Log entry time 07:07:45 on February 06, 2011

Entry number 346090

This entry is a followup to: 346088

Yesterday when I was there, most of my time was spent reviving the
L-HRS DAQ, so I was slowed down by that.  Not complaining, though.
 
Anyway, below is my suggested plan.  I'll be in by 1 pm, but if a
qualified person wants to get started, this will make us more efficient. 

1. The bigbite DAQ worked ("test1" config) last time I was there. 
First thing is to establish it still does, as a baseline. Assuming
it still does, one should use it to establish that all 4 branches of the
TS work.  Simply move branch 1 cable to branch 2, then 3, and check in
each case the DAQ works.  (Obviously, reprogramming the TS in each
step.)  This proves that we have a working Trigger Supervisor on all
4 of its branches, which is not obvious at this point.

2. Checking the RS485 cables: The individual pins should be continuous,
and not shorted to adjacent.  So, for example, pin 4 should go to pin 4, 
but not to pin 3, nor to pin 5.  Did you check for shorts ?  It can 
happen.  In my experience, a newly made RS485 has a fairly high chance
of being bad, and it's always a wrongly made, or misaligned, connector.

3. If we have functioning TS and good cables, the rest should be easy,
since we already know we had 4 working DAQs that ran from local TS.  
A next thing to check is that we know all the "hardware roc numbers"
(the jumpers) -- even though we think we do.  And to check that the
rules for resistor termination on the RS485 are respected: 
no terminators except for the LAST module on the chain.  This could
easily be the problem for the HRS and will be the first thing I'll 
check when I come in (together with noting the hardware roc numbers). 
I think one can survive randomized resistor termination for short cable
runs, but for long runs like we have now, it matters more.  

After 1, 2, and 3 the DAQ should work.  Hopefully by 5 pm.

- Bob M.