Back to forums

Need to take bias longer than 0 seconds on the QHY Minicam

QHYCCD miniCAM8M Bias Frames Calibration
10 replies•295 views
Noah Tingey avatar
Noah Tingey avatar
Below are two bias frames taken on my new QHY Minicam8M. The top one has 0.25 seconds of exposure, and the bottom has 0 seconds of exposure. See the gradient in the bottom one? The right side is darker.



Also, when zooming in, notice how the noise pattern in the 0 second image is funky:



Lastly, here is the histogram of the 0.25 second image:

And the histogram -- at the same zoom level & horizontal position -- of the 0 second image:


Assuming that my minicam isn't defective and this behavior can be seen in all minicams, then just a PSA that you should take your bias frames with an exposure length longer than zero seconds. I have found that 0.25 seconds is sufficient, but I don't know if that's the ideal length. Not sure if this has been documented elsewhere yet, so I decided to make a post about it.
Well Written Helpful Insightful Engaging Supportive
Iosif Bodnariu avatar
Iosif Bodnariu avatar
A bias frame is taken at the shortest exposure of the camera, specified by the manufacturer.

Your camera have 11μs-900sec.

miniCAM8 (IMX585) | QHYCCD Astronomical & Scientific Camera
Jan Erik Vallestad avatar
Jan Erik Vallestad avatar
Or better yet scrap the BIAS frames and do darkflats. I don't see why one would shoot BIAS with a cooled camera anyway.
Yilia Yu avatar
Yilia Yu avatar
Hi, the QHYCCD technician suggests that bias should be taken within 20 ms.
Stelios_Stergiou avatar
Stelios_Stergiou avatar
Jan Erik Vallestad:
Or better yet scrap the BIAS frames and do darkflats. I don't see why one would shoot BIAS with a cooled camera anyway.

because they watching too many lazy YouTubers… 😉
TiffsAndAstro avatar
TiffsAndAstro avatar
Jan Erik Vallestad:
Or better yet scrap the BIAS frames and do darkflats. I don't see why one would shoot BIAS with a cooled camera anyway.

because they watching too many lazy YouTubers… 😉


Lol me too
Noah Tingey avatar
Noah Tingey avatar
Iosif Bodnariu:
A bias frame is taken at the shortest exposure of the camera, specified by the manufacturer.

Your camera have 11μs-900sec.

miniCAM8 (IMX585) | QHYCCD Astronomical & Scientific Camera

This is not true.

Many cameras enter different modes depending on exposure length, which can mess with your bias frames. In the QHY driver, this is called "Stream mode."
Well Written Insightful Concise
Noah Tingey avatar
Noah Tingey avatar
Jan Erik Vallestad:
Or better yet scrap the BIAS frames and do darkflats. I don't see why one would shoot BIAS with a cooled camera anyway.

Assuming that you're dithering, why would you take darks (or darkflats) at all with a camera that has zero amp glow?
Well Written Engaging
andrea tasselli avatar
andrea tasselli avatar
Noah Tingey:
Assuming that you're dithering, why would you take darks (or darkflats) at all with a camera that has zero amp glow?


As far as darks go because is better this way and costs nothing and for darkflats because you'll get a precise calibration with your flats.
Charles Hagen avatar
Charles Hagen avatar
Iosif Bodnariu:
A bias frame is taken at the shortest exposure of the camera, specified by the manufacturer.

Your camera have 11μs-900sec.

miniCAM8 (IMX585) | QHYCCD Astronomical & Scientific Camera

As has just been experimentally confirmed, the shortest possible exposure will not produce a result that will adequately correct the real bias signal of longer exposures. The point of taking calibration frames isn't to conform to convention, it is to correct your exposures. Shortest possible exposures here will invariably produce uneven flat field correction across the frame and will introduce a gradient that did not exist in the lights. There is no reason to do this. In Noah's 0.25s example, there is absolutely not going to be any significant dark current so there is nothing wrong with using that as a bias. This is precisely the reason why for years I have suggested taking 0.2s+ bias exposures - there's nothing to lose and you can avoid issues with stream modes and other short-exposure variability.
Well Written Helpful Insightful Concise
Charles Hagen avatar
Charles Hagen avatar
Jan Erik Vallestad:
Or better yet scrap the BIAS frames and do darkflats. I don't see why one would shoot BIAS with a cooled camera anyway.

With even an elementary analysis of the signal components associated with bias and dark flats, you will find that they are effectively identical. There is no significant additional contribution from dark current at reasonable temperatures (even uncooled with the 585) that could change the calibration frame. Dark flats on cameras that have near-zero amp glow and very low dark current are effectively just bias. There's no reason to spend any additional time taking new dark flats all the time to exposure-match your flat frames. This also does nothing for the primary use for bias (with this calibration routine) being light correction. Most people do not take darks with these types of sensor, and for good reason.
Helpful Insightful Concise