Error with polcal

[sourabhpaul]

Hi,

I am trying polcal for the first time with caracal pipeline. While the crosscal steps completed without errors, I am getting an error (please see attached logs) in polcal where the process fails due to a missing gaintable. The calibrators used are:

gcal: ['J1058+0133']
fcal: ['J0408-6545']
bpcal: ['J0408-6545']
xcal: ['J1130-1449']

Could someone review this and suggest a solution? Thanks!

1680626188_sdp_l0-obsinfo.txt
config.txt
log-caracal.txt
log-polcal-gain_xcal-20250201-202003.txt

[o-smirnov]

The previous setmodel step looks correct, but this is a little suspicious:

# 2025-02-02 00:37:31   INFO    Calibrater::solve         Found good G Jones solutions in 0 solution intervals.
# 2025-02-02 00:37:31   WARN    Calibrater::solve       Spectral window(s) 0,
# 2025-02-02 00:37:31   WARN    Calibrater::solve+        could not be solved due to missing (pre-)calibration
# 2025-02-02 00:37:31   WARN    Calibrater::solve+          in one or more of the specified tables.
# 2025-02-02 00:37:31   WARN    Calibrater::solve+          Please check your results carefully!
# 2025-02-02 00:37:31   INFO    Calibrater::solve       No output calibration table written.
# 2025-02-02 00:37:31   INFO    calibrater::solve       Finished solving.

Can you check and/or plot the calibration tables that you're applying?

[sourabhpaul]

There were three calibration tables being applied in this gaincal task, attached are the plots.

2025-02-02 00:30:21	INFO	gaincal::::+	##### Begin Task: gaincal            #####
2025-02-02 00:30:21	INFO	gaincal::::	gaincal(vis="/stimela_mount/msdir/1680626188_sdp_l0-cal.ms",caltable="/stimela_mount/output/crosscal-1680626188_sdp_l0-1pcal.Gpol1",field="J1130-1449",spw="",intent="",
2025-02-02 00:30:21	INFO	gaincal::::+	        selectdata=True,timerange="",uvrange=">150",antenna="",scan="",
2025-02-02 00:30:21	INFO	gaincal::::+	        observation="",msselect="",solint="60s",combine="",preavg=-1.0,
2025-02-02 00:30:21	INFO	gaincal::::+	        refant="m008",refantmode="flex",minblperant=4,minsnr=3.0,solnorm=False,
2025-02-02 00:30:21	INFO	gaincal::::+	        normtype="mean",gaintype="G",smodel=[],calmode="p",solmode="",
2025-02-02 00:30:21	INFO	gaincal::::+	        rmsthresh=[],corrdepflags=False,append=False,splinetime=3600.0,npointaver=3,
2025-02-02 00:30:21	INFO	gaincal::::+	        phasewrap=180.0,docallib=False,callib="",gaintable=['/stimela_mount/output/crosscal-1680626188_sdp_l0-1gc1_primary.B1', '/stimela_mount/output/crosscal-1680626188_sdp_l0-1gc1_secondary.K1', '/stimela_mount/output/crosscal-1680626188_sdp_l0-1gc1_secondary.F0'],gainfield=['nearest', 'nearest', 'nearest'],
2025-02-02 00:30:21	INFO	gaincal::::+	        interp=['linear', 'nearest', 'linear'],spwmap=[],parang=True)

[o-smirnov]

Looks reasonable. Also check for other obvious things like the reference antenna not being flagged in the polcal scan.

[francescaLoi]

Can you also check the flagging fraction on J1130-1449? Some plots of all the correlations would be useful too, you can use the inspect worker for this purpose. I suspect that there are no data there. If this is true, setting extendflags: False could help. Can you try and let me know?

[sourabhpaul]

Flagging looks okay to me. I have submitted a job with extendflags: False, and will update when it finishes.

[sourabhpaul]

Hi @francescaLoi, got the same error with extendflags: False .