short/long exposure

[Fieldsy]

Hi Guys another question .asi178mm not cooled in the spec it stats up to 1000s exp time .so that means 10 mins is easily possible 600seconds ,does that qualify as a short or long expo ? The shop I purchased the 178 from is saying this camera is no good for dso and non point getting any narrow band filter because the cam cant take the long expo needed with nb filter.?

[Star101]

The advert I read for this camera shows the following---

Features of the ZWO ASI178MM Monochrome Camera
Recommended for high-magnification Lunar and Solar imaging, microscopic imaging and to image some deep sky objects
Resolution: 3096X2080 at up to 30fps (at 14bit ADC) or 60fps (at 10 bit ADC)
Long time exposure up to 1000s!

 

So, yes, it can be used for "some" DSO's but the cooled version would be better.

As the cooled version advert says ---

This camera will mostly be used for high magnification Lunar, Solar and planetary imaging and for microscopic imaging, but due to the active cooling it can also be used for deep sky imaging very well as it supports up to 1000seconds exposure time. The active cooling will ensure reduced noise levels that helps to achieve higher quality deep sky images. 

 

You can try imaging DSO's. But without cooling, there will be a lot of noise to contend with and this most likely will make your images less favorable than a cooled camera.

[vlaiv]

In world of CCDs  - 600s exposure is considered about right for most types of imaging. In world of CMOS sensors it is considered long exposure.

Depending on your setup you could see real benefit of going with 10min exposures with CMOS camera (over shorter exposures), but with some setups you won't see much impact when going longer. It depends on how read noise compares to other noise sources.

This brings me to the point - non cooled camera is going to have host of problems when attempting very long exposure by CMOS standards. It will certainly not be feasible since dark current noise is going to swamp read noise very quickly. You will have problems with heat in such long exposure as well as dark current. You will not be able to properly calibrate your subs - due to lack of set point temperature.

You can try NB imaging with such camera, just be aware of these restrictions.

[Knight of Clear Skies]

The camera certainly can be used for DSO imaging, for example there is a image of M51 taken with an uncooled one in this thread. It's a low read-noise camera so long exposures are not necessarily required.

However, it's a small pixel camera so best suited to working at short focal lengths, tracking is going to be challenging with a 130pds and EQ5. You might get the best results with exposures in the 15-30 second range.

[RolandKol]

I am sure you can use it for DSO,

I have managed to use my 224MC for it!  Noisy... But usable!  The main issue is Field Of View...

It will be very narrow... It will catch only the small parts of the main Nebulas,

but is it the Galaxy season now, so if you have it already, - you have nothing left to do, - hunt them down, M51 should fit perfectly in it's FOV on PDS.

Other thing, -  results.... They will not be so smooth as produced with Cooled cameras and especially in the warmer periods, so I you will be made to spend more time fighting noise in post processing 

and most likely will not have satisfying image at the end.

FOV/Pixel size and speed, makes it a nice planetary cam or expensive guider.

For DSO, - I would not go for longer than 180sec with this one.

Even with my ASI1600 I usually do not go above 300s (trees, London LP and etc), - I would not bin, simply shoot as much subs as possible.

Lots of short exposures will have Almost the same result as Long ones,  - just more data to carry and process... On another hand, "U F O's" visits are not so painful  , plus guiding issues are not so visible.

 

[Fieldsy]

Well looks like I’m not getting out until next week at the earliest uk weather ?☁️☁️?

[cjdawson]

I've got an ASI290MM uncooled.   You can do long exposures, but as others say the problem is that the camera will generate noise when it gets hot.  Which will happen when taking long exposures.

This is why the Cooled version of the cameras exist, they can keep the imaging sensor cool, and help to keep the noise down.

You certainly can take images with the uncooled cameras, just be aware that heat could be an issue.     I'm still thinking about imaging the Ring Nebula with my 290mm, as well as my 1600mm-Cooled.   I sense that the image will be better on the 290mm as the exposures can be short, and the nebula will fill the field of view (8" F10 scope, or F6.3 if I add my flattener)

[RolandKol]

P.S. 180s in Narrow Band!!!!

LRGB 30sec max... Maybe R channel 60sec.

[Adam J]

7 hours ago, Fieldsy said:

Hi Guys another question .asi178mm not cooled in the spec it stats up to 1000s exp time .so that means 10 mins is easily possible 600seconds ,does that qualify as a short or long expo ? The shop I purchased the 178 from is saying this camera is no good for dso and non point getting any narrow band filter because the cam cant take the long expo needed with nb filter.?

You can't do long exposures (for CMOS >2mins) without a cooled camera and certainly not 10min exposures.

That is not to say you cant get good results with a uncooled camera, you just need the correct approach.

You need to set high gain for narrow band imaging, I would go with Gain = 200 to give a read noise of 1.4e and try 60 second exposures. You will need to collect at least 100 sub images per channel to get a nice result at this exposure length 200 would be best. You might want to try 2x2 bin given your focal length of 650mm.

