Документ взят из кэша поисковой машины. Адрес оригинального документа : http://www.naic.edu/~tghosh/a2186/shuk_plots/b0_54675.desc Дата изменения: Mon Dec 1 18:39:55 2008 Дата индексирования: Mon Apr 11 11:45:18 2016 Кодировка: Поисковые слова: южная атлантическая бопнбмйс

Hi Tapasi,
Please find attached the plots below for a few of the days that I
retrieved using my own independent code. They show the same problems as
the quicklook plots (I have also attached a clean day 54697). Another
peculiarity I noted was that the break and shift in lambdas are not same
for all beams (assuming there was one common error we would expect lambdas
to shift the same way for all beams) so it seems Desh's code is
calculating erroneous positions that depend on the beam number in some
way. I am attaching only the beam 0 plots. Since the other beams use
offset in AZ and ZA to plug into a formula to calculate the positions I
did not do that. Those values have been different for the pilot region and
A2184 and since Jeff left I dont know which values to use. I guess you can
tell me the correct values to use for when we talk in the telecon
tommorow.

I also got another thought so I plotted the time series data as well and
you can see that on bad days there are break shifts in time series data as
well which seem to exactly correspond to the break shifts in the pointing
data. So something is going wrong somewhere in some calculation which
corrupts the RA DEC as well as AST values.

Hope this helps.
If you want me to I can try to take a look at Desh's code as well.

Sukhpreet


> Hi Shukpreet,
>
> I know we aught to complete the first stage reduction to meet our
> commitment to ZOA group. However, the position problem still lingers on.
> Mean while the data has been backed up to make room for GALFACTS main. I
> am going to ask Arun (in the
> next mail) to restore the 8 days of the D field that we still need to
> reduce. Meanwhile, I have a big favor to ask.
> If you can read in the position info of all the beams in any stand-alone
> program (IDL ?) from the .spec files, and
> plot just one days data, (say MJD 54698), and establish that the spec
> files didn't have wrong position info, I will feel
> more confident. I hope it will not be a big deal for you. So, while Arun
> gets the data back on disk, it will be nice if we could
> resolve the matter.
>
> nay thanks,
> Tapasi
>
>