Sensor analysis: sensor linear explanation

[ippiu]

Hi Robin.

I did multiple sensor analysis of my new Asi533: starting from 5-6 ms didn't work. 3ms stalled and didn't finish. Finally around 2-2,5 ms worked. It also worked around 1,6 ms too.

So i did in a row 3 sensor analysis starting between 1,6 and 2,4 ms just to see the final results. Almost all parameters were the same in all 3 analysis, as it must be (i suppose).
But up in the window there is a number in percentage next to sensor linear to: this number varied from 94,9% to 99,8%.
What does it mean? How can i interpret this number?

Thanks

RAW16.jpg (69.76 KiB) Viewed 1665 times

RAW16_1.jpg (69.95 KiB) Viewed 1665 times

RAW16_2.jpg (68.12 KiB) Viewed 1665 times

[admin]

Hi,

The linearity test was something I added when I was suspecting that a particular camera may not have been responding linearly to light. SharpCap measures the exposure needed for half maximum histogram peak and then measures the brightness at exposures between about 2% of that exposure and about 220% of exposure (this should be an exposure range that gives histogram peaks from about 1% up to full saturation and a bit beyond).

SharpCap then looks at the linearity of the data that it accumulates during this procedure and flanks the percentage ADU level where the data starts to move away from a nice straight line. This is an estimate of the point at which the sensor is beginning to feel the effects of saturation. The criteria is quite strict, so it's possible for a small random deviation or small variations in the illumination level to trigger lower readings for this calculation. To be honest I would not worry unless you had a sensor that was consistently giving readings under 90% – as you tend not to image with the details you are interested in above 90% saturation.

Hope this helps, Robin

[ippiu]

Thanks, Robin!
Excellent explanation...

[philwest558]

Hi Robin, my Altair GPcam 3 178M tests at 99.1% however my Altair 290M Gpcam3 tests at 79% (I've done lots of tests and they all came out about the same).

Does this mean I should be wary of getting near that saturation level, say keep it in the 60-70% region rather than any higher? Does 79% suggest an issue with the camera thanks.

Regards Phil

[admin]

Hi Phil,

unless you are doing things like photometry or variable star measurements, I wouldn't let it worry you. The threshold that SharpCap uses as a deviation from linear response to trigger the percentage value is pretty small (1-2%) - an amount that really isn't going to cause any issues if you are imaging just for the pictures.

cheers,

Robin

[philwest558]

Thanks Robin - much appreciated. Regards Phil

[oopfan]

@philwest558,

I completely agree with your 79% linearity ceiling on your 290M GPCAM3. I have a GPCAM2, and it has the same characteristic, however, I successfully practice Photometry just as long as I keep the max pixel value at or below 50,000 ADU through exposure adjustment. In cases where the star is still too bright at 1ms, I spread the light over a wider area by defocusing.

Interesting bit of math:
50,000 ADU / 65,536 ADU = 76.3
That is very close to your 79.

Brian

EDIT: Maybe one day you will want to give Photometry a try. It's addictive. My advice is to respect the linearity limit. If you are measuring the light curve of a 2-hour variable star, remember that if the star is rising in the East, it will grow brighter as it approaches the meridian. Therefore, begin your run with an exposure that places it at a maximum of 40,000 ADU, so by the time it reaches the meridian it will be 50,000 ADU. If you forget to do this, then the shape of the light curve will suffer.