East/West Rotated Flats

[Warhen]

Hi Juan et al, Back in January Juan pointed out that 'Actually, you can define two "filters" with Add Custom to group frames by rotation. This way you can calibrate and register everything in a single script run.'

Just to be sure I did this correctly- I first grouped Lights by E and W, then created E and W Flat groups. IOW, the BPP can't read the 'Pierside' field in the FITS header as EAST or WEST, but the East and East/West and West custom filters (Flats and Lights) will be matched to one another- correct? Thank you!

[Mark de Regt]

Any update on this?

[sreilly]

I had though I had an understanding of this but seems each program involved in acquiring the data have their own idea as to what side of pier is. ACP sees this as facing east while TheSk X, which I use to determine when the object transits, to determine which flat to use. After asking Bob Denny, of DC Dreams (ACP), this is the reply I got:

"I would use the FITS info PIERSIDE, which is ... you guessed it ... reversed from ACP's "looking" sense. Pier side is just that "side of pier" which is opposite "looking" when the weights are down at least. For the flats, if you're having ACP put the East/West into the file names, those are "looking". I apologize for this, the "looking" seemed to be the right convention back in 2000."

That said, I sometimes find that using my west side flats on my east "looking" (pre-transit) works better but then sometimes not. So far it seems to be trail and error. The flats are generated using ACP and the  autoflat script and the FlatMan XL el panel.

-Steve

[oldwexi]

Hi,
i missing the point why a "Flat E" should be different from a "Flat W" or a "Flat Zenith"
when using a flat foil?
Please, if anyone can explain this to me i would appreciate that.
Now i am not sure anymore if a Dark W differs from a Dark E or not.

Gerald

[sreilly]

Hey Gerald,

I can't answer this with a mathematical equation but as it was explained to me years ago, in a perfect world it wouldn't make a difference. But of course it is far from perfect and an imaging train, at least mine, is not perfectly centered so the results are going to be different. Considering that I have a camera (STL-11002M, filter wheel (FW10), maybe a AO-L, rotator (Pyxis 3"), and spacers connected to the rear cell of my 12.5" RC (OGS), if any one of those is slightly off center then the rotation would also be off center. Take for example centered, taking an exposure rotating the camera 360 degrees would result in a centered star not moving which would be great. Now take that same image in a system that isn't perfectly centered and that star becomes a moving target that if traced out would slowly move off center during rotation.

I took my master east flat and calibrated it using no dark or bias but did use my master west flat. The results are below. Now if I subtract my Master East from the Master East of course I get a clear image but not so using the master west.

[Warhen]

Just to back Steve up, Gerald- a precision rotator will match dust motes up perfectly regardless of what side of the pier lights/flats were taken. At issue is what can be a dramatic difference in field illumination as Steve shows. And agreed Steve, per tests which corroborate it, we want to use the pierside field in the header.

So, dear Juan- we're still not sure if my original description of assigning E/W filters via Add Custom is right for BPP or not. :>) Of course, this isn't an issue when calibrating manually- just use the matching side lights/flats in independent runs.

No worries on darks/biases Gerald! But you knew that. :>))

[oldwexi]

Hi Steve!
Aha, i did not think about using a rotator.
It is clear, if you rotate the camera when changing the pier side you need "other" flats too.
I did not expect that when changing the pier side that the camera is sometimes also rotated 180 degrees...

Thanks for your explanations!

Gerald

[sreilly]

Well I finally added something that might help. I'm hoping that software will eventually start reading pierside in the fits but then taking flats the way I do won't help either come to think of it as the mount doesn't move from its parked position, only the rotator but then maybe Bob Denny (ACP) can write to the pierside reflecting E/W flats for the flats.