[evlatests] Subarray 2 problems

[Claire Chandler]

Hi Walter,

Note that although Vivek specified sub1=EVLA and sub2=VLA, the online 
system renamed them to the expected sub1=majority, sub2=minority, so that 
when filling them into AIPS the original naming was reversed.  So in this 
sense I think the system probably behaved sensibly, and explains what you 
see.

Note that my tests of subarrays with spectral lines on October 13 showed 
similar problems on change of correlator mode, but those tests included 
multiple scans with the same correlator settings and after the change of 
mode subsequent scans were OK.  It might just on a change of correlator 
mode that you're getting multiple, conflicting, information from the 
executor.

Claire

On Sun, 18 Nov 2007, Vivek Dhawan wrote:

> The files sent to operations had "special instructions"
> 61140a.obs = K band said "Sub 1 = EVLA"
> 61140b.obs = Q band said "Sub 2 = VLA"
>
> Somewhere along I realized sub 1 was not in the majority, but did
> not change the instructions to match, semi-delibrately testing
> another part of user blunder space.
>
> According to the operator log, the files were started together,
> failed, and then were renamed and started 30s apart.
>
> Walter Brisken wrote:
>>
>> Hi Vivek,
>>
>> Can you confirm one thing for me (something else may come up later too.)
>>
>> In the data you loaded into AIPS, which subarray is sub 1? (ie VLA or
>> EVLA?)  Looking through the documents I claim that sub 1 is VLA and sub
>> 2 is EVLA (like your note about minority/majority subarray suggests),
>> but I see the executor setting EVLA up for Kband and VLA for Q band
>> which is inconsistent with your table below.  It would be handy to know
>> which leg of this triangle inequality is broken.
>>
>> I have already found some weird stuff.  Probably one broken thing in
>> idcaf and lots of confusing stuff coming from the executor.  More later...
>>
>> -Walter
>>
> _______________________________________________
> evlatests mailing list
> evlatests at listmgr.cv.nrao.edu
> http://listmgr.cv.nrao.edu/mailman/listinfo/evlatests
>