Difference between revisions of "Logbook Camera-Assembly"

From my_wiki
Jump to: navigation, search
(→‎Enter comments in reverse time order)
(→‎Enter comments in reverse time order)
 
(17 intermediate revisions by the same user not shown)
Line 10: Line 10:
 
! style="width: 200px;" | Date !! style="width: 200px;" | Who did it? !! style="width: 200px;" | Action !! style="width: 500px;" | Comments !! style="width: 200px;" | Documents
 
! style="width: 200px;" | Date !! style="width: 200px;" | Who did it? !! style="width: 200px;" | Action !! style="width: 500px;" | Comments !! style="width: 200px;" | Documents
 
|- valign="top"
 
|- valign="top"
−
| 2018-06-XX
+
| 2018-07-XX
 
  ||   
 
  ||   
 
  ||  
 
  ||  
 
  ||   
 
  ||   
 
  || [ drive]
 
  || [ drive]
 +
|-
 +
| 2018-07-16
 +
||  Elena, Javi, Jordi, Julien, Leyre, Daniel, Camilla, Cristobal
 +
||
 +
* Data Taking:
 +
* CaCo:
 +
|| 
 +
* Data Taking:
 +
** Data taken with new Firmware
 +
* CaCo:
 +
** Check calibrations
 +
** Further tests of state machin
 +
|| [https://docs.google.com/document/d/1zLK3F6V0PM1mUASPDR6fQHF6Hxhvt7pOiHcA5ADH23o/edit?usp=sharing drive]
 +
|-
 +
| 2018-07-13
 +
||  Elena, Javi, Jordi, Julien, Leyre, Daniel, Camilla, Cristobal
 +
||
 +
* Data Taking
 +
* CaCo
 +
|| 
 +
* Data Taking:
 +
** Data taken with new Firmware
 +
* CaCo:
 +
** Check calibrations
 +
** Further tests of state machine
 +
|| [ drive]
 +
|-
 +
| 2018-07-12
 +
||  Juan, Laia, Oscar, Elena
 +
||
 +
* Intelligent Relay
 +
* Data taking:
 +
* PDB:
 +
* Update dragon firmware
 +
|| 
 +
* Intelligent Relay
 +
** Two relays are missing because we miss components for it.
 +
** Other are working properly
 +
* Data taking:
 +
** Data taken with Camera Server at PIC though long fibers:it works
 +
** Tests of new EVB. Development on going
 +
* PDB:
 +
** Relays for the Ethernet switch that were getting down sometime at startup, were replaced by 6A relays instead of 3 by MAES company.
 +
* Update dragon firmware
 +
|| [https://docs.google.com/document/d/1RMYT61-SeUcEwnWx2YVpsG0YxKM6E7eyzv1elob4_nE/edit?usp=sharing drive]
 +
|-
 +
| 2018-07-11
 +
||  Daniel, Jordi, Fernando, Javi, Elena, Mauricio, Michele, Daniele, Cristobal, Camila, Merve, Oscar
 +
||
 +
* Calibration Box
 +
* Trigger:
 +
* CaCali
 +
* Data taking
 +
*CaCo:
 +
|| 
 +
* Calibration Box
 +
** Rate scans showed the box to be operative and that we can control it
 +
** Communication with TIB did not work. To be understood.
 +
* Trigger:
 +
** L0 and L1 scans checking all channels and adders show few problematic modules for which the mezzanine was exchanged
 +
** Module 189 could not be made operative at L1 even changing the mezzanine or the module. The module does not provide problems at data taking.
 +
* Gain calibration sent to full camera with CaCali. It runs, although debugging is needed to make sure results are ok.
 +
* Data taking
 +
** Check fibers between PIC and IFAE workshop.
 +
** Data taking taken with Osaka at workshop and Okinawa running ZFitWritter at PIC successful.
 +
*CaCo:
 +
** State machine with TIB, UCTS, ECC operative.
 +
 +
|| [https://docs.google.com/document/d/1K81WBWxLnX5y_AYt2HCpiW36F-Yk8EIT3rLAS4iTxxU/edit?usp=sharing drive]
 +
|-
 +
|2018-07-10
 +
||  Camilla, Daniel, Cristobal, Michele, Elena, Oscar, Maurizio
 +
||
 +
* Calibration Box
 +
* Back Plane calibration
 +
* Cabling and labeling
 +
|| 
 +
* Calibration Box
 +
** Laser Control from CaCo
 +
** Installations inside Shielded Room
 +
** Flashing the full camera and test with rate scans
 +
** Connection to TIB through optical fiber
 +
* Back Plane calibration with full camera:
 +
** Up and down worked but for 2 modules
 +
** For clock there are about 30 module for which it did not work
 +
* Cabling and labeling inside camera finished. Only final routing for Stereo otical fibers is missing.
 +
|| [https://docs.google.com/document/d/1If6QdJmWqiRutMSqGF5pkmwmBKNM3p7lsIVxHqb8mvA/edit?usp=sharing drive]
 +
|-
 +
|| 2018-07-09
 +
||  Daniel, Cristobal, Michele, Elena, Oscar, Juan, Mauricio, Michele
 +
||
 +
* Modules
 +
* BP calibration
 +
* Data taking:
 +
* Calibration Box:
 +
* Trigger:
 +
|| 
 +
* HV setting: tried to run the test asked by Shunsuke, but it did not work.
 +
* Preparation for the BP calibration: learnt how to quickly equalize the gain response of the pixels.
 +
* Preparation for data taking: Daniel and Elena tested the procedure for data taking. At first try it did not work as the EVB was not re-compiled in Osaka.
 +
* Calibration Box:
 +
** Integration calibration box
 +
** Communication with ClusCo
 +
* Trigger:
 +
** L0 and L1 rate scans using nominal HV provided by Shunsuke to check problematic trigger lines.
 +
|| [https://docs.google.com/document/d/1rFXgDUSFvTh7baChp5zbce_thsN3xCmmLdDHFbONNMs/edit?usp=sharing drive]
 +
|-
 +
| 2018-07-06
 +
||  Léa, Oscar
 +
||
 +
* EVB
 +
|| 
 +
* EVB
 +
** Test TIB random trigger: same strange feature than yesterday when writing data. Not possible to test without writing, TIB crashed.
 +
** Test new EVB:
 +
** Periodic trigger :
 +
*** 4 outputs to okinawa, not recording -> failed
 +
*** 1 output to okinawa, not recording  -> succeeded : 1.15 GB/s on one 10 Gb link (what is expected)
 +
*** 4 outputs to locahost (osaka), not recording -> succeeded : 450 MB/s per zFitsWriter  bound on unsused cores
 +
*** 2 outputs to locahost (osaka), not recording -> succeeded : 1.5 GB/s per zFitsWriter
 +
*** 1 output to locahost (osaka), not recording -> succeeded : 1.03 GB/s per zFitsWriter
 +
*** 1 output to locahost (osaka), recording -> succeeded : 250 MB/s per zFitsWriter
 +
*** 1 output to okinawa (osaka), recording -> -> succeeded : 500 MB/s per zFitsWriter (what we expect even better when we write to the Okinawa disk)
 +
 +
** Only one interface working from Osaka to Okinawa at a normal writing speed (p7p1). For this interface as expected saturation 1.15 GB/s for a 10 Gb link. The 3 other interfaces no time to investigate from where the problem is coming from. We should put one by one the other fiber on p7p1 to see if this is really the fiber the issue or something else
 +
** Data taken on Osaka with on zfitswritter with the new EVB with and without the gain selection
 +
*** 750 Hz saturation without gain selection
 +
*** 1500 with gainselection
 +
 +
** Random trigger :
 +
*** 1 output to okinawa (osaka), recording -> succeeded. Still the same issue that EVB receive less that the camera trigger sent by the TIB. We observed it already yesterday and this morning with the EVB 2 (current EVB). At 900 HZ for the pedestal rate,, busy rate=200 and camere trigger TIB=EVB rate. From 950 Hz for the pedestal rate, busy rate going down to 50 and from this rate the camera trigger of the TIB is higher than the one received by the EVB. After discuting with Luis Angel, it could be due to the firmware as the TIB crash of this morning. A new firmware will be sent soon.
 +
