[daip] CLCOR OPCODE='CLOC' bug?

[Leonia Kogan]

Walter,

I think both RATE and DCLOCK correct the phase identically,
although the units are different in CL table

DPHASE(t) = ...+ (RATE+DCLOCK)*t

So if CLCOR corrects DCLOCK it must not correct RATE although.
Otherwise the relevant phase correction will be doubled.

Leonia
Walter Brisken wrote:
> Hi AIPS,
> 
> I'm about to mail 5 VLBA users that should be notified about unwanted 
> clock jumps in some DiFX correlated data.  I was going to propose a fix 
> using CLCOR with OPCODE='CLOC' since all the needed parameters are easily 
> determined based on the correlator jobs that were run.  Note that this 
> application would make use of CLCORPRM(7) = 1 (the option to add 
> offset and rate corrections to the CL table).
> 
> 
> In my testing of the correction I noticed that the following table columns 
> got updated:
> 
> CLOCK (1 & 2)
> DCLOCK (1 & 2)
> REAL (1 & 2)
> IMAG (1 & 2)
> DELAY (1 & 2)
> 
> The adjusted values appear consistent with the values I supplied.
> 
> Conspicuously not changed were the RATE (1 & 2) columns.  My understanding 
> is that these columns should have CLCORPRM(1) (appropriately scaled from 
> ns/day to sec/sec) added to the existing RATE columns.
> 
> 
> Can someone more knowledgeable than I look into this?
> 
>  	Thanks
> 
>  	Walter
> 
> _______________________________________________
> Daip mailing list
> Daip at listmgr.cv.nrao.edu
> http://listmgr.cv.nrao.edu/mailman/listinfo/daip