What is wrong with my Flats

[Peco4321]

I have always struggled to get flats that make a positive difference to my images.  I tend to not bother but I am now wanting to get rid of that round patch on every image that is a bit lighter to give me better results, especially as I am getter better at polar alignment and getting 45 sec exposures now.  I have a 150p on eq3-2 with dual motors and image with Canon 1100d, taking about an hours worth of data, adding bias and dark frames.  My method for flats is to leave the camera attached after each session, then the next day take it out and perch it on a garden chair pointing skywards at a bright spot with a white pillow case over the end.  In Manual mode, I adjust the exposure time only, until the graph peaks in about the middle, then fire off about 20 images.  Attached is a converted image in jpg format for upload only, I always image in RAW.

Stacked images using flats now give me something with a light patch at the top and also I seem to loose any colour that was previously there, so any advice welcomed.

[PeterCPC]

It might help if you could also post a stacked image so we can see the "light patch". I always aim for about 1/3 of the way across the histogram not 1/2.

Peter

[Peco4321]

Here's is my Eastern edge of Veil Nebula, I added flats today but feel my other version without flats is better as it has more colour.

[PeterCPC]

That looks like a light gradient to me rather than an issue with flats. when you say more colour - did you stack in DSS and, if so, did you move the saturation slider?

Peter

[Peco4321]

Thanks Peter, yes stacked in DSS and I usually slide saturation to about 17.  If it's a gradient issue, is the process to remove it only possible in  imaging software that is purchased as I only use GIMP and PS Express.

[PeterCPC]

I am not a GIMP or PS Express user so I cannot advise. I usually use PI and Paintshop Pro to remove gradients. I would push the saturation a little bit higher. Also with only 45 secs exposure you would need lots of exposures - the more the merrier really. When I have imaged the Veil I would get 2 hours minimum of lights and more if the clouds permit.

Peter

[alacant]

Hi. Looks as though you're quite a way off centre. Here's a quick stretch. Is the secondary positioned correctly?

[bobro]

We've been here before - see the image below from your scope from early July. Why not recollimate your scope and start from there?

[michael8554]

If you think your collimation is correct,  then it might be camera/focuser sag.

Also if you Dither you only need to take Bias at fastest shutter speed, and use these as Darks, and take Flats on the 1100D's AV setting.

Michael

[Peco4321]

12 hours ago, bobro said:

We've been here before - see the image below from your scope from early July. Why not recollimate your scope and start from there?

I've tried all advise before and no improvements. Maybe I should uncollimate and start again. 

[happy-kat]

try and see if the trial of gradient xterminator works in ps express because you might find it does support plugins.  

[bobro]

2 hours ago, Peco4321 said:

I've tried all advise before and no improvements. Maybe I should uncollimate and start again. 

The replacement SkyWatcher 130P mirror in my scope has a central doughnut mark - presumably your scope has a similar mark. I put a piece of white paper with a small hole on the back of my phone and took a photo from about 2 feet away looking down the focus tube of my scope. With the scope collimated and the main image central the mirror centre mark is central too. It looks a bit strange in the photo as you can also see the phone camera aperture in the paper which is slightly off. If you look by eye the mirror central mark will line up with the reflection from your pupil with everything else central.

How does your scope look when you look into the focus tube?

[Peco4321]

I've just put it away after a bit of fiddling, typical!  I've just tried re-collimating and adjusting the secondary mirror and a couple of questions/issues that may be affecting this situation. 

1. Is the position of the laser on the ring on the primary mirror acceptable, I can not get it bang on middle, see 1st picture. 

2. When I collimate and then rotate the whole focus and laser collimater attachment by loosening from the screws nearest the scope (if you know what I mean) collimation is lost, see 2nd and 3rd pics. I assume this is having a major influence as if it is collimated in one position, then I twist the camera for better composition, it will be out. 

[alacant]

4 minutes ago, Peco4321 said:

collimation is lost

Hi. Mmm. I don't think the laser is true. Do you have a Cheshire sight tube with cross hairs? They're great for getting close...

[newbie alert]

Just a thought..when I do my flats I put the camera into AV mode ..

[bobro]

As the laser is locked in the focus tube with only 2 screws it will tend to move to one side/twist if the fit is a bit loose. I don't have a laser - the eyeball works pretty well. Can you remove the laser and look where the centre spot is - it may tell you a lot?

[Peco4321]

Thanks everyone I'll have a look later.  

[bobro]

Just realised looking at the image above of the laser shining on the centre spot that the background is way out - it should show the centre of the secondary, whereas it shows to one side of the secondary.

Best to start again with collimating the secondary. No point collimating the primary if the secondary is out.

[Peco4321]

I think no it's a trick of the angle @bobro as it looked ok, just the way I zoomed in

[bobro]

Ah  yes - I thought it had been taken looking straight down the focus tube - I guess it was taken from the front of the scope.

[Peco4321]

Just now, bobro said:

Ah  yes - I thought it had been taken looking straight down the focus tube - I guess it was taken from the front of the scope.

Correct 

[ollypenrice]

Can you do daytime flats successfully with a Newt whose tube is open at the bottom? I don't know, I've never tried, but I'd have expected light ingress at the bottom of the tube to make a complete nonsense of the flats. I'd have thought the flats would need to be done in the dark using a light panel.

I would try taking a image out of the Milky Way, maybe in the Square of Pegasus, with no nebulosity and a low star count, then take a set of flats. Give the image of the sky a big stretch, give the flats a big stretch, and see if they seem correspond. (Put either the flats or the image into negative mode to make this easier.)

Olly

[Peco4321]

Now forgive me banging on about this again, but looking at this recent M42, it looks like the lighter patch is quite central so I'm thinking collimation is ok. So now all I have to do is perfect my flats, correct??  

[alacant]

On 2/9/2017 at 16:12, Peco4321 said:

I have a 150p on eq3-2 with dual motors and image with Canon 1100d, taking about an hours worth of data, adding bias and dark frames.

Is it still this but with flat frames too? Can you post an example light frame and flat frame? 

[bobro]

1 hour ago, Peco4321 said:

Now forgive me banging on about this again, but looking at this recent M42, it looks like the lighter patch is quite central so I'm thinking collimation is ok.

Keep banging away! (I'm doing that trying to improve RA guiding - on my 3rd stepper motor!) To my eyes the lighter patch isn't central. As @alacant requests - a flat frame will show collimation.