 +
** The buffer saturation problem (crash when output cut) is fixed but not when input data rate is much higher than output capability (when no relaxation is possible on outbound side) : to be fixed.
 +
 +
** Runs with and without gain selection.
 +
*** Problems in data with and without gain selection(some events written multiple times): to be fixed.
 +
 +
|| [https://docs.google.com/document/d/1WMGSliY6aLJ93Z-72TH4o_QFo20-l6u1F69qHUPy9zY/edit?usp=sharing drive]
 +
|-
 +
| 2018-07-05
 +
||  Léa, Oscar Juan
 +
||
 +
* EVB
 +
|| 
 +
* EVB
 +
** Osaka to Okinawa connection: fiber 3,4,5 and 7. The fiber 8 from Osaka to the first patch panel seems down. The fiber 6 presente one green light the other one orange so it can not work for 10 Gb but we didn’t identify from where was the issue.
 +
** Still issue to write data from Osaka to Okinawa with a super low rate of writing for some interfaces. The issue is stil unclear, tests will be perform tomorrow
 +
** Data taken with TIB random trigger. Two Strange features: 1) thebusy rate increase until a input rate of 900 Hz (busy at ~ 200 HZ). Then it drops to 100 Hz for an input rate of 1000 Hz. 2) For high input rate, from 2000 Hz, the EVB receive less than the camera trigger. Test without writing will be perform tomorrow to try to see where the problem is coming from.
 +
|| [https://docs.google.com/document/d/1LIFhVVJuH9wOlVtQFkyUHYL7_y8lp5RW-gbsBd8iado/edit?usp=sharing drive]
 +
|-
 +
| 2018-07-04
 +
||  Oscar, Merve, Nadia, Jean-Luc, Juan
 +
||
 +
Oscar, Merve, Nadia, Jean-Luc, Juan
 +
||
 +
* ECC
 +
** ECC control of the shutter works
 +
** Intelligent relay is tested with Standalone and ECC, it works properly. We are able to read 0,1,6 relays.
 +
** In the ECC, relay temperature is read as Current!
 +
** In the ECC, it can not read the 7th relay, it will be implemented in the next version.
 +
** New executable generated and load in the crio
 +
** New image generated
 +
|| [https://docs.google.com/document/d/1-n-MtU9kvklenvwfkKWJZyxGPYldt99Ls-BTEYADHaU/edit?usp=sharing drive]
 +
|-
 +
| 2018-07-03
 +
||  Nadia, Jean-Luc, Merve, Daniel, Oscar, Carlos
 +
||
 +
* ECC
 +
|| 
 +
* ECC :
 +
** Shutter incorporated inside ECC: open, close and stop from OPCUA working
 +
** Able to manage the switch of the shutter inside the PDB
 +
** Verification of the registers of the PDB with MAES team
 +
** Ask for a modification of the modbus when the ECC is disconnected
 +
** New version of the executable and FPGA incorporating changes mentioned above and fixing software reboot small issue
 +
|| [https://docs.google.com/document/d/1Qc2AlfCDxHIEQmMIjaaTfmn6mjMLpADOOYn7kpwDrLQ/edit?usp=sharing drive]
 +
|-
 +
| 2018-07-02
 +
||  Nadia, Jean-Luc, Merve, Daniel, Oscar, Cristobal
 +
||
 +
* ECC
 +
* CaCo
 +
||
 +
* ECC:
 +
** Temperature sensor 7 fixed
 +
** Test of the transition data point → ok
 +
** Validation of the procedure of the ECC image on the target
 +
** New version of MOS to solve “warning” problem
 +
* CaCo:
 +
** Implementation and test of the heartbeat with CaCo → ok
 +
** Work around fix to catch TIB data points from CaCo console
 +
|| [https://docs.google.com/document/d/1Csgiq6PQN8Xiu0RMF7bm1xz6wUPQoxlPucuruqXsZK4/edit?usp=sharing drive]
 +
|-
 +
| 2018-06-29
 +
|| 
 +
Dirk, Julien, Taka, Seiya, Shunsuke, Elena and Lea
 +
||
 +
* Dragon:
 +
* Camera cabling:
 +
* DataTaking:
 +
|| 
 +
*Dragon:
 +
* Camera cabling: Take out from the shielded room the data fibers that go to PIC. They will be used to connect Osaka to PIC. Now the data switches are connected to Osaka and then 4 fibers connect Osaka to Okinawa.
 +
* DataTaking:
 +
**  Checked that the old and new EVB is still working on Osaka.
 +
** Tried to write the data to Okinawa running the EVB on Osaka: for old EVB crashed, for new EVB it didn’t crash but we didn’t receive any data… We use only 4 optical fiber from Osaka to Okinawa. We change  the optical fiber 6 to and it solves the issue of receiving data. So we identify one bad optical fiber or one bad connexion for this fiber 6. Then we receive data but with a super low rate for writing so there is some issue with the new EVB…development are in progress. Test has to be done if possible in the following week. For general Datataking, we will use current EVB (EVB2) on Osaka, writing on Osaka.
 +
 +
|| [https://docs.google.com/document/d/1c5C0TQRWCcwbXsdFCROts4Sv9avGjQkxA_m4hAJuE4o/edit?usp=sharing drive]
 +
|-
 +
| 2018-06-28
 +
||  Dirk, Julien, Taka, Seiya, Shunsuke, Elena and Lea
 +
||
 +
* Dragon
 +
* Modules
 +
* Data Taking
 +
||
 +
* Dragon:
 +
** Intelligent relays: Removed all the relays and tested for full functionality. Four of them can be re-inserted after minor reparations. Other 4 need to wait for some pieces to be ordered and then repaired.
 +
* Modules in the camera: We disconnected the last 6 modules to make the turn on of the camera faster.
 +
* ClusCo: Implemented a init6_wait to slow down the sending of the commands to L0 and L1. Tested also init6 VS init6_noSCB because init6 gives some timeout when it is executed. For now it seems that thr SCB commands induce the timeout. Further investigation is needed.
 +
* DataTaking:
 +
** Some run with the new EVB with the random trigger of the TIB. Seems to work but there is some issue of event with id 0 that are inserted periodically (this is an EVB issue). Also be careful, we are using a 4 Zfitswritter therefore for one run there are 4 files and each file containing 1 event over 4.
 +
** Cabling to link the Camera to Okinawa. We succes to receive data with the EVB from the 6 switches. Issue: 2 interfaces show a lot of CPU using and we couldn’t go more that 2 Khz whereas on Osaka the limit is 18.8 kHz. Is it due to the length of the optical fiber between the camera and PIC? Or is it only one bad optical fiber? Or somethingelse?
 +
|| [https://docs.google.com/document/d/15pUAxvVEhejMdvLJhTtZpr6xxm47VOl6VrRs5tzi3Ko/edit?usp=sharing drive]
 +
|-
 +
| 2018-06-27
 +
||  Dirk, Julien, Taka, Seiya, Shunsuke, Elena and Lea
 +
||
 +
* Dragons
 +
* Modules Reparation
 +
* Data Taking
 +
||
 +
*Dragon:
 +
** We modified the firmware so that some important signal from FPGA to DRS4 are also output to some pins which we can touch with oscilloscope. This way, we can understand better what is really happening. However, no signal was output. Probably There were some mistakes in the firmware. We made another test.
 +
