Minutes of FT Meeting, held at Liverpool, Wednesday 24/6/98. ============================================================ Present: Graham Houghton, Jim Kay Graham Stokes (DL) Steve Burke, Rob Henderson (Lancaster) Erwin Gabathuler, Tim Greenshaw, Steve Maxfield (Liverpool) Don Clarke, Derrick Hill, Dave Sankey (RAL) Completion of planar chamber design. ------------------------------------ Thickness of walls: Don reported that Geoff Tappern has yet to confirm that the MWPC support walls are a necessary piece of the FT Tank structure, but was himself convinced that this is the case. The current drawings show one of the walls of the new planar chambers to be 10mm thick Nomex (plus two 0.5mm G10 skins). There was some discussion as to whether the supporting structure should be separated from the chamber, say by providing a 10mm Nomex support to which is attached a separate chamber (with two walls of thinner Nomex). This would mean that extra material would be introduced. The thick support walls could have skins of some material other than G10 to reduce the amount of extra material. Tim reported that the FT group have been asked to look at the possibility of supporting some of the load arising from the magnet to be inserted within the FT through the FT itself. This may be possible through the Nomex stiffening walls mentioned above, but would probably be better solved by taking the load through the forward end wall of the FT Tank. If the load were to be taken through the Nomex stiffening, significant re-design of the inner tube of the Tank may be necessary, and of the connections between the stiffening walls and this tube. In this case the separation of the new planar chambers from the stiffening walls would be desirable. Tim will continue discussions with DESY about the magnet support problem, and may visit DESY with Jim to discuss this in more detail in the coming weeks. A decision on whether the support structure is integrated into the new planar chamber wall will be made when the situation w.r.t. the magnet suppport has been clarified. Thickness of end pieces: FEA has indicated this must be increased to 15mm for the new planar chambers due to the additional load caused by the extra wires; the thickness in the existing chambers is 10mm. No. of cells: The current chambers are 26 cells wide. The central 6 cells are split at the beam-pipe leading to a total number of cells of 32. It was proposed that the new planar modules be 24 cells wide, with a total of 30 cells. Dave pointed out that more efficient utilisation of the FADCs etc. is achieved if the total number of cells is 32 or 28. It was agreed that we move to the smaller number (increasing the cell width by about 8%). The width of the chambers should thus be 22 cells, with 28 cells in total. Slightly larger cells have the advantage that the proportion of the cell in which there are problems due to the proximity of wires or walls decreases. The electrode spacings should be increased accordingly (no extra electrodes introduced) but their widths should remain the same. Orientation of new planar chambers: There was some confusion as to the orientation of the existing chambers as one of the drawings is incorrect, two different coordinate systems (reference axes) were being used and the sense with which rotations are measured is also a matter of dispute! The situation is as follows; given are the wire directions, looking from the interaction point (z=0) towards the FT (in the +ive z direction). First planar chamber seen has vertical wires (phi = 90^o). Second chamber seen has wires at 60^o to the vertical, measured clockwise (phi = -30^o or 150^o). Third chamber seen has wires at 60^o to the vertical, measured anti-clockwise (phi = -150^o or 30^o). Rob reported that the optimal orientation for the new planar chamber (i.e. the fourth chamber) is with wires at 90^o to those of the third chamber, i.e. at 30^o to the vertical, measured clockwise (phi = 120^o). The current situation is described in: http://hep.ph.liv.ac.uk:80/~sjm/atlas/planar.html Module orientation: It was decided that the orientation of the modules is to stay as it is. Changes are very difficult to implement, require a lot of time during the critical phase of the project, and do not bring significant improvements (see discussion on 25/6/98). Forward trigger: Don Clarke showed plans for the forward trigger. This is to be built of scintillator (already purchased) and wavelength shifter (WLS) which is expected to arrive in the next few weeks. The trigger will be located at the front end of the FT (outside the FT Tank). It is proposed that the track trigger have 32 fold segmentation (2-fold in r 16-fold in phi) and consist of a single layer. This may be modified to allow higher segmentation and/or a further layer depending