Pacman HOO (with problems )

[simmo39]

Hi all, this is my latest target but not without problems. I have finally had enough of Flat frames! no matter how I take them or mix them I get glowing rings around my targets or just to much noise. This Pacman images is in HOO with darks only applied. At the moment my thoughts on flats is unprintable! Rant over..... Here is my Pacman, 25 x 300s in Ha and only 13 x 300s Oiii. Although far short of the Oiii subs that I would have liked I think its going in the right direction. Any advice on flats would be helpfull.

ASI 1600mm pro -20

Thanks for looking in.

[vlaiv]

Could you post your master flat, and any pointers on how you created it (number of flats, did you use flat darks, what is average adu / length of exposure, ...)?

[simmo39]

1 minute ago, vlaiv said:

Could you post your master flat, and any pointers on how you created it (number of flats, did you use flat darks, what is average adu / length of exposure, ...)?

Hi Vlaiv, Ill try and dig out of the recycle bin. I have been using APTs flats aid set to 18000 adu. Also taking dark flats at the same settings......

Got it, here it is

[vlaiv]

Flat looks ok, and it matches pretty well with red glow distribution in image. Why are you having issues with it?

You said that no matter how you take them or mix them you have issues. There aren't very many ways to use them - you should use them in calibration step by making master flat (really only one way to do it ). So I wonder if you do it like that.

[david_taurus83]

I always struggled taking them on a Newtonian as you get stray light into the focuser drawtube. The only light should be what's reflected off the primary to the secondary. How do you take flats?

[alexbb]

Flocked on inside, some flocking material surrounding the focuser tube to prevent light entering around the focusing ax.

And something to prevent light entering through the back of the OTA around the primary.

[simmo39]

17 minutes ago, vlaiv said:

Flat looks ok, and it matches pretty well with red glow distribution in image. Why are you having issues with it?

You said that no matter how you take them or mix them you have issues. There aren't very many ways to use them - you should use them in calibration step by making master flat (really only one way to do it ). So I wonder if you do it like that.

Im just sorting a stretch of a Ha image to show the problem.

 

8 minutes ago, david_taurus83 said:

I always struggled taking them on a Newtonian as you get stray light into the focuser drawtube. The only light should be what's reflected off the primary to the secondary. How do you take flats?

Hi david Im using a LED light panel To take them with.

[simmo39]

Here is an extreme stretch of the Ha data to just to show the problem.

[simmo39]

3 minutes ago, moise212 said:

Flocked on inside, some flocking material surrounding the focuser tube to prevent light entering around the focusing ax.

And something to prevent light entering through the back of the OTA around the primary.

Hi Moise, I have already done a similar thing to my scope but not half as neat as that!

[alexbb]

Just now, simmo39 said:

Hi Moise, I have already done a similar thing to my scope but not half as neat as that!

It seems I forgot to quote David, sorry for that!

[simmo39]

26 minutes ago, vlaiv said:

Flat looks ok, and it matches pretty well with red glow distribution in image. Why are you having issues with it?

You said that no matter how you take them or mix them you have issues. There aren't very many ways to use them - you should use them in calibration step by making master flat (really only one way to do it ). So I wonder if you do it like that.

Hi I cal the falts using the dark flat as the Master dark .

[vlaiv]

So, flats are over correcting. Let me just think what in calibration routine can cause over correcting flats.

One way it can happen is if you don't use flat darks, or flat darks are "too weak" - meaning they have lower value than they should (shorter exposure, change in gain, whatever). It can also be due to using flat dark library (you shoot them once, but shoot your flats each session - if you don't pay attention to correct temperature - flats are "warmer" than flat darks).

Another case when it may happen is if proper darks are "weaker" than they should be - (have less dark signal in them than lights - can happen again due to gain, temperature - in this case lights are shot at higher temp than darks)

Is any of above the case?

And more important - did you use bias at all and how?

