| ID |
Date |
Author |
Category |
Subject |
|
67
|
Thu Feb 6 00:45:25 2020 |
Dinko Ferencek | Cold box tests | Lost at least one Peltier in the coldbox |
During an overnight (Feb. 4-5) FullQualification run with modules 1545, 1547, 1548, and 1549 the coldbox lost the ability to maintain the temperature at -20 C. As can be seen from the attached temperature log, the problems already started during the temperature cycles where the last 2 cycles were noticeably longer than the first 3 and finally during the second Fulltests at -20 C the temperature started rising and stabilized around -10 C. Based on these observations it looked like at least one Peltier stopped working.
Silvan took apart the coldbox and will install new Peltiers. Based on measurements Andrey did with the old Peltiers it indeed looks like one Peltier stopped working but the remaining 3 also appear to be at different stages of degradation. |
| Attachment 1: TemperatureCycle.pdf
|
|
|
66
|
Wed Feb 5 16:18:30 2020 |
Dinko Ferencek | Module transfer | Module location tracking added to the assembly spreadsheet |
| To keep track of the module location, two columns were added to the module assembly spreadsheet https://docs.google.com/spreadsheets/d/12m2ESCLPH5AyWuuYV4KV1c5O_NEy7uP5PQKzVsdr3wQ/edit?usp=sharing that contain the dates when a given module is shipped to ETHZ and when it is returned to PSI. |
|
65
|
Wed Feb 5 16:07:03 2020 |
Dinko Ferencek | Module assembly | Module assembly spreadsheet migrated to Google Sheets |
Excel file stored on CERNBox containing the module assembly spreadsheet has been replaced with a Google Sheet https://docs.google.com/spreadsheets/d/12m2ESCLPH5AyWuuYV4KV1c5O_NEy7uP5PQKzVsdr3wQ/edit?usp=sharing. This change should make common editing of the spreadsheet easier.
The new spreadsheet has also been linked from the main L1 replacement web page http://cms.web.psi.ch/L1Replacement/ |
|
64
|
Tue Feb 4 19:52:22 2020 |
Dinko Ferencek | Cold box tests | Blue coldbox setup for reception tests commissioned |
Today I continued where Matej left off with commissioning the blue coldbox setup for reception tests. All the necessary software (pXar and elComandante) Matej already installed (just had to add the BB2 section in testParameters.dat). What was a problem in communication with Keithley (Model 2400) which was fixed after changing the BAUD rate in elComandante/keithleyClient/keithleyInterface.py to 19200 (value set in Keithley's communications settings) and changing FLOW-CRTL in Keithley's communication settings to XON-XOFF.
The reception test was successfully run for modules 1545, 1547, 1548, 1549, and 1550 after cap gluing. |
|
63
|
Tue Feb 4 19:51:10 2020 |
Dinko Ferencek | Module assembly | Cap gluing |
| Today protective caps have been glued to modules 1545, 1547, 1548, 1549, and 1550. |
|
62
|
Tue Feb 4 19:35:38 2020 |
Dinko Ferencek | Module transfer | 7 modules prepared for transfer to ETH: 1536, 1537, 1538, 1540, 1541, 1542, 1543 |
The following modules have been prepared today for transport to ETH for x-ray qualification:
1536, 1537, 1538, 1540, 1541, 1542, 1543
The FullQualification tar files for these modules have been copied to the common CERNBox folder shared with ETH. |
|
61
|
Mon Feb 3 17:09:36 2020 |
Andrey Starodumov | Reception test | 2 modules RT failed: 1544, 1546 |
Feb 03
- RT failed
-- 1544:
--- all ROCs are programmable
--- ROC14 no reliable Threshold found, ROC15 no threshold at all
-- 1546
--- all ROCs are programmable
--- no threshold found for all ROCs |
|
60
|
Mon Feb 3 15:53:32 2020 |
Andrey Starodumov | Reception test | 5 modules RT: 1545, 1547, 1548, 1549, 1550 |
Feb 03
- RT is done and OK for all modules. All 5 modules graded A. |
|
59
|
Mon Feb 3 15:35:09 2020 |
Anrey Starodumov | Full test | 4 modules FT: 1539, 1541,1542, 1543 |
Jan 29
- Reception test for 1539 OK
- Cap gluing
Jan 30
- Reception test for 1541, 1542, 1543 OK
- Cap gluing
- FT:
-- 1539: C/B/B. First -20 is C due to INCOMPLETE TEST: DTB stuck
==> Must be retested at -20C and then files merging
-- 1541: C/C/C. Electrical Test is C due to Trimbit test failure in ROC13. Threshold is reasonable (see attachment)
==> Manually regraded to B
-- 1542: C/C/C. Electrical Test is C due to RelGain of 0.200 instead of <0.1 (too wide Gain distribution).
