Darks and stuff

[piprees]

As a general rule, the number of darks that I would take would have been about the square root of the number of lights, 6 for 36, 8 for 64 etc. Since I've been guiding, my exposure rate has gone up from a max of 120 secs to usually above 300 with a 30 sec pause between exposures. I now seem to be getting far more noise that I can't seem to get rid of. I'm thinking it's possibly a problem with the camera, (Canon 40D), but could it be that I need a greater proportion of darks with these higher exposure times? How many darks per lights do other folk do at 300+secs?

Kind regards,

Phil

 

[ecuador]

You don't have to be stingy with darks, especially with a Canon that records the temperature of the sensor so you can build a dark library. Mostly during cloudy nights I make darks with the exposure and ISO settings I use for lights and put them in separate folders, then when I get the lights I start the program Dark Master, point it to the lights and the darks and it gives me a DSS list containing the appropriate darks based on temperature, usually dozens of darks. I haven't heard about going for the square root of lights before, that sounds like way too few.

[piprees]

Thanks Ecuador, I'll give it a try and use more.

What's Dark Master? Sounds good.

P.

 

[ollypenrice]

I'm not a DSLR imager but I don't think the recorded temperature is at all reliable. On long exposures the chip temperature varies considerably, so what is recorded? Tony Hallas and many other good DSLR imagers don't use darks at all. They use a master bias as a dark and run a large 12 pixel dither between subs. Even using set point cooled CCD I have long since given up on darks. I use a master bias as a dark and a bad pixel map to get cleaner results than I did with darks.

Olly

[Pompey Monkey]

3 hours ago, ollypenrice said:

I'm not a DSLR imager but I don't think the recorded temperature is at all reliable. On long exposures the chip temperature varies considerably, so what is recorded? Tony Hallas and many other good DSLR imagers don't use darks at all. They use a master bias as a dark and run a large 12 pixel dither between subs. Even using set point cooled CCD I have long since given up on darks. I use a master bias as a dark and a bad pixel map to get cleaner results than I did with darks.

Olly

I completely agree with Olly on this point: A good master bias, dithering (with a statistical outlier rejection algorithm in stacking), and a bad pixel map gets my system noise well under the sky noise threshold where I live. I'm not even sure that the bad pixel map does an awful lot of extra good once dithering is employed.

Now my weak point is my (lack of) processing skills ... 

[ollypenrice]

9 hours ago, Pompey Monkey said:

I completely agree with Olly on this point: A good master bias, dithering (with a statistical outlier rejection algorithm in stacking), and a bad pixel map gets my system noise well under the sky noise threshold where I live. I'm not even sure that the bad pixel map does an awful lot of extra good once dithering is employed.

Now my weak point is my (lack of) processing skills ... 

Indeed I use the BPM out of habit, too. I'm not sure it does much! You are right to point out that dither goes with outlier rejection stacking, AKA Sigma Clip.

Olly

[piprees]

Thanks guys,

however...

3 hours ago, ollypenrice said:

A good master bias, dithering (with a statistical outlier rejection algorithm in stacking), and a bad pixel map gets my system noise well under the sky noise threshold where I live.

 

12 hours ago, Pompey Monkey said:

They use a master bias as a dark and run a large 12 pixel dither between subs. Even using set point cooled CCD I have long since given up on darks. I use a master bias as a dark and a bad pixel map to get cleaner results than I did with darks.

Whoa!!! 'fraid you lost me there, sorry.

P.

[John78]

28 minutes ago, piprees said:

Thanks guys,

however...

 

Whoa!!! 'fraid you lost me there, sorry.

P.

Explained here...

http://ideiki.com/astro/usersguide/dithering_and_guiding.htm#id_5

Assuming you are using APT and PHD2 for your guiding.

[ollypenrice]

40 minutes ago, piprees said:

Thanks guys,

however...

 

Whoa!!! 'fraid you lost me there, sorry.

P.

I got credited with Paul's reply in your post here! The quotation thing is pretty diabolical for me on the new forum, too.

Olly

[piprees]

Ok, thanks again, this is going to take some time.

An even steeper learning curve but, hey, on the positive side, it may keep the old dementia at bay!!!

P.