[mmaimcal] Re: ALMA total power integration times

[Min Yun]

Hi Fritz,

I wish I could give you a cut and dry answer, but the situation 
might be a little more complicated.  The main scientific use of
the total power measurement would be on-the-fly (OTF) mapping, and 
there we do not really want to average anything.  Ideally what one
wants to do is record the total power output from each antenna
along with the exact AZ and EL encoder positions, basically as fast
as you can.  And the data set accumulated this way will be
averaged only during the re-gridding of the data onto the final
image -- maybe we will come up with a new software that combines
the mosaic interferometer data and this sort of OTF data all at
once and produce a nice image someday.  

Now this is the most common scientific observing mode I can think of, 
and I don't see any immediate problem with the situation you
describe.  There might be other odd
situations where people might do something that contradicts
what I just said.  Pulsar observation or VLBI may have requirements that
are pushing the limit -- I am not an expert on those areas.
There may be engineering work that may be testing the limit
you outline below.  I forward your question and my answer to the
rest of the imcal group in case someone can think of a scenario
where this may pose a real problem.  Cheers!




					-- Min
					



> X-Sender: fstauffe at zia.aoc.nrao.edu
> Date: Thu, 17 Aug 2000 12:28:50 -0600
> To: myun
> From: Fritz Stauffer <fstauffe at cv3.cv.nrao.edu>
> Subject: ALMA total power integration times
> Mime-Version: 1.0
> 
> Min,
> 
> I am looking at the software to control the total power data collection.
> I have some questions about integrating total power data.
> 
> The total power data will be read out every 2 milliseconds.  That's how
> the hardware works.  But, user integrations could be seconds long or
> longer.  Is there anything wrong with just summing the 2 millisecond readouts
> to produce the longer integration time?  If summing is okay, then what is a
> maximum total power integration time?  Summing 16 bit data every 2 
milliseconds
> would overflow a 32 bit word in 65,000 sums or about 130 seconds.
> 
> Thank you,
>   fritz stauffer
>