For broad band imaging (LRGB) you will want a lower gain, I would start with Gain = 100 and 30 second subs max per channel, maybe use gain 50 for brighter targets like M42 for better dynamic range. Again 100 to 200 subs per channel dependent on the target. Again you might want to try 2x2 bin.

If you want to try a more advanced DSO lucky imaging technique in LRGB then that camera is suitable with your F5 scope. Try 5 second subs at gain 200 for detailed Luminescence of galaxies using 1x1 bin. You will need at least 2000 subs for this technique to work. 

Not having cooling is not ideal as previously discussed when you were choosing a camera, however your budget dictated this choice, you should still get some very nice results if you start out using the above guide lines. As you get to know your setup you can optimize your setting further. 

Final word of advice! YOU MUST TAKE DARK FRAMES WITH THIS CAMERA, take them at the end of the night (due to lack of temperature set point) to match the temperature of the light frames, if not you will get amp glow. I would take 20 as a minimum at the same exposure length and gain as your light frames. 

I recommend trying Astro pixel processor too.

Adam

 

 

 

 

[Fieldsy]

This is just a mine field of possibilities. Adam thanks for the recommended settings will def give me a start point and something to work with .

At least I'm not completely blind thanks to everyone for all the advice  now i just need clear skies any advice on achieving that one ? ?

[Adam J]

 

36 minutes ago, Fieldsy said:

This is just a mine field of possibilities. Adam thanks for the recommended settings will def give me a start point and something to work with .

At least I'm not completely blind thanks to everyone for all the advice  now i just need clear skies any advice on achieving that one ? ?

No problem.

As discussed previously, you will be limited to the smaller DSO's, why not start with a globular cluster? Great target with this camera! Quite bright so try 30s exposures at gain 50.

In Ha things like the Crescent and planetary nebula are your targets of choice with your field of view and pixel scale. They are bright and will give great detailed images even with short exposures.

And in the near future you have the likes of M81, M81, M101, M108....etc.etc. Which are ideal targets for short exposure high gain imaging with a small pixel camera like the ASI178mm

Just play to its strengths and avoid its weaknesses in terms of targets and you will have to give it a miss on hot night in the middle of summer as there are limits to having no cooling.

For now get a LRGB filter set like the ZWO set for the ASI1600 (not the cheaper one) and a baader 7nm Ha filter, ignore the guy who said there was no point.

Also get a focus mask for the 130pds.

Good luck cant wait to see your images.

Adam

 

8 hours ago, Star101 said:

The advert I read for this camera shows the following---

Features of the ZWO ASI178MM Monochrome Camera
Recommended for high-magnification Lunar and Solar imaging, microscopic imaging and to image some deep sky objects
Resolution: 3096X2080 at up to 30fps (at 14bit ADC) or 60fps (at 10 bit ADC)
Long time exposure up to 1000s!

 

So, yes, it can be used for "some" DSO's but the cooled version would be better.

As the cooled version advert says ---

This camera will mostly be used for high magnification Lunar, Solar and planetary imaging and for microscopic imaging, but due to the active cooling it can also be used for deep sky imaging very well as it supports up to 1000seconds exposure time. The active cooling will ensure reduced noise levels that helps to achieve higher quality deep sky images. 

 

You can try imaging DSO's. But without cooling, there will be a lot of noise to contend with and this most likely will make your images less favorable than a cooled camera.

As one last point:

Yes its not cooled but neither is the Hypercam 183m (that fan achieves very little temp reduction even according to Altair when I spoke to them) and one of the chaps who posts on here achieved this with it in narrow band using 300 second ha subs....

https://www.astrobin.com/386092/B/?nc=all

Just look at what you get with 1500x 5 second exposures at high gain!

https://www.astrobin.com/full/392619/B/?nc=all

That is amazing detail in M81!

So yes that is the IMX183 but the IMX178m is more sensitive than the IMX183m and not by an insignificant margin, this is due to its higher quality 14bit A/D and hence lower read noise of only 1.4e at gain 200 in comparison to 1.9e for the equivalent setting on the IMX183m which has only a 12bit A/D. Both chips have very low dark current and so if you are going to get a uncooled camera these are the ones to go for.

As such it should be possible to achieve the same result using the 178 as with the 183 but with shorter exposures ~60s or so. That will result in less thermal noise in comparison to other sources.

The downside is a smaller chip.

Lots will depend on ambient temperature mind you, if its 24c on a mid summer night then that is pushing it too far even for shorter exposures.

