5 replies to this topic

[tackettbr]

In the last month, I finally upgraded my flat strategy from a 20 dollar tracing board to Spike A Flat. I had used it with my C 9.25 with good results, but have struggled with it on my RASA 8 combination with my ASI 294 MC pro. I take my flats with the panel resting on the dew shield. Every flat I take leaves a horizontal banding effect. I have not had this banding with the tracing board when I have retried it. 

 

This is present at flats anywhere from 0.02s to 0.8s. I have tried all different dimmer settings with resultsing smaller and larger bands, but still bands. I have also rotated the flat panel to 45 and 90 degrees with the exact same pattern. So I don't believe it is any wiring in the panel. I have also rotated the USB3 and cooler power cable with no change. I have removed the cooler power cable and moved the USB 3 at 0 and 90 degrees with the same horizontal pattern. I also tried adding 1 -3 layers of white T- shirt layers persistent banding. 

 

Finally, the banding tends to not be in the same place (walking banding) each sub flat.  Any suggests or thoughts would be greatly appreciated. 

 

Here are several examples in final image, master flat, and individual flat. (The flat frames are severely compressed jpgs. They look fine other than the banding at full fits resolution)

 

 

Attached Thumbnails

• 
• 
• 

[pyrasanth]

I have the 20" square Spike A flat panel that I use with my C14 with great results.

 

When you say you have tried all settings for the levels is this with the basic control or the USB dimmer that gives you 1023 different levels?

 

I suspect that your panel is too bright for the RASA. I can see you have tried really low exposure settings however what ever the sensitivity of your OTA you still need to be getting exposures in the 1-2 second range as you will not be able to adequately fill the wells on the sensor. 

 

Try a simple test. Buy some nuetral density gel film and use several sheets to get the panel light really low so you are exposing in the 2-3 second range and see if that makes a difference. I am waiting for my 11" RASA V2 to arrive from Celestron and I will be testing with my panel so I'd be interested to see if I can emulate your result and see if there is a fix.

 

I doubt it is the panel as these are well proven. I've seen similar artefacts when a sensor is being overexposed in some regions but the overall exposure is low.

Edited by pyrasanth, 24 January 2021 - 04:55 PM.

[tackettbr]

 

Thanks for the reply. I will have to try the neutral gel film.

 

I should clarify, I have set the dimmer ranging from absolute lowest setting up to 60% with 10 steps in brightness levels in between and then adjust the exposure length to get the histogram with 15-30K ADUs.

 

With my C9.25 I normally have my flat exposures between 1.2 s and 1.5 and they have been great allowing me to use dark flats at the same exposure length.  

Edited by tackettbr, 24 January 2021 - 05:03 PM.

[pyrasanth]

Thanks for the reply. I will have to try the neutral gel film.

 

I should clarify, I have set the dimmer ranging from absolute lowest setting up to 60% with 10 steps in brightness levels in between and then adjust the exposure length to get the histogram with 15-30K ADUs.

 

With my C9.25 I normally have my flat exposures between 1.2 s and 1.5 and they have been great allowing me to use dark flats at the same exposure length.  

Even at F11 on a 2x2 bin my panel is set in the low region of the 1023 range of possible settings with the USB dimmer. If you don't have the dimmer you may find it a useful upgrade to get a much more granular range of brightness levels. 

 

The RASA is a very fast instrument so the use of panels is problematic to get a sufficently low brightness to prevent your sensor from producing odd effects. I might have to use ND 1.2 gels on my panel for the RASA 11. I'm not sure yet.

[tackettbr]

Even at F11 on a 2x2 bin my panel is set in the low region of the 1023 range of possible settings with the USB dimmer. If you don't have the dimmer you may find it a useful upgrade to get a much more granular range of brightness levels. 

 

The RASA is a very fast instrument so the use of panels is problematic to get a sufficently low brightness to prevent your sensor from producing odd effects. I might have to use ND 1.2 gels on my panel for the RASA 11. I'm not sure yet.

I finally was able to obtain neutral density gel film after the first package was damaged in transit. I was able to remove almost all of the  horizontal banding using 6 sec flat exposures and equivalent darks. Thanks again for the help and suggestion. CS! -Brandon 

Attached Thumbnails

• 

Edited by tackettbr, 20 February 2021 - 04:55 PM.

[pyrasanth]

I finally was able to obtain neutral density gel film after the first package was damaged in transit. I was able to remove almost all of the  horizontal banding using 6 sec flat exposures and equivalent darks. Thanks again for the help and suggestion. CS! -Brandon 

I'm glad my suggestion helped. I think the banding was caused by over exposure in some areas and not enough light in other parts of the sensor. Long subs tend to even this out.

 

I find flats very problematic. I created a set of masters last december and they worked fine. I copied these to a storage area for later use. When I retried them with the new PI WBPP script they would not correct the very same lights that originally corrected fine- the result looked like no flat had been applied. It was lucky for me that I kept the original subs that made the flats in the first instance.

 

I put all these subs into the WBPP script and made new master flats and the subs corrected fine. Thinking back the dust motes looked inverted- it is very possible the flats got copied into the fits array inverted as there is a box to alter the flat top down order which sets where the start x-y coordinates are- I wonder if this was the issue- a data mismatch somewhere.

 

It is really weird that the slightest setting change, even under the hood  of a new update, can make significant changes to the expected results. I always see flats as a nuisance but alas an essential processing step.

Edited by pyrasanth, 23 February 2021 - 06:35 PM.