Minutes of the Meeting of the Technical Board ============================================= November 28th 2001, 9:30 - 11:45 News from Hera (J. Spengler): ----------------------------- During Nov. break our magnet was taken into operation for the first time. Everything went fine except a problem which was a consequence of the power crash two weeks ago. The filters of the water cooling of the power supply for our magnet were blocked (when switching on again the huge water pumps after the power crash a sieve of the ion exchanger in hall South was destroyed resulting in lots of small pearls distributed all over the water system). During the week-end our magnet was included in the proton optics and the target run took place with magnet on. However, afterwards the power supply tripped 6 times. The reason is not understood so far. It might take some time to find the origin of this problem because it shows up only after a few hours of operation. The puzzle concerning the proper position of the final focus quads at H1 and Zeus could not be solved up to now. After readjusting the bridges (which support the quads) the situation became even worse. Right now they can fill only 0,2 - 0,3 mA of positrons and cannot ramp them to lumi energies, otherwise the synchrotron radiation becomes much too high. Any scheduling of the planned lumi test run is not possible. Therefore we should prepare an emergency scenario and ask for a few dedicated shifts to progress on our detector commissioning. The Xmas shut-down starts on Dec 21th in the morning. Hera will resume operation on Jan 16th. There will be no cooling water available on Jan 2nd which means that we have to switch off all detector power and trailer 1st and 2nd floor. A survey of our detector has been scheduled for Jan 2nd and 3rd because the results of the survey done in June are not conclusive. Target (M. Funcke): ------------------- Michael reported on the first target run which took place on previous Saturday. Single wire operation was established for all 8 wires. Target steering by hand and by automatics was exercised and emergency situations like beam-loss or HV-loss were tested. Besides a small bug showing up when changing from one wire to a different one the control software worked fine. This bug has been fixed by now. The only real flaw was that data logging did not work which was due to a data base problem. The calibration of the hodoscope counters seems to be quite reasonable. However, the first run can not be used to calibrate the charge integrators due to lack of data (DB problem). The background conditions during this test run were just excellent. But it is too early to draw any conclusions from that because Hera was running without positrons, which can have an influence on collimator settings etc. In summary, the target is ready for single wire operation. The next steps can be done in parallel with other subdetectors. Vertex (G. Sciacca): ----------------------- According to Francesco, they couldn't make any progress since last meeting due to lack of beam-time. No activities are scheduled for the Christmas shutdown. As a minimal program for the next 3 weeks the VDS group asked for 3 fills of at least 16 hours each in order to test pot movements, make a timing scan and collect minimum bias data together with OTR, Rich and Ecal to obtain the first alignment. Inner Tracker (W. Gradl): ------------------------- During the access days MS14- was installed and MS06+ was removed from the magnet. A lot of work was devoted to optimize the trigger links. (This was the last contribution of Stefan Keller, who left directly after the November break . The slow control is now integrated in the common slow control. The readout speed has been increased to 20 MHz. During the Christmas break, the ITR group plans to install the still missing three TC stations, to remove MS06- and maybe MS03+- and to continue their work with the trigger links. Outer Tracker (H. Kapitza): --------------------------- Herbert started his report with the achievements of the November shutdown. According to the decision of Nov 16th MC8+ was removed and they started to open MC2-6. This work has to be finished in December. The gas system was modified accordingly. The readout could be improved. The 70 TDCs mentioned in the previous meeting produce no readout problems any longer. They observe occasional dropouts of certain crates after 50 - 70k events. This effect is under study. A configuration file without MC chambers was prepared. All ASD-8 are checked and the installation of the trigger link boards was completed. The amount of unstable optical links could be reduced substantially. In TC1 it is below 5%. The HV training started on Oct 20th. 13 groups had to be disconnected during the first hour, another 3 groups during the total training of 70 hrs. The OTR plans for the "lumi test run" are as follows: The thresholds need final adjustment which requires a target run. The synchronization can be checked simultaneously. Afterwards they need about 500k min. bias data at 5 MHz to continue with OTR debugging and provide initial calibration and alignment constants. The analysis of these data takes at least 2 weeks. Therefore it is important to collect these data before Christmas. Inner high-pt (R. Mizuk, I. Tichomirov): ---------------------------------------- The HV problems reported during the previous meeting disappeared by themselves. This is attributed to changes in weather conditions. Shorts appeared in PT3+. These problems are known since some time, therefore the chamber was removed and will be replaced by a spare chamber over Christmas. Roman explained that the wires are lost due to bad soldering points. They found out that the temperature control of the soldering iron didn't work properly. A test pixel chamber for aging studies has been installed behind ECAL. During the first target operation a charge of 36 mC was collected. Igor reported on his measurements of the water and oxygen concentration when using different tubes (see pdf-file for details). Outer high-pt (A. Schwartz): ---------------------------- During the Nov. break PT2+ has been completed and all +x FED crates are now fully cabled including the patch panels which were built to reduce the problems when connecting the 25 pair twisted pair cables to the FED boards. A first HV test of PT1- and PT3- since Aug. 2000 was encouraging, because 5 of 8 quadrants hold full HV, the remaining 