Adam

[RolandKol]

yep, it is a good idea to shorten subs by increasing gain, -  completely forgot about it...

but I would avoid Binning as it will result twice smaller images and this is in addition to the already small resolution of ASI178, furher more,  - taking into account the amount of images taken, - majority stacking soft (DSS and even PI) 

will take ages to apply Drizzle (it may take 6 hours or more)

As far as I recall DSS usually failed while processing drizzle for more than 100 subs, so 6 hours wasted in waiting....

But it may depend on the processing power of  the PC.

[vlaiv]

I don't think that using higher gain is going to bring in any gains (ah those puns )

Only benefit of using high gain settings on 178 sensor is to reduce read noise (all gain settings are over unity for this camera as it has smaller full well depth than 16bits). Using gain of about 110 (can't be certain from the graph, but I'm aiming for e/ADU of about 0.25) will bring it down to about 1.7e. Max that you can have with that camera is 1.4e - that is small difference, especially in the light of the fact that dark current noise is going to be dominant component.

Quick search online gives dark current value of about 0.032/e/px/s at -10C, and if we take doubling temperature of 6C, at +20C it is going to be around 1e/px/s. 60s exposure is going to generate ~7.75e of dark current noise so total noise combined from dark current noise and read noise is going to be ~7.93 (vs ~7.87e if we go with the highest gain setting). We have not added LP noise yet. Really no point in going very high gains as you decrease well depth (and you want your well depth since you have non cooled camera and dark current will take up some of the space together with any LP).

Bin x2 is a must at this focal length. Natively (without binning) you will be imaging at 0.76"/px - that will be grossly oversampling for 130mm scope, average mount and average seeing conditions, I would personally think of binning it even 3x3 as it is unlikely that you will achieve resolution of 2"/px.

Drizzle is something that will simply not work for the most of amateur imaging setups. Some benefits might be seen when one has very undersampled data - but that is generally very rare (4 -5"/px on larger scope).

 

[RolandKol]

@vlaiv but what's the point to bin, as even x2 will end up with the image in the size of the postcard... Bin x3 and will be closer to the post stamp!

[vlaiv]

1 minute ago, RolandKol said:

@vlaiv but what's the point to bin, as even x2 will end up with the image in the size of the postcard... Bin x3 and will be closer to the post stamp!

There are two ways to view the image (well, there are more ways than that, but for the sake of argument ... ) : fit to screen and 1:1 (or 100% zoom).

For image displayed, when you bin your data - it will not change FOV, so image will depict same relation of object imaged to surrounding space. When you record your image at high resolution (small "/px) if you are over sampling you will not record any additional detail - image will just look blurry when viewed 1:1.

Best SNR is obtained if you sample image at proper sampling resolution (in comparison to over sampling). Smaller image when scaled to fit the screen will look exactly the same as over sampled larger image, but it will have better SNR (less noise, easier to process) and will look rather good (sharp) when viewed 1:1

This is why it's good to bin your data (if over sampled) - you don't loose detail, if you want a big image that looks blurry - just scale it up, you will get the same result, but when binned it will provide better SNR.

Actual detail recorded in the image is very tightly related to star FWHM, and there is advantage to record image in full resolution for CMOS sensors as you can decide how much to bin after you measured your FWHM in resulting stack. Good sampling resolution is at FWHM / 1.6.

For example, I just had above discussion with one of the members here in private messages about this, so I worked with my old data to provide examples. Last year I gathered some data on M51 and it was average to poor night in terms of conditions (windy and seeing was not so good). I imaged with my RC8" and ASI1600 - it gave me ~ 0.5"/px natively. At the time I processed image by applying bin x2. Couple days ago I did FWHM measurement on it and it turned out that I can actually bin it 4x4 for resolution of 2"/px without loss of data!

Resulting image looks better than the old one when viewed 1:1 and they look the same when scaled to fit the screen. I had to do quite a bit of denoising on the old one, but none on bin x4 version because of increase in SNR.

Here is quick comparison of the two:

Processing is a bit different in terms of stretch (left one is new and background is kept at lower value). These two images contain same data. If you scale up left one - you will get exactly the same data as the right one. Right one simply will not look sharper, but left one properly sampled looks both sharper and has better SNR. Difference between two images is just that I took same linear data that I used for right one, and applied bin 2x2 to get linear data for left one (well, not regular bin, but custom binning that I developed that preserves a bit of sharpness - that is separate topic altogether and is related to pixel blur).

Point is - if you display binned version to fit the screen and regular version to fit the screen - you will get same amount of detail. When viewed 1:1 binned version will look sharp and non binned version will look blurry, but binned version will have higher SNR and be smoother and less noisy. And this last part is very important one.

Btw, 178 sensor will produce nice sized image when you bin both 2x2 and 3x3. It might be considered a bit smaller by today's standards (full res is 3000x2000px), but just 4-5 years ago 1000x670px was considered very decent pixel count in astronomy. Some of the most popular CCD cameras had that sort of resolution like Atil 314L+ - 1300x1000 (crop due to framing and stacking and you are very close to 1000x670)