87 replies to this topic

[james7ca]

Two days ago I started a thread showing some of the more-or-less "raw" or lightly processed images that I had taken with the ZWO ASI178MM-Cool camera (see below for that link) and here I'm going to post some of the images that have gotten more processing. I don't want to post these images in the "raw" thread because I've done some processing that has modified the images to the point where they may or may not be representative of how this camera will perform (given that the images shown below may be better or worse than what the camera can really provide -- some artistic license has been applied to the following images -- noise reduction and artifact hiding to name just two).

 

Here is the link to the "raw" thread:

 

  http://www.cloudynig...l/#entry6989569

 

Anyway, below is a processed sample of the Eskimo Nebula that was done in red light using 458 subs that were each exposed for 4 seconds (total integration time just over 30 minutes). This sample was also drizzled at 2X to increase the image size and then reduced to a final scale of 0.6 arc seconds per pixel. The linear image on the 2X drizzle had a measured median FWHM of 1.84 arc seconds (or three pixels at the final reproduction scale).

 

The imaging scope was a Tele Vue NP127is with a 0.8X reducer producing an e.f.l. of 528mm at f/4.2. There was no guiding used during the capture, so this is an example of the so-called lucky imaging technique for DSOs. There doesn't seem to be a great amount of red light in the Eskimo Nebula (at least in comparison to the blue/green region), so I think I over stretched this image somewhat to try and gain more detail in the nebula itself. I'm working on an RGB version which I hope will look better than this single channel result.

 

Note, for some reason the JPEG compression is clipping the background in this image, I don't know if it is the JPEG or just the conversion to 8 bits or the change in the color space (or all three).

 

Lastly, this was taken with the camera's thermal regulation turned OFF (i.e. no cooling). The sensor temperature was reported to be around 47F with an ambient air temperature in the mid-30s.

Attached Thumbnails

• 

Edited by james7ca, 11 January 2016 - 06:50 PM.

[Jon Rista]

Single channel or not, that is an amazing amount of detail! Those stars look amazing as well. Looking forward to a color version.

[james7ca]

The red channel had by far the best FWHM which makes sense given seeing conditions. The stars are pretty good except for that brightest member right below the nebula. That's aways been a problem when I photograph this object, it's an 8th magnitude star and it kind of overwhelms the field. In the final RGB composite I'm going to use a star mask to see if I can reduce that to something less than it appears here. The eccentricity measurements were quite good on the drizzled image, I had a median of 0.33 (red channel, center 1280 x 1280 pixel region of interest).

 

Planetary nebulae are going to be one of my main targets in the future because they have such high surface brightnesses and there are so many of them to photograph even under my red zone skies. In fact, the ASI178MM is probably the best current sensor for that type of work given my current scopes.

[Raginar]

Nice job James.  Looks like you got over the issues you posted about.

[james7ca]

Nice job James.  Looks like you got over the issues you posted about.

If you are talking about the apparent problems I was having while using the camera cooling then you should note that these subs were taken without using the camera's temperature regulation. In any case, I no longer think that there was anything wrong with the camera (still not 100% sure, but closing in on that belief).

[rkayakr]

ames

  I think that you hit it with planetary nebula and an ASI178 camera. Attached is a Saturn Nebula I took last November with an ASI178MC attached to an unguided C8, 100 6s subs at 350 gain and simple processing (no up/down scales).

 

Edited by rkayakr, 11 January 2016 - 07:27 AM.

[james7ca]

rkayakr, what makes the ASI178MM so interesting for planetary nebulae is that I can work near to my seeing limits even with a relatively short focus and fast refractor. Thus, I can image at f/5.2 (prime focus) with my NP127is and still have an image scale of 0.75 arc seconds per pixel. Then, note that 3 x 0.75 arc seconds is equal to 2.25 arc seconds which is pretty close to the size of the Airy disk on the NP127is (2 arc seconds in green light). That means that I can get good sampling on the stars and then add in a dither and drizzle and everything starts to converge at around 2 arc seconds or better which is probably about as good as I can do given average seeing conditions.

 

Thus, it's the pixel sampling that makes the ASI178MM so interesting for use with my refractors. It should also be a good match on a Hyperstar system if I ever decide to go that route with my C9.25.

Edited by james7ca, 11 January 2016 - 06:48 PM.

