Warning: No correlation between the master dark and target frames (channel 0)

[Jevans]

Hi Juan. Here are some single subs. All were sampled with a QSI 583 WSG CCD binned 2x2 at -20c using Nebulosity 2.5.2 in 16-bit format. All subs measure, integrate and calibrate fine in Nebulosity 2.5.2, AIP4WIN V2 and CCDSoft 5.0.216 leaving no fixed, hot or pixels.

https://www.dropbox.com/s/4gpsgslz7flpwoe/NGC%202903%20-20c%202x2%20%205%20min%20Dark%202014_01_04_005_060654.fit

https://www.dropbox.com/s/a6l1m9gg4p8na7d/NGC%202903%20-20c%202x2%205%20min%20Blue%202014_01_06_007_034210.fit

https://www.dropbox.com/s/jqzpukyd513xbdv/NGC%202903%20-20c2x2%20blue%20flat%202014_01_04.fit

https://www.dropbox.com/s/ikthakwwtrhfy6u/NGC%206888%20-20c%202x2%20Bias%202013_05_02_123_055241.fit

Thanks for your willingness to help

[pfile]

as i suspected, the bias frame is f32… that's most likely the problem here.

rob

[mschuster]

Yes, the single bias you posted is in floating point format, and when read in PixInsight results in way too high statistics. The single dark on the other hand is 16-bit. All raws should be 16-bit.

Also, the FITS headers reports X and Y pixel size of 5.4 microns. This is wrong, it should be 10.8 corresponding to the 2x2 binning.

Also, the FITS EXPTIME reports a 5 sec dark exposure not 5 minutes. The dark url says 20 min, but it downloads as a 5 min file name.

Also, note this issue with some QSI hardware: when taking a long sequence of short exposures (e.g. bias), the TEC can drift due to a firmware issue. During these captures, double check temperature. If you see a drift add a 5 or 10 second pause or delay between exposures. My QSI suffers from this problem.

Regards,
Mike

[Jevans]

I guess I have a few issues to work through... I am aware of the X/Y pixel size in the header... It has been an issue with CCDInspector... I have also seen the TEC drift and will heed your advice there. Time to clean out the old bias library, capture some new subs and try again after I check my firmware version.

There is some TEC drift (.01c) in the bias and mismatched exposures in the dark subs which would definitely explain the inability to correlate.

Thanks

[Jevans]

mschuster, pfile - many thanks.

Images calibrate well now.

Appreciate your willingness to help.

Regards,

Jayme

[pfile]

no problemo, i'm glad it's working for you.

too many ins and outs in this hobby.

rob