Trying to get organized (calibration frames)

[gnomus]

I'm trying to streamline my working a little bit.  As a beginner, however, I don't want to go off down some wrong pathway.  I have a couple of questions:

1)  Bias Frames: I am in the process of shooting 200 frames at 1/4000 of a second on my EOS 600D.  I will shoot three sets: 400 ISO; 800 ISO; and 1600 ISO.  I will then stack each of these sets so that I will end up with a master bias frame for each of the stated ISOs (I'm guessing that will cover most eventualities).  That should mean that I will have 3 bias masters that I can use for the foreseeable future.  Will this work?

2)  Flat frames:  So far I have been using a flash gun fired through a flash 'softbox'.  I am somewhat concerned that this is giving me a hotspot in the middle of the softbox.  I wonder if it might be worthwhile getting some sort of light panel to use for all my flats from now on.  It would need to be big enough for both of my scopes (an 80mm and an 8").  I see many different light panels at various prices.  I don't want to buy something that is useless (for the sake of saving a few quid).  Do people have any recommendations as to makes/models of light panels that might be suitable for the job?

3)  Dark Frames:  This is an increasingly confusing subject.  If I am reading these forums correctly, there seems to be differences of opinion as to whether dark frames are needed at all.  That notwithstanding, could I also build up a library of dark frames?  Let us say that I decide that all of my "lights" will be shot at either 2 minutes or 5 minutes.  Could I then monitor the daily temperature and shoot a number of darks covering a range of temperatures?  So, for example, I could note that it is 5 degrees centigrade tonight.  I could set up and take (lets say) 30 exposures at 5 minutes and 30 exposures at 2 minutes, for each of my ISOs.  When stacked these would give me master darks.  I could then use the master frames the next time I image when it is 5 degrees.  Over time I could build up a set covering every temperature in 1 degree intervals.  (Indeed, would it need to be this fine?)  Perhaps I could start sticking the camera in different parts of the house (garage, utility room, fridge!!!) to build up a library of different temperatures quite quickly.  If I am to do this, I could take as many dark frames as I like.  Is 30 a reasonable amount?  Should I take more?  Should I take less?

Do folks have any thoughts on the above?

Thanks, as usual, in anticipation of your advice.

[frugal]

Bias Frames: That is exactly what I do. There is a great article on the Blackwater Skies blog about the improved quality of master Bias images given a large number of frames (http://www.blackwaterskies.co.uk/2013/09/pixinsight-dslr-workflow-part-1-bias.html). I think mine was made up from 330 bias frames. I keep a single bias master for all of my images and then regenerate it every 6 months or so.

Flat Frames: I used to have real problems with this, but I now use my tablet and hold it tight to the end of the scope. Fortunately I only have an 80mm scope so it fits nicely. I have an art package on the tablet so I can show a flat white image across the whole screen. I do slide the tablet across the front of the scope whilst I am taking the flats, so that I can smooth out any hot spots from the backlighting. As I use a DSLR I can just keep the ISO and focus the same, and stick it into AV mode and quickly take 40 or so frames at the end of an imaging session. I produce a separate set of flats for each imaging session.

Dark Frames: This is dark voodoo and appears to involve the sacrifice of chickens whilst reciting the names of famous astronomers backwards. Some people insist that the dark frames are not required, some insist that the are, some say it depends on the camera. Some say that the dark frames must have exactly the same temperature and duration as the lights, some say that the processing software will scale them anyway so it is irrelevant... I tend to have a set for each duration and ISO that I use, and not fret about the temperature. If the image looks rubbish when processed with the darks I will have another go without them.

[Marky1973]

Hi Gnomus - this is what I did for my Flat frames

http://stargazerslounge.com/topic/227341-flower-pot-flats-the-test/?hl=%2Bflowerpot+%2Bflats#entry2449850

I'm still playing around and it may not be the best solution, but it was cheap!

[Astrosurf]

I've tried creating darks in different temps but the best, I find, are ones taken the same night, even if the temp has dropped by the end. When I tried the fridge method, it was awful. Others have had success with master darks at various temps but I was told it's the sensor temp and not the ambient temp that applies.

Alexxx

[lensman57]

Hi,

Let me confuse you even more.

1- Your Bias frame acquisition is fine so long as you do not end up with Bias frames taken at 25C ambient and image at-5C as it is at some UK parts at the moment. 200 Bias is excellent to give you a clean and almost noiseless master bias.

2- Your Flat frame acquisition is a total waste of time and effort, not only the light source is incorrect ( someone suggested a Laptop screen which is better but not ideal as it will not cover the 8" evenly, if you did go for this with the 80 bring up note pad to have a white background ) but also the position and the orientation of the focuser. The only correct way of getting flats is to do them at the end of the imaging run with the scope at the same position as imaging and using an illumination light source such as well balanced flat panel . There is also a wrong way of doing flats with a DSLR and a right way regarding the exposure but you can ask for this if and when you are ready.

A flat library is a viable proposition only  if the imaging sensor, filters and the optics are in a very clean state and are permanently

housed in an observatory without interference ( taking the camera on and off and so on ).

