Spectral Dispertion of low elevation stars

[Gama]

This post is more for those that have seen it or know of it.

For the past week, ive been struggling to find the cause of some issues ive been having lately with low elevation images (West horizon).
For a couple of nights, ive been pulling my hair out not being able to pinpoint what was wrong with some images i was getting with my setup.
Of course i havent had this issue before mainly because of small focal length scopes, but due to the longer focal length of the new scope (3454mm), and add to that the warm weather causing atmospherics, it reared its ugly head.
What the problem is for others (beginers) is that the stars are no longer round as you go lower into the horizon.
They start to elognate, just like tracking error, but doesnt matter if you expose for 10 sec, 1 sec or .1 sec, its still elognated.
I thought my scope had a mirror shift or flop issue, or something very wrong, but checking collimation showed text book results. Secondly, visually, the stars were round in the same location. I even used a laser collimator and checked for laser dot shift while i slewed in every direction. Everything was rock solid.
Then i thought it was the mount, and after a few minutes of cursing, i figured maybe it wasnt. The accuracy claim was +/- 5 arc seconds P-P without PEC, so you can imagine my cursing, sorry Mathis Mounts ... Because if the mount was at fault, i would still be able to image a round star at 0.01 sec even if the drive was off..
Plus on other nights, the amount of dispertion was different, and gave smaller errors, so i looked up the web for an answer and behold, i found what the problem was, "Spectral Dispertion of stars at low elevations".
Finally my panic attack was quashed by a statement by Roland Christen (Astro Physics) stating this exact problem.
Anyone else fall into this trap ?, as i really havent read anything on this much, if at all.

Theo.

[jase (Jason)]

Theo, when you say low, what are you talking? I typically apply the 30+ degree rule imaging and stay well away from blue filtered data until the target reaches 60+ degrees in elevation.

[Gama]

Below 30, but you can barely just start to make out at 40 deg.
Its very prominant in the red thru to IR, which is the real bandit.

I'll have to start using my UVIR filter, rather than clear.

Theo

[jase (Jason)]

There isn't a huge difference between Red and IR. Technically, red starts are ~550nm and ends at 850nm. IR kicks in at 700nm onward so there is an overlap window of 150nm (based on UBVRI filters). My guess is that its happening across the entire spectrum, not just red filtered data. In any case, yes, UV/IR blocking is recommended... if not mandatory for RGB as the IR signal poisons the data making it difficult to balance the colour correctly.

[Gama]

Actually you would be suprised how IR does effect it.
Heres a quote from the results of the article :
"CCD sensors are particularly prone to this problem because, as well as recording the visible spectrum, they are sensitive deep into the infra-red".

Full article is here : http://www.astrocruise.com/articles/selong/selong.htm

The issue is compounded because the Sony CCD in the QHY-8 is very sensitive ti IR, even deep IR.
Recomendations are to use RGB only, and leaving out the luminance when its poor seeing.

I like using a clear filter, as it provides that extra signal for galaxies when imaging.

Theo