** We modified the code so that confusion between synchronous trigger register and asynchronous trigger register does never happen. This solved the problem for some module, but did not for other modules.
 +
* Modules in the camera: Inserted back  module in position
 +
** (5, 5), C008.125_02 L0 00158 L1 00176 10.1.7.37
 +
** (6, -2)  1--67    L0 00264    L1 00135  10.1.6.135
 +
* DataTaking:
 +
** With 37, 61, 91 and 263 modules: no problem of missing paquet or missing modules anymore. It came from the two noisy modules and the busy one
 +
** 0 pb fixed if we use synchronous clock inside the dragon for the trigger
 +
** Data acquisition (no writing) with the random trigger of the TIB
 +
** Data acquisition (no writing) with the new EVB reaching a maximum of 18800 HZ.
 +
* Modules reparation:
 +
** Mezzanines of Italy C008.125_02 fixed so we replace it at the same place
 +
** Japan 1--67 that was the busy module with the IP 10.1.5.10 change it IP to 10.1.6.135
 +
|| [https://docs.google.com/document/d/1UCaoYl1S2ZKqqyyw-VZa3PX458iZzdE0thRSBGyvMQo/edit?usp=sharing drive]
 +
|-
 +
| 2018-06-26
 +
||  Dirk, Julien, Taka, Seiya, Shunsuke, Elena and Lea
 +
||
 +
* We remove 2 dragons
 +
* Data taking
 +
* We change the module  10.1.5.10
 +
|| 
 +
* We remove 2 dragons this morning because of mezanine issue:
 +
** Type - n° - L0 - L1- old IP- new IP-xclusco-yclusco
 +
** Italy c008.102_02 L0 00538 L1 00600 10.1.6.2 16 8
 +
**bItaly C008.125_02 L0 116 L1 154 10.1.6.25 10.1.7.37
 +
* Data taking:
 +
**bFirst test with 19 modules. One module couldn’t achieve the transition from busy to non-buzy at the EVB connection: 10.1.5.10. Also Use of nc (netcat) to connect to the module instead of the EVB. Reading dragon register of the busy state: it stayed busy. Same protocol for 10.1.5.11 with nc and it doesn’t stayed busy when nc connect. Module 10.1.5.10 has to be remove to be study in details…
 +
** Test with 18 modules: same problem than in the logbook of the 20180625. At a rate of 2khz, when the buffer of the modules if full, it is not emptied again. For this test, all the 18 modules  remains busy. We test data taking with legacydaq and everything works fine, no busy module remains
 +
** TODO: Investigate with the version of the EVB of IAC
 +
** Inversion module
 +
** Test the EVB2 without writing to disk with 19 modules. We add a value of max=82.5 khZ with a maximum of 10 modules on one link  (10 modules*1344 koctet (size of a pacquet)*82.5 khz=1.1 GBytes/s=10 GB/s)
 +
**Test the EVB2 without writing to disk with 37 modules (maximum 16 on one switch). Max value=58.5 HKz that is the theoretical value you can obtain one one link (16 modules*1344 koctet (size of a pacquet)*58.5 khz=1.25 GBytes/s=10 GB/s)
 +
* We change the module  10.1.5.10 by one of the two we removed  in the morning for which the problem was fixed:
 +
** Type - n° - L0 - L1- old IP- new IP-xclusco-yclusco
 +
** Old: Japan 1--67 L0 00264 L1 00135 10.1.1.17 10.1.5.10  - 8 - 8
 +
** New: Italy c008.102_02 L0 00538 L1 00600 10.1.6.2 16 8
 +
 +
|| [https://docs.google.com/document/d/1a9qZsoKXF9rJzvH2zI5HY7TlTVnvJRZE0l9fk1X__3c/edit?usp=sharing drive]
 +
|-
 +
| 2018-06-25
 +
||  Dirk, Julien, Taka, Seiya, Shunsuke, Elena and Lea
 +
||
 +
* Dragon firmware test:
 +
* Mezzanines test:
 +
* Data taking:
 +
||
 +
* Dragon firmware test:
 +
* Mezzanines test:
 +
* Data taking:
 +
** Only one module not powered (10.1.5.138) and one out of range of IP that will be change (10.1.6.128)
 +
** Running with 263 modules: 0 issue remains + some modules are not sending data (around 7 but there are changing). It changes depending of the runs. The one that doesn’t receive any L1 external trigger (2 yesterday) are part of these modules but there are also that have a L1 external trigger and that are not sending data. First test, 7 are not sending data, we remove then from the data taking. Then 7 others are not sending data.
 +
** One module in a busy state from the begining. Before EVB connected to them all the modules are busy, when we connect before having a pulse, one remains busy. The busy propagation of this one is cut. So here, there is a problem for one module from buzy to non-busy state.
 +
** Going to higher rate (2Khz): we acquired data until the time where one module remains permanently busy. The buffer is maybe not emptied again and it leads to a full buzy state of the camera. We can not receive anymore data. Before with data taking at 300 Hz, the buffer was never full, we didn’t see the issue. Here, it’s a problem of going from non busy to busy state. The problem in the transition of busy state (busy to non-busy and non-busy to busy) doesn’t happened on the same module.
 +
** Test with only 19 modules: same issue going to 2Khz of one module remaining busy(10.1.5.10) already problematic in the data taking with 263 modules. Removing this one, test with 18 gives exactly the same feature: the first module, when its buffer is full, remains busy. With 19 modules, we didn’t see the issue we saw for 163 modules of the transition at the EVB connection from busy to non busy.
 +
**Test with 17 modules: they all remain busy when they reach the limit of their buffer
 +
|| [https://docs.google.com/document/d/1IzAzCfcB8KwtWR-FuxGq4uL8vD3hDh-WoSYbPce2U3c/edit?usp=sharing drive]
 +
|-
 +
| 2018-06-21
 +
||  Taka, Seiya, Shunsuke, Oscar, Elena, Cristobal, Laia, Daniele
 +
||
 +
* Dragon zero issue
 +
* L0 rate scans:
 +
* Calibrations
 +
* CaCo communications to subsystems
 +
|| 
 +
* Dragon zero issue
 +
** We checked today that  all counters, Stopcell IDs etc were healthy.
 +
** We managed to adjust the input voltage  below the maximum rating of the ADC. We didn't need to change the firmware. We just changed the configuration.
 +
** We can safely do the DAQ test next week.
 +
* L0 rate scans:
 +
** With HV off, pixels show crazy rates some times. Re-configuring does not change it but turning off/on modules yes.To be understood.
 +
* Calibrations
 +
* CaCo communications to subsystems
 +
 +
|| [https://docs.google.com/document/d/1lp9n3g-mqQ8DIjk9GXDKU8M4kDj8qLcDtgzV2uc2boo/edit?usp=sharing drive]
 +
|-
 +
| 2018-06-20
 +
||  Taka, Seiyam Shunsuke, Elena, Daniele, Cristobal, Oscar
 +
||
 +
*Zeros sent from Dragon
 +
*L0+L1 calibrations
 +
*Check L0 behavior
 +
||
 +
* Zeros sent from Dragon
 +
* We pinned down the origin of the problem (thanks largely to the past study in Kyoto).
 +
** The problem is happening between DRS4s and ADC. Analog input to the ADC is already out of the range. But DRS4 input looks healthy.
 +
** The old firmware we used in IAC doesn't cause this problem.
 +
* L0+L1 calibrations
 +
* Check L0 behavior
 +
|| [https://docs.google.com/document/d/1efrHK6pndw928Sl8KSQZbCx6YlvbM1LYVbo44dUJQfw/edit?usp=sharing drive]
 
