[evlatests] Antenna 22 polarization at X-band ...

[Bob Hayward]

The receiver reads back with the F320 module (rather than the F317) 
which says it's an old timer. Even if it was a new receiver, the 
polarizer is a waveguide phase-shifter in front of the OMT so it's hard 
to think of a way inside the Dewar to cause this effect.

So I'm still thinking it could be a piece of metal. Bits of debris do 
seem to magically appear from out of nowhere inside a corrugated feed. 
I'm betting gravity has something to do with it.

In the past when this has happened, the effect was not usually this 
pronounced - say a few 10's of percent, not 100. But an appropriately 
sized piece might do it.

We could ask Gordon to model it in HFSS, or we could just drop the 
receiver and take a look...

-Bob


Rick Perley wrote:
>    .... are we *sure* it's an old X-band receiver?    Receivers have 
> been known to migrate mysteriously in the past .... (and Ken and I are 
> pretty sure it worked find a few months ago ...)
> 
>    Rick
> 
> Bob Hayward wrote:
>> Except this is an old X-Band receiver with a Septum Polarizer. This is 
>> not a typical failure mode.
>>
>> If it was a C-Band receiver, say, with a bad connection on one of the 
>> linear polarization output from OMT to the 90 deg hybrid coupler, it 
>> would make sense (i.e., degrades the Trx by a factor of 2 and makes it 
>> linear polarized). So I'm puzzled...
>>
>> We should probably drop the receiver to see if there is any debris on 
>> the vacuum window (such as metal shards from the feed).
>>
>> -Bob
>>
>>
>> Ken Sowinski wrote:
>>> On Thu, 23 Dec 2010, Rick Perley wrote:
>>>
>>>>    In calibrating the X-band 'Hercules A' demo project data, (taken
>>>> last week) I find that antenna 22 is 100% polarized!  This invariably
>>>> means that only one of the two linearly polarized outputs from the OMT
>>>> is actually being used ...  A cable is disconnected?
>>>
>>> This is surely the explanation for the poor amplitude
>>> eresponse from ea22 at X band that we have noticed for
>>> many weeks.
>>>
>>> _______________________________________________
>>> evlatests mailing list
>>> evlatests at listmgr.cv.nrao.edu
>>> http://listmgr.cv.nrao.edu/mailman/listinfo/evlatests