3- There is nothing more damaging to your calibration process than Darks that are not correctly matched to your Lights and we are talking about temperature here. As you can not control the imaging TEMP of the DSLR sensor by set point cooling ( CCDs win hands down here ) it is best to either not bother at all  or just take a minimum of 3~5 Darks at the end of the imaging session and hopefully the Temp will be within 2C or 3C variance of the lights for which the stacking software with " Dark Optimisation " enabled can compensate for. I have seen noise analysis tests that showed  a minimum of 3 Darks that were correctly matched to the lights removed up 85% of the Dark Current noise from the image. So it is best to take a correct  few at the end rather than take a lot which are not a match and useless. Crazy ideas like putting your DSLR in the fridge and setting the camera to take Darks may work but it is only a shot in the dark ( no pun intended ) .

Good calibration frames help a great deal to extract detail out of the master light but done incorrectly they can also do a lot of damage so take your time.

Regards,

A.G

PS: In my opinion it is a better idea to image with extreme dithering enabled between sub frames and using a large number of subs , atleast 16, and a Sigma Clipping routing during the stacking to combat the noise rather than taking and applying mismatched Darks.

[thohem]

I also renew my bias frames only twice a year, so you're good.

For flats I use this:http://www.teleskop-austria.at/shop/index.php?lng=de&m=2&kod=FFB20&skod=%2046&om=Astro-Fotografie (I know it's german but for me it's the best buy for that low price)

I do 10-50 flats after the session, depends if I want to go to sleep

I personally don't use darks because with dslr you can't quite match temperatures. I always dither to counter hotpixels

[thohem]

I also renew my bias frames only twice a year, so you're good.

For flats I use this:http://www.teleskop-austria.at/shop/index.php?lng=de&m=2&kod=FFB20&skod=%2046&om=Astro-Fotografie (I know it's german but for me it's the best buy for that low price)

I do 10-50 flats after the session, depends if I want to go to sleep

I personally don't use darks because with dslr you can't quite match temperatures. I always dither to counter hotpixels

[gnomus]

...The only correct way of getting flats is to do them at the end of the imaging run with the scope at the same position as imaging and using an illumination light source such as well balanced flat panel... . 

Firstly, thanks everyone for their replies.

Lensman: I was getting awful results with the flashgun and my C8 - hence my desire to do something different.  If I have to spring for a "well balanced flat panel", then so be it.  Would you have any specific recommendations?  I can find LED flat panels across a range of prices, but I don't have enough knowledge to determine whether or not they are suitable.

I appreciate that I cannot build a flat library.

I have PHD2 and APT, so I think that I am dithering.  Not sure about the correct settings though.  

Pleased to hear the news about the darks.  I have to say they were the most boring part of this process!  I may do as you suggest and just try a few darks, and see whether the result looks better with or without them.

Any suggestions for that flat panel would be most welcome.  I'm probably going to have to buy something due to my being "un-handy".

Thanks again

[lensman57]

Firstly, thanks everyone for their replies.

Lensman: I was getting awful results with the flashgun and my C8 - hence my desire to do something different.  If I have to spring for a "well balanced flat panel", then so be it.  Would you have any specific recommendations?  I can find LED flat panels across a range of prices, but I don't have enough knowledge to determine whether or not they are suitable.

I appreciate that I cannot build a flat library.

I have PHD2 and APT, so I think that I am dithering.  Not sure about the correct settings though.  

Pleased to hear the news about the darks.  I have to say they were the most boring part of this process!  I may do as you suggest and just try a few darks, and see whether the result looks better with or without them.

Any suggestions for that flat panel would be most welcome.  I'm probably going to have to buy something due to my being "un-handy".

Thanks again

I use an Arora Panel from Gerd Neumann, I think it is a 10" one but I am not too sure about the size. The one suggested by Thohem, the Lacerta one also looks interesting as it has a corrected RGB components. In any case these will cost you more than they should as they need to be imported from Germany or Austria.

As for Dithering with a DSLR you should set the dithering distance to MAX possible. Make sure that the server mode is enabled in PHD2 and I am sure that you know what to do in APT to enable dithering.

A.G

[gnomus]

...

As for Dithering with a DSLR you should set the dithering distance to MAX possible. Make sure that the server mode is enabled in PHD2 and I am sure that you know what to do in APT to enable dithering.

A.G

Thanks again

I have the following settings in APT:

Are you saying that I should change the third item down to '5' - that is the maximum allowed as far as I can tell.  I have PHD2 server mode enabled.  Do I need to set "Autodithering" (second item down) to "On"?

When I set up a plan, there is a box in the upper right corner, which says, "Don't Dither this Plan".

I assume that I leave this unchecked (which is the default).  

Do I need to do anything more to get dithering working?

[lensman57]

Thanks again

I have the following settings in APT:

APT.jpg

Are you saying that I should change the third item down to '5' - that is the maximum allowed as far as I can tell.  I have PHD2 server mode enabled.  Do I need to set "Autodithering" (second item down) to "On"?

When I set up a plan, there is a box in the upper right corner, which says, "Don't Dither this Plan".

dith2.jpg

I assume that I leave this unchecked (which is the default).  

Do I need to do anything more to get dithering working?

Turn auto dithering on and change the 3rd item to 5, watch the PHD graph while dithering is taking place and the random movement of the axis should be substantial.

A.G

[glowingturnip]

the latest version of PHD2 also helpfully writes the word 'dither' onto the graph

[gnomus]

Turn auto dithering on and change the 3rd item to 5, watch the PHD graph while dithering is taking place and the random movement of the axis should be substantial.

A.G

Thank you.  With any luck I might be able to try it out tonight.