|-
 
|-
 
| 2018-06-19
 
| 2018-06-19

Latest revision as of 09:21, 17 July 2018

Enter comments in reverse time order[edit]

Follow the module installation evolution in: drive


Date Who did it? Action Comments Documents
2018-07-XX [ drive]
2018-07-16 Elena, Javi, Jordi, Julien, Leyre, Daniel, Camilla, Cristobal
  • Data Taking:
  • CaCo:
  • Data Taking:
    • Data taken with new Firmware
  • CaCo:
    • Check calibrations
    • Further tests of state machin
 drive
2018-07-13 Elena, Javi, Jordi, Julien, Leyre, Daniel, Camilla, Cristobal
  • Data Taking
  • CaCo
  • Data Taking:
    • Data taken with new Firmware
  • CaCo:
    • Check calibrations
    • Further tests of state machine
 [ drive]
2018-07-12 Juan, Laia, Oscar, Elena
  • Intelligent Relay
  • Data taking:
  • PDB:
  • Update dragon firmware
  • Intelligent Relay
    • Two relays are missing because we miss components for it.
    • Other are working properly
  • Data taking:
    • Data taken with Camera Server at PIC though long fibers:it works
    • Tests of new EVB. Development on going
  • PDB:
    • Relays for the Ethernet switch that were getting down sometime at startup, were replaced by 6A relays instead of 3 by MAES company.
  • Update dragon firmware
 drive
2018-07-11 Daniel, Jordi, Fernando, Javi, Elena, Mauricio, Michele, Daniele, Cristobal, Camila, Merve, Oscar
  • Calibration Box
  • Trigger:
  • CaCali
  • Data taking
  • CaCo:
  • Calibration Box
    • Rate scans showed the box to be operative and that we can control it
    • Communication with TIB did not work. To be understood.
  • Trigger:
    • L0 and L1 scans checking all channels and adders show few problematic modules for which the mezzanine was exchanged
    • Module 189 could not be made operative at L1 even changing the mezzanine or the module. The module does not provide problems at data taking.
  • Gain calibration sent to full camera with CaCali. It runs, although debugging is needed to make sure results are ok.
  • Data taking
    • Check fibers between PIC and IFAE workshop.
    • Data taking taken with Osaka at workshop and Okinawa running ZFitWritter at PIC successful.
  • CaCo:
    • State machine with TIB, UCTS, ECC operative.
 drive
2018-07-10 Camilla, Daniel, Cristobal, Michele, Elena, Oscar, Maurizio
  • Calibration Box
  • Back Plane calibration
  • Cabling and labeling
  • Calibration Box
    • Laser Control from CaCo
    • Installations inside Shielded Room
    • Flashing the full camera and test with rate scans
    • Connection to TIB through optical fiber
  • Back Plane calibration with full camera:
    • Up and down worked but for 2 modules
    • For clock there are about 30 module for which it did not work
  • Cabling and labeling inside camera finished. Only final routing for Stereo otical fibers is missing.
 drive
2018-07-09 Daniel, Cristobal, Michele, Elena, Oscar, Juan, Mauricio, Michele
  • Modules
  • BP calibration
  • Data taking:
  • Calibration Box:
  • Trigger:
  • HV setting: tried to run the test asked by Shunsuke, but it did not work.
  • Preparation for the BP calibration: learnt how to quickly equalize the gain response of the pixels.
  • Preparation for data taking: Daniel and Elena tested the procedure for data taking. At first try it did not work as the EVB was not re-compiled in Osaka.
  • Calibration Box:
    • Integration calibration box
    • Communication with ClusCo
  • Trigger:
    • L0 and L1 rate scans using nominal HV provided by Shunsuke to check problematic trigger lines.
 drive
2018-07-06 Léa, Oscar
  • EVB
  • EVB
    • Test TIB random trigger: same strange feature than yesterday when writing data. Not possible to test without writing, TIB crashed.
    • Test new EVB:
    • Periodic trigger :
      • 4 outputs to okinawa, not recording -> failed
      • 1 output to okinawa, not recording -> succeeded : 1.15 GB/s on one 10 Gb link (what is expected)
      • 4 outputs to locahost (osaka), not recording -> succeeded : 450 MB/s per zFitsWriter bound on unsused cores
      • 2 outputs to locahost (osaka), not recording -> succeeded : 1.5 GB/s per zFitsWriter
      • 1 output to locahost (osaka), not recording -> succeeded : 1.03 GB/s per zFitsWriter
      • 1 output to locahost (osaka), recording -> succeeded : 250 MB/s per zFitsWriter
      • 1 output to okinawa (osaka), recording -> -> succeeded : 500 MB/s per zFitsWriter (what we expect even better when we write to the Okinawa disk)
    • Only one interface working from Osaka to Okinawa at a normal writing speed (p7p1). For this interface as expected saturation 1.15 GB/s for a 10 Gb link. The 3 other interfaces no time to investigate from where the problem is coming from. We should put one by one the other fiber on p7p1 to see if this is really the fiber the issue or something else
    • Data taken on Osaka with on zfitswritter with the new EVB with and without the gain selection
      • 750 Hz saturation without gain selection
      • 1500 with gainselection
    • Random trigger :
      • 1 output to okinawa (osaka), recording -> succeeded. Still the same issue that EVB receive less that the camera trigger sent by the TIB. We observed it already yesterday and this morning with the EVB 2 (current EVB). At 900 HZ for the pedestal rate,, busy rate=200 and camere trigger TIB=EVB rate. From 950 Hz for the pedestal rate, busy rate going down to 50 and from this rate the camera trigger of the TIB is higher than the one received by the EVB. After discuting with Luis Angel, it could be due to the firmware as the TIB crash of this morning. A new firmware will be sent soon.
    • The buffer saturation problem (crash when output cut) is fixed but not when input data rate is much higher than output capability (when no relaxation is possible on outbound side) : to be fixed.
    • Runs with and without gain selection.
      • Problems in data with and without gain selection(some events written multiple times): to be fixed.
 drive
2018-07-05 Léa, Oscar Juan
  • EVB
  • EVB
    • Osaka to Okinawa connection: fiber 3,4,5 and 7. The fiber 8 from Osaka to the first patch panel seems down. The fiber 6 presente one green light the other one orange so it can not work for 10 Gb but we didn’t identify from where was the issue.
    • Still issue to write data from Osaka to Okinawa with a super low rate of writing for some interfaces. The issue is stil unclear, tests will be perform tomorrow
    • Data taken with TIB random trigger. Two Strange features: 1) thebusy rate increase until a input rate of 900 Hz (busy at ~ 200 HZ). Then it drops to 100 Hz for an input rate of 1000 Hz. 2) For high input rate, from 2000 Hz, the EVB receive less than the camera trigger. Test without writing will be perform tomorrow to try to see where the problem is coming from.
 drive
2018-07-04 Oscar, Merve, Nadia, Jean-Luc, Juan

Oscar, Merve, Nadia, Jean-Luc, Juan

  • ECC
    • ECC control of the shutter works
    • Intelligent relay is tested with Standalone and ECC, it works properly. We are able to read 0,1,6 relays.
    • In the ECC, relay temperature is read as Current!
    • In the ECC, it can not read the 7th relay, it will be implemented in the next version.
    • New executable generated and load in the crio
    • New image generated
 drive
2018-07-03 Nadia, Jean-Luc, Merve, Daniel, Oscar, Carlos
  • ECC
  • ECC :
    • Shutter incorporated inside ECC: open, close and stop from OPCUA working
    • Able to manage the switch of the shutter inside the PDB
    • Verification of the registers of the PDB with MAES team
    • Ask for a modification of the modbus when the ECC is disconnected
    • New version of the executable and FPGA incorporating changes mentioned above and fixing software reboot small issue
 drive
2018-07-02 Nadia, Jean-Luc, Merve, Daniel, Oscar, Cristobal
  • ECC
  • CaCo
  • ECC:
    • Temperature sensor 7 fixed
    • Test of the transition data point → ok
    • Validation of the procedure of the ECC image on the target
    • New version of MOS to solve “warning” problem
  • CaCo:
    • Implementation and test of the heartbeat with CaCo → ok
    • Work around fix to catch TIB data points from CaCo console
 drive
2018-06-29

