ZWO ASI178MM-cooled – gain settings

I have been looking through the internet for information on what gain settings people use for astrophotography with their CCD/CMOS imagers. It does look like there are certain settings which can be useful for the older sensors, but how about for the Sony sensor of the ASI178MM-cooled? I decided to find out about the difference between a few of the gain settings on this camera, as this is worth knowing for people who are genuinely interested in using it for guided deep sky imaging.

Test 1 (short exposure):

I had set up the camera with the Samyang 135mm F2 lens and Astronomik SII filter for taking flat frames – I used a white t-shirt to cover the lens and adjusted the exposure time at each gain level to get the histogram to peak at about the same level. Below are the results.

For Gain 0, I required 100ms to get the histogram to peak @ 112.

Gain 0, 100ms exposure
Gain 0, 100ms exposure

At Gain 100 (or 19%), it took 33.0ms to get the histogram peak at 112.

Gain 100, 33ms exposure
Gain 100, 33ms exposure

At Gain 200 (or 39%), it took 11.0ms to get the histogram peak at 112.

Gain 200, 10ms exposure
Gain 200, 11ms exposure

At Gain 300 (or 58%), it took 3.5ms to get the histogram to peak at 112.

Gain 300, 3.5ms exposure
Gain 300, 3.5ms exposure

From the above test, it looks like the exposure time required goes down to about a third for every increase of gain by 100. The curve of the histogram does spread out wider with each increase in gain as well.

Test 2 (longer exposures):

For this test I had set up to image the black surface of my home-made dew shield, using the Samyang 135mm lens with Astronomik SII filter with an exposure time of 5 mins @ F22 (smallest aperture for the lens), being lit by the compact fluorescent bulb in my dining room. I then boosted the exposure until the top shoulder reaches the 1/4 mark of the histogram.

5 min sub @ Gain 100, requires + 3.59 EV boost to bring top shoulder to 1/4 mark on histogram
5 min sub @ Gain 100, requires + 3.59 EV boost to bring top shoulder to 1/4 mark on histogram
5 min sub @ Gain 200, requires + 2.53 EV boost to bring top shoulder to 1/4 mark on histogram
5 min sub @ Gain 200, requires + 2.53 EV boost to bring top shoulder to 1/4 mark on histogram
5 min sub @ Gain 300 - already passed1/4 mark on the histogram
5 min sub @ Gain 300 – top shoulder already passed1/4 mark on the histogram by a fair bit

From the above results, it would suggest that Gain 200 has over +1 EV gain in exposure compared to Gain 100, whereas Gain 300 has probably +3 EV gain compared to Gain 200.

Good use for maximum gain setting (510):

Now I do tend to use the maximum gain of 510 for adjusting focus of the stars with a Bahtinov mask, but I also found that it could be useful for aiding the correct placement of the deep sky object on the sensor. Below was a screen grab from Sharpcap showing the Rosette nebula captured in H-alpha, using Gain 510 and 11.4s of exposure – how useful is that? It did look much brighter on the screen that night, and in fact could be boosted with the FX settings of Sharpcap.

Framing of Rosette nebula using maximum gain 510 and 11sec exposure
Framing of Rosette nebula using maximum gain 510 and 11.4 sec exposure

Banding an issue with the IMX178 sensor?

Now recently while imaging the Rosette nebula with 5 min subs @ 0 gain, I noticed that there was a central division with one side of the image brighter than the other. And on the darker side was some banding noticeable – now the Sony IMX178 is not supposed to have such a problem, which made me think that something wasn’t right. I did notice the image being split into a bright and dark side in the Gain 100 images as well, but did not notice any banding.

It appears that since I updated to the newest version of Sharpcap version 2.7, I have not been able to replicate this problem, so hopefully that has been resolved. Someone on the ZWO forum did mention about situations when the image would appear to have a brighter and a darker side, and apparently this problem has been fixed with Sharpcap in the newest release – so hopefully the banding will not reappear – although to be fair, from my tests of the various gain settings I will not be imaging with zero gain anyway.

Single 5min sub @ 0 gain taken with Sharpcap - shows central division into brighter and darker parts of the image. Fixed pattern noise noted on the darker sider
Single 5min sub @ 0 gain taken with Sharpcap – shows central division into brighter and darker parts of the image. Banding is present on the darker sider (but not present in dark frames)

I have to say that I am pretty impressed with the images I have managed to get with 5 min subs @ gain 100, more so than the 2 min subs@ gain 300. Looks like I will have to do a further test of 5 mins subs @ gain 100 vs 3 min subs @ gain 200 vs 2 min subs @ gain 300, with each group having the same total length of exposure, to see which is the best way to image for me.

Boon