Affiliation
American Association of Variable Star Observers (AAVSO)
Sat, 03/07/2020 - 12:22
I would like to do linearity tests on a remote telescope/CCD. They are having trouble with their flip flap. Is it possible to do this test using twilight or do I need to have a constant light source. (Pretty sure I know the answer there...) Has anyone done testing on a Moravian G2-8300 gen1 (Kaf-8300) ?
Anyone have experience with this CCD?
Thanks for any help
Jim in Sweden
Hello! I've used Landolt standards (or Henden) - taking different exposures and plotting time vs. ADU count. I would think a standard field, like M67 would work as well, since a single exposure would give standard stars with many different ADU counts
The advantage of lightbox flat, for me, was at the high and low ADU counts. For me and my equipment, accuracy began to decline. Best regards.
Mike
Hi Jim,
Give some more details about the remote installation. Is this in a dome (translucent or opaque), a single rolloff, or part of a larger telescope farm rolloff with many telescopes inside? Do you have an available webcam with remotely switchable NIR LED illumination?
Using twilight is really hard, because the illumination changes rapidly, there is a gradient in the sky, and stars will be present. It is vaguely possible if you take "guard" exposures between each linearity exposure. A guard exposure has constant exposure time, so you track how the illumination changes in these exposures and apply a correction to the linearity exposure. Personally, I wouldn't trust the result.
Usually a KAF8300 sensor is set up in the electronics so that you only digitize to about half full-well, as this is an anti-blooming gate sensor which gets nonlinear towards full well. If this is the case, then you can probably assume that it is linear to something over 60K counts and not bother to do a linearity test.
You can generally use local illumination, such as the webcam night-time LED glow, as your "constant" light source, when the dome or rolloff is closed. These LEDs typically are 860nm, so you need an I filter or a clear filter. You can also use a white card illuminated by moonlight (the moon is approaching full right now), though I'd use the guard exposure technique for that case.
Arne
Thank you for these constructive comments! The telescope is part of E-Eye in Extremadura Spain, a telescope farm with a roll off roof. We might be able to have them install a white screen and use ambient lighting, much like your suggestion.
Passing this info along.
Thanks!
Jim
Hi Jim,
Often you don't even need to install a white screen, if you have ambient lighting. Just point the telescope at a wall or the roof. As long as the light source or the telscope or the surface the light is reflecting off of don't change/move, you can do a linearity test. You don't have to have uniform lighting as you would with a flat. A white screen won't hurt, but it is not necessary except perhaps to get enough light into the telescope (wooden walls are not great reflectors!).
Arne
Arne