SX-694 First Light

[Dom543]

I saw a used SX-694 monochrome Trius on Astromart for an affordable price. I bought it as a reward and compensation for all the work that has been keeping me away from astronomy in the past year. The logic: If you don't have time for astronomy, have at least a new camera. Makes complete sense, doesn't it?

Given the limited time I can spend under the stars, I wanted a camera that I could use immediately. Without a long learning curve, without waiting for adapters from Precision Parts, without waiting for software to be written etc. I wanted a camera that I could use for near-real time EAA today and that would also have potential for something over my existing cameras (SX-825 and Lodestars). The SX-694 fit the parameters, as it has the exact form factor of the SX-825 and thanks to StarlightLive being vertically universal for all SX cameras.

The SX-694 has a 6Mp (2750 x 2200) 16mm diagonal ("1 inch format") Exview CCD sensor with 4.54 micron square pixels. For comparison, the SX-825/Ultrastar has 1.5Mp (1392 x 1040)  11mm diagonal (2/3" format) sensor with 6.45 micron pixels. The Lodestar has a 0.44Mp (752 x 582) 8mm diagonal (1/2" format) sensor with 8.2x8.4 micron pixels. More detailed specifications of the camera can be found here https://www.sxccd.com/trius-sx694 or even more detailed here http://www.highpointscientific.com/telescope-accessories/astro-photography/ccd-cameras/starlight-xpress-trius-sx-694-mono-ccd-camera-usb-hub.

My calculation was that, if I bin the SX-694 2x2, then I should get Lodestar sensitivity (9 micron pixels) and Ultrastar/SX-825 resolution (1.5Mp). Those combined would satisfy my requirement for "something over my existing cameras".

What follows below is a "first light" thread and first light is about looking for fun. Not about writing an evaluation for scientific publications. To me real fun must involve colors. And as the 694 is a higher resolution camera, looking for details is also essential part of a first light fun.

Let the action begin.
The mailman with the Priority mail box pulled up, while I was loading my van for an astro outing. It was urgent to leave to get through traffic and to set up before dark. The computer was already in the van so I didn't have time to download any drivers. I just threw the box in the van and hit the road. The plan was to screw the new camera in the place of the SX-825, connect the same cables and see what happens.

What happened was the best of all possible scenarios. StarlightLive opened, connected to the camera and recognized it. The only thing it failed do was congratulating me to my new camera. That will be my new feature request for v.3.3. I see that there is already a thread for that.
 

Clear Skies!  --Dom

[Dom543]

Eastern Veil Triple NB with SX-694 and no binning

Here red is Hydrogen, green is Sulfur and blue is Oxygen. Yellow is the mix of Ha and Sii. Magenta is the mix of Ha and Oiii.

This is at the full resolution of the camera but I had to reduce the capture to 50% to fit the posting file size limits. At this reduced size the pixel count is about the same as that of the Ultrastar/SX-825. Even to see the image at this reduced resolution, you will probably have to click on what you see on the web page.

These are 3x60sec exposures mean stacked with each of the three filters. I was totally surprised that I got this image with just 60sec exposures. I could have obviously gotten more detail and smoother texture by stacking more frames. But I use a rather stict definition of "near real-time" when it comes to my own work. (Don't worry, I am more tolerant to others.) The optics used was an old manual focus Nikkor 400mm f/2.8 teleobjective.

I aimed at the Veil because it has both colors and fine detail. And also, because earlier this summer I spent some time exploring the potential of the Veil using the SX-825. This post has two captures from that venture for comparison https://stargazerslounge.com/topic/269747-multispectral-captures-with-lodestar-and-ultrastar-mono-cameras/?do=findComment&comment=3036103.

The screen appearance of the captures with the 825 and the 694 are quite similar on these internet postings. Due to the fact that both are reduced to the same size and to similar pixel counts. The lesson is, if the sole purpose is to post on these forums, then there is no point in using a higher resolution camera.

