 1.48*10**29 cm-2s-1 with Specific Luminosity near 3.4*10**29 cm-2s-1 mA-2. Rates at this period: RateVC=70kHz, RateETPD=10.5kHz, RateETPD(bg)= = 6kHz. At 11:20 firstly at 1994 it was proclaimed "Luminosity Run" at HERA TV screen. H1 Shift Crew is preparing for making of actions with first Luminosity Run at H1 area (there is some plan). I asked to write H1LumiCalibrationRun with high mentioned conditions. (Ie current is enough large value - more then yesterday on factor 4-5). This asking was accepted by H1 Run Coordinator. We got 20 min per H1LumiCalibration run with high mentioned beam conditions. Run 76137 was started at 11:24 and was finished at 14:48 due to Lumi' FIC#1 was crashed. The next Run 76138 was started with the same conditions. At 11:52 Run 76138 was stopped for preparing of another tests and actions with H1CDAQ and H1 subdetectors. Thresholds: PD=0.75, PD1=0.5, ET=5.5 VCv=1.75 VCs=1.0 Subtriggers: s86(as eTAG), s93 and s94. HVvcv=970V, HVvcs=670V. At 15:12 HERA-people proclaimed EndOfLuminosityRun and Movable Platforms were moved into Parking Positions. At 15:20 both HERA beams were dumped. During last ep-collisions it were collected next Integral Luminosity values: HERA Produced - 1.089 nb-1 H1 Run Gated - 0.482 nb-1 H1 Data Taken - 0.180 nb-1 These digits were collected from 11:20 up to 15:12 today. Luminosity branch was Off from H1CDAQ. 21.05.1994 ===== FIC#1 ======== H1KFOM === Problems ................. Today situation with building of RunStartRecord from Luminosity branch became critical. It was fixed stable situation of non-coexisting Luminosity branch with 10th and 11th branches at H1CDAQ during RunStart Record building. When 10th, 11th H1CDAQ branches or both from them are involved into H1CDAQ operation it was fixed stable reaction of Bill's VMExi subrouti- ne MEBRecReq - not OK as ordinary but "RunStartRecord not required". Reboot of FIC#1 program did not help (program really does not crash) - if 10th and 11th are Off from H1CDAQ - all is OK - no this problems. Today during proclaimed "Luminosity Run" Luminosity branch was Off from H1CDAQ due to high mentioned reason. Yesterday evening I wrote mail to Bill with question about his diagnose of high mentioned situation. Let's wait his answer. Yesterday evening and today (during running with H1LumiCalibration Run) program at FIC#1 was crashed. This is another problem connected (as seems) with L2Keep-rates. It was seen earlier that with some L2Keep-rates (with large rates ordinary and may be with some 'resonanse' rates) program at FIC#1 is crashed. As seems we achieved of critical line at rapidity of FIC#1. For H1Lumi's FIC#1 maximal L2Keep rate is near 100-110 Hz (with L3Reject absence). Problem under study. At 15:12 when H1 Program for Luminosity Run was finished I put Luminosity branch into H1CDAQ together with all branches - all was OK. It was not fixed any problems with RunStartRecord building from Lumi- branch. At 17:30 after different investigations of Luminosity branch beahaivour at H1CDAQ-enevironment with different L2Keep-rates and full set of H1 branches it was got stable situation without any problems with coexistence Luminosity branch with another branches. Once it was fixed problem with absence RunStartRecordPrepare-Acknolege from MWPC and Lumi-branches. It was new situation - our monitoring tool shows that there are no RunStartRecord interrupts from H1CTRIG. After reboot of H1CTRIG's FIC - all became OK. 21.05.1994 ===== FIC#1 ======== H1KFOM === Temporary Decision........ As seems it was found some decision which can help at problem with RunStartRecordBuilding from Luminosity branch (may be temporary). Today it was fixed twice - after reboot of H1CTRIG's FIC - problem of incompatibility (incoexistence) of Luminosity branch with another ones at H1CDAQ had disappeared.