[rkayakr]

Agreed - most of my ASI178 images were taken with shorter FL refactors like my TS 80/580 with an .8x reducer that yields .9 arc seconds/pixel. I was happy with the experiment at the much longer FL of my old C8 even with the .43 arc seconds / pixel resolution and a decent image size.

I think the other half is having such low read noise that you can stack many short images. This allows you to take unguided subs that that would be very poor if traditional multi-minute subs were used.

[anismo]

That looks excellent James. 

[David Ault]

This is one of the more detailed images I've seen of the Eskimo nebula.  Very well done James!  I would definitely be interested in seeing a version that doesn't have the blacks clipped.  I've had similar problems when pushing custom color profiles into my JPGs.

 

Regards,

David

[james7ca]

Here's what I got from the full set of RGB data. This has about 30 minutes on each channel (458 x 4s) and there were also 100 bias and dark frames in the mix.

 

I had an unusual problem with PixInsight and I ended up spending several hours trying to figure out why the blue channel integrations always came out as it they had been pre-stretched or scaled even when it should have been outputting linear data. I'm not sure where the problem was but I ran lots of tests and I finally decided to start from scratch so I reinitialized the PixInsight preferences and integration caches and then reprocessed all 458 subs for the blue channel and that seemed to fix the problem.

 

There was no cooling or temperature regulation on these subs and I also had problems getting the calibration frames to match the lights. I guess because there was an hour or two between the time I took the LF and DF/bias frames and during that time I reset the camera and changed exposures. That and perhaps a one or two degree change in the air temperature (the reported camera temperature changed from a high of 47.5F for the red channel to an ending low of 45.5F for the DF). The lights and darks used the same gain (330) and exposure (4.0s) but between those sets I took some images using 30 second subs with the cooling turned back on which probably caused the problems with the calibration frames.

 

After all of that I really wasn't in the mood to spend a lot more time on this data set, since in a way it was just something that I started as a quick test of the ASI178MM-Cool camera. I used a 2X drizzle and then reduced that to a final scale of 0.6 arc seconds per pixel (from a 528mm e.f.l. scope at f/4.2).

Attached Thumbnails

• 

Edited by james7ca, 12 January 2016 - 06:23 AM.

[james7ca]

And here is a blue channel on M66 and M65 that was done in 7.5 minutes (15 x 30s) with the camera temperature set point at 0 degrees Fahrenheit. I didn't expect this to come out very well but I wanted to try the camera with cooling turned on and this is the only sequence I got before the clouds rolled in. Same setup as before, Tele Vue NP127is with the 0.8X reducer, e.f.l. 528mm at f/4.2, no guiding.

 

Looks promising, I like the detail (or suggestion of detail) in the cores although it needs a lot more exposure (obviously, 7.5 minutes isn't going to show much). I've also included a crop of M66 that is at the full image scale (0.94 arc seconds per pixel).

Attached Thumbnails

• 
• 

Edited by james7ca, 12 January 2016 - 06:13 AM.

[geo444]

... I had an unusual problem with PixInsight and I ended up spending several hours trying to figure out why the blue channel integrations always came out as it they had been pre-stretched or scaled even when it should have been outputting linear data. I'm not sure where the problem was but I ran lots of tests and I finally decided to start from scratch so I reinitialized the PixInsight preferences and integration caches and then reprocessed all 458 subs for the blue channel and that seemed to fix the problem...

 

That was worth all that work, very nice Blue shades !

 

Was your Asi178mm set to 12-bit or 14-bit output ?

 

( some vendors claim it's a 10-14-bit cam, some say it's a 12-14-bit cam, others say it's a 12 bit cam ? )

 

[james7ca]

...Was your Asi178mm set to 12-bit or 14-bit output ?

 

( some vendors claim it's a 10-14-bit cam, some say it's a 12-14-bit cam, others say it's a 12 bit cam ? )

 

 

I just selected 16 bit in the most recent FireCapture beta which uses the ZWO native support for the ASI cameras. I don't think there is any other option in FireCapture and it seems like the ASCOM driver isn't yet complete for the ASI178MM (because the H-gain and L-gain buttons don't seem to do anything useful, both set the gain to zero). In the ASCOM driver there is also a pop-up menu for "hardware" binning, but it too doesn't seem to do anything (I guess ASCOM doesn't support changing the binning, so it just displays the current setting which must be set with another application).

