Flats not correcting on the red channel

[emyliano2000]

I have no clue what's going on, I even opened the filterwheel to check the red filter and it's all good

[vlaiv]

Ok, here is the think that I don't get.

Circles are in left bottom part of the image:

- in first sub that you posted in this thread

- in master flat

- in calibrated stack

- in sub that you attached

- in single flat that you attached.

However - there is a pair of stars that is in top right corner in all the images except in full stack - they are in bottom left.

It is feasible that you did meridian flip when shooting red channel and that you have some subs with double stars in bottom left and some subs with double stars in top right corner - but calibration issue is only present in bottom left corner - not top right.

I would expect both corners to be affected if some subs were rotated by 180 degrees due to meridian flip

Do you have any idea of why is that - did you do meridian flip at all - are some subs with stars in bottom left corner?

[emyliano2000]

the stack is exactly what came out of APP, with autostretch applied in pixinsight and saved as png, nothing else.

3 minutes ago, vlaiv said:

did you do meridian flip at all - are some subs with stars in bottom left corner?

yes, sgp did a meridian flip during the night

[vlaiv]

Ok, I'm all out of ideas.

Only thing that I could suggest to you is that you leave your camera on telescope and do 120s dark that way - do a single sub for testing.

Compare mean ADU value of that sub to mean ADU value of dark sub that you took in your fridge. If there is significant difference - redo your darks on scope - lights were shot on scope so darks also need to be shot in same conditions (if there is a difference compared to those shot in fridge).

This could resolve the issue. I mentioned before that either light signal is stronger or flats are "weaker" for some reason. We concentrated on flats, but signal can also be "stronger" than it should be - if darks are "weaker" than they should be - having lower ADU number.

Those darks shot in the fridge to have lower ADU values than expected - maybe darks on the scope will have higher mean ADU and this will solve the issue.

LP is generally a bit red in color - and this can explain why only red subs were affected - higher LP background ADU in these subs compared to green and blue.

[emyliano2000]

I wonder if it might be a light leak from the tilt plate 

Edited September 19, 2020 by emyliano2000

[emyliano2000]

@johnrt, did you ever have this problem?

Emil

 

 

[johnrt]

Hello,

Nope, never had a problem with this sort of thing. I used to bin all my RGB 2x2 and shoot a master flat using the Lum filter only at full resolution and then use Pixinsight to register it to the binned images. Sorry I can't be of more help.

Would have to be a fairly bright red light shining directly on to the scope from somewhere to be a light leak.

Edited September 19, 2020 by johnrt

[emyliano2000]

Thanks John. 

My back garden is fairly bright so it might be a light leak. 

I covered the gap on the tilt plate and shot one sky flat to see if there's an improvement and the circle I was taking about is nearly gone, if not completely gone. 

 

[ollypenrice]

1) Do you have red lighting anywhere in the observatory or use a red torch when setting up, etc? Could that have been getting into the tilt plate gap?

2) I don't like the look of the original red flat at all and the arcs certainly shouldn't be there. Whatever is causing them (we hope the tilt plate gap, now fixed) may well have been at a very different intensity between the flats light source and the lights light source, so they wouldn't be expected to calibrate out. If the flats were under correcting the percentage interference by whatever caused the arcs in the flats was lower than the interference during the shooting of the lights.  That would fit (I think) with red light in the observatory getting in through the gap during the imaging run.

I would paint that tape with barbecue or stove paint as well. You need a pigment based paint rather than a dye-based one for blackening optics. Black dye paints can be reflective in IR.

Olly

Edited September 19, 2020 by ollypenrice
Typo

[emyliano2000]

17 minutes ago, ollypenrice said:

Do you have red lighting anywhere in the observatory or use a red torch when setting up, etc? Could that have been getting into the tilt plate gap?

Thank for the reply Olly

I don't use any red lights or anything but I do keem my laptop close to the mount when shooting the flats. I usually control everything from inside but when shooting flats, I use my laptop outside. It's possible that some light from the screen but not highly possible as the screen is usually facing the opposite direction. Other than that I don't know what light it might be. I have a street light right at the end of my garden and to keep the telescopes in the shade I open a large parasol during the night. The lights are turned off between 1 and 5am.

25 minutes ago, ollypenrice said:

 

I would paint that tape with barbecue or stove paint as well. You need a pigment based paint rather than a dye-based one for blackening optics. Black dye paints can be reflective in IR.

I have some blackboard paint from the time I had a newtonian and used it to paint the back of the mirrors and the focuser. Would that work? 

Emil

[emyliano2000]

Hopefully I'll be able to see some stars tonight and shoot a few lights and new flats to see if covering the gap did the trick. 🤞🤞🤞

[emyliano2000]

What on earth is going on? Covering the gap helped a bit but the flats are still not fully correcting. I shot the flats after 1am when it was dark outside, not street lights on and no other lights around to make sure there's no leak but I still get it.

This is the master flat from last night

And this is how the calibrated lights stack looks like

I shot some sky flats this morning but those didn't work either, they actually made it worse

Edited September 20, 2020 by emyliano2000

[emyliano2000]

I think I will put the camera on the TS65 to see if it's the RC causing it or the filters being too far from the sensor.