3 tripped at 50 V below. A quite ambitious schedule for the December break was shown. Alan hopes to complete the connection and tests of all 6 modules. RICH (A. Gorisek): ------------------ Andrej was not able to show his transparencies because he had to attend the alignment meeting. After installation of some RICH veto boards (see Muon pretrig.) they inspected 16 PM outputs. They didn't find any indication for pickup or noise. Afterwards they took data using the LED system. The RICH readout doesn't show any different behavior due to the installation of the veto boards. Concerning the gas system, the delivery vessel with the new C4F10 will be installed temporarily in hall West and a temporary fill line will be added. This solution avoids all the heating problems when having the vessel outside. Safety people agreed. First tests of the gas system have started. However, the cooling compressor has not been turned on yet. TRD (V. Saveliev): ------------------ Valeri reported that the upper four FED crates have been installed. He prefers to shift the installation of the lower four crates to the last moment (Jan 15th 21:00 ??) in order not to disturb other people working in this area. ECAL (M. Bruschi): ------------------ Marco reported that 213 out of 226 readout boards are installed by now.116 out of 128 pretriggers boards are installed. 16 readout and 6 pretrigger boards are expected to be back from CAEN within a few days. Noise tests were repeated with the pretrigger boards being connected to the system. Except the innermost section (Hammamatsu PMTs) the width of the pedestal distributions increased by a negligible amount (inner: 1.4 to 1.45; middle: 0.74 to 0.81; outer: 0.77 to 0.82). The width of the innermost section increased from 1.8 to 2.35 channels (1 ADC count corresponds to 25 MeV). According to Marco, this is acceptable. All cabling work for the energy inhibit card has been completed. Some basic functionality tests were done with LEDs. More work is still needed. MUON (Y. Zaitsev): ------------------ Yuri showed a list of problems which could be fixed during the November access (readout problems in MU2- bottom, HV problems with a few MU4 chambers, low gas flow in MU1.4.2, noisy chambers in MU1+). However, quite some work had to be left over for Christmas. Fixing low voltage problems in MU1+ frame will require huge efforts because 6 chambers have to be removed to get access to the ASD-8 boards in question. The same frame needs more work to reduce the noise level. MUON pretrigger and RICH veto (U. Husemann): -------------------------------------------- All hardware of the Muon pretrigger system is in place. An automatic connection test of the FLT links is being carried out during the state transition (Init --> Config). The online software for the pixel system is in progress and is expected to be ready by mid December. The first boards of the RICH multiplicity veto system (7 base sum cards + 1 FED sum card) were installed and connected. Boards for the second RICH crate are currently being stuffed. 3 veto boards are available. The complete chain was tested in Dortmund. All features worked as expected. The veto boards allow for two adjustable thresholds. Directly after its installation the veto was taken into operation. The Ljubljana group didn't observe any change (increase of noise etc.) in the Rich readout. Using the LED system, BSC and FSC boards were tested. It worked. Afterwards Ulrich presented results of a study which was made to find out whether a veto derived from two Rich crates is representative for the complete event. Using minimum bias events they calculated the correlation between all hits and the hits detected by one Rich crate. For all crates the correlation factor was above 0.85. They came to the conclusion that the RICH veto can be used both as interaction trigger as well as multiplicity veto. The eff. and rejection rate was determined for run 16784 (J/Psi trigger). When requiring less than 200 hits about 98% of the J/Psi events are accepted and 8% of all events are rejected. First Level Trigger (I. Riu): ----------------------------- The installation of hardware for the 2nd TDU needs a few weeks to be finished. The connection test with the Muon pretrigger was finished successfully. The efficiency losses in MU1 and TC2 are still under investigation. Right now they are studying in detail the MU1 geometry. The online FLT software is almost complete. The Formatter needs more work. Imma expects everything to be ready in February. Second Level Trigger (P. Kreuzer): ---------------------------------- Concerning MUON-SLT, the efficiency of the tracking algorithm (tube chambers) could be improved by a factor of four. Tests on existing J/Psi data are ok. Tests on minimum bias data to measure the rejection rate are still ongoing. Efficiency tests of pixel tracking are planned for this week. The hit request and hit preparation for MUON-SLT required quite some efforts. L2Formatter had to be adapted, hit preparation for tube chambers is ready, for pixels is completely new and needs more work. The package SLTsim is available to simulate the trigger settings of next year, whereas FLTsim has some problems which have to be understood (description of 2nd TDU). Fourth Level Trigger (J. Hernandez): ------------------------------------ Before the 3rd reconstruction of data starts, several selected runs were used as a test bench. The situation with J/psi --> mu+ mu- is not conclusive, whereas the e+ e- signal looks better. They plan to increase the statistics before final conclusions can be drawn. The distribution of all CnA constants to 200 SLT nodes corresponds to about 2.2 GBytes of data which have to be distributed almost simultaneously. To speed up this process, 14 intermediate DB memory caches were introduced. A dedicated logging and archiving PC has been ordered to avoid any interference with offline. Data Acquisition (V. Rybnikov): ------------------------------ Vladimir described the new scheme of target data distribution and the target data record going to SLP (see pdf-file for details). All parameters booked by users for data quality purposes are being saved to DB every 10 sec. A new tool allows to show variables as a function of time. The long standing problem of low input rate to the ECAL calibration node was fixed. 4 spare SHARC boards became available by not using the OTR MC2-8 chambers any longer. J. Spengler December 3rd, 2001