[Pafgbt] BYU backend testing in GB

Fri May 3 14:05:57 EDT 2013

Bill,

Good to hear the trip has been good so far. I've inserted some answers 
below.

Karl

On 5/2/2013 10:59 PM, Bill Shillue wrote:
> Karl,
>
> Thanks for the slides.  We have met many excellent people here at CSIRO, and I am sure that Brian will have a great interaction when he comes.
>
> I am preparing my presentation and will include some of your slides.  I would like to know a little more  about the feeds.  If I understand correctly, there are three versions of the BYU dipoles: kite, GBT1, and GBT2.  We currently have kite installed.  GBT1 has been replaced by GBT2 before it was ever used, and GBT2 represents a more optimized version.
Correct. GBT1 was an interim design of which we only fabricated one.

We do have several earlier designs in addition to these, two single pol 
and one dual pol, but they were all for previous ambient temperature PAFs.
>
> Questions:
>
> 1. Why is there such a large change in the angles P3 and P5?  Do you attach any significance to these angles: ie does increase in P3 generally mean wider bandwidth?  What does p5 do electromagnetically?  I might guess that it allows you to increase the physical length of the dipole while having less coupling to an adjacent element?  But it is not clear how the angle affects the   single element dipole pattern.  What would happen if p5 was 98 degrees instead of 82 degrees?  Or do questions like this have no place since the computer optimization determines all...?
These are hard questions to answer. When we design the dipoles, we are 
using software optimization and tools of our own development that 
includes bandwidth, field of view, impedance matching, sensitivity, 
survey speed, and element spacing (i.e., cryostat size) into account. 
It's quite complex. It is noteworthy to me, though, that our designs 
have progressively become more like the square patches used by CSIRO for 
the checkerboard array as we have pushed for higher bandwidth.
>
> 2. The last plot is a nice way to show the optimization.  There are three cases:
> (a) kite at 20m spacing
> (b) GBT2 at 20m spacing
> (c)GBT2 at wider spacing
Correct.
>
> And the plot seems to show that
>   (a) has higher Tsys/Aeff but good FoV
> (b) has low Tsys/Aeff but less FoV
>   (c) has low Tsys/Aeff and good FoV
>
> Is that summary correct?  20 K seems very low.  Can  you explain how you get that?  What is the Aeff in that case?  Can you say what causes the large Tsys/Aeff of "Kite" at the low end of the band (1.2 GHz) while at 1.5 GHz it is so good (even better than GBT2?!)
Yes, this is also correct. These curves include spillover and aperture 
efficiency, but are primarily driven by impedance matching. For an 
array, the relevant impedances are active impedances, which move around 
the Smith chart as a function of beam scan angle. We try to keep the 
active impedances as close to 50 ohms as possible over the full field of 
view and operating band.

The performance of the kite at the high end is mainly due to the active 
impedances passing directly through the Smith chart, and moving away at 
the low end of the band.  The kite is inherently narrow band than the 
GBT2 dipole, and is tuned to a higher center frequency as well. There is 
also a modest trend for aperture efficiency.