-- 1543: B/B/B (-20/-20/+10). Electrical Test is B due to mean noise >200e. |
| Attachment 1: 1542ROC13-TrThr.pdf
|
|
|
58
|
Mon Feb 3 11:28:38 2020 |
Anrey Starodumov | Full test | 4 modules FT: 1536, 1537,1538, 1540 |
Jan 28: Reception test for 1536, 1537, 1538 OK
Then cap gluing
Jan 29
- Reception test for 1540 OK
- FT:
-- 1536: B/B/B (-20/-20/+10). Electrical Test is B due to mean noise >200e. IV: 3umA for -20 and +10, slope >300
-- 1537: C/B/B. First -20 is C due ROC10 >200 TrimBits failure, noise >200e. IV: A
-- Trimbit test fails but Trimbit distribution is OK and Threshold reasonable (attachment)
==> Manually regraded to B
-- 1538: B/B/B (-20/-20/+10). Electrical Test is B due to mean noise >200e. IV: <0.2umA for +10, slope ~1
-- 1540: B/B/B (-20/-20/+10). Electrical Test is B due to mean noise >200e. IV: <0.8umA for +10, slope ~2.7 |
| Attachment 1: 1537ROC10m20-TrThr.pdf
|
|
|
57
|
Thu Jan 30 13:51:03 2020 |
Matej Roguljic | Cold box tests | Fulltest failure - psiAgente |
During the first fulltest@-20 of the full qualification procedure for modules: 1539, 1541, 1542 and 1543; an error message popped up "psi Agente: self.Testboards[0] failed - the following boards are busy: TB1(DTB...:M1541), TB2(DTB...:M1542), TB3(DTB...:M1543). This means that the first fulltest for module 1539 was not executed.
Second fulltest seems to be working fine (so far)
Cause of this problem is not clear. |
|
56
|
Mon Jan 27 17:50:48 2020 |
Matej Roguljic | Cold box tests | Failing modules at -20 |
| Friday, 24.1., it was noticed that the modules are graded as 'C' in similar fashion. Module is graded A/B at +10 degrees and then graded C at -20. The reason for grade C was that the trimming procedure was not successful on at least one of the ROCs on each module. This problem was further investigated on 27.1. Wolfram suspected that the problem lies in zero suppression mode since the pixels with failed threshold comes in groups of four pixels. The corresponding DAC 'ctrlreg' was set, as recommended, to 17, during previous tests and after changing it to 9, the problems at -20 seem to be gone. This will be further investigated. |
|
55
|
Fri Jan 24 18:22:00 2020 |
Dinko Ferencek | General | L1 replacement web page updated |
This week several new links were added to the main L1 replacement web page located at the following URL
http://cms.web.psi.ch/L1Replacement/
The links point to other pages or information related to the replacement module qualification. |
|
54
|
Fri Jan 24 18:16:09 2020 |
Dinko Ferencek | Module assembly | Protective cap gluing |
| Today I additionally practiced protective cap gluing by adding a protective cap to module M1536. |
|
53
|
Fri Jan 24 18:12:00 2020 |
Dinko Ferencek | Software | MoReWeb code updates |
MoReWeb code updated to include the ROC wafer info in the XrayCalibration and XRayHRQualification pages:
https://gitlab.cern.ch/CMS-IRB/MoReWeb/commit/4de8ea39050600367ae0b8e959fcc00f29be45d8
In addition, two BB2 plots with wrong axis labels were fixed:
https://gitlab.cern.ch/CMS-IRB/MoReWeb/commit/3cfa18fc3aab1a29859ba9f0f81aa0ed4c59d5c8
https://gitlab.cern.ch/CMS-IRB/MoReWeb/commit/bb6627e6ab3f4d08eb54970a7e3d3b751983e15a |
|
52
|
Thu Jan 23 15:15:45 2020 |
Dinko Ferencek | Software | Trimming Vcal value changed from 35 to 40 |
Trimming Vcal value changed from 35 to 40 in the testParameters.dat files stored in
/home/l_tester/L1_SW/pxar/data/tbm10d_procv3/
/home/l_tester/L1_SW/pxar/data/tbm10d_procv3_test/
/home/l_tester/L1_SW/pxar/data/tbm10d_procv4/
/home/l_tester/L1_SW/pxar/data/tbm10d_procv4_test/
This is the value we will use for the FullQualification. |
|
51
|
Thu Jan 23 14:54:45 2020 |
Dinko Ferencek | Cold box tests | Keithley exchanged |
Yesterday during an attempt to run the FullQualification, elComandante would lose control over Keithley after performing the IV curve measurements and would go in the subsequent Fulltest with HV off. This happened both between the two sets of tests at -20 C and between the tests at -20 C and +10 C. Today we exchanged the old Keithley with a new one and we observe that the communication with this new Keithley is much smoother without any error codes and warning sounds produced by Keithley.
