Do I have a problem with my 1600mm pro (PROGRESS REPORT 2 )

[simmo39]

11 minutes ago, steppenwolf said:

Absolutely, this is what I have been saying all along?

 

Thanks, Ill try tomorrow night without the filter and see what I get. 

[simmo39]

Had a quick test last night I think I have reduced but not eliminated the problem. but one thing that may also help is my flats. A few of you mentioned over correction, does that mean i need to lower the ADU level or increase the level? on the test I did last night I found I couldnt get rid of some of the dust bunnies.  I have to say Im still not convinced that there isnt a problem with the camera but will try any more suggestions.

[coatesg]

Your flats are definitely overcorrecting - the bright dust doughnut, and bright corners are a dead giveaway - it's making it hard to tell what else is happening (that could be it). What's your process for taking flats and then reducing your data?

[simmo39]

10 hours ago, coatesg said:

Your flats are definitely overcorrecting - the bright dust doughnut, and bright corners are a dead giveaway - it's making it hard to tell what else is happening (that could be it). What's your process for taking flats and then reducing your data?

Hi, for the taking of the flats i use APT flats aid. I have tried it at 20000ADU also at 25000 and as low as 18000 I have been doing some more research on the net and it seems I may need to lower it around 15ADU. Ill try that next session, was going to try it last night but I seem to have been hit with man flu and my wife wouldnt let me out to play! lol.  For the processing Im using PI with dark flats in the dark frame slot to subtract that from the flat and then use master dark and the flat with the data subs.

[coatesg]

The APT flats tool should work fine - if the pixel level counts are in the linear region of the histogram (ie somewhere near the middle/two thirds) then there shouldn't be any issue - the only considerations are to ensure the level is well above the noise floor, and so that there's no saturation.

You could try generating the master flat prior to use in BPP :  https://www.pixinsight.com/tutorials/master-frames/#Master_Flat_Generation_I is long winded an possibly not required (in terms of bias reduction, etc.Also, you have to be careful with CMOS data in all situations - you do not want to let PI Optimize Dark frames as they don't scale properly - the dark lengths must match exactly. 

Even just doing an image integration on the flats for each filter (mean combination, percentile rejection, multiplicative normalization) without bias subtraction, etc. prior to using this as a master flat in batch preprocessing should give you an idea on whether this resolves it or not. 

 

[simmo39]

28 minutes ago, coatesg said:

The APT flats tool should work fine - if the pixel level counts are in the linear region of the histogram (ie somewhere near the middle/two thirds) then there shouldn't be any issue - the only considerations are to ensure the level is well above the noise floor, and so that there's no saturation.

You could try generating the master flat prior to use in BPP :  https://www.pixinsight.com/tutorials/master-frames/#Master_Flat_Generation_I is long winded an possibly not required (in terms of bias reduction, etc.Also, you have to be careful with CMOS data in all situations - you do not want to let PI Optimize Dark frames as they don't scale properly - the dark lengths must match exactly. 

Even just doing an image integration on the flats for each filter (mean combination, percentile rejection, multiplicative normalization) without bias subtraction, etc. prior to using this as a master flat in batch preprocessing should give you an idea on whether this resolves it or not. 

 

Hi thanks for the pointers, when you say the dark lengths must match exactly is that the dark flat length?  Also I think i have been leaving optimize checked. will try again with it not checked later and see what happens.

[coatesg]

If you want to use dark flats then yes they should all match for CMOS cams, but I don't personally think dark flats are necessary unless the flats are quite long (for NB perhaps) - the thermal signal in a short integration on a cooled camera is negligible compared to the actual signal in the flat. 

[Stub Mandrel]

If you are using bias AND darks and dark flats that could cause issues. If you use darks and dark flats you don't need bias.

[simmo39]

3 hours ago, Stub Mandrel said:

If you are using bias AND darks and dark flats that could cause issues. If you use darks and dark flats you don't need bias.

Dont use a bias, Just need another clear night to do another test at 15000 ADU. hopefully that will work.

[simmo39]

Ok, Had a clearish night here last night so more tests. Here is a Ha image of the California Neb. Flats taken at 15000 adu and the filters the wrong way round in an attempt to get rid of reflections. Over all seems a bit better but still there. I think im on the right track but still need to get the Flats sorted better. Any thoughts welcome please.

[coatesg]

Well, that looks a lot better! If you stretch the life out of it, the background still appears a little uneven, but hard to pin down - it could be local conditions, cloud, reflections in the tube/drawtube/camera/etc. Could even be amp glow, but hopefully that's resolved by using darks properly. Also worth noting that nebulosity covers pretty much the whole image here, so you might be seeing faint gas rather than background! (see below)

Looking at heavily stretched flats can help, and also inspecting the scope from the focuser end while looking at a daytime sky to look for reflections. 

[astrosatch]

Just a thought.

Did you calibrate flats with bias? I had simmilar problems with overcorecting flats on my dslr and nothing worked until I calibrated flats with bias. Or with dark flats if they are longer exposures. But not both. If you use bias and dark flat you end up extracting bias twice and that shows as overcorecting or undercorecting on images.

