Minutes of the Calorimeter Commissioning Meeting 14th February 2007 Reminder




Дата канвертавання25.04.2016
Памер23.49 Kb.
Minutes of the Calorimeter Commissioning Meeting

14th February 2007


Reminder: All the slides presented can be accessed through the agenda.

http://indico.cern.ch/conferenceDisplay.py?confId=12279
Pascal presented a request from Richard Jacobson in LHCb Commissioning meeting on the supplying in hardware by each detector of 2 status bits. 24 status bits are foreseen in the ODIN input, passed with each event in the ODIN bank. The foreseen input is 2 signals per detector, LVDS levels. The main use of these bits is to control the processing in the EFF, in case HV/LV … trip.
Rustem presented the PS cabling status and planning. A lot of progress has been made. All bottom cable trays (PS+SPD), PS top cable trays and PS C-side lateral cable tray are installed. For PS C-side, all top cables and bottom services cables have been installed up to the cable chain. With his team he was congratulated for this achievement! A good access was very important for that. Installation of lateral cable trays on A-side and bottom signal cables (C-side) will be done in the next days. Then, when the gallery assembly will be finished, the cable chain will be bent and the cable installation will be completed up to the racks. He hopes that this will be achieved by March 9. After that they will proceed with A-side cables. The complete PS cabling could be finished by mid-April. Then SPD cables installation will start, with the goal to complete it within 4-5 weeks. Rustem stressed that SPD cabling instruction (excel file with all required data as for PS) has to be provided well in advance (and cables delivery).
Marie-Noelle presented an SPD cabling overview. After a description of the SPD cable chain she listed the pieces designed and their expected delivery. The latest one (cable fixing system) will arrive beginning of April, and so in time for SPD cabling installation. In the meantime a lot of work is needed to prepare the SPD cable chain installation. In particular 100 holes have to be drilled in M1 for rails fixation from the RICH platform. This work will be done by Annecy team. Interference with PS cabling, making big use of the RICH platform, has to be evaluated. Installation of SPD top cables will be very tricky. The SPD top cable tray is under design and a prototype will be made to test its insertion under M1. Strong constraints from signal cable lengths are present to fill cable chain. A solution has been found for cable arrangement inside the chain. It fits with the present cable lengths with 4 cables on each side with an extra length smaller than 1 meter. The cable distribution to the rack has to be checked.
Robert showed us the ECAL+HCAL gallery installation sequences. The A-side will be done in the continuity of the C-side and is expected to be completed in week 10 (March 5th). The cable chains will be installed (C-side ready end of February). Then he presented the cable status from patch panel to detector. The optical cables (L0 trigger + data readout, TFC links and splitter box) have not yet been delivered. They are expected end of February. Bolek is keeping track of the delivery status. Due to wrong connection assignment, SPECS cables have been returned to manufacturer for reparation and will be back mid March. Most of the general service cables will be delivered mid March. Preparation of cables and a strategy for cabling from patch panel to fix point has to be defined. Water cooling flexible pipes are installed on the chain and so only after complete cable installation. The fixation of cables (in particular optical cables) at the entry and exit of cable chain has also to be designed. The required manpower has to be found. It appeared that cable installation will not be finished by mid March. As it is important to have final cables in place for the global commissioning test, it was decided to postpone it by around 1 month in order to have them. The situation will be reviewed at the next LHCb week in order to decide if we wait for final cable installation or if delays are too long and if we opt for a temporary solution.
Irina reported on ECAL CW noise study. Some plots were presented after pedestal subtraction and she was asked to show them also before subtraction. Some impacts on ECAL performance have been presented and some possible scenarios to replace them. It is a very long job, around 4 months with permanent access. It was decided before to take any decision to evaluate more accurately the impact on physics and in particular how it compares with the increase of dead materials. Irina and Olivier will investigate what can be done to evaluate this effect at software level and which manpower is needed.
Yuri presented a proposal for CALO online monitoring and of possible implementation of the calibration farm process. It can produce online histograms, updates to CondDb and alarms to ECS. This proposal has to be discussed. It appears that it could be not a problem to have 10.000 histograms. To reduce the number of bins has to be privileged rather than to have them on demand.
Yuri then presented the CALO database status. ECAL/HCAL parts are well advanced and close to be finished. Several access functions, which allow to read connection tables, settings, calibration constants for PM and LED, are provided. An Oracle C interface is used while a C and PVSS API are available. Work has started for PS/SPD.

