"T shirt" flats

[Demonperformer]

I have been considering this as a simpler (less electronic) approach to taking flats and am wondering if they are going to be practical in my situation.

Firstly, I need to dismantle my kit at the end of every session. This includes the removal of the camera & filterwheel from the scope, so clearly the flats need to be taken before I do that.

Secondly, I understand these need to be taken against a "twilight sky". Now for morning (pre-sunrise) sessions that is not a problem, but for evening (post-sunset) sessions, I don't see how that is going to be practical. Before I can take the flats, I need to get precise focus (so that I don't change it when I start to image), but until it is dark enough to get some decent star images on the chip, I'm not going to be able to do that. I can certainly get "approximate" focus (to within about a mm) but precise focus - no. So, short of staying out all night and waiting for morning twilight (not a proposition I fancy at this time of year), I don't see how I can achieve the desired result ... with the possible exception of when the moon (always the friend of the DSO imager!) is around in the evening, when I can use that for focus. My most distant terrestrial target is a pylon about 200m away, which does not achieve perfect astro-focus (although "close").

So is there a way round this? Am I being particularly obtuse? Or will I just need to stick with electronic illumination?

Thanks.

[Ouroboros]

I am in the same position. My solutions include using an LED panel and a white cloth over the end of the telescope. It's powered from a 12V power pack.  

Alternatively, when I turn in for the night, I leave the telescope exactly as I used it when imaging and cover it with a telescope cover.  I then do the flats in the morning, again with the white cloth, and using the sky as the light source. 

[Demonperformer]

Using an LED panel is what I currently do, but am trying to get away from, and leaving it set up is, as previously stated, not an option.

[alan potts]

Not knowing what you have posted I just took flats just before or at sunset with a T shirt, probably explains why my images don't stand up against any of yours, plus I don't really know what I'm doing 90% of the time. I have also used one of my light tables from negative days with a with T shirts over it. As luck would happen I know where focus is which sounds as if it helps.

[alexbb]

Unless you use a petzval where the distance between the sensor and the last lens can change and matter significantly if you have dust specks on it, I really don't think a 1mm focus error makes any difference.

[MartinB]

5 minutes ago, moise212 said:

Unless you use a petzval where the distance between the sensor and the last lens can change and matter significantly if you have dust specks on it, I really don't think a 1mm focus error makes any difference.

Agreed, focus often changes over an imaging session and, if you then take flats during the day focal point will be different again even without moving anything.  

[Ouroboros]

1 hour ago, MartinB said:

Agreed, focus often changes over an imaging session and, if you then take flats during the day focal point will be different again even without moving anything.  

I wonder how significant small changes to focus are with regard to flats? After all, the position/size etc  of dust bunnies on sensor and filters won't change because they are independent of focal position. So, we are maybe only talking about changes in vignetting from the telescope. Vignetting from filters etc should remain constant because camera and filters move together. So I imagine vignetting changes slowly as a function of focal position. After all we're probably talking about sub-mm changes in focal position. I don't know, I'm speculating. Anyone tried an experiment? 

Although not advisable I've successfully used flats obtained on different occasions and for which the focus was probably slightly different.   

Note added in edit. In extremis it's possibly better to use some kind of flat than none, even one imperfectly focused.  But then I'm probably less fussy than some. ?

 

[Demonperformer]

11 minutes ago, Ouroboros said:

Anyone tried an experiment? 

These  are some interesting thoughts. A fairly extreme experiment should be quite easy to perform. I will give it a go - I  should get the chance later today. 

[Adreneline]

A white tee-shirt and an iPad running a 'lightbox app' work really well for me.

I take flats for all the filters I've used at the end of the imaging session. I use the focus position for the last filter to take all of the flats.

For a typical NB imaging session it takes me no more than 10 minutes to take all my flats, typically 25 per filter.

If all else fails I can take them the following morning because I can leave the imaging train intact when I bring the equipment in at the end of the session but it is a bit of a faff connecting everything up again and running up the software.

HTH

[Demonperformer]

Right, here are the results of my experiment:

Star71 >> 16.5mm adapter >> 11mm nosepiece >> ASI 224cool camera. Within the 11mm nosepiece I used the 1.25" filter adapter and placed my Baader UV-IR cut/L filter. The dewshield was extended and my small ELP placed. Used sharpcap flat field generator, 10* exposures. The system was unchanged between these two flats, except that one was taken with the focusser fully in and the other taken with the focusser fully out (33mm difference). The FITS files were then converted to JPGs for upload and easy viewing (original FITS files can be provided if requested).

I don't claim that they are "perfect" flats, but should suit to demonstrate the effect of altering focus.

So, how much difference do you see between them? Enough to matter in practice?

