[evlatests] ea23 returns, slowly

Wed Sep 7 10:09:49 EDT 2011

Bob,

Yesterday, you asked me for help in figuring out this matter, which I have been working on. By raising it on this forum, you have apparently sought additional help from a wider group of experts within the observatory. In the interest of maintaining a fact-based approach, please provide the actual data to this forum so that we may all help you. Excel or csv files will be fine.

Regards,

Dr. Gordon M. Coutts, PhD
Associate Scientist, Research Engineer
National Radio Astronomy Observatory

On Sep 7, 2011, at 2:17 AM, "Robert Hayward" <rhayward at aoc.nrao.edu> wrote:

> Gordon,
> 
> The earlier test that you and Chuck seem to be fixated on was a bit more
> of a jury rig measurement than I would have normally liked because I had
> to kludge together a thermal gap assembly. Had I a full set of replacement
> parts like we have now, the results would likely have been different. You
> ask what my change in methodology was. It is really rather obvious - I
> finally have a full set of interchangeable parts from two polarizers - one
> that is good, and another that is less so.
> 
> The rotary flange does seem to be a generating a lot of iron filings these
> days, but they occur on the outside of the big flange and not on the
> window. When the big flange is removed, however, you have to be careful
> that the iron fillings don't migrate inwards along the top plate to the
> window.
> 
> You can be assured that there were no metal residue on the window during
> my tests on Friday as I had been conscious of this problem over the last
> few weeks in the dozens of pseudo axial radio tests I carried out on our
> 44 new hybrids. Any fragments you saw more than likely came about after I
> removed the rotary flange so I could access the screws on the top plate to
> release the OMT. You may borrow my can of compressed air if you want.
> 
> In any case, I would suggest even if there had been a few metal fragments
> (although there weren't) on the window, it would have been inconsequential
> in the direct comparison tests of the OMTs since neither the window nor
> the rotary flange were disturbed or altered between swapping the OMTs (ie:
> the VLBA OMT was removed in situ from the bottom of the thermal gap
> assembly and then the Arecibo OMT was screwed directly in its place).
> 
> You will note that I purposely never said anything negative in the
> evlatests message about the C-Band OMTs. All I said was that they "may not
> be created equal". I stand by that statement. If you take offense to that,
> or are not prepared to believe what I've been trying to tell you about my
> recent series of tests, feel free to do your own set of axial ratio
> measurements on the two polarizers yourself and then simply swap the OMTs.
> 
> I thought I had already adequately explained what my "pseudo axial ratio"
> measurement technique was when we chatted earlier this morning. I'll give
> it another try later today if you like.
> 
> As for my test data, I showed the more relevant sections to you this
> morning. If I went thru it too quickly, or explained it in a confused
> fashion, you might have spoken up. You now seem to be accusing me of
> something inappropriate. All my tests, as described briefly in my last
> email, lead me to the inescapable conclusion that the discrepancy in the
> axial ratio performance is due to the OMT, and not the thermal gap
> assembly or the hybrid or the phase-trimmed cables. If you disagree, you
> will need to show me some axial ratio tests that prove quite the opposite
> of what I feel my set of tests definitely indicated.
> 
> By the way, because of the visit by Obama's NSF fiscal adviser on
> Thursday, I will need to reassemble both the VLBA and Arecibo receivers
> for Steve Durand's "Dog & Pony" show in the afternoon. I'll probably need
> to start getting things ready on Thursday morning. We can disassemble it
> all again afterward if you need it to do more tests.
> 
> -Bob
> 
> 
>> Bob,
>> 
>> A few weeks ago, you indicated that after swapping various components, you
>> had found that the OMT was not the cause of aforementioned axial ratio
>> variations. I spoke with Chuck earlier today, and his recollection was
>> similar. What has changed in your methodology?
>> 
>> Upon inspection of the test stand, I found metal fragments on the window.
>> This is possibly a contributing factor, and it would behoove you to
>> reexamine your results with a clean test stand.
>> 
>> As you have chosen this venue to raise this subject, it would be
>> beneficial for you to share the actual data. Also, please clarify what you
>> mean by "pseudo"?
>> 
>> Regards,
>> 
>> 
>> Dr. Gordon M. Coutts, PhD
>> Associate Scientist, Research Engineer
>> National Radio Astronomy Observatory
>> 
>> 
>> On Sep 6, 2011, at 9:23 PM, "Robert Hayward" <rhayward at aoc.nrao.edu>
>> wrote:
>> 
>>> Gordon,
>>> 
>>> I thought I did exactly that this morning when I showed you the "pseudo"
>>> axial ratio plots that I had measured on the ellipticity test stand on
>>> Friday which compared the good polarizer from the Arecibo receiver and
>>> the
>>> poor polarizer from the 4th VLBA receiver. I then started mixing and