Andrej

[simmo39]

5 hours ago, coatesg said:

Well, that looks a lot better! If you stretch the life out of it, the background still appears a little uneven, but hard to pin down - it could be local conditions, cloud, reflections in the tube/drawtube/camera/etc. Could even be amp glow, but hopefully that's resolved by using darks properly. Also worth noting that nebulosity covers pretty much the whole image here, so you might be seeing faint gas rather than background! (see below)

Looking at heavily stretched flats can help, and also inspecting the scope from the focuser end while looking at a daytime sky to look for reflections. 

 

Hi Graeme. Thanks for the reply, yes it does look better and the night I did the test shot on was poor seeing and lots of moisture in the air as well as a howling gale! not the best conditions to run a test on, All though It looks better and lowering the ADU value of the flats has helped Im still not totally  convinced that I havent got a problem with the camera. Wednesday I have set a side as scope day and I will give the scope a good look over and also a collimation as that is also not good. On the ADU settings for the flats of the ones found that others are using they are in the 20 to 30000 range and im getting down to the 14000 to 15000 and thats what is nagging my mind that the camera may have a fault.

once again thanks Simmo

[simmo39]

5 hours ago, astrosatch said:

Just a thought.

Did you calibrate flats with bias? I had simmilar problems with overcorecting flats on my dslr and nothing worked until I calibrated flats with bias. Or with dark flats if they are longer exposures. But not both. If you use bias and dark flat you end up extracting bias twice and that shows as overcorecting or undercorecting on images.

Andrej

Hi Andre, Im just using dark flats. Bias is not good with cmos. I find anything less that a 5 sec exposure and a cant get a proper image. again another reason i think i have a camera problem.

Simmo

[astrosatch]

20 hours ago, simmo39 said:

Hi Andre, Im just using dark flats. Bias is not good with cmos. I find anything less that a 5 sec exposure and a cant get a proper image. again another reason i think i have a camera problem.

Simmo

What do you mean bias is not good with cmos? 

[simmo39]

From what I have read, the expert's say that with CMOS camera the short exposures required for bias frames are not stable. I have found with my camera that I can't get a stable exposure for anything below 5 seconds. 

[Filroden]

12 minutes ago, simmo39 said:

From what I have read, the expert's say that with CMOS camera the short exposures required for bias frames are not stable. I have found with my camera that I can't get a stable exposure for anything below 5 seconds. 

Do you have any links that go into this in more detail? I have the 1600MM-C and I've been taking bias with 32micro-second exposures and my flats (with 30k ADU) don't exceed a second other than in narrowband. I've never noticed any exposure related issues with bias, darks, flats or lights.

[simmo39]

48 minutes ago, Filroden said:

Do you have any links that go into this in more detail? I have the 1600MM-C and I've been taking bias with 32micro-second exposures and my flats (with 30k ADU) don't exceed a second other than in narrowband. I've never noticed any exposure related issues with bias, darks, flats or lights.

on the cloudynights website there are several threads regarding this. As for my camera, like i have said im not sure if there is a problem with it, If i take say 25 flats at anthing less than 5 secs i get all sorts of exposures from just pure white to grey and the odd one that look correct.

[Stub Mandrel]

1 hour ago, simmo39 said:

on the cloudynights website there are several threads regarding this. As for my camera, like i have said im not sure if there is a problem with it, If i take say 25 flats at anthing less than 5 secs i get all sorts of exposures from just pure white to grey and the odd one that look correct.

Sounds wrong... if I'm imaging Mars with 5ms frames and my ASI120MC I get all the frames quite clearly with the same exposure.

[MartinB]

In 2010 I bought a new SX M26 which had a fault similar to yours - inconsistent darkening towards the centre of the image.  It also showed up on flats but, as I say, it was variable.  It turned out to be an electrical fault in the camera and SX promptly replaced it. 

[simmo39]

16 minutes ago, Stub Mandrel said:

Sounds wrong... if I'm imaging Mars with 5ms frames and my ASI120MC I get all the frames quite clearly with the same exposure.

Hi.I also have an ASI 120MC and have no probs with that. I will give the camera one more test next clear night and if I still have problems I will get intouch with FLO and se what they say.

[simmo39]

6 minutes ago, MartinB said:

In 2010 I bought a new SX M26 which had a fault similar to yours - inconsistent darkening towards the centre of the image.  It also showed up on flats but, as I say, it was variable.  It turned out to be an electrical fault in the camera and SX promptly replaced it. 

Hi. thanks for he reply, the fault in your camers is similar to what I think i have. as above next clear night will be a good test and ill make a decision then to contact suplier and see what they say.

[Adam J]

5 hours ago, simmo39 said:

Hi. thanks for he reply, the fault in your camers is similar to what I think i have. as above next clear night will be a good test and ill make a decision then to contact suplier and see what they say.

actually that reminds me of ice on the sensor.

[simmo39]

12 hours ago, Adam J said:

actually that reminds me of ice on the sensor.

I have been checking for ice but havent seen any. to try and rule that out on my next session im going cool a lot more slowly to see what happens.

[Laurin Dave]

Have you tried it in the day time/indoors on b&w video/still mode ?  That should work but if you can't get stable images below 5sec exp then maybe not in which case there's a camera fault