Got my DSLR in the fridge

[Richie092]

I had a go at imaging M81/M82 the other night and by the time I got set up and actually imaging it was already late. I got around 20 5min subs but no darks/bias so am trying to address that now.

I have the camera in a camping fridge in the living room at the moment that is down to around 5 degrees, a couple higher than the night i was imaging. I am running darks at around 25 degrees at the moment so still not as low as the 22 I was shooting on the night. How critical is the 3 degree difference?

If it is critical i can leave it in there for another hour or so and hopefully get it down to 22 if I really need to. What do you think?

All i need to do now is order the flat field panel for my lights. Have been waiting on the FLO one but it still looks a way off.

[Bizibilder]

You should be OK with darks that are within +- 5°C of your lights. Ideally the temperature should be the same but you can get away with it! You need to be more careful the warmer the camera was for your lights - at 20°C you may have to be within a couple of degrees either side.

[RikM]

I used to use fridge darks with my DSLR. They always worked out fine for me, just take plenty of them.

[kirkster501]

I though of doing this fridge trick too. Sure I know flats need to be done in situ with the scope. But man, I have not got time to be doing darks in situ as well! So I leave DSLR with pastic cover on connected to computer in garage and go to bed and let it fire off 30 darks or so. Its within a couple of degrees of outside temp so should be OK ?

[Bizibilder]

'Tis a good job for a cloudy night. Build up a library and you will soon be able to pick and choose the best to use. i still use darks taken over a year ago - but you have to remember that camera sensors change with time so you may need to renew your library from time to time.

[Richie092]

All good advice, thank you. I have now built up a large run of about 100 darks so will try to pick a couple of dozen around the temp my lights were.

It takes a lot of explaining to the other half when asked why the fridge is clunking!

[Astrosurf]

Excellent thread. Thank you!

Alexxx

[ollypenrice]

Have any of the DSLR users tried a Defect Map (AKA Bad Pixel Map) I wonder?

Olly

[swag72]

...... All i need to do now is order the flat field panel for my lights. Have been waiting on the FLO one but it still looks a way off.

Why not make your own? You need an EL panel from someone such as Earlsmann, then some 3mm opaque perspex. I bought mine off of EBay. Then we put it together into a kind of picture frame. Sits perfectly on the scope and takes excellent flats. No more waiting around!!

[kirkster501]

Why not make your own? You need an EL panel from someone such as Earlsmann, then some 3mm opaque perspex. I bought mine off of EBay. Then we put it together into a kind of picture frame. Sits perfectly on the scope and takes excellent flats. No more waiting around!!

Great idea! Gonna make one me thinks. Getting fed up of the iPad method - Madame kirkster501 is always on that - and it is not big enough for the RC 8 scope.

I may use a cheap picture frame from IKEA, Wilkos or somewhere. Buy the picture which will have clear acrylic at the front, dump the picture inside, cut the panel to fit and Bob's your uncle...

[Richie092]

Why not make your own? You need an EL panel from someone such as Earlsmann, then some 3mm opaque perspex. I bought mine off of EBay. Then we put it together into a kind of picture frame. Sits perfectly on the scope and takes excellent flats. No more waiting around!!

Not a bad idea actually. I might well have a go at this. It looks like it will sit nicely on the scope.

[Richie092]

Have any of the DSLR users tried a Defect Map (AKA Bad Pixel Map) I wonder?

Olly

Hi Olly

No, I haven't tried that. I just googled it and it looks like it needs more than a 2 minute read. At the stage I am at so far I think I will just familiarize myself with the basics, see what results I get and go on from there.

I have been into this hobby since Christmas and as yet not got a pic worth posting but am hoping to change all that if tonight is clear :-)

[IanL]

I have used a similar process:

- Small mini-fridge (peltier-type cooler), camera on laptop control with external power supply. You can't close the door fully because of the cables so I seal them up using duct tape after putting the camera in.