Evgueni presented first results of optical fibers/LED monitoring damages fast search with digital camera and LED in current mode. This method was applied to SPD bottom A-side (1408 channels). Three problems cases are identified: 11 potential LED problems (low light from both fibres); 3 potential fibre problems (low difference of fiber light) and 1 fiber problem (clear case of different light between the 2 fibres). Some more technical improvements are needed and problem cases need more checks. Some difficulties have been encountered: access (not only to electronics boxes but also to splitters) and removal of electronic boxes cover. It was agreed that the test of the already cabled PS modules would be combined with the VFE installation, and that this needs careful planning.
Pascal presented the status of the PS VFE electronics boards. 105/119 MAPMT blocks (MAPMT + base + radiator + thermoconductor pad + VFE board + connector board) have been assembled and tested. 2 VFE boards (chips) have intermediate problems and have been rejected and 103 are ready for installation (100 are necessary). They have been packed in special boxes for transport. 14 blocks have to be assembled and tested (1 VFE board has been burnt during components assembly). All the regulator boards have been tested successfully. The design for the PS LV patch panel (used to connect cables powering the regulator board to the Wiener power supply) has been presented. It is under fabrication and will be ready within 2 weeks. The status of the PS/SPD cooling system installation was also given. All pipes, radiators inside boxes, distribution circuit and water sensor system (for DSS) along boxes have been installed by Clermont. A first test of the cooling circuit (PS C-side top part) has been performed with success. The lateral pipes (along the lateral cable tray) have to be installed. The one on PS C side was already installed yesterday morning by Rustem’s team. The tests of all the circuits have to be performed. The distribution circuit from the manifold (near to the bunker) to the detectors has to be designed. Pascal presented the procedure tested at lab to install PS VFE electronics and to connect the 8 cable types. The signal cable connection will be checked before optical bundle assembly thanks to a set-up which has been designed. The installation could start beginning of March for bottom part when the cable installation will be completed and access granted.
Xavier presented the status of SPD VFE boards. They have been all tested; 112/120 are ok. Among them 38 have been burn in; the rest will be from Friday. The set-up to test them with MAPMT is ready and under calibration. Due to the late cable installation, it was decided to start the VFE installation before cables. Half of the boards will be ready for installation in the pit on March 12. Xavier described the set-up and method which will be used to test the VFE boards during installation. The regulator boards have been tested in Barcelona and they will undergo a burn in process together with the VFE boards. Then Xavier presented the status of cables. LV cables are at CERN, data cables are already reeled in Barcelona but connectors have to be polished to be plugged on the backplane. The SPD LV patch panel is under designed. A dedicated meeting took place yesterday on this subject. A place on the ECAL platform has to be found for it, it could be installed over the gallery if possible. A decision will be taken in 2 weeks, when the gallery will be in place, to have a better judgment of the available space. Finally, he pointed out that a decision for cables is urgent to be ready on time. They have extra cable length to produce few additional cables but long delays are needed to manufacture them.

Patrick presented an overview of the “March” commissioning test. The baseline of the test is to test the complete HCAL chain (trigger and readout). The LED monitoring system will be used to flash HCAL PMT. A lot of boards are involved up to L0DU and ODIN. Some backup possibilities have been shown in case all the cables would not be in place to perform the test from the control room. But as discussed previously this possibility is disfavored, waiting for final cable installation being preferred. It means that more probably this test will occur in April and is named “Global Calorimeter Commissioning test”! After to have shown the test logics, a minimal test setup has been defined. HCAL C-side FE board 10, 11 and 12 (in crate 23) will be used with the corresponding ECAL boards, 4 validation boards, 3 CROC, 1 selection board and 1 TELL1. The possibility to have an electronic log book and the PANORAMIX event display will be studied. The goal of the test is to have a first slice of the calorimeter operational, to validate a lot of cable connections (SPECS, optical, trigger, …) . The use of “commissioning” software developed during the last common tests will be a good starting point. The behavior of the ECS system in real conditions will be tested. Strategies will be developed for time alignment. Afterwards the test will be extended to a larger area and to other detectors (PS/SPD).
Anatoli showed the status of the calorimeters LV power supply and ECS control. “Middle” voltage PS’s have been bought and tested. A first prototype of the PVSS software for control and monitor these devices was developed. One open question about devices cooling into the rack is under investigation. Then he presented the status of the development of the PMT time alignment procedure. The needed tools are numerous. First prototypes of these have been developed and tested on the test setup in building 156, including PVSS control software. Collected data have been analyzed allowing to test successfully the procedure. The next step is a test in the pit.
Frederic presented the status of the ECAL/HCAL FE board production. 85% of the (260) boards have been checked at LAL and declared ok; 75% are already at CERN. 16 or 32 boards (~10%) will be kept at LAL for the CROC production tests which should be finished in May. 22 boards are still under reparation. The test of the 16 boards extra production is ongoing. 2 ECAL and 1 HCAL racks have been fully tested in the cavern. Noise is in specifications for HCAL and there are few noisy channels for ECAL. A solution has been found on the JTAG problem of the APA PGA. The fix has to be implemented on the FE boards. 2 ECAL racks on C-side have to be tested and a slot has to be found to access the ECAL platform. This could happen only when the PS cabling C-side will be finished (and installation of bottom PS VFE electronics) i.e. end of March.

