Документ взят из кэша поисковой машины. Адрес оригинального документа : http://www.naic.edu/~emuller/EALFA/
Дата изменения: Thu Oct 14 00:54:04 2004
Дата индексирования: Tue Oct 2 00:04:36 2012
Кодировка:
Поисковые слова: чоеыойе рмбоефщ
Leapfrog precursor test 
ALFA - 'dishspot' (pre-cursor for 'leapfrog' mode - we've called it 'dishspot' because it observes a constant range of positions on the
dish, we will re-name this eventually....).
This mode:
1) Moves to the requested RA and Dec, and remembers the AZ and EL.
2) The telescope tracks a source at sidereal rate, for the user-requested
length of time.
3) The telescope moves back to the original AZ and EL, calculates the new
RA, and loops back to 2)

For proper operation, the current mode needs to predict the slew time to the next point. In
theory, this should be accurate to less than 10 seconds

Data taken at around midday on the 16th May (the sun generated
some problems)
10 positions measured, 300 s each pointing.
4096 channels, 100MHz Bandwidth, centred on 4120 MHz.
These data here are for the first polarisation on the first pixel.
(there are no real reasons why any other polarisations on any other pixels
would be much more or less stable.)


For Dishspot:
        Start time (lst hrs):1.6805090
        Finish time (lst hrs):2.6052562
        Number of tracks: 10
        Time per track (min):5
        Total observing time (mins): 55.4848
        Total integration time (mins): 50
	Mean Slew time for between each step (sec)=32.9088
        observing efficiency for dishspot:90%

NB: For Regular ononff, Wait 1 min at each new 'off'
        Total observing time=55.4848+Ntracks
        observing efficiency for onoff:76%
        

Estimates for leapfrog
(expected wait at each new 'on' is MAX 10 sec)
        total observing time=55.4848+Ntracks*0.6
        observing efficiency for leapfrog:81%

Efficiency will improve for longer integration times.

More info on dishspot: 10 positions measured, 300 s each pointing.
RA (hrs) Dec(deg)
1.38993 34.4629
1.47482 34.4674
1.55956 34.4707
1.64392 34.4740
1.72857 34.4767
1.81321 34.4789
1.89806 34.4803
1.98258 34.4808
2.06706 34.4805
2.15136 34.4802

This data is from pixel 0 (centre), pol 0

click here for pixel 1

A mean bandpass is calculated for each of the 10 integrations, and subtracted. The mean bandpass does not to vary very much in shape through each of the sets, only in offset.

mean bandpass (click here for ps copy)

stdev of bandpasses (stdev as a function of channel) (click here for ps copy)

This graph has a number of offset lines, showing increasing temperature uniformly across the band, many lines overlap indicating some stability. As these tests were done during the day, is is impossible to assess the timescale of the stability.
The variation of the mean temperature for each band, as a function of local time is shown below: (click here for ps copy)

The bandpass-subtracted data for each of the ten sets are shown below: (click here for a pdf hardcopy)

Variation of total power of mean bandpassfor (for each of the 10, 300sec integrations) is around 10%.

click here for this image as ps.
click here for this image as ps.
Variations became large after the first 3 sets (i.e. 15 minutes - probably due to the sun)