D0 Calorimeter Electronics Upgrade Meeting Minutes

September 10, 1999

present: M. Tuts, D. Huffman, L. Bagby, D. Elvira, L. Groer, C. Hays
on video: SUNYSB: D. Schamberger, B. McCarthy
           Paris: U. Bassler, B. Olivier, H. Lebbolo, Frederic Machefert (new post-doc - Welcome!)
Meeting time is now changed to 9:00 am, Thursdays in DAB1. Video connection available.
Contact Leslie Groer groer@fnal.gov.

= ACTION ITEM

Pulser System

UB introduced their new postdoc Frederic Machefert who will be working on the pulser interface.

UB met with the Orsay group yesterday.   The pulser system, switches and fanouts have started production.  Expect to receive pieces around Oct 15.
The Orsay group (Giselle, Christoph, Eve, Pierre)will come out in pairs Nov 8-19 to start installation.  The Paris group will come Nov 29-Dec 10. (Note: Thanksgiving is Nov 25).  DS will be at FNAL 2nd week Nov.
We need to make sure there is access to the platform and to the preamp boxes by that time - needs coordination with muon people?  Is there a schedule this work needs to be on?  MT will investigate.
DS asked how they will conduct tests of the installed pieces.  HL replied they can control the pulser via a stand alone PC which directly talks to a pulser box or via using the VME in the 5k test stand - this would require a very long cable (~200m?).  They will start shipping the components in October.

MT asked for a budget code for the pulser power supply monitor boards (15).  UB will follow up.

The VME control card which Paris hoped to send out next week is not quite ready.  DS and UB will be at FNAL for the D0 collaboration meeting Oct 6 and will try out the VME interface then.  DS mentioned that the T&C system would not be able to drive more than one pulser box for at least 6 months.

DS mentioned that there are three controller crates, one each in the CC and ECN and ECS.  The CC is uncabled for the new BLS control fanout crate.  The ECN might be available next week.

FNAL will also purchase and bill Paris for a +/- 15V transformer for the monitor boards.  These are about 10 cm^3 blocks which will go in the pulser PS.  There was a discussion about whether these PS needed an external interlock and a safety relay on the AC power.  DH did not think so but will think about it.

HL needs a NIM pulse generator for simulating the trigger when they do the setups in Nov.  MT will try and locate one at Nevis as they are in short supply here.

DS raised the question of do we need to get on the schedule for "rearranging"  the pulser current cables in the winders.  MT thinks not at the moment.

 BLS Systems DS has received the 0.1 Ohm resistors he needed and is now looking at the BLS motherboard.  He discovered a dead CMOS chip on an old motherboard which gave wierd gains.  There is a missing connection on the new motherboard prototype - it is solderable on the 10 prototypes and is an easy fix for production.
DS is training a new EE grad student to relay out the T&C board.
The daughterboards are not done yet - these still have another month on the schedule.
The m/board tests are on schedule.  There are 48 hybrid shapers on the test board.  DS has enough for the 5k test stand. The 8-12 week delivery on the rest of the shapers has started i.e. delivery about beginning to mid-Nov.  Production time dependent on how fast the company can get parts.  The tester he sent them broke in shipping but is fixed.
DS will ask Chipco to bid on the preproduction daughterboards.

MT reported that Nevis has started machining the copper plating for the BLS backplanes.  $125 for all 60 pieces.  MT asked what preparation work can be done on the platform.  DS mentioned that one of the CC backplanes is ripped apart, hence the old BLS cards will not fit in there anymore.

BM asked when we think we need the calibration software.  DS replied that we are a long way from getting useful information out but when there is a working trigger framework (Nov?) we could start using it and testing it.  MT mentioned there is a possibility of a cosmic ray run Aug 2000 when we definitely would need the calibrations.

 Timing and Control DE has had 2 problems with the design program in trying to change the T&C board layout, one of which he has solved, the other which might be more serious. He has an error and an ACK signal going through a buffer and which are prioritized and put through an encoder to the output.  He is sending errors out of the FPGA on bi-directional pins into the encoder, but the design program does not like this.  He could send out the errors without encoding but this would change the board layout.  DS suggested putting another buffer between the output encoder and the output pin.  DE will try this and DS will take a look at the design this weekend. Power Supplies DH has sent out the monitor board artwork.  Two prototypes will be made.  The PO is hung up in clerical work but should be out soon.  He is finishing the work on the pulser monitor board and writing up documentation.
The PSs are only waiting on the monitor boards as these need to be wired in.  We have 4 steel boxes up on the preamp platforms and these roll in and out with no problems.  The plan at the moment is to put the PS in the steel box before craning it up.  Some harness or hook has to be made though to enable a PS to be removed after installation.
Mike Cherry is working on PS assembly.  Another person could be useful in this, once the monitor boards are here - we need someplace to put them and the PSs. SCAs DH reported the 6" wafers look good - most of the bad chips are on the edges of a wafer.  All the 6" will be packaged UP.  From a message from Pat Liston about what we have on hand:
 
SCA type 4" 6"  Total
UP 7314 8319 to be packaged 
expected yield 7072		 14386 (expected)
DOWN 6196 5223 11419
Total 13510 12295 25805

The ratio of UP/DOWN is 3/2 hence we need 15k/10k.  We will be a little short on the UPs.
Expect packaging to take about 2 weeks.  We need to come up with both a test scheme and person/s to do the testing.  The testing is expected to take two full person-months but probably cannot be done 8 hours/day by one person due to fatigue.

 ICD DS did not talk to LaTech this week and LB has not heard from them.  LB will follow up and get them to explain their last fax a bit better.
The new schedule has the ECS on the platform in Oct 2000.  How will this impact the ICD.  There is a question about cabling the detector end of the ICD pulser cables for testing prior to this.
We should consult with the LaTech group. Preamp production etc. LB has submitted the PO to Bob Cibic for the H-species with a request for MicroEngineering.  CH will gets the parts ready to ship.  We believe we have all we need here.  MicroEng will use a SMT resistor for R6 (1.5 Mohm) and a zero-ohm jumper for the crossover - they will supply these parts.  They will apply a silicon coat.  We should be able to ship the parts next week.  LB expects prototypes (5 each) in 4-6 weeks and then 4-6 weeks for the full production.  We will send the technical info that we have but they have to relay out the board as we do not have electronic files from KTY.  LB is looking at a 10:1 scale drawing of the layout to make sure it is OK.
The H-test jig board has been stuffed - a couple of parts and connectors are still needed.
LB will dummy up some Ha-Hg from the He stock we have in hand to get the current windows for the tests.  She will use ES2 He's rather than the preproduction.  R6 and R37 are for compensation only and will not change the current draws.  Only the D-species changed between ES2 and preproduction.
Donna K has the connectors that need to be polarized.

LB is having some slight problems getting the ZIF socket in place for the second test jig for KTY - should have this done next week.  She has the setup files for the scope that KTY asked for and has tested them at FCC.

LG has been looking at the air temperature in the preamp boxes with the cover closed.  The maximum temperature seems to be about 50 deg C with the box on for a few hours.  He noted that after the power failure on Friday morning, the preamp PS in the 5k test stand came back in the ON position.  LG will next try putting some baffles in the box to redirect the airflow to see if that makes a difference.

We should be getting the 4 preproduction v3.2b motherboards back next week.  Once we sign off on these we can get Metrocircuits going on the full production which is expected to be about 4 weeks.  CH and LG will check the inventory for the motherboard production and start ordering the missing pieces.

This page updated 15-September-1999
Send comments/corrections to Leslie Groer groer@fnal.gov