Dominique addressed some general points. Some Wiener PS encountered failures and it would be nice if someone based at CERN could be contact person to make the link between ESS and Wiener and follow repairs. Grounding of the platform has to be improved and a good protection of the platform electronics has to be settled.
Stephane (Monteil) presented the status of the PS FE board production. A pre-production of 15 boards has been launched in December. Few problems appear on PCB production, only 7 were ok and components mounted on them. They are currently on tests in Clermont. A lot of JTAG problems have been encountered. 2 TRIG-PGA cannot be re-programmed anyway and will be changed on Friday. 3 SEQ-PGA have also funny behavior. This is under investigation. In addition minor problems (solder) has been identified and fixed. Unfortunately, since the meeting the 8 other boards under production have also JTAG problems at the manufactory. Extensive checks are needed, which could delay the start production foreseen end of February… Then Stephane described the first 3 commissioning phases of the PS/SPD electronics. The first phase consists of photographs of optical bundles fired with LED monitoring system. The second one is during VFE installation and uses a dedicated system, different for PS and SPD. The last one is with the complete chain and FE electronics, prior to the global commissioning test. Some services will be needed for that, in particular water cooling and a solution will have to be found with a temporary system.
Frederic reported the CROC status. PRR has been passed on December 19. Some tests have been done since PRR. In particular a CROC has been connected to an ODIN board to test L0 signal and channel B command. It was also used with PS FE boards. Two slots had problems, unfortunately due to lack of availability of these boards, it was not possible to investigate this problem probably due to badly plugged board. Measurements of temperature in the vicinity of the optical mezzanines boards have been performed with a thermic camera. No problems are expected. Some modifications have been done on the PCB and the production has been launched this week. 2 prototypes are available and quite similar to the final production; 5 CROC should be received on March 19th and so 4 will be available for the commissioning test. Firmware development is ongoing. A full DAQ test during the next LHCb week is envisaged.
Bolek discussed the data flow up to DAQ/Online and gave an overview of what is already installed and what will be. Already 16 TELL1 are installed and all should be installed by mid March. In conclusion, DAQ/online should be ready for mid-March tests if we will not have bad surprise with delivery/installation of (short range) cables.
Stephane (T’Jampens) presented the TELL1 status. The TELL1 control software has been updated with Calo specifications with help from Lausanne. Some tests have been done with Optical Generator Board. The error reported last meeting to build MEP for ECAL/HCAL with only 2 out of 4 PP-FPGA used, is still present and has been reported to Lausanne last week. Extensive tests have still to be done, in particular write/read MEP and compare with what is sent. They should be ready for March test.
Olivier (Deschamps) presented the software status for commissioning tests, all pieces of software being essentially in place. He described the Calo readout scheme in condDB. It is ready and checked for ECAL; monitoring system has to be implemented for HCAL. Accessing the condDB, decoding Calo raw data and simulating LED data are ready for commissioning test. Afterwards Olivier discussed monitoring and histograming. A new LHCb project, named ORWELL, has been built. It contains the online tasks to be executed in the Calibration Farm. A lot of histograms are built. They can be enable/disable by options. Timing issue was also discussed. Should be OK to run on a single CPU with a reasonable rate/cell of 10-20 Hz. The main remaining work is to connect to hardware. TELL1 acquisition will benefit from discussion with VELO people who already did the exercise and Online group. Olivier will go to LAPP end of February to work with Stephane and TELL1 data. It has to be verified by users that the provided monitoring tools are all what we need for commissioning tests. Documentation has to be provided. Olivier finished with some questions, in particular how to monitor acquired consecutive events. Discussion with Online people will be needed.

Olivier Callot stressed few points. Calibration and monitoring are quite different purposes. In this spirit, we don’t have to save plenty of histograms; measurements of mean values and sigmas are preferred. Use of Hbook is disfavoured. We have to get in touch with Eric and Monica for analyzer/presenter.

Finally Irina is volunteer to look at Calo event display with PANORAMIX for commissioning. Bravo!
Xavier described ECS requirements for electronics tests. On the hardware side, control PC should be in the barrack ready to be used and cables in place for the commissioning test. Concerning software there is still a lot of work to do; the priority of people was to work on hardware up to now. Separate pieces are ready on CAT or PVSS, TELL1 control software is under study, but is not foreseen to be ready by March … DSS status is not known. Software integration is the forthcoming issue. In parallel work should be done to move from current CAT software to PVSS.
Conclusion: The “March” commissioning test will take place later as it is important to have final cables in place for it. The situation will be reviewed during the next LHCb week in order to decide if we wait for final cable installation or if delays are too long and if we opt for a temporary solution. It will be named “Global Calorimeter Commissioning” test and few phases will be defined: the first seed will be some HCAL cells, then corresponding ECAL cells will be added, then PS/SPD. A planning will be elaborated.

A list of things to check in the cavern before the test can start will be done. It includes cable installation, services, water cooling, DSS, DAQ/Online, … Some people will be identified to take care of each item. Similarly the needs for software and ECS will be reviewed. After the LHCb week, Pascal will give a presentation in the LHCb commissioning meeting in order to present an overview of the test and our needs from other parts (ODIN, DAQ/Online, …).



Some groups expressed interest for commissioning tests in building 156 (CROC, SPD multiplicity board, PS FEB, …) . They will take place in March, most of them during the LHCb week, but a planning has to be built before.


База данных защищена авторским правом ©shkola.of.by 2016
звярнуцца да адміністрацыі

    Галоўная старонка