on the results from the Forward Interaction Timer (FIT), installed this year. The trigger will be supported on a Nomex sheet (or similar). It is proposed that the readout be via Avalanche Photo-Diodes (APDs). Don will start lab tests using scintillator and WLS with APD readout. He pointed out that work is also needed on pre-amps, cabling, bias and low voltage distribution, environmental monitoring, slow controls and logic (to be realised in VME using a board being designed at DESY?). Planar drift velocity calibration: There was some discussion on the possibility of installing a source and trigger scintillator in the walls of the new planar chambers. The benefits of having a direct measure of the drift velocity were stressed. It was suggested that some of this may be able to be coupled with the forward trigger. Radial chamber modifications: The first two and the last radial wires should be discarded (information from Girish Patel). It was decided that we read out the middle 8 wires. The new field formers will be designed in Liverpool. Minutes of FT Meeting, held at Liverpool, Wednesday 24/6/98. ============================================================ Which radial wires should we read out? There was further discussion of the above point, in particular as to how possible segment finding using adjacent radial and planar chambers might be affected by the choice of wires to read out. Rob felt that there was unlikely to be a significant difference between reading out wires 2..9 (of 0..11) or wires 3..10. Orientation of new planar chamber? Rob went through his calculations of the effects on segment linking of the orientation of the new planar chamber. He demonstrated that the ``guesstimate'' of having the wires in the new (fourth) chamber (w orientation) at 90^o to those in the preceding (third) chamber (v orientation) is close to optimal as regards linking performance. This is also the choice giving best segment finding performance (resolution of ambiguities). Loss of the measurements in one of the chambers (x, u, v or w orientations) always reduces the linking performance, but this loss of performance cannot be reduced by re-orientation without compromising the performance when all chambers record hits, expected to be the case for the majority of tracks. Orientation of new modules? Rob's studies above indicated that nothing is to be gained by reorientation of the modules; they will remain as they are. Software needing modification: There was a discussion on the software needing modification. The following table summarises the situation; each item is marked 1 (=> OK), 2 (=> perhaps OK) or 3 (=> definitely not OK). Completion dates for the projects are given. Item Status Date Responsible Code librarian 1 7/98 GDP* Bank structure 1 12/98 SB FTD simulation 1 12/98 SB/MJ/GDP* Patt. Rec. 1 12/99 RCWH Linking 2 12/99 ?[] Kalman filter 2 12/99 SB/JVM Rad. Seg. finding 2 12/99 SM Monit. and Calib. 1 2/00 JVM/GDP* QT and QT display 1 2/00 JVM/DPCS Event display 3 2/00 ? Alignment 3 2/00 ? Zvtx finder 3 2/00 ? CTD FTD linking 3 2/00 ? DST banks 3 2/00 ? * These must be discussed with Girish! [] JVM and RCWH have some ideas on how this may be done. The group estimate that at least one additional person is required to complete the above in the time available. The UK universites will be asked to provide a replacement for Steve Burke, whose contract with Lancaster is running out. Alignment: John Morris outlined the situation w.r.t. the alignment of the current FTD. There is some evidence that the Radials are rotated w.r.t. the Planars. Rob will check his previous work on this point and see if a sensible correction can be made. The overall shifts and rotations deduced by John will also be introduced into the database and some tests done to check that these improve the reconstructed momenta etc. If this is the case they will be used in the forthcoming reconstruction (delaying it slightly if necessary). All possible corrections should certainly be in place for the 1998 running. Readout hardware: Dave Sankey pointed out that the fan-ins to the FADCs should be as the current planar design. These must be constructed. HV hardware: New planar distribution boxes will be necessary. The situation as regards CAEN SY227 crates must be clarified (Don?). Physics with upgraded FT: Tim suggested that the FT upgrade group think about physics that can be done using the FT. This will allow us to exploit the current system as much as is possible, give us benchmarks for testing the performance of the upgraded FT, simulation and software and put us in a good position to rapidly exploit the FT following the upgrade. Dave Milstead and Tim intend to look at D* identification and some of the physics that can be done using D*s (e.g. highish x gluon distribution).