[daip] DTSIM - AIPS

[Eric Greisen]

Alessandra Bertarini wrote:
> Hello Leonia,
> 
> thank you for the reply. In the mean time I also looked into the code and 
> found the 'ADNOIZ' subroutine. Since I was digging into the code I found 
> another strange 'feature'. The routine DTSIM.FOR calculates the 
> the full polarization noise flux (FLUXN) as follows:
> 
> FLUXN(1) = DAMPL1(1) * DAMPL2(1) / SQRT(2.0*BWIDTH*TINT)
> 
> DAMPL* is set and calculated in another subroutine called:
> DTCALS.FOR.
> 
> In DTCAL.FOR is written:
> 
> DAMPL Noise power level in Janskys (RCP,LCP)
> DAMPL Square root (DGAIN * DTSYS)
> DGAIN Point source sensitivity (RCP, LCP) (K/Jy)
> DTSYS System temperature (RCP, LCP) (K)
> 
> But this makes no sense, right? DGAIN * DTSYS is something which units are
> K^2/Jy...
> 
> Should not it be DTSYS/DGAIN [Jy] ?
> 
> I actually simulated the data putting in DTSIM control file 1/DPFU instead 
> of DPFU (as it is written in the help file) and then checked that the 
> fluctuations in flux produced were about the amount expected for that 
> given Tsys, DPFU, bandwidth and AP. 
> The two values agrees well if I use 1/DPFU. 
> 
> Btw about the DTSIM 'feature' of using only one IF instead of three, I 
> guess it is because in the code there is no do loop around the index that 
> is reading the IF number although the index is declared in the program.... 
> but I am no fortran expert, so I might be wrong. 
>  
> Cheers,
> Alessandra
> 
> 
>> Alessandra,
>>
>> Having looked into the DTSIM's codes I found subroutine ADNOIZ which produce
>> noisy  data....
>> I'll look more tomorrow
>>
>> Leonia
>>
>>
>> Alessandra Bertarini wrote:
>>> Hello Leonia,
>>>
>>>  I am trying to estimate the error on the D-term calculation on my VLBI
>>>  dataset and I was thinking to use DTSIM to simulate a UVData set since, to
>>>  my understanding, DTSIM is the only AIPS simulation task allowing for
>>>  D-terms.
>>>
>>>  Is there any way to add gaussian noise to the simulated set?
>>>
>>>  Btw, there might be a "feature" in DTSIM: I have generated a 1.6 Jy
>>>  unpolarized point source with D-terms causing flux in RL and LR using
>>>  three IFs. In IF1 everything looks reasonable (RL and LR are as expected,
>>>  apart from the absence of noise), but in IF2 and IF3 they are set to
>>>  zeros for the whole array. Is that "feature" expected or is it a bug?
>>>  Feel free to pass this e-mail to the "designated aip" :).
>>>
>>>  Thanks in advance.

GAIN is Jy/K.  DTSIM has received serious work in the last week and 
seems to be more useful than before.  The coordinates of standard EVLA 
and VLBA stations were updated (EVLA were very wrong) and concepts of 
default gain and Tsys were introduced for convenience.  The Explain file 
was improved significantly.

Eric Greisen