Sensor Analysis interpretation - Orion Starshoot G16 (same as Touptek ATR3-16000-KPA) - Need Help!

[jasonjeremiah]

I just bought my first dedicated astro cam a few weeks ago and have yet to get any sky time with it. I have been imaging with a modified T4i and D5500 and have produced some great images but decided to take the next step. I bought an Orion Starshoot G16 OSC camera at a decent price, but I'm already questioning my choice as information is very limited and rare on usage of this camera. Luckily, it is the exact same camera as the Touptek ATR3-16000-KPA and there is a little information out there regarding this camera, as well as drivers that, for the most part, seem compatible with the camera. It shares the same Panasonic sensor as found in the QHY 163C and ASI 1600MC, though the software and drivers seem to be vastly different.

I ran a sensor analysis from Sharpcap using the ASCOM driver and the results produced are quite different from anything I've ever seen. I ran the analysis twice and results were exactly the same. When I initially connect to the correct, there is some ASCOM dialog pertaining to bit depth and HCG and LCG (though both were greyed out), as well as black point. I left all settings at the default except for bit depth, which is set at 12 which is the highest for this camera.

The attached JPG and spreadsheet show the results. I posted these results on CloudyNights and one user suggested there was something wrong due to the jump in readout noise at higher gains. Could someone please interpret what I'm looking at here? What does this mean as far as preferred gain and offset values? Thanks for your help!

[admin]

Hi,

the odd thing there is really the read noise for low gain values (less than about 300). I wonder if the sensor data is being clipped at the zero point for low gain, leading to low estimates of the read noise? If your camera driver has an 'offset' or 'black level' adjustment in the ASCOM camera settings, try increasing that and then running the analysis again.

cheers,

Robin

[jasonjeremiah]

You can adjust the black level in the ASCOM settings when initially connecting. Any suggestions as to what setting I should use?

[admin]

Hi,

set the camera to maximum gain (or at least the maximum that the sensor analysis uses - 2000). Then try different black level settings and capture dark frames (1s exposure will do) at 16 bit and look at the histogram. Aim for a black level value that means that the histogram peak is not touching the left hand side of the graph.

You want it to look like this

goodhisto.JPG (16.7 KiB) Viewed 1197 times

Not like this

badhisto.JPG (15.71 KiB) Viewed 1197 times

cheers,

Robin

[jasonjeremiah]

Hi Robin - thanks for the help. I will need to re-run the analysis using the suggested gain settings. In the meantime, I did find this info on Astrobin from two users using the same camera; one is using the Touptek and the other using an Explore Scientific OSC which is a rebranded Touptek. Interestingly enough, the one user was setting the analog gain prior to using the camera through Sharpcap or APT (using the ASCOM driver). This is confusing as I see no reason why gain could not be set from inside the acquisition program of choice using the gain setting. But maybe the appropriate way it to set the gain prior to connecting. Your thoughts?

[admin]

Hi,

if the ASCOM driver is written properly then there shouldn't be any need to set it first - the slider in SharpCap should have the same effect as the options in the ASCOM driver setup window, and both should affect the frames captured after any change.

cheers,

Robin

[jasonjeremiah]

I re-ran the sensor analysis using a black level offset of 90 which resulted in a histogram similar to the one you posted above. This is the result from the new sensor analysis. What does this tell me as far as optimal acquisition settings? The manual does not state any unity gain or preferred gain or all.

[admin]

Hi,

well, unity gain is the point where the e/ADU value is 1, so about gain 500. However, my opinion is that there is nothing 'special' about unity gain...

Anyway, the analysis looks good this time - I would suggest keeping the black level offset at 90 in future as that will avoid any black level clipping.

As for what settings to use, those depend on a number of factors - how much light pollution you have, how long you can trust your mount to track/guide reliably for, etc. The SharpCap Smart Histogram functionality is designed to take these into account and give you a custom recommendation based on your sensor analysis data, sky brightness and your choices for minimum/maximum sub length. See https://docs.sharpcap.co.uk/4.0/#Smart%20Histogram and in particular try out the smart histogram 'brain' window.

cheers,

Robin

[jasonjeremiah]

Robin - thanks for your help thus far. I tried running the Smart Histogram but it was taking a long time to run (15+ minutes) and never posted an optimal time so I cancelled. How long should the Smart Histogram run before giving a result?

Another aside topic - are there any plans to add native support for Touptek cameras? The SDK is available for their CMOS cameras and this covers a wide range of CMOS astro cams as they're rebranded under a variety of different names (Omegon, ES, Orion, RisingCam, Mallincam, etc). I've only been using Sharpcap for polar alignment and NINA for image acquisition as native support exists in NINA. I find it much easier to adjust settings using the native drivers vs the ASCOM driver. Thanks!

[admin]

Hi,

the whole analysis process shouldn't take more than about 5 minutes. It also shouldn't need to use exposure lengths longer than about 2 seconds. Sometimes there can be problems if the camera does not implement exposures accurately - for instance there was a camera that for exposures above 1s would round to the nearest 0.2s, so you could get 1s, 1.2s, 1.4s, etc but nothing in between. This could cause problems when SharpCap wanted a particular image brightness as it might find it could not be achieved (and even if it got close enough, the incorrect exposure would lead to errors).

Feel free to post the log after attempting the analysis (make sure you are using the latest SC 4.0 first) and I will look at what is going on. The log will likely be quite large

Right now I have no plans to add direct support for the TT branded cameras to SharpCap, sorry.

cheers,

Robin