[evlatests] EVLA Tests July 18, 2005

[Mark Claussen]

EVLA Tests July 18, 2005
Summary - EVLA Tests on An 16 only

1.  ACU problems at beginning.
2.  T304s got messed up during ACU reset.
3.  After T304s reset, the ch1-hi-attenuators
    settings required for good T5 readings were
    different than before the reset (this at L band).
4.  No T5 sd at X-band, no matter what the T304
    attenuators set to.
5.  Modcomp problems - had to reboot.
6.  Question about times in EVLA scripts.

Mark

Details follow
---------------------------------------------- 



Today's EVLA tests were not any better than my previous attempts.  
The main thing learned today was that I think I don't know enough 
about running EVLA antennas.  Surely Barry or Ken could have 
managed better.  The good thing is is this will be my last attempt 
for a while, as Barry will be back tomorrow, and Ken later in the week.
I will be on vacation from July 22 through Aug 1, inclusive.

Antenna 16 had 2 IFs with downconverters and digitizers, with all 
other associated equipment.  Two other downconverters and digitizers 
were installed, but the deformatters weren't quite ready.  The 
downconverters and digitizers to make 4 sets were stolen from An 14, 
so An 14 wasn't available at all.

At the beginning, the antenna got ridiculous acu commands and headed 
off toward the limits.  Servo was called and Ken Lakies went out and 
rescued the antenna, and reset the acu.  During the time that Ken L. 
was there, I was playing with the attenuators and watching total power 
at the T5, trying to make sure that I knew what was going on when time 
came to do it for real.  Things actually looked like they were working; 
I estimated that the best HI-attenuator settings were 24 dB on C and 
16 dB on A.  While waiting for Ken L. to report back, however, both the 
T304s went belly-up.  The total and switched power at the T5 went away.  
The T304 A side went into a state where I couldn't talk to it; the C 
side went to where I could no longer control it.  I found Rob in the 
1000 meeting; he called Linda M.  and instructed her on going to the 
antenna to reset the T304s.  

Once Linda reset the T304s, we seemed to be getting good tp and sd from 
both IFs, and we got weak fringes at L band.  However, the 
ch1-hi-attenuators were now reset to 8 dB.  Changing them to 20 or 24 dB 
made things much worse; tp and sd voltages became very low, and variable.  
After we played with this for a bit, we went back to 8 dB on these 
attenuators and found that the tp and sd readings were still variable.
About that time the operator called and said we were getting rfi at all 
antennas (we were at standard band with 12.5 MHz back end filters), so we 
gave up at L band.  

Went to X band.  Here the T5 tp seemed reasonable, but there was no sd 
voltage.  We played with the attenuators in the T304s, but to no avail.  
Needless to say there were no fringes at X band.  As it was clear we 
weren't making any progress, we decided to return to L band.  I realized 
that I needed to make my files longer in time, so I did that, putting a 
scan until 0700 LST in my previous file, remaking the EVLA script, and 
starting them back up.  Just as I tried to begin things, the MODCOMPS 
began having trouble.  DUMP got stopped, and the serial ports to BOSS 
hung up.  The operator eventually had to reboot the modcomps.

As the operator was working to restore things, Vivek and I were inspecting 
the evla scripts, comparing them to the times that that were coming back 
from the ACU, etc. and found that we either didn't know how to interpret 
the script times, or somehow they were approximately an hour wrong.  Though 
that doesn't make sense since we *did* see fringes earlier, which we
wouldn't have done had we been off by an hour in pointing on 3C84.  A curious 
question, to be sure.