In order to properly communicate with the new Keithley, it was necessary to set its communication mode to RS-232 and BAUD rate to 57600 (https://youtu.be/-5RmguqC7xA). Everything else was left unchanged:
BITS = 8
PARITY = NONE
TERMINATOR = <CR+LF>
FLOW-CRTL = XON-XOFF |
|
50
|
Wed Jan 22 11:50:20 2020 |
Dinko Ferencek | Software | Module configuration files for interactive tests |
There are two module configuration folders set up for elComandante
/home/l_tester/L1_SW/pxar/data/tbm10d_procv3/
/home/l_tester/L1_SW/pxar/data/tbm10d_procv4/
These folder can also be used for interactive tests with pXar. However, in that case these folders get filled with pxar.log and pxar.root files. For some reason, elComandante when copying the module configuration files also copies all pxar.log files. To prevent unnecessary duplication of junk files, two new module configuration folders were set up for interactive tests with pXar
/home/l_tester/L1_SW/pxar/data/tbm10d_procv3_pxar/
/home/l_tester/L1_SW/pxar/data/tbm10d_procv4_pxar/
The configurations are identical to those used by elComandante. However, one needs to remember to keep the tests folders in sync with the elComandante folders once they get updated. This can be done as follows
rsync -avPSh /home/l_tester/L1_SW/pxar/data/tbm10d_procv3/ /home/l_tester/L1_SW/pxar/data/tbm10d_procv3_pxar/
rsync -avPSh /home/l_tester/L1_SW/pxar/data/tbm10d_procv4/ /home/l_tester/L1_SW/pxar/data/tbm10d_procv4_pxar/
To check if there are any extraneous pXar files present in the configuration folders, run
find /home/l_tester/L1_SW/pxar/data/tbm10d_procv?/ -type f \( -name 'pxar*.log' -o -name 'pxar*.root' \)
If any, they can be deleted by running
find /home/l_tester/L1_SW/pxar/data/tbm10d_procv?/ -type f -name \( -name 'pxar*.log' -o -name 'pxar*.root' \) -delete |
|
49
|
Wed Jan 22 11:39:48 2020 |
Dinko Ferencek | Software | BB2 configuration |
| When attempting to run the FullQualification this morning, pXar crashed while running the BB2 test. After some investigation, we realized the source of the problem was the BB2 configuration missing from testParameters.dat. The configuration is available in moreTestParameters.dat but by default the mkConfig script does not put it in testParameters.dat. Since we run BB2 in the FullQualification, the configuration needs to be copied by hand every time the configuration files are regenerated. |
|
48
|
Tue Jan 21 12:19:08 2020 |
Dinko Ferencek | Cold box tests | Latest tests of PH optimization |
| Dinko Ferencek wrote: | We did some tests by running the latest version of pXar on modules M1521, M1529, M1534 and M1536, specifically trimming and PH optimization. Of the 4 modules tested, the PH optimization converged only for M1521. We plan to further investigate with Urs' help possible reasons for failed PH optimization.
We also noticed a change in the threshold distribution plot of the latest trim bit test compared to the previous version of the test (see attached plots). |
The problem was traced down to a wrong value of the vcallow parameter in the PH optimization configuration. The value was 10 instead of 50 expected by the test. It turned out that the optimization converged for M1521 because this is a PROC600V3 module for which the configuration files were regenerated for another reason and vcallow was consequently set to the right value of 50. The other 3 modules are PROC600V4 modules and for them old configuration files with a wrong vcallow value were used. |