As much as my inner physicist would like to build the impedance match
good to one part in 10^5 or 10^6 - Aaron EW's experience with calibrating
out the MWA cable reflections at low frequencies makes me think the
reality is we *have* to get (all - not just cable) reflections out of the 
cosmological 
region of k-space.

Jackie

----------------------------------------------------------------------------------------
Jacqueline N Hewitt
Professor of Physics
Director, MIT Kavli Institute for Astrophysics and Space Research
[email protected]
617-253-3071 (Phone)
617-253-3111 (FAX)
----------------------------------------------------------------------------------------

On Apr 13, 2016, at 2:46 AM, Miguel F Morales <[email protected]> 
wrote:

> I’ll chime in here on the other side. As Nichole showed in her last paper, 
> unless you can get the cable reflection below -50 dB it will remove the 
> associated mode (there is no way to calibrate it accurately enough). So 
> unless you are really good at building cable connectors...
> 
> M
> 
> 
> 
> 
>> On Apr 12, 2016, at 9:29 PM, Aaron Parsons <[email protected]> wrote:
>> 
>> I'm with Nima on this.  Better to spend time getting the termination correct 
>> than jiggering with cable lengths.
>> 
>> On Tue, Apr 12, 2016 at 3:14 PM, Nima Razavi <[email protected]> wrote:
>> Guys, I'm not sure we need to go down that path.... You can actually design 
>> a well matched system and use decent cables.  I've been talking to a few 
>> cable manufacturers about an LMR200 equivalent 50ohm cable ~35m length. On 
>> the FE and RX side, I was looking at 3-cable combo in a single sheath:  2x 
>> LMR200 + 1x cat7 for the upfront phase switch/noise source/temp sensor 
>> control.
>> 
>> Nima
>> 
>> 
>> On 12/04/2016 22:49, danny jacobs wrote:
>>> Can one can get any cheaper than we have right now on PAPER?
>>> 
>>> On Tue, Apr 12, 2016 at 2:45 PM, Chris Carilli <[email protected]> wrote:
>>> solution:  buy cheap cables (ie. cable with a lot of intrinsic loss).  
>>> attenuation in the cable cuts down on the return loss dramatically (3 
>>> passes down the cable from the FE instead of 1, and attenuation multiplies 
>>> exponentially. or something like that). 
>>> 
>>> you just have to crank up the power into the cable at the FE. or add some 
>>> gain at the receiverators. 
>>> 
>>> the VLA had a horrendous standing wave problem in the waveguide IF system 
>>> for years, until we put in attenuators. fixed. 
>>> 
>>> cc
>>> 
>>> 
>>> 
>>> 
>>> 
>>> On 04/12/2016 03:32 PM, Aaron Michael Ewall-Wice wrote:
>>>> I think it’s also worth considering that long cables may have 
>>>> sub-reflections in them at delays smaller than the cable length due to 
>>>> inhomogeneities in the dielectrics and bends. I think our reflectometry 
>>>> measurements suggested that sub-reflections may be at the ~-50dB level 
>>>> which could still be a problem for 21cm measurements. That said, we may 
>>>> gain from the fact that these sorts of sub-reflections should be 
>>>> uncorrelated between cables to different antenna elements. 
>>>> 
>>>> cheers,
>>>> 
>>>> -Aaron 
>>>> 
>>>> 
>>>> 
>>>> 
>>>> 
>>>>> On Apr 12, 2016, at 5:25 PM, DAVID DEBOER <[email protected]> wrote:
>>>>> 
>>>>> Let’s put this on the agenda for tomorrow.  We could keep as is, except 
>>>>> put in 25m everywhere we can (so have 25m and 35m)…
>>>>> 
>>>>> Dave
>>>>> 
>>>>>> On Apr 12, 2016, at 2:22 PM, Nithyanandan Thyagarajan 
>>>>>> <[email protected]> wrote:
>>>>>> 
>>>>>> Part of me favors having one cables of one kind (it this is an option) 
>>>>>> than two - better to lose one mode than two besides other complexities 
>>>>>> which we may not fully grasp yet that come with multiple cable lengths. 
>>>>>> 
>>>>>> From: <[email protected]> on behalf of DAVID DEBOER 
>>>>>> <[email protected]>
>>>>>> Date: Tuesday, April 12, 2016 at 1:54 PM