[simmo39]

2 minutes ago, vlaiv said:

So, flats are over correcting. Let me just think what in calibration routine can cause over correcting flats.

One way it can happen is if you don't use flat darks, or flat darks are "too weak" - meaning they have lower value than they should (shorter exposure, change in gain, whatever). It can also be due to using flat dark library (you shoot them once, but shoot your flats each session - if you don't pay attention to correct temperature - flats are "warmer" than flat darks).

Another case when it may happen is if proper darks are "weaker" than they should be - (have less dark signal in them than lights - can happen again due to gain, temperature - in this case lights are shot at higher temp than darks)

Is any of above the case?

And more important - did you use bias at all and how?

Flat darks are done at the same time I do flats, end of session so all temps should be good. Not using any bias frames. Dint think they where any good with the ASI 1600. I find that any exposure of less than 5s can be very inconsistent. Especially when taking flats. I get all sorts of different brightness then.

[vlaiv]

10 minutes ago, simmo39 said:

Flat darks are done at the same time I do flats, end of session so all temps should be good. Not using any bias frames. Dint think they where any good with the ASI 1600. I find that any exposure of less than 5s can be very inconsistent. Especially when taking flats. I get all sorts of different brightness then.

Yes, you are right about bias - you should not use them as they can have similar effect with ASI1600 because they don't match for some reason (that is why I asked).

It's very odd that you get inconsistent results with flat exposures less than 5 seconds. I run my flats at anywhere between 0.01s and 1s (depends on scope / correctors used, and filter) and never had any issues. My LED panel is strong enough to provide 75-85% histogram values in this range.

Only time I had issues with flats was when my flats panel had problem with power jack - it had some sort of break in circuit and that was so fast that I could not see it with naked eye - but showed as flickering on short exposures (half frame bright / half dark - randomly). It was quickly fixed with some soldering - and not one issue since.

Even if you have power issues (changing voltage or something) - that makes your flats be different intensity - that should not produce above problem - it is related to intensity over whole flat (multiplicative) and above issue is sort of additive issue.

Only thing that I can think of that would cause above problem is light leak of sorts - that happens when you take flats and not when you image and take light subs. Do you turn on any local lighting when you finish imaging to take your flats (apart obviously flat panel - anything "external" to the system)?

[simmo39]

4 minutes ago, vlaiv said:

Yes, you are right about bias - you should not use them as they can have similar effect with ASI1600 because they don't match for some reason (that is why I asked).

It's very odd that you get inconsistent results with flat exposures less than 5 seconds. I run my flats at anywhere between 0.01s and 1s (depends on scope / correctors used, and filter) and never had any issues. My LED panel is strong enough to provide 75-85% histogram values in this range.

Only time I had issues with flats was when my flats panel had problem with power jack - it had some sort of break in circuit and that was so fast that I could not see it with naked eye - but showed as flickering on short exposures (half frame bright / half dark - randomly). It was quickly fixed with some soldering - and not one issue since.

Even if you have power issues (changing voltage or something) - that makes your flats be different intensity - that should not produce above problem - it is related to intensity over whole flat (multiplicative) and above issue is sort of additive issue.

Only thing that I can think of that would cause above problem is light leak of sorts - that happens when you take flats and not when you image and take light subs. Do you turn on any local lighting when you finish imaging to take your flats (apart obviously flat panel - anything "external" to the system)?

Nope, all dark till I start packing away. on the flickering light thing, you may have something there. Ill have to look into that.

[wornish]

Like others I read that using bias frames with the ASI 1600MM is not recommended.  I now use dark flats and flats instead with a single master dark.

What target ADU value for your flats are you using ?

I use the CCD Flats aid in APT to create mine and set the target ADU to 20000.

Depending on the filter I use the flat exposure times are all very short.

[simmo39]

2 hours ago, wornish said:

Like others I read that using bias frames with the ASI 1600MM is not recommended.  I now use dark flats and flats instead with a single master dark.

