QHY168c Gain Scaling question

glauria
Posts: 17
Joined: Wed May 30, 2018 7:15 pm

Re: QHY168c Gain Scaling question

Post by glauria » Thu Sep 05, 2019 8:40 pm

Thanks Robin,

At least these measurements confirm where the unity gain setting is. Insofar as the read noise is concerned, I'm not surprised that it doesn't drop off.
If you look at the read noise vs. gain graph that QHY posts for the camera, it's pretty constant over the entire gain range. Its max is at 3.2e- at (minumum) gain equals 1 and 2.8 for maximum gain of 14. Here's the specs. of the camera that QHY has posted: https://www.qhyccd.com/index.php?m=cont ... d=13&cut=1

Now that the sensor analysis is complete, I'll just have to get outside and see how things work with the smart histogram. I haven't been able to do that because the weather in Tucson is pretty poor during the summer and it rarely clears off in the evening for observing. Things should start to improve during the month of September though.

Thanks for your help!
-Gene

glauria
Posts: 17
Joined: Wed May 30, 2018 7:15 pm

Re: QHY168c Gain Scaling question

Post by glauria » Fri Oct 04, 2019 1:02 am

Robin,

Something still isn't making sense to me in regards towards the unity gain value being between 1 & 2 which to me seems way too low. I think I said this in an earlier post, but the generic gain range that QHY posts on their web page for this camera is from 0 to 14 where 10 is the value for unity gain. After running the sensor analysis in Sharpcap, the value for unity gain comes out to roughly 1.8 which isn't achievable since the gain changes by integer numbers only.

What I don't understand here is that I really can't get any values below 1.57 e-/ADU with Sharpcap's gain value equal to 1 which, according to QHY's graph, gives an equivalent gain in their units of 7. Likewise for their maximum gain value of 14, which translates to roughly 0.234 e-/ADU, scales roughly to a Sharpcap gain value of only 6. So I'm afraid that I still don't know how to interpret how Sharpcap's gain translates to QHY's range from 0 to 14. i.e., according to the sensor analysis, the gain using sharpcap from 1 to 6 scales to a value of 7 to 14 using QHY's convention.

Incidentally, has anyone done a sensor analysis for a ZWO ASI071? It's profile should be the same as the QHY 168c since it uses the same sensor.
-Gene

User avatar
admin
Site Admin
Posts: 3289
Joined: Sat Feb 11, 2017 3:52 pm
Location: Vale of the White Horse, UK
Contact:

Re: QHY168c Gain Scaling question

Post by admin » Fri Oct 04, 2019 5:41 pm

Hi,

ZWO have arranged for the gain value for all of their cameras to be measured in tenths of a dB, so value of zero is no gain value of 100 is 10 dB and so on. I have the 071 camera and have run the central analysis on it many times and get a very good curve showing that you can access different gains with fine control of the value and the read noise and e/ADU value changing smoothly with gain as you would expect.

Somehow, QHY have really made the gain for their version of the camera behave very oddly and I suspect that the reason that the values differ from those you are seeing in documentation is that the behaviour has changed with different versions of the SDK. To me it feels as though they have not gone to the effort of making the gain control work in dB and instead just made the raw values sent to the camera's configuration available by the SDK.

Sadly until they get round to fixing this there's really nothing I can do from the SharpCap side to improve the behaviour.

Cheers, Robin

glauria
Posts: 17
Joined: Wed May 30, 2018 7:15 pm

Re: QHY168c Gain Scaling question

Post by glauria » Fri Oct 04, 2019 7:59 pm

Very good Robin,

That will explain the difficulties that I've been running into by using the sensor analysis tool. I'll just have to experiment and see which sharpcap gain settings will work.

Alternatively, I can use another way to capture images and store them into a folder and then load them into Sharpcap for live stacking. In my case, I can use INDI/Ekos to take the images and in there, the gain settingd go from 0 to 14. However, it's much better to control the camera from Sharpcap since I can take advantage of the auto color balance which I can't when using the folder option.

To try and find the unity gain value (as a starting point for gain settings) in Sharpcap, perhaps I can set the camera to unity gain (10) using Ekos, load the image file into Sharpcap, and look at the histogram. Then try and adjust the gain in Sharpcap with the same exposure time to give me the same histogram shape (or close). Maybe that would work?

I don't know how far I will get, but I will forward your previous message to QHY and see if they can change their SDK to make it more compatible for you to work with it. I'm curios though, do other QHY cameras have the same problem or only the 168c?

-Gene

User avatar
admin
Site Admin
Posts: 3289
Joined: Sat Feb 11, 2017 3:52 pm
Location: Vale of the White Horse, UK
Contact:

Re: QHY168c Gain Scaling question

Post by admin » Sun Oct 06, 2019 1:14 pm

Hi,

A lot of their cameras have sensible gain values, although they aren't as standardised as the ZWO ones. I have some cameras where the gain is measured in dB and you can choose a value from 0 to 50 or so. There are other cameras where the gain is measured in 0.1 dB so you can choose a value from 0 to 500 or so. Then there are cameras like yours where the gain doesn't seem to be directly related to the DB value at all and you get ranges from 0 to 255

It would be much easier if they were all consistent!