To show a close-up of the texture and detail at full resolution, I cropped out a part the original unreduced image that fits the file size limit. I believe that this is the part called NGC6995. Again, you will have to click on what you see to get the full resolution image.

This is a stack of three 60sec frames per channel and no binning.

Clear Skies!  --Dom

[stash_old]

Maybe with pics like that it should be a case of "less work more play" 

[Paul81]

Wow! I'm glad all worked well - i'll add some suitable welcome messages from the camera to the next release :-) (your post made me laugh!)

The camera is certainly doing the business! How did you find the processing times? The current latest code baseline (working towards V3.3) has some additional multi-processor / SIMD speed-ups in stacking and the histogram side of the display processing. On my 2013 MacBook Pro Ultrastar stack times are much faster, and down to < half a second in my usual suite of test cases.

The higher resolution gives you plenty of options - like you say you can bin 2x2 and still get Ultrastar resolution - nice :-)

I still feel the EAA field has more to offer in terms of camera resolution and S/W based processing. Its just gonna get better over the coming years!

[Dom543]

Thanks Paul,

I agree that there is still a lot of fun, and more fun, ahead for us in EAA. But when we are enjoying the rosy outlook ahead, we should also remember that the recent EAA revolution, that lead to the current live software supported use of still USB cameras, has been the result of two key developments. One was Nytecam's discovery of the potential of the Lodestar and the other was your development of LodestarLive.

To answer your question, I didn't notice any processing time bottleneck. I use a very old and outdated laptop. I keep it because it has a very nice large and fine resolution screen that they don't make any longer. But its processor is geriatric so my processing times would not be relevance to anybody using current generation computers. But even so, there is no stacking backup or anything that would interfere with enjoying the views.

Download times are noticable longer with the 6Mp sensor. But that's due to USB2 bandwidth. Nothing we can do about. For alignment and focusing with short 1 second exposures I use 2x2 binning.

Thanks again for all the effors that you have put into StarlighLive. This time I particularly appreciated the fact that the software is absolutely future proof, keeps the door to upgrading cameras wide open.

--Dom

[Dom543]

Western and Central Veil Triple NB, no binning

This is again a capture made at the full 6Mp resolution of the SX-964, but to fit posting file  size, the image has beed reduced to 50% (linearly). The resolution of the reduced image is 1.5Mp, about the same as full resolution with the Ultrastar/SX-825/Infinity.

2x90sec exposures per channel mean stacked. Red is H-alpha, green is S-II and blue is O-III. I use the Baader 6 (Ha) and 8.5 (O-III and S-II) nanometer NB filter set. The same Nikkor 400mm f/2.8 ED teleobjective was used as optics.

This Western part of the Veil has extensive Sulphur coverage but the S-II signal is much weaker than in the Eastern part. More spread out but weaker. I tried to stack 60 sec exposures and it went well for the O-III and H-alpha. But when the S-II was added, the entire image got very noisy. Since Sulfur is all over, noise was also all over. Then I switched to 90 sec exposures and got what you see. If signal is weak, then there is a floor on exposure times, one cannot go below that.

For comparison, here is a capture of the same area with the SX-825/Ultraster Sr made last summer https://stargazerslounge.com/topic/269747-multispectral-captures-with-lodestar-and-ultrastar-mono-cameras/?do=findComment&comment=3036938. It uses a different palette but al;so 90 sec exposures.

On these web postings there doesn't appear to be much difference between the full resolution Ultrastar image and the 50% reduced 694 image. Comparing them on a 32" Ultra HD 4K monitor, the two are clearly in different categories. The best I can do here again is to cut out an area from the full resolution image that still fits within the posting file size. Here it is.

To see any of these images in their highest possible posted resolution, you have to click on the image and than click the "Full Size" button in the lower left corner.

Clear Skies!  --Dom

[RobertI]

Congratualtions on the new camera Dom, it looks like the Trius is giving some nice results and gives a nice big field of view with very good sensitivity at 2x2 binning. Wish I could see the original high res image on a 32" monitor!