Dirk, Julien, Taka, Seiya, Shunsuke, Elena and Lea

  • Dragon:
  • Camera cabling:
  • DataTaking:
  • Dragon:
  • Camera cabling: Take out from the shielded room the data fibers that go to PIC. They will be used to connect Osaka to PIC. Now the data switches are connected to Osaka and then 4 fibers connect Osaka to Okinawa.
  • DataTaking:
    • Checked that the old and new EVB is still working on Osaka.
    • Tried to write the data to Okinawa running the EVB on Osaka: for old EVB crashed, for new EVB it didn’t crash but we didn’t receive any data… We use only 4 optical fiber from Osaka to Okinawa. We change the optical fiber 6 to and it solves the issue of receiving data. So we identify one bad optical fiber or one bad connexion for this fiber 6. Then we receive data but with a super low rate for writing so there is some issue with the new EVB…development are in progress. Test has to be done if possible in the following week. For general Datataking, we will use current EVB (EVB2) on Osaka, writing on Osaka.
 drive
2018-06-28 Dirk, Julien, Taka, Seiya, Shunsuke, Elena and Lea
  • Dragon
  • Modules
  • Data Taking
  • Dragon:
    • Intelligent relays: Removed all the relays and tested for full functionality. Four of them can be re-inserted after minor reparations. Other 4 need to wait for some pieces to be ordered and then repaired.
  • Modules in the camera: We disconnected the last 6 modules to make the turn on of the camera faster.
  • ClusCo: Implemented a init6_wait to slow down the sending of the commands to L0 and L1. Tested also init6 VS init6_noSCB because init6 gives some timeout when it is executed. For now it seems that thr SCB commands induce the timeout. Further investigation is needed.
  • DataTaking:
    • Some run with the new EVB with the random trigger of the TIB. Seems to work but there is some issue of event with id 0 that are inserted periodically (this is an EVB issue). Also be careful, we are using a 4 Zfitswritter therefore for one run there are 4 files and each file containing 1 event over 4.
    • Cabling to link the Camera to Okinawa. We succes to receive data with the EVB from the 6 switches. Issue: 2 interfaces show a lot of CPU using and we couldn’t go more that 2 Khz whereas on Osaka the limit is 18.8 kHz. Is it due to the length of the optical fiber between the camera and PIC? Or is it only one bad optical fiber? Or somethingelse?
 drive
2018-06-27 Dirk, Julien, Taka, Seiya, Shunsuke, Elena and Lea
  • Dragons
  • Modules Reparation
  • Data Taking
  • Dragon:
    • We modified the firmware so that some important signal from FPGA to DRS4 are also output to some pins which we can touch with oscilloscope. This way, we can understand better what is really happening. However, no signal was output. Probably There were some mistakes in the firmware. We made another test.
    • We modified the code so that confusion between synchronous trigger register and asynchronous trigger register does never happen. This solved the problem for some module, but did not for other modules.
  • Modules in the camera: Inserted back module in position
    • (5, 5), C008.125_02 L0 00158 L1 00176 10.1.7.37
    • (6, -2) 1--67 L0 00264 L1 00135 10.1.6.135
  • DataTaking:
    • With 37, 61, 91 and 263 modules: no problem of missing paquet or missing modules anymore. It came from the two noisy modules and the busy one
    • 0 pb fixed if we use synchronous clock inside the dragon for the trigger
    • Data acquisition (no writing) with the random trigger of the TIB
    • Data acquisition (no writing) with the new EVB reaching a maximum of 18800 HZ.
  • Modules reparation:
    • Mezzanines of Italy C008.125_02 fixed so we replace it at the same place
    • Japan 1--67 that was the busy module with the IP 10.1.5.10 change it IP to 10.1.6.135
 drive
2018-06-26 Dirk, Julien, Taka, Seiya, Shunsuke, Elena and Lea
  • We remove 2 dragons
  • Data taking
  • We change the module 10.1.5.10
  • We remove 2 dragons this morning because of mezanine issue:
    • Type - n° - L0 - L1- old IP- new IP-xclusco-yclusco
    • Italy c008.102_02 L0 00538 L1 00600 10.1.6.2 16 8
    • bItaly C008.125_02 L0 116 L1 154 10.1.6.25 10.1.7.37
  • Data taking:
    • bFirst test with 19 modules. One module couldn’t achieve the transition from busy to non-buzy at the EVB connection: 10.1.5.10. Also Use of nc (netcat) to connect to the module instead of the EVB. Reading dragon register of the busy state: it stayed busy. Same protocol for 10.1.5.11 with nc and it doesn’t stayed busy when nc connect. Module 10.1.5.10 has to be remove to be study in details…
    • Test with 18 modules: same problem than in the logbook of the 20180625. At a rate of 2khz, when the buffer of the modules if full, it is not emptied again. For this test, all the 18 modules remains busy. We test data taking with legacydaq and everything works fine, no busy module remains
    • TODO: Investigate with the version of the EVB of IAC
    • Inversion module
    • Test the EVB2 without writing to disk with 19 modules. We add a value of max=82.5 khZ with a maximum of 10 modules on one link (10 modules*1344 koctet (size of a pacquet)*82.5 khz=1.1 GBytes/s=10 GB/s)
    • Test the EVB2 without writing to disk with 37 modules (maximum 16 on one switch). Max value=58.5 HKz that is the theoretical value you can obtain one one link (16 modules*1344 koctet (size of a pacquet)*58.5 khz=1.25 GBytes/s=10 GB/s)
  • We change the module 10.1.5.10 by one of the two we removed in the morning for which the problem was fixed:
    • Type - n° - L0 - L1- old IP- new IP-xclusco-yclusco
    • Old: Japan 1--67 L0 00264 L1 00135 10.1.1.17 10.1.5.10 - 8 - 8
    • New: Italy c008.102_02 L0 00538 L1 00600 10.1.6.2 16 8
 drive
2018-06-25 Dirk, Julien, Taka, Seiya, Shunsuke, Elena and Lea
  • Dragon firmware test:
  • Mezzanines test:
  • Data taking:
  • Dragon firmware test:
  • Mezzanines test:
  • Data taking:
    • Only one module not powered (10.1.5.138) and one out of range of IP that will be change (10.1.6.128)
    • Running with 263 modules: 0 issue remains + some modules are not sending data (around 7 but there are changing). It changes depending of the runs. The one that doesn’t receive any L1 external trigger (2 yesterday) are part of these modules but there are also that have a L1 external trigger and that are not sending data. First test, 7 are not sending data, we remove then from the data taking. Then 7 others are not sending data.
    • One module in a busy state from the begining. Before EVB connected to them all the modules are busy, when we connect before having a pulse, one remains busy. The busy propagation of this one is cut. So here, there is a problem for one module from buzy to non-busy state.
    • Going to higher rate (2Khz): we acquired data until the time where one module remains permanently busy. The buffer is maybe not emptied again and it leads to a full buzy state of the camera. We can not receive anymore data. Before with data taking at 300 Hz, the buffer was never full, we didn’t see the issue. Here, it’s a problem of going from non busy to busy state. The problem in the transition of busy state (busy to non-busy and non-busy to busy) doesn’t happened on the same module.
    • Test with only 19 modules: same issue going to 2Khz of one module remaining busy(10.1.5.10) already problematic in the data taking with 263 modules. Removing this one, test with 18 gives exactly the same feature: the first module, when its buffer is full, remains busy. With 19 modules, we didn’t see the issue we saw for 163 modules of the transition at the EVB connection from busy to non busy.
    • Test with 17 modules: they all remain busy when they reach the limit of their buffer
 drive
2018-06-21 Taka, Seiya, Shunsuke, Oscar, Elena, Cristobal, Laia, Daniele
  • Dragon zero issue
  • L0 rate scans:
  • Calibrations
  • CaCo communications to subsystems
  • Dragon zero issue
    • We checked today that all counters, Stopcell IDs etc were healthy.
    • We managed to adjust the input voltage below the maximum rating of the ADC. We didn't need to change the firmware. We just changed the configuration.
    • We can safely do the DAQ test next week.
  • L0 rate scans:
    • With HV off, pixels show crazy rates some times. Re-configuring does not change it but turning off/on modules yes.To be understood.
  • Calibrations
  • CaCo communications to subsystems
 drive
2018-06-20 Taka, Seiyam Shunsuke, Elena, Daniele, Cristobal, Oscar
  • Zeros sent from Dragon
  • L0+L1 calibrations
  • Check L0 behavior
  • Zeros sent from Dragon
  • We pinned down the origin of the problem (thanks largely to the past study in Kyoto).
    • The problem is happening between DRS4s and ADC. Analog input to the ADC is already out of the range. But DRS4 input looks healthy.
    • The old firmware we used in IAC doesn't cause this problem.
  • L0+L1 calibrations
  • Check L0 behavior
 drive
2018-06-19 Oscar, Taka, Seiya, Shunzuke
  • Cabling
  • Power up
  • Zero problem in Dragon
  • Cabling:
    • Ethernet control cables for TIB and switches routed.
  • Power up:
    • Humidity at 50%
    • Power up procedure followed from completely off. Managed to power all but two modules after 5th trial
    • We decided to disconnect one bus bar to not have this two modules in unstable state.
  • Zero problem in Dragon Data
    • We configured only the central backplane. All other BPs do not receive/send Busy or Trigger or Clock. Then we injected the test pulse to the central module. L1 is generated there and send it to TIB, and TIB send back to the central BP. Then we took data with DragonDaqM, a.k.a legacy LST Daq. With 100 k events, we didn't find any zero sending.
    • We configured only the central 7 modules. Then we injected the test pulse to the central module. Just like 1), L1 is sent to TIB and comes back. And then propagated to 6 neighboring modules. We took data from the 7 modules with DragonDaqM. And then, the zero problem Lea reported us yesterday is reproduced. All the 7 modules sent only 0s after 500-200 events, depending on the modules. Interestingly, the central one also send zeros.
    • So, this 0 problem has nothing to do with Daq program. Most probably the problem originates from the Dragon firmware/hardware. And this is obviously a very serious issue.
 drive
2018-06-18 Oscar, Seiya, Cristobal, Shunsuke, Daniel, Taka, Laia, Juan, Alex
  • Power up:
    • At beginning of the day, with already cold camera, all modules with power cables went on at the third attempt
    • Measurement of the voltage and intensity provide by Power Supplies at power up could finally no be done. Pending.
  • Relays from master to vertical bus bar were unmounted for their modification to solve the problem of blowing fuses.
  • The five modules that were not communicating could be recovered. The problems was the Ethernet communication. After solving it, all modules power up without problem.
  • Condensation of cooling system
    • The condensation was appeared in cooling pipes of camera front. The reason was that camera has cooled while front window was opening.
    • That’s because TV crew came us to interview about Spain and Japanese science relationships, then we opened camera window for it.
    • We wiped it with croth,We will wait until the room temperature be decreases and close the window.
  • Distance between dome and shutter was measured with a Vernier caliper: 65.7+-0.5 mm
  • CaCo state machines tested with ECC, TIB and UCTS. It is working, although the control form the GUI needs to be improved since it is not stable.
  • The data taken with the EVB previous days shows data for all modules going to everything 0 after ~hundreds of events (not the same number for each event). It reminds a problems seen at IAC ad that was found to be linked to how the firmware is compiled. It will check if it is also observed with the Legacy-DAQ.
 drive
2018-06-17 Oscar
  • Power up
  • BP Calibration
  • L0 and L1 scans
  • Power up:
    • First trial with Chiller on an camera at 24 deg, 35 modules off
    • Second trial, just going with ECC to safe and then back to ready (1 minutes difference), all on except standard 6 failing from the beggining and actually not powered
  • BP Calibration:
    • 259 modules on
    • Noisy modules at L1 , removed from calibration
    • Several modules with No L1, rate scan to check for the good DT for each module. All modules set at 1000 V, hence gain difference could be large. Need to set default HV according to IAC QC.
    • Three modules do not get L1 ext: 178, 194,208, to be investigated.
    • Calibration run at 1KHz for 231 modules, the BP calibration does not succeed for those modules for which de DT is close to the noise. Need to have default HV.
    • Full camera takes on hour with flasher at 1 KHz. Fast test at 9KHz that confirms that is goes 9 times faster.
  • L0 and L1 scans done with HV off and with HV on and flasher at 9 KHz.
 drive
2018-06-14

