Hello All,

I was imaging M13 a few nights ago and have encountered an issue calibrating the frames.
I have included a screen shot below. The images are from top left, clockwise. A calibrated light, the master flat, the histogram for the master flat (centered on .3139) and the uncalibrated light.

The calibration seems to have gone a bit odd around the edges of the dust bunnies.
Have you seen this before?

I have processed the images in Pixinsight.

Thanks for any help,

Michael.

Capture by Michael Murphy , on Flickr

Hi Michael,

I never saw anything like that before.


I used to use PI all the time for everything - now I use APP for stacking and calibration.....so I may be a bit rusty on this.

What would be of great help though would be the workflow you are using....and perhaps a screen shot of the Image calibration window:  This one..... 

Hi Brendan,

The calibration screen looked like the screenshot below.

I took the flats two days after the lights and I was talking to someone who suggested that the camera might have moved slightly between taking the lights and flats or that the filter wheel did not move back to exactly the same position when I took the flats.

For the next imaging run I will take the flats immediately after the lights so see if I can rule out the camera/filter wheel moving between the two.

Michael.

In truth it is very hard to see on the posted image - exactly what the problem is.....though it is indeed evident that there is a problem. Quite possibly the filters are slightly misaligned between the 'shoots' - that is more likely than the camera shifting.....but then again we don't exactly what transpired in those two days.

When I was using PI for calibration, I was lucky not to have to use Darks......so I Bias calibrated my Flats when making a Master Flat. I then just calibrated my Lights with just that Master Flat. 

In your calibration screenshot you are using both Bias & Dark calibration frames .....but as far as I know the Dark frames also contain the Bias.....so you should try it again without any Master Bias on that page.

So, I would try it in stages.....If you don't have FlatDarks - then calibrate the Flats with the Bias to make a Master Flat.
Make a Master Dark from darks with the same exposure time, temperature and ISO speed of the light frames.

Then use the Master Dark & Master Flat (ONLY) in that Image Calibration window.

I am not too well up on this....so I am open to and accept correcting if I am wrong. If nobody chimes up....then while you are waiting on your next imaging run....give the above a go....see if it makes any difference.
Brendan.

Wow that really is an odd one. Like Brendan i'm using APP so can't be much help. I'd love to know what caused it.

Just something else that may be of help.....the actual steps/settings for creating the Master Flat....see below.

This is from a Doc I got a long time ago....Jim's PixInsight Cribsheet. Search for that on the internet and you will find the guys website.....he will send out the full document if asked.

Kinch wrote:

In your calibration screenshot you are using both Bias & Dark calibration frames .....but as far as I know the Dark frames also contain the Bias.....so you should try it again without any Master Bias on that page.

Thanks Brendan,

I followed Warren Keller's instructions for creating the calibration frames.
The master flat was calibrated against the master bias and master dark.  PI calibrated the master dark to match the exposure of the flat frames.

I'll try what you suggest and PI's script to do everything to see if that makes any difference.

Michael.

A quick followup.

I compared some of the lights and flats using dynamic crop and it does look like the camera moved between the two sets of images.  I'll see if i can use Pixelmath to 'move' the position of the dust bunnies on the flats.

Well at least you have a definite answer to the oddity.....best of luck trying to align things.

Brendan.

I tweaked the master flat using pixel math and moved each pixel 15px down and to the right.
The new image (on the right) is better but I guess I'm going to have a look at my imaging train to figure out why things are moving.

Good news is you figured out the problem - once that is known, the fix is more straightforward …. to a degree. I am sure you will figure out what actually moved through some trial & error testing.  Good luck with that.

Brendan.

I took advantage of the clear skies last night in Dublin to have another go at M13.
I checked the camera and it was a little loose so I removed the camera and filter wheel and put them back on again.

The result is below, it's a basic process.

Lum 20 120s 1x1, RGB 20 by 30s 2x2.

That's looking great. Nicely done.