[evlatests] EVLA phase jumps.

[Rob Long]

We've had issues in the past with the 128 MHz cable length needing to be 
changed due to phase jumps on individual antennas (usually in A/C or 
B/D), but this is definitely antenna specific.  The other issue we have 
seen is that the VLA gets confused about time (through the CMP) and it 
appears that the EVLA is jumping when it is actually the VLA jumping.

The EVLA central timing is compared against the new GPS receiver (which 
is also our NTP server) 1PPS signal.  If the GPS were jumping, we would 
catch it in the comparison at the L350.  If the L350 were jumping, we 
would still catch it in the comparison to the GPS.  When we were still 
using the IAT clock, we saw the CMP get confused about timing several 
times and give us the same effect.

Rob

Vivek Dhawan wrote:
> Went looking for time offsets, found nothing amiss. But found
> this week's Horror Show, which is EVLA phase jumps.
> 
> All EVLA antennas jump at the same time. No VLA antenna does.
> 
> Jumps happen roughly 1/hour. Not every project is affected,
> but many are.
> 
> The phase difference between IF 1 and 2  is mostly a multiple
> of 60 degrees, though other values are seen.
> 
> I seem to recall something about a cable needing its length
> adjusted when this happened. Jim, Rob or Barry might remember
> what I'm on about? I also recall that was an antenna- specific
> thing, this current malady affects all EVLA antennas.
> Perhaps the L350 or something else connected to the timing is
> jittery?
> 
> The last good project I can find is AM979, 2009 march 10,
> UT 06:00-13:00. after that (Wednesday) was TSTMAINT, MAINT,
> PN3DB, APEVLA(Rick), EMTST(Emmanuel), BASEL, POINT. The last
> 2 were definitely afflicted, and it appears to be still
> going on now. So my guess it broke last Wednesday, that's
> already a 120 hours of mangled data.
> _______________________________________________
> evlatests mailing list
> evlatests at listmgr.cv.nrao.edu
> http://listmgr.cv.nrao.edu/mailman/listinfo/evlatests