[mmaimcal] Re: [alma-sw-ssr] Band switching times revisited...

[Darrel Emerson]

I'm sorry that the wording in my note was ambiguous.

   Altogether, there will be 3 bands ready for operation.  One will be
in use, typically one will be the 90 GHz band for pointing, and there's
one other.  So, it would be possible to observe at one high frequency,
switch in 1.5 seconds to another high frequency band, and still have the
90 GHz pointing system ready.

   It MAY be that we can do better than that.  It's all a matter of
cryogenic loading.  We won't know if we can do better until after
receivers are built.

		Cheers,
			Darrel.

Mark Holdaway wrote:
> 
> I think this is quite bad.
> 
> My understanding is that there are two hot frequency bands at any time
> (though the wording at one place stated "two standby frequency bands"),
> and one of them is usually 90 GHz as it is required for pointing and
> fast switching phase calibration.
> 
> If my understanding is correct, this is clearly bad for scheduling,
> dynamic or otherwise.
> 
> If you are doing a project at 230 GHz and you then need to change to 650
> GHz  (ie, if the wether gets good, or if you have just scheduled it that
> way), you have something like 15 minutes while 650 is warming up and 230
> is powered down, so in this 15 minutes of 650 GHz weather your only
> choice is to observe at 90 GHz, clearly a waste of 650-GHz-quality time.
> 
> I think this overhead will put strong limitations on how often ALMA
> changes frequencies and will eat away some part of the efficiency gains
> which are made with dynamic scheduling, which seeks to be very agile.
> Furthermore, it would likely impact low elevation and very demanding
> observations, both of which require different atmospheric conditions
> at a given frequency than the generic high elevation observations
> at the same frequency, and therefore would likely be observed during
> the same conditions that might be used for higher frequency generic
> zenith observations.
> 
> If there is a current observing band plus two others as standbys,
> I think we can live with that.  I assume these specifications are
> well ossified?
> 
>         -Mark