What target ADU value for your flats are you using ?

I use the CCD Flats aid in APT to create mine and set the target ADU to 20000.

Depending on the filter I use the flat exposure times are all very short.

Yep, you seem to be doing the same as me except that I aim for 180000 ADU.

[wornish]

2 hours ago, simmo39 said:

Yep, you seem to be doing the same as me except that I aim for 180000 ADU.

Why so high ?

The camera ADC is 12 bit which is rounded up to 16 bit.

Full exposure i.e. White = 65520

I  think thats the right number but I might be wrong.  APT selects it knowing what the camera is.

[vlaiv]

8 minutes ago, wornish said:

Why so high ?

The camera ADC is 12 bit which is rounded up to 16 bit.

Full exposure i.e. White = 65520

I  think thats the right number but I might be wrong.  APT selects it knowing what the camera is.

I think it's typo, correct value is 18000 not 180000 as written in one of posts above.

Although camera is 12bit - subs are stored as 16bit and usual way to do this is to pad values with zeros so real 12bit values are stored in HSB part (Hi significant bits - or bits 15-4, bits 3-0 contain zeros). This translates to multiplication by 16 (every bit shift is in fact x2, so four shifts is two to the power of 4 = 16).

While I think 18000 ADU is a bit low value, I myself would go for ~50000ADU (expressed in these "shifted" values) - it should not matter for quality of flats - they will be a bit more noisy, but not by much, about x1.5 times. Important thing is not to get clipping and to choose linear region (but ASI1600 has very good linearity anyway, so just avoid clipping and choose higher values if you want small SNR boost in your flats).

[carastro]

I can't really help with this as I  have no experience of the CMOS cameras, except to say that one of the members of my local group has also had lots of problems with his OSC ZWO camera and has virtually given up all forms of calibration to eliminate the problems.  He's pretty technical and knows what he is doing, and did not have these problems when he had a DSLR.

So basically I wonder whether we are trying to apply old CCD/DSLR methods to a new style of camera where it doesn't work the same.

Just a thought.

Carole 

[simmo39]

Hi sorry for late replies, have been out trying to get Sii data for this and have had another go at getting flats.

54 minutes ago, wornish said:

Why so high ?

The camera ADC is 12 bit which is rounded up to 16 bit.

Full exposure i.e. White = 65520

I  think thats the right number but I might be wrong.  APT selects it knowing what the camera is.

As Vlaiv said the ADU was a miss print its set to 18000.

[simmo39]

1 minute ago, carastro said:

I can't really help with this as I  have no experience of the CMOS cameras, except to say that one of the members of my local group has also had lots of problems with his OSC ZWO camera and has virtually given up all forms of calibration to eliminate the problems.  He's pretty technical and knows what he is doing, and did not have these problems when he had a DSLR.

So basically I wonder whether we are trying to apply old CCD/DSLR methods to a new style of camera where it doesn't work the same.

Just a thought.

Carole 

You could be right there. Ill give it another go with the Sii data I just got but failling that ill be just using darks.

[vlaiv]

Out of interest, what sort of software are you using for calibration / stacking?

If you want, here is small tutorial on how to calibrate your subs using ImageJ - maybe give it a go and see if you get anywhere with it - use your software to stack calibrated subs after (I'll be posting about registering / aligning and stacking with ImageJ in next few days so you can try that as well if you like). Here is link to topic:

 

[simmo39]

9 hours ago, vlaiv said:

Out of interest, what sort of software are you using for calibration / stacking?

If you want, here is small tutorial on how to calibrate your subs using ImageJ - maybe give it a go and see if you get anywhere with it - use your software to stack calibrated subs after (I'll be posting about registering / aligning and stacking with ImageJ in next few days so you can try that as well if you like). Here is link to topic:

 

Thanks Vlaiv, I will give it ago and see what happens.