Oscar, Laia, Lea, Javi, Otger Gustavo Martinez Camilla Maggio Joaquim Palacio Cristobal Pio

  • Labeling
  • Data taking
  • BP
  • Labeling: Advancing on Ethernet cable cabling
  • Power up of camera:
    • First trial with chiller off and camera at 24 degrees: all modules up
    • After few minutes when temperature at 26 degrees and chiller already operative: ~30-50 modules do not power up.
    • Disconnecting 2 bus bars, 3 trials all of them successful. Not understood why not all camera can be powered up at the same time.
  • Data taking:
    • Fix the busy. There was 5 modules remaining busy even when the EVB was connected and that the camera trigger was at 0 Hz. Cutting the propagation of the busy bit putting the bit 6 of the register 24 to 0 for those modules fix the problem.
    • We acquire data for the full camera and now the trigger and abs counter are ok. It was due to the fact that the CPB busy was put at 0 and the busy of certain module was not take into account
    • Network test ok, the data pass only in the 6 sata switches. Try to increase the frequency of the pulse to 8000 and it stops the data flux, although it worked at 3000 Hz.
  • BP
    • After some configuration issues we finally could test UP BackPlane Calibrations:Coarse UP calibration worked fine
 drive
2018-06-13 Gustavo Martinez

Camilla Maggio Joaquim Palacio Oscar

  • Power up:
  • BP Calibration:
  • Power up: not all modules power up
    • Removing TIB and UCTS did not help
    • Switching on bypassing the ECC neither.
  • BP Calibration:
    • Flasher with all leds on is too much light, we worked with only the central led on
    • Full backplane calibration with 9 clusters in column 0, CIEMAT software ok
    • Full backplane calibration with 26 clusters in columns 0, 1, 2, CIEMAT software ok (after some trials) -> 1 cluster removed previously for 0 L1 rate
    • Moving to 70 clusters, we found bad TDC measurements -> to be understood
 drive
2018-06-12 Elena, Leyre, Gustavo
  • L0+L1 calibrations
 [ drive]
2018-06-11 Lea, Cristobal, Quim, Camilla, Daniel K., Leyre, Elena
  • DAQ
  • Connectivity Test
  • BP Calibration
  • ECC
  • L0+L1
  • DAQ:
    • Data taken based on external triggers for the full camera
    • Still remains 8 modules that doesn’t present any L1 external trigger
    • Remains the issue that we put the busy of the CPB to 0 (output and input)
  • Connectivity test:
    • Done with 8 modules (central ones)
    • Launched for all the camera (results to understand)
    • (PENDING) Rate-window optimization, to reduce the total time (~45min current)
  • BP Calibration:
    • Test running for 10 modules (column 10)
    • Crazy modules (crazy L0 rates) seem to annoy the test
    • Aim of running it for full camera (avoiding crazy modules)
  • ECC:
    • image of the cRio inside the camera done and sent to Nadia (LAPP) for investigation
  • L0+L1:
    • Follow up on the investigation on the gain calibration problems. Discussion on the model for the calibration distribution inside the camera.
 drive
2018-06-08 Cristobal, Daniel, Nadia, Jean-Luc, Lea, Dirk, Julien, Cristobal, Camila

ECC operative CaCo to subsystems communication Data taking BP calibration

  • ECC:
    • ECC version that allows to pass from safe to ready
    • Image disc could not be done, will be done by Daniel next week
  • CaCo:
    • Communication with TIB, UCTs and ECC tried but for all of them the datapoints need to be updated.
    • Sending tests from CaCo still show unsolved configuration issues.
  • DAQ:
    • Taking data from several modules, but not from external trigger. Discussion followed by mail.
    • Modules are not al powering, it takes several trials to at least have only 1-2 missing.
  • BP calibration
    • Did not manage to get trigger from the flasher. Need to investigate
 drive
2018-06-07 Oscar, Daniel, Lea, Camila, Quim, Cristobal, Elena, Luis Angel, Otger, Leyre, Nadia, Jean-Luc
  • ECC operation
  • BP configuration
  • TIB & UCTS Configuration
  • L0 + L1 Calibrations
  • Cabling: The 18 power cables that were disconnected from the camera for testing outside with dummy boards the solution for the Intelligent relay problem are back in the camera.
  • ECC: Nadia and Jean-Luc installed a CRIO executable generated in the CRIO inside the camera that works. It seems the problem is related to producing the executable in a different CRIO
  • BP configuration::
    • BP configuration correctly charged when initialising ClusCo (init6)
    • Camera trigger distributed to full camera from a single module with pulse injection. Trigger distributes. Still, in the full camera, busy needs to be disabled.
    • Trigger distribution works in the two module setup (for which the BP firmware was updated). Hence, it is likely there are modules in Busy in the camera.
    • A couple of noisy modules were removed from trigger and busy distribution but it did not solve the problem.
  • TIB & UCTS Configuration:
    • TIB and UCTs in the camera configured. Trigger distributed through TIB.
    • Busy rate reaching correctly TIB
  • External UCTs and TIB working fine
  • L0 + L1 Calibrations
 drive
2018-06-06 Oscar, Daniel, Lea, Camila, Quim, Cristobal, Elena, Luis Angel, Otger
  • ECC debugging
  • Preparing Data taking
  • Work on calibrations
  • Cabling: 18 power cables (from columns 17 and 18) were disconnected from the camera for testing outside with dummy boards the solution for the Intelligent relay problem.