>>> matching components from the two to find out what was needed to make the
>>> VLBA polarizer better.
>>> 
>>> The polarizer from the 4th VLBA receiver (i.e., the T-gap + OMT +
>>> phase-trimmed cables + hybrid) was always poor unless I swapped in the
>>> OMT
>>> (and only the OMT) from the Arecibo polarizer. With this OMT, it looked
>>> equally fine no matter which hybrid I used. This tells me that it is the
>>> OMT out of the VLBA receiver that is different from the Arecibo OMT (and
>>> thus "not created equal").
>>> 
>>> On top of that, I have tested all of the hybrids remaining from our 46
>>> units - 3 of them in the 2nd VLBA receiver, 20 of them in the 3rd VLBA
>>> receiver and the last 21 in the Arecibo receiver. I have found no
>>> anomalous hybrids.
>>> 
>>> I also played with introducing gaps in the RF tree to see if I could
>>> make
>>> the good axial ratio in the Arecibo receiver worse. This includes
>>> forcing
>>> unrealistically large gaps between the Top Plate and the T-Gap assembly,
>>> and between the T-Gap and the OMT, plus making the gap in the T-Gap
>>> assembly too big. None of this changed the crossover points seen on the
>>> axial ratio.
>>> 
>>> Since we've not had any spare OMTs to play with, the first real chance
>>> I've had to do a proper good vs. bad set of mix & match measurements was
>>> when I reluctantly tore the Arecibo receiver apart to do this test.
>>> 
>>> I honestly don't know what exactly is different with the OMT out of the
>>> 4th receiver, I only know that it is different than the OMT that was in
>>> the 1st VLBA receiver and the one from the Arecibo receiver. The OMTs in
>>> the 2nd, 3rd and 4th VLBA receivers are unfortunately less good in a
>>> circular polarizer.
>>> 
>>> I'm hoping you can tell us exactly what is different and hopefully what
>>> we
>>> need to do to make them all perform identically good.
>>> 
>>> -Bob
>>> 
>>> 
>>>> Bob,
>>>> 
>>>> Please substantiate claim (4) with data.
>>>> 
>>>> Regards,
>>>> 
>>>> Dr. Gordon M. Coutts, PhD
>>>> Associate Scientist, Research Engineer
>>>> National Radio Astronomy Observatory
>>>> 
>>>> 
>>>> On Sep 6, 2011, at 4:32 PM, Bob Hayward <rhayward at nrao.edu> wrote:
>>>> 
>>>>> Chuck Kutz wrote:
>>>>>> The Ku receiver is waiting on Hayward for test.
>>>>>> 
>>>>> 
>>>>> And Hayward has been...
>>>>> 
>>>>> 1) testing & debugging the first L-Band Solar Receiver, slated to be
>>>>> installed on Ant 24 in less than 2 weeks.
>>>>> 
>>>>> 2) testing & debugging the latest upgraded VLBA C-Band receiver,
>>>>> slated
>>>>> to be shipped to Brewster on Sept 19th.
>>>>> 
>>>>> 3) testing & debugging the new improved EVLA-style C-Band receiver,
>>>>> which has a new thermal gap assembly from the VLBA installed in it
>>>>> (that
>>>>> will knock about 5K off the T(rx) at the low-end) and which needs to
>>>>> be ready by Sept 19th when we have a couple of engineers visiting from
>>>>> Arecibo who want to look at the receiver (and hopefully buy one).
>>>>> 
>>>>> 4) sorting out annoying axial ratio problems in the C-Band polarizers
>>>>> (it seems all OMTs may not be created equal).
>>>>> 
>>>>> 5) preparing for Thursday's "Dog & Pony" show.
>>>>> 
>>>>> 6) and writing stupid memos like this.
>>>>> 
>>>>> I'll try to fit the Ku-Band S/N 15 receiver testing in when I can find
>>>>> the time over the next few days - unless someone wants any or all of
>>>>> the
>>>>> above dates to slip.
>>>>> 
>>>>> -Bob
>>>>> 
>>>>> 
>>>>>> -----Original Message-----
>>>>>> From: evlatests-bounces at nrao.edu [mailto:evlatests-bounces at nrao.edu]
>>>>>> On
>>>>>> Behalf Of Ken Sowinski
>>>>>> Sent: Tuesday, September 06, 2011 3:18 PM
>>>>>> To: Ken Sowinski
>>>>>> Cc: evlatests at aoc.nrao.edu
>>>>>> Subject: Re: [evlatests] ea23 returns, slowly
>>>>>> 
>>>>>> On Tue, 6 Sep 2011, Ken Sowinski wrote:
>>>>>> 
>>>>>>> X and L band receivers are not yet installed
>>>>>>> S band works
>>>>>>> C band works, but the polarizations are backwards
>>>>>> 
>>>>>> C band cables have been reversed; pointing will be done tonight.
>>>>>> 
>>>>>>> Ku band rotation setting is unknown
>>>>>> 
>>>>>> A rotation setting has been provided, but there seems to
>>>>>> be no receiver.  the downconverters show no power at Ku band.
>>>>>> 
>>>>>>> K band works, but is very weak, probably because of poor pointing
>>>>>>> Ka and Q bands not examined
>>>>>>> 
>>>>>>> If C band cables are swapped today we can do pointing tonight.
>>>>>> _______________________________________________
>>>>>> evlatests mailing list
>>>>>> evlatests at listmgr.cv.nrao.edu
>>>>>> http://listmgr.cv.nrao.edu/mailman/listinfo/evlatests
>>>>>> 
>>>>>> 
>>>>>> _______________________________________________
>>>>>> evlatests mailing list
>>>>>> evlatests at listmgr.cv.nrao.edu
>>>>>> http://listmgr.cv.nrao.edu/mailman/listinfo/evlatests
>>>>> 
>>>>> _______________________________________________
>>>>> evlatests mailing list
>>>>> evlatests at listmgr.cv.nrao.edu
>>>>> http://listmgr.cv.nrao.edu/mailman/listinfo/evlatests
>>>> 
>>> 
>> 
>> 
> 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listmgr.nrao.edu/pipermail/evlatests/attachments/20110907/637719f3/attachment.html>