Help with RANSAC errors

allen456 said:

M82_again - Google Drive

drive.google.com drive.google.com

I use ASIAir+ for everything, for at least the last two years.
Click to expand...
Can you add a single raw dark sub to that folder?

Although not as severe as the last samples, your darks still have higher backgrounds than your lights. If the camera settings (gain and offset) are the same in both cases, that strongly suggests that your darks aren't being acquired in the dark. That you have some kind of light leak. How do you take darks with your setup (since this camera doesn't have a mechanical shutter)?
 
Last edited:
cloudbait said:
Can you add a single raw dark sub to that folder?

Although not as severe as the last samples, your darks still have higher backgrounds than your lights. If the camera settings (gain and offset) are the same in both cases, that strongly suggests that your darks aren't being acquired in the dark. That you have some kind of light leak. How do you take darks with your setup (since this camera doesn't have a mechanical shutter)?
Click to expand...

M82_again - Google Drive

drive.google.com drive.google.com
 
allen456 said:

M82_again - Google Drive

drive.google.com drive.google.com
Click to expand...
Well, your individual dark has a significantly higher background and mean than your individual light, which should be impossible if the two were taken with the same camera settings (which the header seems to suggest they were). Unless your dark was taken under conditions where some light could reach the sensor.

Also hard to explain is why the mean for your dark sub is so much higher than the mean for your master dark.
 
Last edited:
The dark image is (with high uniformity) almost exactly 20% brighter than the light image background. It is much too uniform to be a light leak. This must be a difference in camera settings.
I note that the dark and light were taken three months apart. The slight differences in wording of the FITS header suggest that there may have been some change to the capture software or camera driver. The light was captured yesterday; could you capture another dark with exactly the same software and settings as yesterday's light?
 
fredvanner said:
The dark image is (with high uniformity) almost exactly 20% brighter than the light image background. It is much too uniform to be a light leak. This must be a difference in camera settings.
I note that the dark and light were taken three months apart. The slight differences in wording of the FITS header suggest that there may have been some change to the capture software or camera driver. The light was captured yesterday; could you capture another dark with exactly the same software and settings as yesterday's light?
Click to expand...
Some kind of driver change is definitely possible. But a light leak can be very uniform if its coming from the open end of the OTA (covered with something not completely opaque) or from a camera capped with plastic.
 
cloudbait said:
or from a camera capped with plastic.
Click to expand...
If anyone takes darks with a capped camera this is a real risk. I think a lot of folks don't get the message - it's not about the cap being leaky, its about it being simply transparent; it looks opaque and black at visible wavelengths, but it can be transparent (just like clear plastic) at infrared wavelength within the sensitivity of the camera.
 
fredvanner said:
If anyone takes darks with a capped camera this is a real risk. I think a lot of folks don't get the message - it's not about the cap being leaky, its about it being simply transparent; it looks opaque and black at visible wavelengths, but it can be transparent (just like clear plastic) at infrared wavelength within the sensitivity of the camera.
Click to expand...
Yup. But it does happen. Something has elevated the background of the darks here. Could be a sensor offset not reflected in the numbers the driver provides, or it could be actual photons. If the latter, it has to be from a diffuse source, not some kind of internal reflection. A non-opaque cap on the camera or on the telescope could cause that.
 
I considered this so I took a dark with just the cap on and checked the Mean in Statistics. Then I put the camera (still capped) in one of those bags that are light proof that film photographers used when they needed to open a camera and didn't have a lightroom handy. I only took one dark because I knew that the camera would heat up pretty quickly but the Mean for that shot matched previous one exactly.
 
allen456 said:
I considered this so I took a dark with just the cap on and checked the Mean in Statistics. Then I put the camera (still capped) in one of those bags that are light proof that film photographers used when they needed to open a camera and didn't have a lightroom handy. I only took one dark because I knew that the camera would heat up pretty quickly but the Mean for that shot matched previous one exactly.
Click to expand...
You just took this dark today? Can you upload it?
 
allen456 said:
Yesterday. I added the to the M82_again folder. They were both 60sec darks. The first was capped. The second was capped and in the black bag.
Click to expand...
The level doesn't look unreasonable, but the dark is 60s, which is hard to compare with the 180s light.
 
Your new 180s darks are a good match to your 180s light. I don't know what had happened with your old dark.
 
The trouble started when I was using an ASI183MC Pro, which was badly over sampling (although it did fine for last years galaxy season). After trying to resolve this for almost a week, I switched to the ASI294MC, with much larger pixels and much less noise, but continued, at first, to have failures. But for the last three sessions I've been having clean results from WBPP.
 
You must log in or register to reply here.
Back
Top