[mmaimcal] Re: mmw instrumental phase

[Min Yun]

Hi Bill,

This question occurred to me when I heard that magic number of
"0.5 degree per GHz" for the VLA before.  If this were true for the 
mm arrays, one would expect 50 degree rms at 100 GHz and 
complete decorrelation at 230 GHz.  Well, I could not tell you
if there was any LO noise in the OVRO system when I observed
there previously -- the OVRO data is completely dominated
by the atmospheric phase noise.  When I asked Dave
Woody, his answer was "very small", but he did not
offer a number (maybe he measured this at one point).
So, a reasonable working requirement would be "much less than
than the atmospheric phase noise", which is spected to be
15 micron rms in path length delay (equivalent to 2 degree rms
at 115 GHz, 4 degree rms at 230 GHz, etc., scaling with 
the frequency).  Is this an achievable target for the 
electronic noise for the system?  We probably need not kill
ourselves to do much better than this.



				-- Min


> From wbrundag at zia.aoc.NRAO.EDU Wed Jun 30 15:51 MDT 1999
> Date: Wed, 30 Jun 1999 15:52:50 -0500
> From: Bill Brundage <wbrundag at aoc.nrao.edu>
> X-Accept-Language: en
> MIME-Version: 1.0
> To: "Dhawan, Vivek" <vdhawan at zia.aoc.NRAO.EDU>, myun at zia.aoc.NRAO.EDU,
>         cwalker at zia.aoc.NRAO.EDU, thelfer at zia.aoc.NRAO.EDU
> CC: "Butler, Bryan" <bbutler at zia.aoc.NRAO.EDU>,
>         "Battle, John" <jbattle at zia.aoc.NRAO.EDU>,
>         "Thacker, Skip" <sthacker at cv.nrao.edu>,
>         "D'Addario, Larry" <ldaddari at cv3.cv.nrao.edu>
> Subject: mmw instrumental phase
> Content-Transfer-Encoding: 7bit
> 
> Vivek, Craig, Min, Tamara,
> 	A discussion of the ALMA LO system yesterday raised the question
> of just how good (or poor) is the instrumental phase
> stability/noise of existing millimeter arrays.
> 	Bryan Butler estimates from 6mm and 1.2 cm observations that the
> VLA instrumental phase stability is close to 0.5 degrees per GHz
> of observing frequency per 10 minutes of integration between
> phase calibrations.
> 	If anyone has a number for the VLBA, Caltech array or BIMA
> array, it could be helpful to the ALMA LO engineering effort.
> 	Thanks.
> ...Bill
>