Cannot Integrate Bias Frames

[averen]

Hey All,
I'm following the tutorial below about creating master calibration frames:
http://pixinsight.com/tutorials/master-frames/en.html

I have the image integration module exactly like what is pictured and have 100 bias frames loaded.  Whenever I attempt to run the integration I receive an error about "Zero or insignificant signal detected: empty image?"

I used the same settings and stacked the darks just fine, but the bias are having issues.  These were taken with a Qhy9 Mono at -20C and the exposure length was as short as possible.  This happens with the 1x1 and 2x2 binned sets. 

Any idea what I could be doing wrong?

Thanks!
Jared

[sreilly]

I'm not familiar with your camera but I suspect that you may be taking the bias frames incorrectly. What software are you using to control the camera with? Both MaxIm and CCDSoft you tell what type of frame you are taking such as bias, dark, and flat. The bias frames are taken using the very shortest exposure possible for that camera. It may also be that the that information is written into the FITS header and PI isn't reading it.

These are just my guesses. I have no issues creating master frames taken with either of my SBIG cameras controlled by either program above.

[averen]

I think you might be correct.  The bias frames where taken a long while ago and they were taken with Nebulosity.  Looking at the info the mean ADU is 0, Max is 352 stdev is .2. 

I'm going to try taking them again here shortly and see if that takes care of the issue. 

Thanks!
Jared

[Nocturnal]

Either your offset is too low or somehow the FITS files aren't being read by PI as Neb intended them. In other words when Neb saves an ADU of 2000 perhaps PI sees a -20 or something like that. Neb can be used in Maxim compatible mode that generally works fine. See if Neb's stats match PI stats for the same file.

[averen]

The 2x2 bias frames just finished and they look much more "normal".  I'm not entirely sure what happened with the last set of bias frames as they were taken about a year ago. 

Anyways, the issue seems to be taken care of now.  I just used Nebulosity and Sequence Generator and captured another 100 of each.  Thankfully bias frames are fast!  I believe I took the dark data at the same time as well.  Hopefully the dark data is ok.  I don't really want to take that again but I may just for sanity sake.

Thanks!
Jared

[Nocturnal]

I recommend tracking down the problem so you know what went wrong and can avoid it next time. Plus others might benefit too

[averen]

Ya, that's a good idea.  Although I just hope to magically avoid it in the future!  Whatever it was it occurred while I was capturing the bias data and that was about 10 months ago so I'm not really sure what could have happened!  Like the error message said the images were pretty much empty.  Back then I was just learning the camera so it's pretty likely that I messed something up with the gain and offset whenever I went to acquire the bias data. 

Whatever the case it seems to be working now.  I'm redoing the darks too just for peace of mind.  That will take about 6 hours total for 25 darks at 5 minutes and another 25 at 10 minutes. 

When I took all the calibration data last time I hooked the camera directly up to my laptop which is not how I have it hooked up in the field.  It's connected through a hub and that goes into a box with power cables and likely all kinds of electrical noise.  I know it's not the best environment...but it's highly convenient which is what I need since I'm mostly a mobile observer/imager.  I've duplicated this now so that the bias and darks will more closely mirror the environment that the lights and flats will be taken in. 

Jared

[cfranks]

I too am having a problem with integrating Bias (and Dark) frames as per the same tutorial.  The bias frames shown here are today's integration of 30 frames together with a similar 30x integration from 18 months ago (different originals).  The one today is very coarse compared to the other and the Dark integration (not shown) is even worse.  I am using the current version of PI but the 18 month old was with the then current version (1.6.something?).  Has the Integration module been changed recently?  Due to equipment failure, I haven't done any processing for 8 months.

Sorry, I can't attach .jpg's of the Integrations as they are just convert to blank images.  The .fit files are 8Mb.

Charles

[Billhinge]

I had this problem today using a DSLR, looking at the iso 800 @ 1/4000s bias RAW frame in adobe lightroom I had to increase the exposure 3 or 4 stops to see something

I fixed the problem by creating 64 bit fits rather than 32 bit