SharpCap Forums

Hi Robin,

I did the gain test that you suggested (sorry, it took a while since I was on vacation). First test was setting the gain at 1 and adjust the exposure to put the histogram at 50%.

Gain: Exposure Time: Mean count:
1 315 ms 34535
25 12.4 ms 35515
255 1ms 31453

I also did a test where I started at a gain of 255, and adjusted the exposure to get the histogram at 50%. Then I increased the exposure by a factor of 10 and adjusted the gain to put the histogram back at 50%. Here's what I found:

Gain: Exposure Time: Mean count:
255 0.9 ms 32365
33 9 ms 33842
3 103 ms 34850

Hopefully, these measurements will help give a better idea on how the gain is scaled.
-Gene

Hi Gene,

Yes, I did see your original results and it seems like the gain is actually measuring the multiple of brightness being applied. For instance going from gain 1 to gain 25, the exposure reduces by a factor of almost exactly 25. Again, going from gain 25 to gain 255 the exposure reduces by about a factor of 10.

I will try to update the sensor analysis code to be aware of this possibility for qhy brand cameras.

Cheers, Robin

Great Robin,

I'm glad that I the data I provided helped to give you a better understanding on what's going on. I'm not sure if other QHY users ran into the same situation when they performed the sensor analysis for their cameras. I'm sure that once you make this change, my results will make a lot more sense.

Thanks,
-Gene

Hi,

The problem I have here is that I have seen other qhy cameras which also have a gain range of 1 to 255 where the gain doesn't work in the same way – on those cameras gains above about 150 are ridiculously high (and to be truthful no use to anyone). This is why the sensor analysis only works on the range of gains from 1 to 100 on your camera – since going to higher values was causing problems on those other models.

I think what I will do is up the maximum gain measurement to 150 in the case that the ranges from 1 to 255 and also make a change to try and get a more even spread of values throughout the range.

Cheers, Robin

This is timely. I just purchased SharpCap Pro & a QHY168c. Please advise how you would proceed at this point, i.e. SharpCapPro patch, etc..

Thanks/Matthew

Hi,

Always use the latest version of SharpCap unless you have a very good reason not to. As far as we understand now the gain value for this camera is effectively a multiplier, so gain one is the minimum, gain two will make the image twice as bright and so on.

The sensor analysis run the gain up to a maximum value of 100 currently, which is probably sufficient for most general purposes as you be unlikely to go past that point for deep sky imaging. In the very latest version it should pick a better range of values between gain 1 and gain 100 with more values at low gain to give a better curve.

Cheers, Robin

Thanks Robin,

I'll download the latest version of Sharpcap and run another sensor analysis to see if there is much of an improvement. I'll post the graph of the results to see what it's like by taking more samples at the lower gain levels.

-Gene

Hi,

I've also been trying to determine what the best gain and offset settings are for my QHY168c. I've attempted to do a sensor analysis 6 times but keep getting errors. I'm getting stuck after it runs the read noise analysis. While it's running the "Measuring relative gains" it gives me multiple notifications that the exposures are getting long because of low light levels but then I get an error message saying "Minimum exposure reached. Reduce the light level and try again" (see screenshot). What can I do? Should I adjust the light level while it runs the "Measuring relative gains?"

glauria wrote: ↑Wed Aug 21, 2019 11:35 pm Thanks Robin,

I'll download the latest version of Sharpcap and run another sensor analysis to see if there is much of an improvement. I'll post the graph of the results to see what it's like by taking more samples at the lower gain levels.

-Gene

Robin,

Here are the results after running v3.2.6086.0
Gain Value e/ADU Read Noise (e) Full Well (e) Relative Gain Rel. Gain (db) Dynamic Range (Stops)
1 1.56856841380373 2.85293338653619 25699.4248917603 1 0 13.1370023926713
2 0.789591995077765 2.87094177631257 12936.6752473541 1.9865556180686 5.9620145721433 12.1376551963785
3 0.531042377831612 2.89534183115944 8700.59831839313 2.95375374788094 9.40748571393744 11.55316521318
6 0.274499939416344 2.99671419720202 4497.40700739738 5.71427599269749 15.1392242491491 10.5514962383141
10 0.16946113830352 3.0813996204543 2776.45128996487 9.25621313244272 19.3286669214195 9.81544057548719
15 0.112158753226932 3.0521713968989 1837.60901287006 13.9852518744571 22.913405848476 9.23377814242049
25 0.0669372171343297 3.04362282232868 1096.69936552886 23.4334273361847 27.3967162493925 8.49316279801563
39 0.0434602217672655 3.08514523224894 712.052273434877 36.0920480848809 31.1482305459974 7.85050094110386
63 0.0268736824464538 3.07737267934428 440.298413202699 58.3681978429687 35.3235256877085 7.16063866563038
99 0.0165749037885223 2.98288932265333 271.563223671149 94.6351444217689 39.5210489846947 6.50843385744856
100 0.0164323022294419 2.97967240794788 269.226839727177 95.4563999555284 39.5961010269756 6.49752470567135

As you mentioned, it looks like the latest version of the tool takes smaller steps in the gain and I'm getting a curve that looks more "typical". However, I'm still not sure that I understand the results. I am trying to find the gain setting that gives me unity gain and from the table, it looks like it's somewhere between 1 & 2. In using the camera, I'm not sure I trust this value because I know that this value is way too low. In using this value, my images are way too dim.

Hi,

You're quite right that the unity gain value is somewhere between gain 1 and gain 2. The only thing that is a little unusual about those results is that the read noise isn't dropping off significantly as you go to higher gains which you would usually expect. The fact that the read noise doesn't drop off means that there is no real advantage in going to higher gain values on this camera.

There's no real reason that I can see to make me expect that images at unity gain would be of any particular brightness level, so I'm not sure that that conclusion applies in this case.

Cheers, Robin