>>>>>> To: danny jacobs <[email protected]>
>>>>>> Cc: hera <[email protected]>
>>>>>> Subject: Re: cable lengths
>>>>>> 
>>>>>> I was going to suggest the same thing.  We could have an ‘active’ long 
>>>>>> cable (i.e. a pre-post-amp right at the balun).  And how long do we need 
>>>>>> — 150m is already uncomfortably long…
>>>>>> 
>>>>>>> On Apr 12, 2016, at 1:50 PM, danny jacobs <[email protected]> 
>>>>>>> wrote:
>>>>>>> 
>>>>>>> What about making most of them short and then some of them really long?
>>>>>>> 
>>>>>>> On Tue, Apr 12, 2016 at 1:49 PM, Adam Beardsley 
>>>>>>> <[email protected]> wrote:
>>>>>>>> My concern would be pushing any cable reflection contamination to 
>>>>>>>> modes in the window. Currently only a small subset of MWA antennas 
>>>>>>>> have 150 meter cables, but the reflection line shows up very clearly. 
>>>>>>>> 45 meters puts the delay right at 0.2 h/Mpc, which is where we do a 
>>>>>>>> lot of our sensitivity forecasts. I doubt moving to 0.3 would change 
>>>>>>>> sensitivity much, but just food for thought.
>>>>>>>> 
>>>>>>>> -Adam
>>>>>>>> 
>>>>>>>> On Tue, Apr 12, 2016 at 1:26 PM danny jacobs <[email protected]> 
>>>>>>>> wrote:
>>>>>>>>> The 35m length has been worrying me a little lately. Would there be 
>>>>>>>>> any downsides like having a different spectral response for the 
>>>>>>>>> longer cables?
>>>>>>>>> 
>>>>>>>>> ~D
>>>>>>>>> 
>>>>>>>>> On Tue, Apr 12, 2016 at 1:24 PM, DAVID DEBOER <[email protected]> 
>>>>>>>>> wrote:
>>>>>>>>>> Hi - I was thinking about the analog cable lengths a bit.  We have a 
>>>>>>>>>> spec at 35 m (30m+the vertical part to the feed).  This puts the 
>>>>>>>>>> delay at about 0.15h/Mpc.  Would it be advantageous to have, say, 
>>>>>>>>>> most of the cable lengths at 25m and a smaller subset at 45m?
>>>>>>>>>> 
>>>>>>>>>> Dave
>>>>>>>>> 
>>>>>>>>> 
>>>>>>>>> 
>>>>>>>>> -- 
>>>>>>>>> ================================================================
>>>>>>>>> Daniel C. Jacobs                                                      
>>>>>>>>>      KE7DHQ
>>>>>>>>> National Science Foundation Fellow
>>>>>>>>> Arizona State University
>>>>>>>>> School of Earth and Space Exploration
>>>>>>>>> Low Frequency Cosmology
>>>>>>>>> Phone:           (505) 500 4521
>>>>>>>>> Homepage:     http://loco.lab.asu.edu/danny_jacobs/
>>>>>>>>> MWA:   mwatelescope.org
>>>>>>>>> HERA:   reionization.org
>>>>>>>>> PAPER: eor.berkeley.edu
>>>>>>> 
>>>>>>> 
>>>>>>> 
>>>>>>> -- 
>>>>>>> ================================================================
>>>>>>> Daniel C. Jacobs                                                        
>>>>>>>    KE7DHQ
>>>>>>> National Science Foundation Fellow
>>>>>>> Arizona State University
>>>>>>> School of Earth and Space Exploration
>>>>>>> Low Frequency Cosmology
>>>>>>> Phone:           (505) 500 4521
>>>>>>> Homepage:     http://loco.lab.asu.edu/danny_jacobs/
>>>>>>> MWA:   mwatelescope.org
>>>>>>> HERA:   reionization.org
>>>>>>> PAPER: eor.berkeley.edu
>>>>>> 
>>>>> 
>>>> 
>>> 
>>> 
>>> 
>>> 
>>> -- 
>>> ================================================================
>>> Daniel C. Jacobs                                                           
>>> KE7DHQ
>>> National Science Foundation Fellow
>>> Arizona State University
>>> School of Earth and Space Exploration
>>> Low Frequency Cosmology
>>> Phone:           (505) 500 4521
>>> Homepage:     http://loco.lab.asu.edu/danny_jacobs/
>>> MWA:   mwatelescope.org
>>> HERA:   reionization.org
>>> PAPER: eor.berkeley.edu
>> 
>> -- 
>> Dr Nima Razavi-Ghods
>> Senior Research Associate
>> 
>> Cavendish Astrophysics
>> University of Cambridge
>> JJ Thomson Avenue
>> Cambridge CB3 0HE
>>  
>> Tel: +44 (0)1223 766763
>> Fax: +44 (0)1223 337563
>> 
>> 
>> 
>> -- 
>> Aaron Parsons
>> 
>> 510-406-4322 (cell)
>> Campbell Hall 523, UCB
> 

Attachment: smime.p7s
Description: S/MIME cryptographic signature

Reply via email to