CACAL Does Not Work

Before running CACAL, you should always check that the phases, gains and delays - especially the delays, are stable, i.e., the noise on the delays should be less than a nanosecond at 128MHz bandwidth and less than ten nanoseconds at 4MHz bandwidth. If this is not so, do not run CACAL.

It is possible that you have a hardware problem, however, if the amplitudes are reasonable, the problem is more likely to be that your source isn't strong enough for running the array calibration. Possible solutions include:

• finding a stronger calibrator (at 20 and 13 cm PKS0407$-$658 is stronger than PKS0823$-$500, however, PKS0407$-$658 is resolved at higher frequencies);
• using less channels (but the same bandwidth) for your calibration - this means using a different correlator configuration just for your calibration (e.g use FULL_4_32-128 instead of FULL_4_1024-128).

(Remember to use the correct correlator configuration for you amplitude and bandpass calibration.)

If you realise that running CACAL has corrupted your data, it is possible to return to the default delays (and phases and gains) by running cacal, choosing I(nteractive) mode and then to R(eset) calibration data.

A failed noise diode may also result in a failure of CACAL to calibrate.

CHECK the /dsd and /sdo pages in CAMON for any values in reverse-video (i.e., values out of range).
If so halt observing:

	CAOBS$>$ stop
	CAOBS$>$init		To restart any failed diodes.

If this doesn't work, then get the Duty Astronomer or local staff.

Also,

• CHECK that your chosen bandwidth matches your correlator configuration;
• CHECK that Tsys enabled in the CAOBS window. If not CAOBS$>$ enable tsys
• CHECK the correlator programs (Section 7.6).