Interesting stats for ASI1600 @ 250 Gain

codemonkey (Lee) · #1 03-09-2016, 04:13 PM

Been doing some playing around, as I thought I could get sky limited with very short subs (as in 6sec lum*) and high gain on the ASI 1600.

Ran the Basic CCD Parameters script in PI and got some interesting numbers.

Apparently setting the gain to 250 gets me a true gain of 0.269 which is about what I expected, but the read noise is coming in at 1.17x.

* For the record, I now think I was way off with the expected sub length and I think I'm actually looking at about 114s lum @ 250 gain, and I've been underexposing lately.

Atmos (Colin) · #2 03-09-2016, 05:07 PM

114s seems a bit long at 1.17e- read noise, unless you have the blackest of black skies.
I am sky limited at about 150s up at Heathcote at 4.6e-.

codemonkey (Lee) · #3 03-09-2016, 05:16 PM

Quote:

Originally Posted by Atmos

114s seems a bit long at 1.17e- read noise, unless you have the blackest of black skies.
I am sky limited at about 150s up at Heathcote at 4.6e-.

Yep, looks like you're right. I thought the basic formula was 10*RN*RN, where RN was the read noise in ADU. Turns out it's 10*RN*RN/gain where RN is the read noise in electrons.

That gives me a target of 51 ADU, which I'd reach in about 30 seconds -- my brief test last night just after dark gave me ~ 1.6 ADU /s in lum @ 250 gain.

Atmos (Colin) · #4 03-09-2016, 05:27 PM

Yep, you're right now

With such a low well depth at gain 250, are you going to have saturation problems at 30s?

codemonkey (Lee) · #5 03-09-2016, 05:45 PM

I would absolutely have saturation problems with stars at that sort of gain; I would need to be doing HDR integration with separate stack for the stars, but that doesn't bother me as it's fairly trivial to do.

I'm trying to get subs down to a point were I essentially don't need to guide. Unfortunately the Avalon has drift in RA of about 4"/min, and I'm sampling at under 1"/px, so that's pretty tough to do. I think I might be able to play with some settings to sort out that drift, I haven't confirmed that theory though.

Shiraz (Ray) · #6 03-09-2016, 11:02 PM

Lee, have you used software that records in only 12 bits?

I tried working out your optimum sub length, but your sky signal of 1.6ADU/sec seems a bit lower than I would have expected for 16 bit data at high gain (eg from SGP).

codemonkey (Lee) · #7 04-09-2016, 08:07 AM

The data was captured in 16 bit and output as fits as per normal. You're 100% right though, that 1.6ADU/s was in 12bits, not 16. In order to get the correct numbers out of the CCD Parameters script you need to set it to 12 (which is what the camera's actually capable of, of course), and then I used 12 bit statistics to keep everything consistent.

glend (Glen) · #8 05-09-2016, 07:08 AM

Lee, Jon Rista has been doing some high gain narrowband imaging testing with the ASI1600 with impressive results. He settled on using a Gain setting of 200 and 90" subs. It is written up in the Cloudy Nights imaging forum, "ASI1600 high gain experiment". Worth a read.

http://www.cloudynights.com/topic/54.../#entry7414496

What i found in reading the thread is we are both amending workflow to take advantage of the cameras strengths, for example, dialing up the gain setting for frame and focusing on faint targets, before settling back to capture settings for the run.