MCM-116 was one of the first three MCM's we received. It was tested at the time (early June 1999?), but did not officially do through the QA tests. However, it was tested enough to know that it was working. Because of time pressures, this was considered sufficient testing and the MCM/output cable combination was attached to a Silicon detector/cable combination in June 1999.

On 18 and 19-Aug-99, this MCM was put through the official QA program. 32 channels did not work. Although we did not make records (or even have a defined QA procedure) during the early tests, we (Sangkoo and I) are certain we would have seen this problem. The 5th block of 32 channels did not respond to Vcal input correctly during the QA tests. It was very noisy (100-200 channels sigma) and did not seem to respond to changes in Vcal. The preamp spy channels show that the preamp was OK. The AMU spy line showed large event-to-event fluctuations (i.e. noise). The ADCs seem to work -- they appeared to correctly digitize the bad input signals.

One one occaison, we noticed that the 4th block of channels did not seem to work properly, but this was not reproducible and we will assume they are OK.

On the second day of testing, the 6th block of 32 channels did not always work. We did not work hard on an exact dignosis, but it seems to behave like the 5th block.

Normally, we would reject this MCM as a grade C or D, but it is already attached to a Si detector, so it is not obvious what to do. I think that the surface mount components have been changed since the original (June) preliminary tests of this MCM. That is one possible source of problems. Perhaps this assembly has been damaged in our handling -- although we have no clear evidence to support this idea.