ECC:

  • Several versions of the CRIO executable were tried, with independent access to PDB, PSB and fans.
    • For none of them, the 24 V control signal that needs to reach the solid state switch powering up TIB and UCTS were reaching the PSB.
    • Nadia and Jean-Luc are coming on Thursday morning
  • Preparing trigger distribution:
    • Trigger is distributed when charging the BP configuration from a file
    • BP configuration does not seem to be charged by default by ClusCo
  • Calibrations with CaCo
  • L1 Calibrations
  • TIB and UCTS: configuration of external spares
 drive
2018-06-05 Quim, Camila, Lea, Daniel, Oscar, Elena, Leyre
  • ECC: Degraded version tried, again without success
  • Trigger distribution: Working for partial camera
 drive
2018-06-04 Oscar, Daniel, Leyre, Cristobal
  • ECC: Several executables tried without success
  • Busy mask:Script to set modules not distributing its busy available at:
 drive
2018-06-01 Cristobal, Leyre, Elena, Carlos D, Carlos D, Gustavo M., Alvaro, Quim
  • Connectivity test (Cristobal)
  • Thermal tests (Carlos Di & Alvaro)
    • Incremental study of the shielded room temperature
 drive
2018-05-31 Cristobal, Leyre, Elena, Carlos D, Carlos D, Gustavo M., Alvaro, Quim
  • Connectivity test
    • backplane network connections.
    • Preliminary tests of the backplane network time calibration procedure with light.
  • Thermal Test
    • chiller performance under high pressure (water flowing resistivity)
 drive
2018-05-30 Cristobal, Leyre, Elena, Carlos D, Carlos D, Gustavo M., Alvaro, Quim
  • Connectivity test (Gustavo, Carlos De & Cristobal)
  • Thermal Test
    • cooling system optimization
 drive
2018-05-29 Cristobal, Leyre, Elena, Carlos D, Carlos D, Gustavo M., Alvaro, Quim
  • ClusCo in BB:
    • Debugging of calibrations and XML files for calibrations.
  • Thermal test (see picture)
    • the water distribution system
    • incremental power up
    • FIRST FULL CAMERA POWER-UP!!!
 drive
2018-05-28 Oscar, Carlos, Quim, Alvaro, Daniel, Elena, Cristobal
  • Preparing for Thermal tests:
    • Chiller control
    • Additional temporal sensor inside camera installed
    • Decrease temperature of camera with 0 modules switched on
    • Switch on 35 modules and monitor temperatures. To switch them on, the ECC communication to PSB and PDB is by passed since the ECC still does not go to ready. Modules switched off at 20:18.
    • Modules are left off and everything else on, during the full night starting at 21:15
  • ECC:
    • Several version of the CRio executable tried but still not able to stay in ready.
  • Black Box
    • Put 2 modules in the BB. For now only one has the mezzanine, the other will get a mezzanine when one will become available. The operative module (ID 1- 21) has the IP 10.1.6.173.
    • Used by Cristobal, Leyre and Elena for the calibration tests.
 drive
2018-05-25 [ drive]
2018-05-24 Oscar
  • Cabling
    • 0.5 m SFTP CA6 RJ45 cables (2) installed between TIB and UCTS
    • All power cables installed, except for the two modules that do not fit to the backplane. Spare cables stored in CTA cupboard in the workshop.
 drive
2018-05-23 Léa, Otger, Oscar, Daniel
  • Switches configuration:
    • Disable the Spining Tree for the 6 switches
    • Change IPs of the switches:
10.1.4.21        Data switch 1
10.1.4.22        Data switch 2
10.1.4.23        Data switch 3
10.1.4.24        Data switch 4
10.1.4.25        Data switch 5
10.1.4.26        Data switch 6
  • Plug the switch to each Osaka port
Switch 1 : P1P1
Switch 2 : P1P2
Switch 3 : P2P1
Switch 4 : P2P2
Switch 5 : P3P1
Switch 6 : P3P2
  • Check we can ping the modules from Osaka for each switch
  • ECC IP changed


 drive
2018-05-22 Daniele, Seiya, Quim, Elia, Lea, Elena, Leyre, Cristobal, Otger
  • Module Validation:
    • Spear mezzanines finished (production ongoing)
    • Number of modules installed = ~240
    • Validating modules without mezzanine
  • L0 & L1 calibrations:
    • Leyre <-> Elena transfer knowledge
  • Network re-configuration:
    • “Fake”-cacooperator pc should have external+internal network now
    • Some “time-outs” with Clusco. Otger is investigating is network is stable.
 drive
2018-05-21 Seiya, Quim, Oscar

Bad modules investigation (Seiya)

Bakplane calibration test:

  • Light flasher installed in the shielded room
  • L0 triggers through all the field of view
 drive
2018-05-18 Quim, Quentin, Oscar, Daniel, Lea, Elena, Daniel M., Cristobal, Seiya, Daniele, Camilla drive
2018-05-17
  • Module installation
  • Cabling
  • Power up (periods of ~1 minute)
  • Module sensor monitoring
  • Cabling:
    • Half of the Back Planes power cables installed. Only first vertica bus bar connected to master bus bar.
    • Spares monomode fibers for data and control installed. Right part routed. Figure 1 shows connection of Switch to patch panel.
  • Power up of camera:
    • One full vertical bus bar powered. All modules communicating. Temperature of the BP increases even with rear funs on.
 drive
2018-05-16 Léa, Quim, Elia, Quentin, Laia, Seiya, Daniele, Daniel Module Installation progressing drive
2018-05-15 drive
2018-05-14 Daniel K., Lea J., Seiya, Elena M., Leyre N., Quim P. Module validation Network problems related to the network configuration (+ IFAE network problems) delay the module validation process. drive
2018-05-13 Oscar
  • Ethernet data cabling
  • Cabling Ethernet for modules:
    • Full camera cabled
    • Switch 1, 2, and 3 … labelled and tested
    • Switch 4, 5 and 6 … installed and routed, but they need to be labelled and log were each cable is plugged in:link
  • Cabling optical fibers: switches connected to the optical fibers in the camera patch panel:
    • Switch 1 - Fiber 8
    • Switch 2 - Fiber 7
    • Switch 3 - Fiber 6
    • Switch 4 - Fiber 5
    • Switch 5 - Fiber 4
    • Switch 6 - Fiber 3
    • Switches also connected to main switch, which is connected to fiber 1. Routing of fibers optics and installing spare ones still needs to be done.
 drive
2018-05-12 Elena, Daniel, Seiya
  • Module installation
  • Installation of modules up to ID C008.074_02, to see the progress refer to : link
  • Starting adaptation of the “2 modules dark box” set-up in a “4 modules dark-box setup”
 drive
2018-05-11 Elena, Otger, Javier, Quentin, Oscar, Laia, Elia, Daniele, Daniel, Seiya
  • Shielded room closed
  • Module installation
  • Power up 3 modules test
  • Shielded room closed;
  • Modified the Osaka network interface to be able to talk to all the modules when connected to the black box:
em1: inet 10.1.4.12  netmask 255.255.0.0  broadcast 10.1.255.255
  • Installation of 14 modules, so now there are 34 modules in the camera.
  • First 3 modules connected to vertical bus bars and powered. All good: respond to ping and check out ok for a fast mezzanine test.
  • Reconfiguration of the network. A scheme of the new scheme is here: link
 drive
2018-05-10 Léa, Quim, Elia, Quentin, Seiya, Daniel, Elena, Laia
  • Connectivity configuration
  • Module installation
  • Ethernet data cables labelling/installation
  • Connectivity:
    • Connection to the port p1p1 fixed (coming from switch configuration, just change the switch entry of p1p1)
    • Osaka port configuration:
      • P1p1 p1p2
      • P2p1 p2p2
      • P3p1 p3p2
      • fdl
    • Optical fiber cable:
      • 1 to p1p2 (from 1.10.5.128 to 1.10.5.255)
      • 2 to p1p1 (from 1.10.5.1 to 1.10.5.127)
      • Copper cable (black) to p2p1 (from 1.10.6.1 to 1.10.6.127)
  • Front side camera:
    • light installation (help module installation)
    • We installed 17 modules today, so 20 modules is in camera now.
    • 3 Dragons had a problem at DragonQC before installation, Seiya will check if this problem is related with the analysis or module.
    • Some mezzanines had a problem at mezzanine check. But we repeated the check, then sometimes it works successfully and sometimes not.
  • Back side camera:
    • Backplane row/column identification
    • Ethernet cables being checked. One bad cables already routed have been exchanged (~30 minutes)
  • Black box:
    • added the cooling through a pinguino cooler.
 drive
2018-05-09 Daniel K., Lea J., Elena M., Quim P., Oscar B., Gustavo M., Joan B., Elia D.S.
  • Power studies
  • Module installation
  • Power studies (Oscar, Elena, Joan, Gustavo, Manel, Laia)
    • The switching option to rump-up is discarded
    • The use of an inductance or a mosfet to rump-up is being evaluated
    • Dummy boards simulating the Dragon and backplane charge are being produced and components to correctly evaluate the possible solutions purchased.
    • Cables done and installed to bypass the intelligent relay and power modules with the power supply enable.
  • Module installation started (documentation developed by Daniel, Seya, Quim, Elia, Oscar)
    • Validate the modules
    • Change dragon IPs
    • Install safety the modules

