Thoughts on which imaging rigs to concentrate on

[Demonperformer]

Gina, Are bias frames temperature-dependant? I thought one set of bias would apply to all subs, but maybe I'm wrong?

 

[Gina]

On looking at last night's captures I see that I have 193 120s OIII subs of the Heart and Soul Nebulae but about half a dozen frames may be marred by the shadow of my weather station mast.  Whether any others are marred I haven't checked as I only checked a few overall and narrowed it down to where the tracking stopped due to the mount hitting its end stop.  With luck I may have well over 6 hours of OIII data

[Gina]

7 hours ago, Demonperformer said:

Gina, Are bias frames temperature-dependant? I thought one set of bias would apply to all subs, but maybe I'm wrong?

I don't know, I'll look it up.

[Gina]

Opinions seem to vary from no to yes with only for flats in between.  I'm still reading various info on Google...

[russellhq]

44 minutes ago, Gina said:

Opinions seem to vary from no to yes with only for flats in between.  I'm still reading various info on Google...

Take a set of cold ones then a set of room temperature, and compare the 2. Probably quicker than googling for consensus

[Gina]

That's a good idea - thanks

[Gina]

Just taking a set of flats for the latest setup.  Here's a sub, histogram stretched black to white.  There are dust bunnies even if they didn't show up in the light subs.

[Gina]

Taken biases and darks for calibration - same temperature.  Dark with same exposure.  All with gain of 53.  Flats were taken with double layer of white T shirt over end of dew shield with a plain grey misty sky and camera pointing towards Polaris.  3nm OIII filter and focus untouched from last night.

Statistics for a single frame selected at random :-

1. Flat - black 29377 white 40178
2. Dark - black 16 white 367
3. Bias - black 16 white 398

Doesn't look like the darks and biases would be very significant but I'll use them anyway.

Edited November 1, 2016 by Gina

[Gina]

As a test I shall take a few bias frames at ambient temperature and compare.  I'll also see how low the temperature will go and do some at that.  I can also test at my usual lights imaging gain of 500.

[Gina]

Two bias subs compared for temperature difference.  One at -27°C and the other at +7°C.  Both with a gain setting of 501 = 50dB.  At this gain it is clear that bias frames are essential for calibration.

1. -27°C - black 368  white 15392
2. +7°C - black 352 white 14240

The result is that temperature does not change the bias frames significantly.  In fact if there were to be any difference I would expect the warmer set to show brighter pixels whereas these two are the opposite way round.  Clearly the difference is due to random variation.

Conclusion :- Temperature does NOT affect bias data for this camera.  Gain most definitely does (as expected).

[Gina]

I'm now capturing darks for last night's 120m light subs.  The ambient temperature has dropped since last night and I can easily reach the -25°C to -27°C temperatures of last night's OIII subs.  Set-point = -26°C.

[Gina]

Here are the two histograms for the bias samples at -27°C and +7°C.

[Gina]

Weather doesn't look at all promising for tonight but at least I have plenty of data to play with

Edited November 1, 2016 by Gina

[Mick J]

Can you list the software you are using for capture Gina (I may have missed it in here somewhere ),  it's interesting watching your progress as usual and I think will help others with this camera.

Edit:  Gina I have linked the last couple of pages into this thread

Edited November 1, 2016 by Mick J

[Gina]

I'm using APT (AstroPhotography Tool) for capture and SharpCap for framing and focussing.  I run these on a Windows 7 laptop.  For looking at the subs to check all is well (or not) I connect to the laptop in the observatory with TeamViewer 11 and copy files to my Win 7 desktop PC indoors.  FITS Liberator is used to convert FITS files to TIFF and then Photoshop to view these, do a bit of processing and save as PNG for upload here when appropriate

I'm planning to migrate to Linux and the RPi for control and KStars/Ekos client and INDI for the server on the RPi but ATM I'm struggling with getting all the software set up.

Edited November 1, 2016 by Gina

[feilimb]

Just now, Gina said:

I'm planning to migrate to Linux and the RPi for control and KStars/Ekos client and INDI for the server on the RPi but ATM I'm struggling with getting all the software set up.

Hi Gina

FYI KStars / Ekos 2.7 is now available on multiple platforms, including Windows natively. I have setup Indiserver myself recently on an Odroid U2 (an ARM based device similar in ways to the Raspberry Pi) but it should now be possible to use the 'client side' ie. Kstars & Ekos on Windows.

Details of the release are here:

http://knro.blogspot.ie/2016/10/kstars-270-is-released.html 

Edited November 1, 2016 by feilimb
Typo

[Gina]

That's interesting - could be useful though my plan is to move away from Windows where possible.  Thank you

Edited November 1, 2016 by Gina

[Gina]

OK so the required calibration files are :-

1. One master bias per gain value. 
2. One master flat per optical system but may need renewing if the rig is dismantled and re-assembled.
3. Master darks needed for all permutations of exposure, gain and temperature.

[Gina]

Right then... data processing... using PixInsight BPP (Batch PreProcessing) to get a master flat for my present imaging rig. 

Required items :-

1. Flat frames exposure 0.1s, gain 50, temperature -15°C
2. Dark frames with exposure 0.1s, gain 50, temperature -15°C
3. Bias frames for gain of 50

It would also seem that PI needs some Lights to run BPP but since we don't want a master light from this process just a few lights from anywhere should do.  The required items above were capture in the same session this afternoon.  100 subs of each.

As a result of the above processing I shall have a master flat I can use with recent lights and also a master bias for a gain of 50 and a master dark for 0.1s exposure, gain 50 and temp -15°C.  This means that if I use the same settings next time I want a master flat I can use the master files for dark and bias and not need to re-capture subs.

What I shall do is make a folder for all the master files with the parameters in the file name.

Edited November 1, 2016 by Gina

[Gina]

Phew...  I think I'm beginning to get organised at last

[Gina]

Now processing biases, darks and flats as above plus about 10 frames from a random set of lights with BPP in PixInsight.

[Gina]

Finished   Now have the master flats file to use with my lights plus master dark and bias for next time I take flats.

Next process will be to take my OIII Heart and Soul lights plus the appropriate darks and biases I captured this afternoon.  The master dark and bias from the flats processing only apply to those exposure, gain and temperature parameters.

Edited November 1, 2016 by Gina

[Gina]

Before running the next BPP I need to copy the wanted files from the obsy laptop SD card.  I can also delete no longer wanted files, which includes all the bias, darks and flats files I've used to create the masters.  I can also delete all the calibrated files as they too are no longer required.  Also the master light is useless and can be deleted as lights were only processed to satisfy PI for producing the master flat and dark/bias files.  Keeping all these no longer wanted files was my mistake in the past and one reason I have filled my main HD.

The only wanted files from the flats BPP are master flat, master dark and master bias the latter two being useful for later flats.  The "source" subs amount to 10GB and likewise the calibrated versions also 10GB plus the calibrated lights and master light (amounting to 3.6GB - 11 lights plus master, all 32MB files).  In total 23.6GB of files can simply be deleted.  Care is needed not to delete the wrong files but these unwanted files must not be kept!

[Gina]

PixInsight BPP running and with 100 biases, 100 darks and 193 lights, this is going to take a while   Even with a master flat.

[Gina]

Done the biases now doing the darks.