DØ Calorimeter Electronics Upgrade Meeting Minutes

October 28, 1999

present:: M. Tuts, D. Huffman, L. Bagby, P. Liston, L. Groer, B. McCarthy

Meeting time is 9:00 am, Thursdays in DAB1. Video connection available.
Contact Leslie Groer groer@fnal.gov.

= ACTION ITEM

Pulser System

FM has been working on the pulser interface card (PID). U. Bassler sent out a powerpoint file summarizing a meeting held Oct 8. DS will look at the specifications in this document and comment on them. There was a question of ensuring that the write bit was being latched correctly to ensure proper communication with the EPICS system.
FM estimates that once the design is finalized that the coding should take only a few weeks. If the decision is made to ramp the pulsers locally, then that logic will take a little longer to implement.

HL reported that the pulser systems are in and are being tested. Two have tested fine so far. They will test the rest and ship them out to FNAL, probably be the end of next week.
The pulser power supplies have been ordered. They expect 15 PS to arrive mid-November and to ship these to FNAL by end Nov. HL will try and get a production module of the PS to FNAL sooner so that we cna understand what work needs to be done on these when they arrive. Some parts still need to be ordered. DH and HL will stay in contact about this.

MT asked what the French groups need when they get to FNAL.
The Orsay group (Nov 8-19) will need access to the preamp boxes, the 5k test stand setup for initial testing, their PC for downloading parameter files into the pulser, a pulse generator for generating the trigger signals. They would like a cable to go from the 5k to the platform if possible - this is probably over 200ft (60m). The better solution might be to bring the PC down to the platform level. There are no VME crates within 300ft of the platform.
FNAL will locate a pulse generator for the trigger (NIM pulse, -1V, 75 Ohm).
HL will bring a long 75 Ohm cable.
MT placed the order for the 5 twisted-pair cable. This will be used to daisy-chain the pulsers.

DH said that the prototype pulser PS will be available and that we should be able to talk to it through the 1553.

BLS Systems DS has been getting pricing on the BLS motherboard fabrication and assembly. These are 8-layer 9U boards (280mm) and some quotes are rather steep ($400/each). Accura in West Chicago might be able to do them for $100. DS has a final quote on assembly of the 1250 boards.
The SCA daughterboard had been left out of the original cost estimate but some savings might be found in other parts of the BLS production (shapers might have ~$50k available - being checked).
DS in Rome Nov 4-10. Hopes to have 1/2 crate worth by time he gets back and will bring these to FNAL.
There have been a few problems with the shaper hybrids. The ground is injecting noise into the preamp (the RF ground is actually an RF source). The ground has been removed from the test jig at MTI for testing. There is probably a week delay in starting the production of the 5k pieces (Oct 30 original start date).
The BLS control board might be over budget as no money set aside for test-points, LEDs etc on the front-panel. Total expected to be about $30k.

DS reported that D. Elvira has finished testing the last glitch in the Timing + Control RAM, where the RAM was changing randomly. He has started on the final command - timeout. Expect this to take a week. The design is almost done but will need to be integrated in the rest of the system. DE may have time to work on this.

Preamps LB told MicroEngineering to start on the preproduction samples (5 each) of the Ha-Hg preamps. Expect delivery in 4-6 weeks ie. around end Nov.
The H-test jig is working and the hand-made Hx's have been tested. The currents are a little low for the windows set by the Spice model. There is no test pulse going into the preamps in this test-jig, so this is expected. The resistor R3 on the preamps should be withinb 1% and she will check that this is the case. The jig should be done in a couple of weeks. Shua has been working on it.

KTY shipped A,B and G preamps last week. We should have these in a week.
D,E,F are supposed to ship today.
KTY requested 500 more connectors. LB sent 1000 to Korea which arrived today. Presumably this is for the C production.
KTY will send 10 samples of the I-species and measurements on C8 and C9.

LaTech has not responded to the questions about the test point data that was sent yet.

It was decided that we probably will do the motherboard stuffing with the preamps here at FNAL as we are running out of money and there is a question about too much handling of the components. Annie and M. Gao are making good progress on 6-inch SCA testing, so we should have help available.

MT has ordered a PC to be used in the 5k test area with a new Bit3 board that DS bought.

SCAs MG finished testing lot 02-20 of the 6-inch UP SCAs - yields were about 87% giving about 1800 good. The 02-03 (6000 chips) yields are about 59% with about 1000 tested. These are not failing with offsets so PL does not think this is a packaging problem. PL and DH looked at the testing of the wafers and there were some wafers which did have a large distribution of failures.

PL reported that 2 of 49 4-inch chips failed in the 7.5 week burn-in. There were power glitches during this time. 1 was a hard failure, the other starts in the tester but then stops. Another 49 chips have been loaded and are being tested.

ASAT has determined that they cannot package the 4-inch wafers due to the damage caused to the pads (cratering) during bonding. PL and DH will try and get more details on what the pad make up actually is. We have about 69 wafers of the 4-inch left. It might be worth trying other companies to see if they can do the packaging. Accurel in CA could maybe tell us more about the pad compostion. MT will start the discussions with SupaTech (nee Orbit) about getting the 8 6-inch wafers that they still have.

Power Supplies The PS monitor boards are being assembled at Chipco and should be back next week. The read out boards are being done by Bob Jones group and the first should be back next week as well. The front panels are being silk-screened at MCI in Addison - should also be done next week.

DH reported that the PS production is going along fine - they need the boards which will be in next week. They are still waiting on the wire that was ordered. One other problem is that the epoxy holding the Al piece to the PS manifold that holds the diodes seems to crack easily during assembly due to warping of the plate. These might need to be reepoxied. The failure mode is that there is about a 0.3-0.4V drop with 20 amps through each diode generating 6-8W of heat. If the Al plate comes away from the cooling plate these can get very hot and turn the wire "crispy". The suggestion was to place two thermal sensors on either end of the Al piece - there are spare analog and digital channels available to read these out. A software interlock can be used with these sensors. The Vicors have internal, local thermal interlocks, but all the rest are in software.

This page updated 4-November-1999
Send comments/corrections to Leslie Groer groer@fnal.gov