All info can be found in: link

  • Osaka communication problem (Lea and remotely people from Marseille)
 drive
2018-05-08 Oscar Blanch, Laia Cardiel
  • Cabling:
  • Labelling of Ethernet cables done for columns 6 and 7
  • Routing of Ethernet cables for column 5 and 6
 drive
2018-05-07 Seiya Nozaki, Laia Cardiel, Lea Jouvin, Elena Moretti, Oscar Blanch, Quim Palacio, Laia Cardiel Module validation before installation developed
  • We checked the procedure of validation test
    • Scan IP; sudo /usr/local/bin/arp-scan -I p1p2 –l; At first, we could not scan IP because we used p5p2 as Interface option.
    • Change IP (~10 seconds)
    • Change Dragon Firmware(~3min)
    • DragonQC(DAQ) (~4min)
    • DragonQC(Analysis) (~4min); Analysis should be done during Mezzanine check
    • Mezzanine check with Laia’s script(~2min)
  • Seiya unified the 1 to 4 procedure with single script.
  • Camera Cabling:
    • Ethernet cables labelled for modules rows 1 to 6 at the switch end.
    • Ethernet cables for rows 3 and 4 routed and connected. Row 5 half done.
 drive
2018-05-04 Daniel Kerszberg, Elena Moretti, Juan Boix, Oscar B,, Joaquim P., Chiller installation,
  • Chiller installation
  • ECC activities
    • Sensors monitoring done with OPC-UA client connected to ECC for temperature, water pressure and light. ECC and sensors behaved as expected.
  • Work on the cacoctaop station:
    • CaCo installed
    • ClusCo installed
  • Set up for power distribution check:
    • Created a 7 modules set up to check if the solution of turning on and off the relay works with more than 2 modules.
    • We find out that there are problems with the solution find out yesterday, but we understand how to correct them
    • The set up is left in the first part of the electronic lab, we will check better on Tuesday morning
  • Backplane ethernet connection:
    • Cross-checked switch 3 connections to backplanes from columns 1 & 2.
    • Not plugged yet, waiting for a label machine
    • Cable 03XX_DATA_B0206 shows no continuity (Quim: XX to be filled)


  • Optical fibres:
    • Started installation from top of the camera structure to the rack out of the shielded room


 drive
2018-05-03 Nadia Fouque, Carlos Díaz, Miguel Polo, Antonio López, Seiya Nozaki, Oscar Blanch, Laia Cardiel, Javi Gaweda, Juan Boix, ECC activities, Accessing camera front, Module validation before installation
  • ECC actvities:
    • Shutter tests, use Carlos Standalone program, Shutter Open and Close correctly;
    • Shutter program will be included in next ECC version program.
  • Accessing camera front:
    • Three module racks are now outside the shielded room to allow the frontal door to be opened;
    • Camera was moved inside the multitool to be able to open the frontal door, the hydraulic system was mounted (Figure 1) and the door was open (Figure 2), when leaving the door was left close;
    • Shutter was open successfully, when leaving the shutter was left close, it needs standalone labview program to run for the moment.
  • Module validation before and after installation:
    • Working on modified IAC test;
    • Two modules set-up for validation before installing in the camera: read temperatures, set some registers and check correctly done, L0 and L1 rate scans as well as short run with {HV and Pulse Injection off, Pulse Injection on, HV on};
    • For validation in the camera (every ~30 modules installed), we add rate scans and runs with flasher.
  • Preparing chiller installation:
    • Chiller connected to power;
    • Water pipes pre-installed;
    • Water to fill the chiller available.


 drive
2018-05-02 Oscar Blanch, Laia Cardiel, Nadia Fouque, Jean Luc Panazol, Daniel Kerszberg,

Validation of ECC already using executable in CRIO

  • Optical fiber for the slow control has been connected both inside the camera and outside from the patch panel to an external switch.
  • MAES came and the relay problem was solved by adjusting the potentiometer.
  • ECC activity:
    • Validate the camera security with the light sensors
    • Validate the installation of a new executable on the cRIO
    • Writing a procedure
    • Validate the procedure to stop and restart the the camera
    • Writing a procedure
 drive
2018-05-01 Nadia Fouque, Jean-Luc Panazol, Oscar Blanch, Merve Colak, Daniel Kerszberg ECC set up to have same state than in CIEMAT

Control of the PDB with standalone labview ⇒ OK. Control of the ECC state machine with the Main ECC program ⇒ OK. Installation of the Main ECC executable into the cRIO ⇒ OK. Run Main-ECC.exe all night without problem.

drive
2018-04-27 Javier G., Carlos Díaz, Miguel Polo, Laia Cardiel, Otger Ballester, Sebastian Grinspun, Oscar Blanch, Merve Colak, Daniel Kerszberg Camera arrived at IFAE from CIEMAT Camera was unload and introduced in the shielded room

Together with the camera, the chiller (as well as pipes) and the hydraulic system to open frontal door arrived to IFAE. Carlos Díaz and Miguel Polo open the back door of the camera. 3-Phase and 1-Phase power connected to the camera Ethernet RJ45 cable connected to main switch inside the camera long enough to be connected to a computer outside the shielded room (See Picture 4) which will be needed for ECC validation from May 1st to May 3rd.

drive
2018-04-25 Javi Gaweda, Noel Space around the Shielded Room is prepared for the camera arrival;

Wall of the shielded room open to allow to enter the camera.

drive
2018-04-24 Quim, Elena, Daniel, Quentin and Lea Clean shielded room drive
2018-04-23 Léa, Quentin, Elena Dark-Box external setup of 2 modules drive
2018-04-20 Dirk Hoffmann, Julien Houles, Quim drive
2018-04-19 Dirk Hoffmann, Julien Houles, Oscar Blanch, Takayuki Saito Received TIB (+ power cable) drive
2018-04-18 Léa, Dirk, Julien, Elena, Joan drive
2018-04-17 Cristobal, Quim, Léa, Julien IFAE switch replaced from Marseille's one drive
2018-04-16 Lea, Quim & Cristobal Taka visits IFAE.

Fiber-switch connected.

drive
2018-04-13 Camilla, Lea, Quim & Oscar

Power connection mounted and tested. Communication with 10 modules checked.

drive
2018-04-12 Camilla, Lea, Quim & Oscar Vertical bus bar arrived

Problem with the plugs cable from the bus bar to the back plane; Joan and Alex are building new cable (cables have been plugged manually and fixed with tape by Lea&Quim); Mezzanines attached; 3-phase power supply wired, but not yet plugged;

drive
2018-04-11

Camilla, Lea, Quim & Oscar

The set up for 10 modules is organized; Larger switch (to allow for 10 modules); 1 3-phase power supply (2 spares) but not yet wired; drive
Retrieved from "https://www.lst1.iac.es/wiki/index.php?title=Logbook_Camera-Assembly&oldid=870"