[daip] Re: APCAL bug

[Jim Ulvestad]

Hi Amy & Lorant,

I figured it out.  The sensitivity entered for GBT is about 1.7E+00, a 
factor
of almost 20 more than a VLBA antenna, as expected.

HOWEVER, the gain curve is wrong.  The gain column of the GC table
includes polynomial coefficients for each antenna, which are interpreted
as

GAIN = SENS *(A0 + A1*sin(el) + A2*sin^2(el)+...)  or something like
that.  For all the VLBA antennas, A0=1.0E+00, and all other coefficients
are zero.  But for the GBT, A0=0.0E+00 !!!!

Therefore, at some point, somebody made an incorrect entry in the
polynomial coefficients for the GBT gain curve, at least at S band.
Since the leading term is zero, the total gain is very low.

At X band, the first coefficient is 9.590E-01, almost unity, and
therefore appears to be correct.

Cheers,

Jim

Amy Mioduszewski wrote:

>Hi Jim,
>
>It looks like there is something wrong with the TY table, the
>Tsys for the left and right circular are different by a factor
>of ~5.6, and since the gain curve is identical for the left
>and right this give different amplitudes.  The left looks like it
>is correct, it ends up giving gain of about 50 for GB as opposed
>to about 20 for the other antennas and the RCP of GB.  I think it
>is a coincidence that the gains for the VLBA antennas and GB RCP
>are similar.  Do a POSSM on stokes 'LL' and see if it makes more
>sense, I can't try this b/c I don't have the data only the tables.
>Lorant checked the Tcals in the monitor data and the RCP and LCP
>Tcals are different for GB, so it looks like a problem with the
>monitor data.
>
>Cheers,
>
>Amy
>  
>