Cheers, Robin

glauria
Posts: 17
Joined: Wed May 30, 2018 7:15 pm

Re: QHY168c Gain Scaling question

Post by glauria » Mon Oct 07, 2019 8:05 pm

Very good Robin,

So this brings me all the way back to the beginning of this post where I am trying to get an idea how QHY's 0 to 14 scale maps to Sharpcap's 1 to 255. Since it looks like the sensor analysis tool doesn't tend to work very well since the gain isn't related to decibles, it's hard for me to figure what value of gain in the Shapcap 1 to 255 scale gives me unity gain of 1e-/ADU as a reference point. Right now, I have no idea what gain settings to use with Sharpcap. I'm just doing it by trial and error.

The only way I know what gain I'm using in my case is that I use Ekos to control the camera where I can set the gain to 14 which puts the images into a folder and then load them into Sharpcap for live stacking. It's a bit more cumbersome to do it this way because the folder organization gets more complicated when using 2 different software packages do to your imaging with.

-Gene

bwells
Posts: 28
Joined: Sat Sep 08, 2018 1:01 am

Re: QHY168c Gain Scaling question

Post by bwells » Mon Oct 21, 2019 1:05 pm

Robin:

I installed the latest version of SharpCap to try with my 165C camera. I found that my sensor analysis came much closer to working than anything previously. I noticed the gain and offset ranges on the SharpCap sliders are more in line with what I see from the native QHY drivers, whereas before, SharpCap gave me a much shorter range of values. I also noticed the temperature adjustment works now, even when it sets the value to 255, the camera continues to cool.

I suspect both of these work because QHY fixed bugs in their drivers.

The problem I have now with SharpCap and my QHY165C is that the Sensor Analysis starts at a gain value of 1, but only goes up to a gain value of 500. But Unity gain is at 2800. I see the gain slider in SharpCap goes up around 4200. Here are two different analyses, the first at a brighter light level than the 2nd.
SharpCap_Analysis_SensorAnalysis.JPG
SharpCap_Analysis_SensorAnalysis.JPG (185.7 KiB) Viewed 384 times
What are your thoughts? Can I use this analysis and how would it help/hurt the smart histogram?

Thank you
Bryan

User avatar
admin
Site Admin
Posts: 3289
Joined: Sat Feb 11, 2017 3:52 pm
Location: Vale of the White Horse, UK
Contact:

Re: QHY168c Gain Scaling question

Post by admin » Mon Oct 21, 2019 5:53 pm

Hi Bryan,

It looks like qhy have changed the mapping of the gain values to actual gains in their SDK and because of that SharpCap is mistakenly cutting off the gain measurements at 500 thinking that that value is equivalent to 50 dB. I will have to work out another way of picking the right maximum gain for cameras like yours.

Cheers, Robin

bwells
Posts: 28
Joined: Sat Sep 08, 2018 1:01 am

Re: QHY168c Gain Scaling question

Post by bwells » Mon Oct 21, 2019 6:46 pm

Robin:

Perhaps something I could set in a config file, on the command line or through your scripting tool. Looking forward to using SharpCap to its fullest!

Thanks
Bryan

User avatar
admin
Site Admin
Posts: 3289
Joined: Sat Feb 11, 2017 3:52 pm
Location: Vale of the White Horse, UK
Contact:

Re: QHY168c Gain Scaling question

Post by admin » Mon Oct 21, 2019 8:26 pm

glauria wrote:
Mon Oct 07, 2019 8:05 pm
Very good Robin,

So this brings me all the way back to the beginning of this post where I am trying to get an idea how QHY's 0 to 14 scale maps to Sharpcap's 1 to 255. Since it looks like the sensor analysis tool doesn't tend to work very well since the gain isn't related to decibles, it's hard for me to figure what value of gain in the Shapcap 1 to 255 scale gives me unity gain of 1e-/ADU as a reference point. Right now, I have no idea what gain settings to use with Sharpcap. I'm just doing it by trial and error.

The only way I know what gain I'm using in my case is that I use Ekos to control the camera where I can set the gain to 14 which puts the images into a folder and then load them into Sharpcap for live stacking. It's a bit more cumbersome to do it this way because the folder organization gets more complicated when using 2 different software packages do to your imaging with.

-Gene
Hi Gene,

I have just been thinking about this again and wondered if I was missing a trick here. I wonder if the gain values on this camera accept fractional values – for instance setting a value of 1.5 or 2.62. Would you be able to try out a test version for me if I make a version where the gain value can be a fraction to see if it gives you smoother control of the gain of the camera.

Cheers, Robin

Post Reply

Who is online

Users browsing this forum: No registered users and 1 guest