As far as why the minimum is 20 K, that is based on a fairly large set 
of parameters for  amplifier noise performance (Tmin, Zopt, noise 
resistance Rn), spillover due to integrated power that misses the dish 
and illuminates the ground, and additional fixed noise terms due to 
scatter and sky noise. Although we've tried to get them in the ballpark, 
we don't worry too much about getting the fixed terms exactly right, 
since the optimization essentially ignores those terms. We just try to 
get the LNA noise, aperture efficiency, and spillover noise to be 
correct, and the other terms are just parameters. Additionally, there 
does seem to be a mysterious Tsys contribution of a few Kelvin for 
cryogenic systems that models just don't predict well, which we haven't 
included.
>
> Thanks,
>
> Bill
>
>
> -----Original Message-----
> From: Karl Warnick [mailto:warnick at ee.byu.edu]
> Sent: Tuesday, April 23, 2013 5:12 PM
> To: Bill Shillue
> Cc: Brian Jeffs
> Subject: Re: BYU backend testing in GB
>
> Bill,
>
> Per your request in yesterday's telecon, attached are slides for your talk in Australia. Have a great trip, and please pass on our greetings to the group there.
>
> Karl
>
> On 4/19/2013 2:49 PM, Bill Shillue wrote:
>> ok for 4.30 pm
>>
>> One other issue I should bring up now...I was just talking to Bob Simon and he mentioned that he was expecting dipole parts from you.  He was wondering where that was at (knowing you have been traveling), but if he gets these very soon he might have time to get the work started before we travel next Thursday.
>>
>> Bill
>>
>> -----Original Message-----
>> From: Karl Warnick [mailto:warnick at ee.byu.edu]
>> Sent: Friday, April 19, 2013 3:43 PM
>> To: Bill Shillue
>> Cc: Brian Jeffs
>> Subject: Re: BYU backend testing in GB
>>
>> Sure..
>>
>> On 4/19/2013 1:40 PM, Bill Shillue wrote:
>>> Can we push this meeting until Monday 4.30 pm EDT 2.30  pm MDT
>>>
>>> A meeting I had at 3 pm got pushed to 3.30 ...
>>>
>>> Bill
>>>
>>>
>>> -----Original Message-----
>>> From: Karl Warnick [mailto:warnick at ee.byu.edu]
>>> Sent: Thursday, April 18, 2013 12:01 PM
>>> To: Bill Shillue
>>> Cc: Brian Jeffs
>>> Subject: Re: BYU backend testing in GB
>>>
>>> Works for me. My number is 801-422-1732, and I can conference Brian in (if the time works for him as well).
>>>
>>> Karl
>>>
>>> On 4/18/2013 9:57 AM, Bill Shillue wrote:
>>>> how about Monday 4 pm EDT 2 pm MDT?
>>>>
>>>> I may be away from my office so if you provide a number I can call you there.
>>>>
>>>> Bill
>>>>
>>>> -----Original Message-----
>>>> From: Karl Warnick [mailto:warnick at ee.byu.edu]
>>>> Sent: Thursday, April 18, 2013 11:55 AM
>>>> To: Bill Shillue
>>>> Cc: Brian Jeffs
>>>> Subject: Re: BYU backend testing in GB
>>>>
>>>> Sounds good. Either Friday or Monday should work for a call.
>>>>
>>>> Karl
>>>>
>>>> On 4/18/2013 9:51 AM, Bill Shillue wrote:
>>>>> I can update you at the very end of this week (Friday between 4-5 EDT) or early next week about GB PAF status if you like.
>>>>>
>>>>> Interesting comments on ASKAP.
>>>>>
>>>>> Bill
>>>>>
>>>>>
>>>>>
>>>>> -----Original Message-----
>>>>> From: warnick at ee.byu.edu [mailto:warnick at ee.byu.edu]
>>>>> Sent: Saturday, April 13, 2013 3:05 AM
>>>>> To: Bill Shillue
>>>>> Cc: Brian Jeffs
>>>>> Subject: Re: BYU backend testing in GB
>>>>>
>>>>> Bill,
>>>>>
>>>>> Thanks for the update. I hope your trip to CSIRO is worthwhile. I met with Carole Jackson here in Gothenburg (who is leaving CSIRO, by the way), as well as a number of other members of the worldwide PAF community. There were rumors of performance problems with the ASKAP front end at the meeting. The BYU/NRAO collaboration, while small in terms of total numbers of people involved, has been quite successful in terms of basic performance metrics such as sensitivity. I'll be interested to hear what you find after your visit.
>>>>>
>>>>> Keep us posted on the cryo LNA problems and schedule for GBT tests.
>>>>>
>>>>> Karl
>>>>>
>>>>>> Our team will be visiting ASKAP 4/25--5/11.  Tony Beasley is
>>>>>> encouraging us to be more outward looking, so this will be a good
>>>>>> opportunity for us to see the amazing work that they have done in Australia.
>>>>>>
>>>>>> Testing could in principle take place in GB after that, the latter
>>>>>> half of May.  However, there may be some issues.  The most
>>>>>> critical is that we appear to have some failed amplifiers.  This may not in itself prevent all
>>>>>> the tests, but we have scheduled GBT test time in June.   If we can we
>>>>>> would like to meet that date.  If the amplifiers have failed as we
>>>>>> suspect, we have been discussing whether we could repair
>>>>>> amplifiers and recover in time for the GBT test.
>>>>>>
>>>>>> My current thinking is that we will likely have to postpone the GBT test.
>>>>>> This is all fairly late breaking.  We will know more by the end of
>>>>>> this week I think.
>>>>>>
>>>>>> I think it would be worth having a conference call with you to
>>>>>> discuss more, maybe next week?
>>>>>>
>>>>>> Bill
>>>>>>
>>>>>>
>>>> --
>>>> Karl F. Warnick
>>>> Department of Electrical and Computer Engineering Brigham Young
>>>> University
>>>> 459 Clyde Building
>>>> Provo, UT 84602
>>>> (801) 422-1732
>>>>
>>>>
>>>>
>>>>
>>>>
>>> --
>>> Karl F. Warnick
>>> Department of Electrical and Computer Engineering Brigham Young
>>> University
>>> 459 Clyde Building
>>> Provo, UT 84602
>>> (801) 422-1732
>>>
>>>
>>>
>>>
>>>
>> --
>> Karl F. Warnick
>> Department of Electrical and Computer Engineering Brigham Young
>> University
>> 459 Clyde Building
>> Provo, UT 84602
>> (801) 422-1732
>>
>>
>>
>>
>>
> --
> Karl F. Warnick
> Department of Electrical and Computer Engineering Brigham Young University
> 459 Clyde Building
> Provo, UT 84602
> (801) 422-1732
>
>
>
>
>

-- 
Karl F. Warnick
Department of Electrical and Computer Engineering
Brigham Young University
459 Clyde Building
Provo, UT 84602
(801) 422-1732