 

I've been told by ZWO that there is a new driver coming out fairly soon so maybe that will enable some more options.

Edited by james7ca, 12 January 2016 - 07:14 AM.

[geo444]

 

I just selected 16 bit in the most recent FireCapture beta which uses the ZWO native support for the ASI cameras. I don't think there is any other option in FireCapture and it seems like the ASCOM driver isn't yet complete for the ASI178MM (because the H-gain and L-gain buttons don't seem to do anything useful, both set the gain to zero). In the ASCOM driver there is also a pop-up menu for "hardware" binning, but it too doesn't seem to do anything.

 

I've been told by ZWO that there is a new driver coming out fairly soon so maybe that will enable some more options.

 

.

Ok. I'm not surprised that all the drivers are not yet fully updated...

... the Asi178mm has a brand new CMos sensor : the Sony imx178LLJ

I can't even find the flyer.pdf !... it seems not yet published ?

 

Hardware binning is not possible with CMos sensors...

... but Software binning is ok... at half the improvement :

http://zwoug.org/vie...php?f=21&t=2237

http://zwoug.org/vie...php?f=21&t=2275

- see Sam's posts

 

Could this be the good setting for Longer FL ?... or Shorter Exposures ?

... or simply the way to stop Hard Disks congestion ?

 

Edited by geo444, 12 January 2016 - 09:07 AM.

[wenjha]

 

...Was your Asi178mm set to 12-bit or 14-bit output ?

 

( some vendors claim it's a 10-14-bit cam, some say it's a 12-14-bit cam, others say it's a 12 bit cam ? )

 

 

I just selected 16 bit in the most recent FireCapture beta which uses the ZWO native support for the ASI cameras. I don't think there is any other option in FireCapture and it seems like the ASCOM driver isn't yet complete for the ASI178MM (because the H-gain and L-gain buttons don't seem to do anything useful, both set the gain to zero). In the ASCOM driver there is also a pop-up menu for "hardware" binning, but it too doesn't seem to do anything (I guess ASCOM doesn't support changing the binning, so it just displays the current setting which must be set with another application).

 

I've been told by ZWO that there is a new driver coming out fairly soon so maybe that will enable some more options.

 

Hi James

L-gain and H-gain is the same because the ASI178 provide 14bit ADC

so the system GAIN is always below 1e/ADU if output at 14bit

anyway, you can try to set the gain to 300 to get a lowest readnoise image for lucky imaging

Attached Thumbnails

• 

[james7ca]

Here is something a little different. One reason I purchased the ASI178MM is that it seemed a good match for pixel scale with my short focus refractors. In fact, I suspect that I'll use this camera more often with my SV80 (480mm focal length f/6) than any other scope. So, tonight there was a crescent moon and I brought out the SV80 on an AVX mount to see what could be done at the prime focus when imaging the moon at just over one arc second per pixel.

 

Well, that was my plan but high clouds closed in before I could really settle into any good session with the camera. In any case, since high resolution work was out of the question I wondered how well a 2X bin would work to photograph the earthshine. So, I reduced the gain to the minimum, selected 16-bit capture, and set the bin to 2X and shot some 1 second subs through the clouds.

 

The results, while not great, seem promising since a stack of 30 subs produced a seemingly noiseless image with quite a bit of dynamic range and absolutely no posterization. Unfortunately, the clouds really reduced the contrast and detail but I think with clear skies that this camera could produce some really high dynamic range shots of the earthshine.

 

Anyway, below is the image, stacked and processed in PixInsight (30 subs, each exposed for 1 second) with some final adjustments in Photoshop CC 2015. Note how the full moon just fits within the field of the camera when mated to a 480mm scope.

Attached Thumbnails

• 

[james7ca]

I had a break in the clouds and even though the moon was pretty low in the sky and I was able to get this shot by stacking 60 subs with 16-bit capture, gain zero, and an exposure of 2 seconds. This was without binning at the full 6MP resolution but here reduced by just over 50% to meet the posting guidelines on CN.

 

Considering the relatively advanced phase of the moon this turned out pretty good (IMO). I'd imagine with clearer skies and with the moon higher in the sky and with an IR filter this combination of scope and camera could do some very nice shots of earthshine. Perhaps a lot of if but...

