[evlatests] crossed-hand delays

[Bob Hayward]

All of the EVLA receivers have a single noise diode that is split 
between the two circular polarizations. While the noise from the diode 
would be correlated between LCP & RCP, there is no guarantee that the 
noise is split equally (due to imbalances in the cal splitter, cal 
couplers and connecting cables) or that the path lengths would be 
identical (we often uses different lengths of cables between the 
splitter and cal couplers to keep the Dewar layout tidy - had this been 
a requirement, we could have ensured (more-or-less) identical path 
lengths, but the phase imbalance arising in the cal splitter and cal 
couplers would still be there).

-Bob H.


Bryan Butler wrote:
> ahh, right, what i was missing was that it's not the fact that the noise 
> source 'couples equally' into the two polarizations at the same strength 
> that is important.  sorry...
> 
> 	-bryan
> 
> 
> Ken Sowinski wrote, On 11/18/10 13:48 PM:
>> On Thu, 18 Nov 2010, Bryan Butler wrote:
>>
>>> the noise sources are different for each polarization.
>>>
>>> -bryan
>> No, Bob is right. I determined the crossed-hand delays by
>> looking at the crossed-hand autocorrelations just as he
>> suggested. I checked a few crossed-hand cross correlations
>> as well, and they agreed.
>>
>> Long ago, I did this with the Modcomps and was able to verify
>> that crossed-hand delays optimized this way gave the correct
>> answer with a strongly polarized source (3C286), and conversely
>> that this method gave the correct answer when delays were
>> optimized in continuum using 3C286.
>>
>>
>>> Bob Sault wrote, On 11/18/10 13:34 PM:
>>>> With the switched noise source measurements, presumably the source
>>>> couples into both polarizations equally, and so cross-delays and
>>>> RL phase could be measured using this mechanism (ie correlating the
>>>> R and L of the noise source on one antenna)?
> _______________________________________________
> evlatests mailing list
> evlatests at listmgr.cv.nrao.edu
> http://listmgr.cv.nrao.edu/mailman/listinfo/evlatests