- It didn't get cold enough so I started using four freezer packs as well which gets things nice and frosty for up to 24 hours.

- Pre-cool the camera for a couple of hours.

- Then I use APT to take darks on a 10 minute exposure and 8 to 12 minute pause on a repeating cycle at the correct ISO (400 in my case).

- With a bit of tweaking of the pause I find I can get darks of a consistent 8, 9 or 11C temperature for a 12 hour period, after which it slowly drifts up by a degree every couple of hours until I stop (never had a single one 10C for some reason).

I then stack these in PixInsight to make a master dark. PixInsight calibration does dark frame scaling, whereby it uses a statistical method to match the noise in the light frame to the noise in the dark frame. This seems to work pretty well and means I don't have to match temperatures between lights and dark

I did a bunch of tests using using temperature matched master darks of between 10 and 30 frames on a light frame stack of 50 images (depending on how many darks I had for each temp). I then tried a 50 frame master dark at 9C on all light s with dark scaling turned on, and finally I used 109 frame master dark made from the 8, 9 and 11C exposures together with dark scaling.

I ran noise evaluation routines on a sample of each set of individual calibrated frames and the final stack for each test. Bottom line was that for an EXIF temperature range of 9 to 11C 109 frame master dark with dark scaling I got both the best individual frames and the best full stack. If you don't use PI for stacking you could easily perform similar tests with your own software; I suspect it would be much more important to match temperatures between darks and lights if you don't have dark scaling available.

PI has a noise evaluation script, but a simple way to evaluate noise is to calculate the standard deviation of either the whole image or the same part of the background of the image if you prefer. Provided you keep all other things equal and only vary the darks from test to test, standard deviation should give you a reasonable approximation of relative amounts the noise in your test images (smaller values are better than bigger ones so just find the process that gets you the smallest StdDev).

Don't forget Bias is also important. I am now using a master bias of 330 frames (takes a while to stack that one) and it did improve the noise considerable over my original 50 frame bias.

[Richie092]

Thank you for taking the time to post that. Very informative.

Good advice re the pause. My temps were slowly rising I bet a 5 min pause would stop that.

I can't tell you how tempted I am at the moment to put the DSLR bag in the bag and order the 314L+ set up!

[RikM]

I can't tell you how tempted I am at the moment to put the DSLR bag in the bag and order the 314L+ set up!

That's pretty much what I did. Canon 1000D --> SXV-H9 mono + filters and it was the best move ever. I love the CCD

[Richie092]

That looks like a lovely camera.

Going off topic here but I have read and re read making every photon count. I love it but is there an equivalent book on ccd imaging/processing. Something that talks in our language with decent step by step processes?

[swag72]

Getting shot of my DSLR and buying the 314L+ was the best move I ever made in imaging. Go on, you know you want to !!

Sadly no processing book that I've ever found that explains everything in an easy to understand fashion seems to exist. If you find it, let me know!!

[Richie092]

Oh no, as much as I wanted to hear that... I really didn't want to hear that!

I have it all in my "wish list" on my FLO account but so far have held myself back from pressing the add to basket. I have spent days reading every thread on here that even mentions the 314L+

Maybe Steve Richards has got another book in him?

[Tim]

Here you go:

1) Put CCD on telescope

2) Focus it

3) take exposure

4) take more

5) Stack them up

6) enjoy learning to process them.

There, that should get you going

[Richie092]

Here you go:

1) Put CCD on telescope

2) Focus it

3) take exposure

4) take more

5) Stack them up

6) enjoy learning to process them.

There, that should get you going

Lol, ok that's as good a starting point as anyone needs I guess.

[steppenwolf]

Maybe Steve Richards has got another book in him?

He has and he is on page 69 of 200+!

[swag72]

He has and he is on page 69 of 200+!

That's excellent news Steve as it was only page 38 last year!!

[Richie092]

Cool. Can we have it in installments? :-)