Would someone mind please checking my flats?

[BrendanC]

Hi all,

I'm not even remotely sure I'm doing the right thing with my flats.

I've tried two methods: one with the laptop white screen against the scope, using a Flats plan in APT; and one with the same method, but using a white t-shirt between the laptop and scope.

However, I have zero idea whether either method has 'worked', or how I would even check this.

So, would anyone who even slightly remotely has an inkling of what they're doing, please take a look at an example of each one and let me know what they think?

They're here: https://1drv.ms/u/s!AqovBuVZMwj3iZl0sdWvAf1KkFaXEA?e=uN3cAA - one RAW sub taken directly from the screen (.CR2 format, from a Canon EOS1000D), and one using the t-shirt method, and the master TIF for each method, created in DSS as a result of stacking 50 of each.. The t-shirt versions seem quite dark to me, but I know that the actual colour of a flat isn't that important. Still, it would be good to know whether one, or both, or neither, are any good!

Thanks!

Edited September 6, 2020 by BrendanC

[tooth_dr]

Hi Brendan. I can’t see these on my phone, but if the exposure was correct for each flat then the histogram should be in the same place for both.  Was the exposure time with the t shirt longer than the one without?

[BrendanC]

Hmm, I hope I've shared them ok using OneDrive...

I used the AV exposure setting on the camera because that's what APT uses (and I've seen it recommended elsewhere too, not least by AstroBackyard). My understanding was that this means the correct exposure time is automatically calculated without needing to manually check the histogram.

The exposure time was 1/500s for each.

I've only just started taking flats. The past few times I've used them, they've just produced weird results in StarTools. So, I think I must be doing something wrong, but I'm 'flying blind' because I don't know what a 'good' flat or master flat should look like! 

Edited September 6, 2020 by BrendanC

[Adreneline]

11 hours ago, BrendanC said:

I'm 'flying blind' because I don't know what a 'good' flat or master flat should look like!

Well I'm not sure that I do either, however, I've looked through my past flat library and these are taken with a Canon 70D (AV mode) on a 6"RC and an iPad as a light source (barely big enough) with two sheets of white paper over the end of the OTA.

This is just a single frame linear ....

and stretched:

They seem to work for me.

HTH

Adrian

[BrendanC]

Thanks for sharing. I can see that mine are probably similar to yours. But if anyone could check mine I'd really appreciate it.

[tooth_dr]

13 hours ago, BrendanC said:

The exposure time was 1/500s for each

The exposure time can’t be the same for T-shirt versus no T-shirt!

[BrendanC]

OK, well, in that case my entire flats approach is wrong! Which is totally why I need help with this.

APT tells me to set my camera to 'AV' mode. AstroBackyard also says AV is the perfect solution to taking flats. So, that's what I've done. 

Canon's AV algorithm just selected the same exposure time, it would seem. In other words, don't blame me, blame 'the algorithm'!

All I need is for someone, anyone, to take a look at either flat and tell me if they're ok.

Edited September 7, 2020 by BrendanC

[tooth_dr]

1 hour ago, BrendanC said:

OK, well, in that case my entire flats approach is wrong! Which is totally why I need help with this.

APT tells me to set my camera to 'AV' mode. AstroBackyard also says AV is the perfect solution to taking flats. So, that's what I've done. 

Canon's AV algorithm just selected the same exposure time, it would seem. In other words, don't blame me, blame 'the algorithm'!

All I need is for someone, anyone, to take a look at either flat and tell me if they're ok.

Histogram without Tshirt is just about ok - you ideally want the histogram peaks centred

 

 

But with T shirt is underexposed:

[BrendanC]

Thank you! 

I think I might just abandon the AV exposure technique and get more hands on with the histogram. 

Really appreciate the help here.

[happy-kat]

Es your ISO the same as your lights? AV mode works for me on the 1100d histogram peak is in the middle. Make sure before you select AV that the lens is set to MF as switching to AV can loose focus (when using a camera lens)

[BrendanC]

Yep, ISO is definitely the same (800 - I shoot everything at that, makes calibration library so much easier and it's supposed to be good for signal/noise ratio for my EOS1000D), and I'm using a 130PDS scope, so no lens to factor in here.

I just think that it's 'one of those things' - AV mode isn't quite cutting it. It's no real problem just to do this manually, I've already set up a plan in APT to take a series of shots at different exposures, then I'll see which are best for the histogram, and iterate until I've hit the sweet spot.

[tooth_dr]

17 minutes ago, BrendanC said:

Yep, ISO is definitely the same (800 - I shoot everything at that, makes calibration library so much easier and it's supposed to be good for signal/noise ratio for my EOS1000D), and I'm using a 130PDS scope, so no lens to factor in here.

I just think that it's 'one of those things' - AV mode isn't quite cutting it. It's no real problem just to do this manually, I've already set up a plan in APT to take a series of shots at different exposures, then I'll see which are best for the histogram, and iterate until I've hit the sweet spot.

AV mode should work!  I’ve used it on my 1000d with APT too, and never failed to get histogram centres. Is the exposure +\- setting turned down a stop or two? 

Edited September 7, 2020 by tooth_dr

[BrendanC]

Good point, I'll take a look at some point over the next few days after I've broken the rig down (I want to keep it all as is over the next few days because, shock horror, we might be getting some good nights!)

I still think that, once I've figured out the best exposure time for the 'laptop against the scope' method, it's not such a hardship to stay in Manual. But it would be good to fix it, as you say, just because, well, it's better when things work the way they should. Having said which, at pretty much every step along this torturous AP path, hardly anything has worked the way it should...

Having said which, I just realised, if it was turned up/down, then that wouldn't account for why the t-shirt and screen methods had the same exposure time.

Edited September 8, 2020 by BrendanC

[tooth_dr]

1 hour ago, BrendanC said:

Having said which, I just realised, if it was turned up/down, then that wouldn't account for why the t-shirt and screen methods had the same exposure time

No it wouldn’t. It sounds like the system didn’t realise you changed the brightness of the panel and just went with the original 1/500.