[daip] 31DEC13, NOIFS problem

[Eric Greisen]

On 06/30/2016 06:23 AM, 정다운 wrote:
> To do FPT(Frequency Phase Transfer), phase calibration, I need to use
> SNCOR (OPCODE 'XFER').
> When I use SNCOR (OPCODE 'XFER') to transfer low frequency phase to high
> frequency, I'm concern that it will make a few problems if I do not use
> NOIFS.
> If I transfer 21.55 GHz phase to 43.10GHz and I have 4 IFs with 16MHz
> bandwidth, IFs will be like this.
>
> IF1                 IF2                   IF3                IF4
> 21.55GHz    21.566GHz    21.582GHz   21.598GHz  (real)
> 43.10GHz     43.132GHz    43.164GHz   43.196GHz (transferred)
>
> 43.10GHz    43.116GHz    43.132GHz   43.148GHz   (but, real)
>
> I think that there will be differences between real IF frequencies and
> transferred IF frequencies.
>
> However, If I use NOIFS, I will have an IF with 64MHz bandwidth.
> IF1
> 21.55 GHz  (real)
> 43.10 GHz  (transferred)
>
> 43.10 GHz (real)
>
> I think there will be no problem with IF frequencies, so I concluded
> that I needed to use NOIFS.
>
> You can access my data by Filezilla
>
> My IP address :
> 165.132.28.151
> port  : 22
> data :
> /home/kvnyonsei3/KVNFPT/K13035A.FITS

This tool is unfamiliar to me but we tried it and the connection was 
refused.  Apparently I actually have to have an account set up or 
"anonymous" has to be allowed in without a password - which is not a 
wise thing to allow.  Can you find some other way to make the data 
available?

I do not understand your concerns about the frequencies since I do not 
know the detailed frequency structure of your data.  I do know that 
phase transfer is used with the Korean VLB data with very considerable 
success using a mix of AIPS and ParselTongue in a pipeline.  How does 
this differ from what you need?

I have taken a look at SNCOR and it supports BIF and EIF.  Can you not 
run it 4 times changing the PHASPRM(29) and (30) appropriately for each
IF?  This would be more accurate (??) than doing a single phase for all 
IFs.

I would still like to find out what is wrong with NOIFS - have you 
looked at the output of NOIFS with a more preise tool like PRTUV?

Eric Greisen