On-The-Fly (OTF) Discussion


Contents


Total Power (Single Dish) OTF:

From the ALMA Production Antenna technical specifications...
5.4.2 On The Fly Total Power Mapping

During on the fly total power mapping the antenna will scan across a
target source ranging from one arcminute to one degree in size, then turn
around and scan back across the source, at speed up to 0.5 deg/s.

For design purposes, in on the fly total power mapping profile to be
considered here, the antenna shall scan at a rate of 0.5 deg/s on the saky
across a target source on one degree in size, then turn around at a
distance of 1 arcmin, settle within 0.8 sec time and scan back across the
source in the opposite direction with a 2 arcsec RMS accuracy within
primary pointing conditions (as provided by the encoders reading,
corrected with any metrology).  This performance shall be attained up to
60 degrees.

Problems with the way this specification is written...

  • Why is OTF restricted to EL<60 degrees?
Mark Holdaway: OK, so it starts to break down above 60 deg -- that probably means we spin the motors faster, and it takes longer to turn around, but we should still be able to do 0.5 deg/s on the sky up to quite high -- like 75 or 80 deg? Perhaps the pointing accuracy wouldn't be as good, but as you know, for TP, it is all about pointing KNOWLEDGE rather than pointing at where you SPECIFIED. The 1 arcminute and 0.8 s turnaround seem fine, and contain the equivalent information about the acceleration.

Jeff Mangum: This is where the screwup has happened. To an antenna vendor, this says that the TPOTF and MOTF modes only need to be possible at EL<60. Also, it is not strictly true that we only need to know where the antenna was pointed during these scans. If you don't at least Nyquist sample, you can't grid the data properly. So it is better to say that you must at least Nyquist sample while at least knowing where you are positioned.


Mosaic (Interferometric) OTF:

From the ALMA Production Antenna technical specifications...
5.4.3  On The Fly Interferometric Mosaicing

In on-the-fly interferometric mosaicing, the antenna will scan at a rate 
of up to 0.05 deg/s on the sky across a target source, ranging from one
arcmin to one degree in size and then turn around and scan back across the 
source in the opposite direction.

For design purposes it shall be considered that during the scans across 
the source the antenna shall follow the commanded path to within 1 arcsec 
RMS, within primary operating conditions (as provided by the encoders 
reading, corrected with any metrology).

Problems with the way this specification is written...

  • Is 0.05 deg/s fast enough?
Mark Holdaway: OK -- some numbers. If you have a source 1 deg square, at 300 GHz you want to sample about once every 7", so you need 514 scans across the source, and each scan will take 20 s -- so the whole observation will take 2.86 hours. This sounds OK -- though we would rather be able to go faster and repeat -- WE PROBABLY WILL BE ABLE TO GO FASTER, though perhaps with larger pointing errors -- we'll find out, I'm sure.

  • Is the 1 arcsec pointing error correct?
Mark Holdaway: The 1 arcsec pointing error doesn't sount correct -- why not the 0.6 arcsec pointing error specification? That applies to OTF mosaics too, or at least it SHOULD.

Jeff Mangum: I agree, but I think that I just found from where this has crept in. The prototype antenna technical specification specified 1 arcsec RSS pointing accuracy. This might be a hard one to change. How important is this?

-- JeffMangum - 30 Sep 2004
Topic revision: r1 - 2004-09-30, JeffMangum
This site is powered by FoswikiCopyright © by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding NRAO Public Wiki? Send feedback