 

[Demonperformer]

I'm now wondering if I made a tactical error in using the 224 rather than the 1600. With a much smaller chip, vignetting would be much reduced and so any potential differences would also be minimised. I may repeat the exercise with the 1600 this afternoon to see what difference  (if any) I get.

[glowingturnip]

I set up and tear down too, but at the end of the session I leave the camera in the OTA and carry the two together (very slowly and carefully, considering how expensive it all got) back inside, remount it inside and shoot my flats in the morning.

[Demonperformer]

Unfortunately,  as per OP, that is not an option for me. 

[Jkulin]

5 hours ago, Adreneline said:

For a typical NB imaging session it takes me no more than 10 minutes to take all my flats, typically 25 per filter. 

Are you taking your flats at 25,000-30,000 ADU, it takes me a couple of hours to do mine and will probably take even longer now that I have switched over to 3nm.

[Adreneline]

9 hours ago, Jkulin said:

Are you taking your flats at 25,000-30,000 ADU, it takes me a couple of hours to do mine and will probably take even longer now that I have switched over to 3nm.

Hi,

Yes, I aim for 32,000 ADU. Using my Atik428ex with 2" Baader Ha, OIII and SII filters with my Samyang 135mm at F2 the respective exposure times in SGPro are 0.75, 0.7 and 0.7 seconds.

If I use the 428 with 1.25" Ha filters and the SW-ED80DS-Pro the exposure time is 5 seconds to achieve 32,000 ADU - according to SGPro. I also use LRFB filters with the ED80DS but off the top of my head can't remember the exposure times - I've not used it for some time now.

Adrian

[geordie85]

I reuse my flats as I leave my camera, Flattener and scope all connected and as long as I haven't taken the camera off I never have any issues. Even though I still have to refocus at the beginning and during my imaging runs. 

I don't think a slight change in focus would matter. I've used the same flats for months at a time before without any trouble. 

[Scooot]

On 29/11/2018 at 09:47, Adreneline said:

A white tee-shirt and an iPad running a 'lightbox app' work really well for me.

HTH

Hi Adrian 

Whats the name of the Lightbox app you use?

[ollypenrice]

Like other contributers I don't think 1mm of focal difference will significantly affect the flats.

Perhaps the best experiment would be as follows:

1) Make a set of flats at perfect focus and a second set with focus altered by 1mm. 

2) Make masters of both.

3) Subtract one set from the other and see how uniform the result is when stretched.

(When taking beginners through the AP steps my standard demo is to subtract a flat from a flat to show what they do.)

Olly

[Demonperformer]

OK, I think this could be a case of my OCD exerting its authority. When people tell me flats have to be taken with everything exactly the same as used for the lights, I tend to cling on to the word exactly in a way that would make a limpet appear weak-willed.

The subtraction idea certainly seems to be a reasonable way forward, and I may try that with the 224 flats I did above ... if I can work out how to do it in GIMP!

Thanks.

[Adreneline]

7 hours ago, Scooot said:

Whats the name of the Lightbox app you use?

Hi Richard,

The app is called 'Light Pad'. I bought it some time ago to look at negatives and 35mm slides. I like it because the whole screen is 'white' - no borders or other annoying features - and I don't get any banding which I used to get when I used a Huion light pad.

I see it is no longer available in the App Store but there is an app named LightBox Extra which seems very similar.

HTH

Adrian

[Scooot]

4 minutes ago, Adreneline said:

Hi Richard,

The app is called 'Light Pad'. I bought it some time ago to look at negatives and 35mm slides. I like it because the whole screen is 'white' - no borders or other annoying features - and I don't get any banding which I used to get when I used a Huion light pad.

I see it is no longer available in the App Store but there is an app named LightBox Extra which seems very similar.

HTH

Adrian

Thank you. I’ll have a look. Ive been using an image of a white piece of paper to date, which seems to work.

[Demonperformer]

Well, assuming that I have done it right, the result is attached. Looks pretty horrible to me, but that was with a full 33mm difference in focus. Still, I guess it shows that focus does make a difference. I guess the next thing is to get the gear out and try it with 1mm difference. Watch this space ...

[coatesg]

33mm is a lot of difference, but I suspect that 1mm will not be as prominent. What's the actual differences in values after subtraction? (Assuming the above is stretched)

[Scooot]

1 hour ago, ollypenrice said:

 

(When taking beginners through the AP steps my standard demo is to subtract a flat from a flat to show what they do.)

Thanks for this. I’ve just tried it, a very interesting & illuminating experiment. 

[Demonperformer]

Yes, stretched as per Olly's post. I can't tell you what the precise values are, but they were all nicely stacked at the bottom end of the histogram - the unstretched version looked just a uniform black. When I get the chance I will redo the experiment using closer values and also using the bigger chip camera (which will likely increase vignetting).