QSI WSG-8 upgrade

Slawomir (Suavi) · #1 12-10-2015, 09:03 PM

Just a quick update from Paddy in Brissie....

I decided to send my QSI back to the US for fixing of "Sony glow" - amp glow (within warranty) and while it was there I thought it would be nice to upgrade to a WSG-8 cover (I originally got a slim version of the camera). QSI sorted all out within a few days and today I got my camera back. Although my camera is only 1yo, they also repurged CCD chamber with Argon as part of the service.

Did a quick 15-minute dark-frame test and amp glow is totally gone and the dark frame is very even. The mean ADU has dropped from 514 (before fix) to 505 (after the fix), but most importantly StDev for the entire single 15-minute dark dropped from 25 ADU to 14 ADU (both at -20C, measured with PI Statistics tool). Looks like I won't be needing darks anymore, only bias and flats.

Maybe later I will measure read noise as well and compare to the value before the fix (4e).

As for the wheel, it looks VERY solid and well engineered, as well as the OAG. Nice to have a camera, 8-position wheel and OAG all in one cover.

Now just need to make it to school holidays and then take a few new shots with the upgraded camera

RickS (Rick) · #2 12-10-2015, 09:23 PM

Nice to hear about a good customer experience, Suavi!

SimmoW (SIMON) · #3 13-10-2015, 06:13 AM

Ha, that's scary, so your pics will look even better now?! :-)

gregbradley · #4 13-10-2015, 09:50 AM

That's great. I'd like to see a 10minute dark to compare it to my Trius 694.

QSI really seem to be one of the top companies to deal with and their products are outstanding.

Its a shame they are limited to smaller sensors though.

I wonder if they can squeeze in the new KAI 16200 when its in production without a new body design? QHY has one coming out but not 100% about QHY. A lot use them. I wonder if they are OK or a bit of a compromise?

Greg.

Slawomir (Suavi) · #5 13-10-2015, 05:52 PM

Quote:

Originally Posted by RickS

Nice to hear about a good customer experience, Suavi!

Sure is! But I actually have found that majority of people in this wonderful industry are nice and helpful, but must admit that QSI was very fast in dealing with my upgrade and with fine-tuning the camera.

Quote:

Originally Posted by SimmoW

Ha, that's scary, so your pics will look even better now?! :-)

Will do my best Simon

Quote:

Originally Posted by gregbradley

That's great. I'd like to see a 10minute dark to compare it to my Trius 694.

QSI really seem to be one of the top companies to deal with and their products are outstanding.

Its a shame they are limited to smaller sensors though.

I wonder if they can squeeze in the new KAI 16200 when its in production without a new body design? QHY has one coming out but not 100% about QHY. A lot use them. I wonder if they are OK or a bit of a compromise?

Greg.

As requested, attached are jpgs of re-scaled (30% original size) old single 15-minute dark frame at -20 C and surface plot (after significant stretching to highlight Sony glow) and a new single 15-minute dark also at -20 C with the same stretch applied plus surface plot of the stretched dark.

I double checked the numbers and actually Average Dev for the old dark (entire sub) was 20.0 ADU, while for the new one is 11.2 ADU. Not sure why I got different numbers for the same subs yesterday - I verified the values above with STDev for smaller areas in the subs.

Paul Haese · #6 13-10-2015, 07:10 PM

On my second QSI. I had the second one to lob in Australia (583) and then about 2 or 3 years ago I bought the 683 with WSG 8. Their cameras are great. Bullet proof and not ever caused me any grief. I think I have pulled 2700 full frames out of the camera and something like that again in sync frames which are binned.

Like Greg I would like to see them with a larger sensor and there was development going until the funds for the development were pulled. Maybe in the future that will happen.

Kevin and his team give great service and a good product from the start. Nice to hear they are continuing the tradition.

I can say right now if they ever develop a bigger camera I will be buying it as my next camera for the FSQ.

Slawomir (Suavi) · #7 16-10-2015, 09:42 PM

Just a quick update on readout noise.

Used Basic CCD Parameters script in PI to check the readout noise, and for three different sets of test frames it was just under or at 4e, the same as I had before sending the camera to QSI for amp glow removal. So the readout noise is unaffected by fine tuning, which is good news.

EDIT:
The abovementioned script suggested that gain has changed from 0.17 to 0.28 after fine-tuning the camera at QSI. I did not trust this value so I performed a more thorough analysis as per the article by Craig Stark - Measuring your Camera, and my analysis resulted in the attached graphs.

Needless to say, gain has really been altered (the gradient of the line of best fit) consequently suggesting full well to be at around 17000e (0.27*64000), in line with the value from the script as well. Not sure whether that is a realistic number since the CCD remains linear up to only around 30 000 ADU. QSI currently claims that well capacity is 14 000e for this camera. Also, today's analysis confirmed that read noise is 4.07e.

I am not quite sure whether anyone will be willing to read my waffle, nevertheless it has been fun measuring my camera and sharing my findings

RickS (Rick) · #8 17-10-2015, 11:37 PM

I think it's a useful practice to check the camera parameters and make sure they look reasonable, Suavi.

I've done a full Photon Transfer Curve analysis a couple of times and found it very interesting and informative.

Cheers,
Rick.

gregbradley · #9 18-10-2015, 08:59 AM

As requested, attached are jpgs of re-scaled (30% original size) old single 15-minute dark frame at -20 C and surface plot (after significant stretching to highlight Sony glow) and a new single 15-minute dark also at -20 C with the same stretch applied plus surface plot of the stretched dark.

I double checked the numbers and actually Average Dev for the old dark (entire sub) was 20.0 ADU, while for the new one is 11.2 ADU. Not sure why I got different numbers for the same subs yesterday - I verified the values above with STDev for smaller areas in the subs.[/QUOTE]

Thanks for that. Those darks look very very clean.

Greg.

Slawomir (Suavi) · #10 18-10-2015, 09:05 AM

Quote:

Originally Posted by RickS

I think it's a useful practice to check the camera parameters and make sure they look reasonable, Suavi.

I've done a full Photon Transfer Curve analysis a couple of times and found it very interesting and informative.

Cheers,
Rick.

Thank you Rick.

So true, I have learnt quite a lot about my camera (and calibration frames too) by measuring its performance.

Since I am hoping not to use darks anymore, the last piece of puzzle (for now) was to subtract master bias from a single dark and see what's left.

So in PI I created a super bias from 250 bias frames and subtracted it from a single 15-minute dark . All frames were captured at -15C.

The mean for dark minus super_bias was 12.9 ADU and AvDev was 7.9 ADU.

Attached is scaled down (30% original size) and stretched dark minus super_bias frame for reference (and surface plot for un-stretched difference with an added value).

Time to do some imaging!

Slawomir (Suavi) · #11 18-10-2015, 09:07 AM

Quote:

Originally Posted by gregbradley

Thanks for that. Those darks look very very clean.

Greg.

No worries Greg