Attached Thumbnails

• 

Edited by james7ca, 14 January 2016 - 01:36 AM.

[james7ca]

Here is a shot done with my Stellarvue SV80 at prime focus (480mm, f/6) with the ZWO ASI178MM-Cool camera. This is a single sub taken using an exposure time of 32ms and with a camera gain setting of zero. Processing done with PixInsight and Photoshop CC 2015. The original/capture image scale of 1.03 arc seconds per pixel was increased to 1.35 arc seconds to meet the CN posting guidelines (image size is 1600 x 936).

 

Edited by james7ca, 15 January 2016 - 07:33 AM.

[xb39]

Here is a shot done with my Stellarvue SV80 at prime focus (480mm, f/6) with the ZWO ASI178MM-Cool camera. This is a single sub taken using an exposure time of 32ms and with a camera gain setting of zero. Processing done with PixInsight and Photoshop CC 2015. The original/capture image scale of 1.03 arc seconds per pixel was reduced to 1.35 arc seconds to meet the CN posting guidelines (image size is 1600 x 936).

 

 

 

...oh....my....god!!! The level of detail is amazing!
Though I have already the ASI224MC-cool and the ASI174MM-cool, I think I also should must have it! Would fit great with my WO Star71.

[james7ca]

 

Here is a shot done with my Stellarvue SV80 at prime focus (480mm, f/6) with the ZWO ASI178MM-Cool camera. This is a single sub taken using an exposure time of 32ms and with a camera gain setting of zero. Processing done with PixInsight and Photoshop CC 2015. The original/capture image scale of 1.03 arc seconds per pixel was reduced to 1.35 arc seconds to meet the CN posting guidelines (image size is 1600 x 936).

 

 

 

...oh....my....god!!! The level of detail is amazing!
Though I have already the ASI224MC-cool and the ASI174MM-cool, I think I also should must have it! Would fit great with my WO Star71.

 

I've got another 160 subs that I will probably stack and drizzle and then sharpen using wavelets in RegiStax. I'm not sure how much better it could get, but with that low of a gain setting the individual subs look almost completely noiseless and the dynamic range is good even with a single sub. Of course, I suspect that it would be better to shoot at the critical sampling, which would be at about f/11 or with a 2X barlow on the SV80.

 

[EDIT]

Here is the thread where I posted the "final" (and higher resolution) version of the above capture of the crescent moon done with the ASI178MM-Cool

 

 http://www.cloudynig...-6#entry7003576

[/EDIT]

Edited by james7ca, 17 January 2016 - 10:31 AM.

[mdine1us]

Here is a shot done with my Stellarvue SV80 at prime focus (480mm, f/6) with the ZWO ASI178MM-Cool camera. This is a single sub taken using an exposure time of 32ms and with a camera gain setting of zero. Processing done with PixInsight and Photoshop CC 2015. The original/capture image scale of 1.03 arc seconds per pixel was increased to 1.35 arc seconds to meet the CN posting guidelines (image size is 1600 x 936).

 

Wow. THAT is a beautiful image! 

 

best,

Marty

[james7ca]

Marty, thanks. The "final" image that I posted in another thread is better, particularly if you use the link from there that goes to a higher resolution version.

[mdine1us]

Yep, that full res version is even more impressive! Nice work James!

[james7ca]

Well, I finally got some time to use the ASI178MM-Cool on a decently sized DSO (M106 and NGC 4248 on the upper right). This was taken with the last quarter moon about 50 degrees away from the target, probably didn't hurt that much but it certainly didn't help my already red/orange zone skies. This is plain luminance done with 30 seconds subs with a total integration time of 105 minutes. The scope was a Tele Vue NP127is working at 660mm, f/5.2. I used a temperature set point of -25C on the camera with the "Lowest Read Noise" gain setting.

 

This will probably be a lot more interesting with color added, maybe I'll be able to get some RGB data in the next week or so. However, the detail looks pretty good (IMO), the median FWHM was 2.18 arc seconds and the median eccentricity was 0.42. I think the reproduction scale shown here is about 1.5 arc seconds per pixel. It's still a bit noisy, probably needs two to four times more integration time (for the luminance) and of course a darker sky would help greatly.

Attached Thumbnails

• 

Edited by james7ca, 30 January 2016 - 12:15 PM.