Best Gear for small DSOs

[Tommohawk]

22 hours ago, Astroboy239 said:

Small dso we use rasa astrograph and a c14 with Barlow 

 

Varad

Hi Varad. RASAs look fabulous but typically very short EFL - I'm not sure this would help my situation.

C11 even without with barlow would certainly be good, but I want to keep the weight down a bit. I would certainly be interested to hear anyones experience using SCTs - especially the C9.25 - for small DSOs.

Seems to me the issue with SCTs would be that if the target approaches the full FOV - which is the whole point of the exercise - then field correction becomes an issue. The corrector typically available for non - EDGE SCTs is also a reducer - which is unhelpful. On the other hand, if opting for an EDGE type, there doesn't seem to be an option for a reducer to use for larger targets.

Question - Is there an SCT corrector which doesn't reduce?

[PhotoGav]

Would be worth chatting with this person: 

Not a very encouraging title!! 

[ollypenrice]

1 hour ago, Tommohawk said:

Yes that makes perfect sense - I thought you were suggesting otherwise in your previous post. In practice I probably wont use a barlow - maybe just the C9.25 with reducer.

BTW Olly, can I ask what kit would you use to image smaller DSOs? Assume a CCD/astro CMOS like the ASI1600 and a budget of say approx £1500 for the scope. Actually it would be interesting to know what you would use even without the budget constraint.* 

Also assume EQ6 mount.

These camera makers drive me nuts. They don't give the chip size in mm other than as a diagonal when the software used to model FOV asks for width and height. Curses!

This camera has tiny pixels so would need to be binned 2X2 at even moderate focal lengths. On the other hand it will give good resolution at short FL. I don't see it as being a good bet for long FL.

I would be reluctant to ask an EQ6 to autoguide at better than 0.5," and 0.7" would be a safer bet. I'd double that to define the finest pixel scale I'd use for imaging, so no less than 1 to 1.4"PP. FLO have a calculator for scope-camera combinations.

What are you calling small DSOs? Something like M27? My ideal setup for this would be a FL of 2.5 metres and a pixel scale of about an arcsecond per pixel with an excellent mount. At one time I ran a rig belonging to Yves Van den Broek. 14 inch optimiszed Dall Kirkham (FL 2.5 metres) with SXVH36 giving 0.66"PP. I suspect bigger pixels would have been better suited but the camera wouldn't bin. This lot ran on a Mesu and worked nicely. Here's our M27. 

With the camera you mention and a short FL refractor on an EQ6 I can't see why you wouldn't get super results though.

Olly

[derrickf]

2 minutes ago, ollypenrice said:

I suspect bigger pixels would have been better suited but the camera wouldn't bin. This lot ran on a Mesu and worked nicely. Here's our M27. 

I am intrigued to know why you think bigger pixels would have improved this image. At 0.66 arcseconds/pixel you are able to critically sample stars with an FWHM of around 2 to 2.2  arcseconds (depending on which interpretation of the modified Nyquist theorem you use). It is not unusual to have better than 2.2 arcsecond seeing at my unremarkable site so I'm surprised if it is worse where you are.

Derrick

[ollypenrice]

41 minutes ago, derrickf said:

I am intrigued to know why you think bigger pixels would have improved this image. At 0.66 arcseconds/pixel you are able to critically sample stars with an FWHM of around 2 to 2.2  arcseconds (depending on which interpretation of the modified Nyquist theorem you use). It is not unusual to have better than 2.2 arcsecond seeing at my unremarkable site so I'm surprised if it is worse where you are.

Derrick

Perhaps I'm wrong but my thinking is that we we were probably not resolving any more at 0.66"PP than we would have resolved at the 0.78"PP we'd have had with the 11 meg Kodak chip. I'm not well up on the behaviour of stars because I'm always more concerned about nebular detail. It's not that the image would have been better. It's rather that it's more efficient to use the largest pixels you can without compromising reachable resolution.

Olly

[derrickf]

The use of stellar FWHM as a yardstick has nothing to do with a desire to image the stars themselves but is simply a convenient way of evaluating the ability of an optical system to properly discriminate features (whether they be stellar or diffuse)   - this is analogous to specifying photographic lens resolution in lines/mm - few people actually photograph card with fine lines inscribed on them :-). Stellar FWHM of a critically sampled image is also a good indicator of seeing and correlates well with seeing measurements made more rigorously using Hartmann masks etc.

The whole point of assessing the platescale of a system is to define the size of the smallest features that can be critically sampled; FTAOD when a feature (doesn't matter if it's a star or a fine dust lane or streamer) is critically sampled it means that there is sufficient data in the image to correctly re-construct the maximum level of detail of the  imaged feature that can be captured by the optical system.  If you increase the effective pixel size (and hence platescale) you reduce the ability to image fine detail. Clearly the factor which is out of our control which is the ultimate determinant of max resolution is the seeing so we aim to ensure that we can capture all the detail possible by matching our platescale to that which will properly "sample the seeing" - ie a platescale of seeing/3.3

 

Derrick

[Tommohawk]

Right  - hanging on by my fingertips here!

Derrick - wasnt quite sure how we arrived at "platescale of seeing/3.3"

For anyone else struggling to keep up this might be useful.

BTW Olly I'm guessing the 14" ODK with Mesu wasn't the sub £1500 option!? Did give an excellent M27 though - very nice.

And yes I was thinking M27 or smaller, say to Ring neb  (2'30")

Gav - thanks for the pointer to that C9.25. Interesting that he's looking to trade for an 8" RC. Actually I've got one lined up very locally at a good price which should work OK hopefully. That said, it looks like trying to match the C9.25 to a big chip is going to give lots or probs in the periph. I'll probably still run with it for lunar and planets, and experiment with DSOs at full chip size but not expect too much. 

Maybe an RC then. (But will I be able to sleep in my bed at night for fear of the astigmatism??)

[derrickf]

36 minutes ago, Tommohawk said:

Derrick - wasnt quite sure how we arrived at "platescale of seeing/3.3"

See this page if you can take in any more without your head exploding :-)

Unless I missed something, you started this thread looking for solutions to imaging small scale DSOs - in that case you will not need to worry about using a large detector with any scope you choose since the DSO will typically only occupy a small part of the real estate.

FWIW I would go with an RC in your shoes - astigmatism is not an issue if you collimate properly.

 

Derrick

[Tommohawk]

20 hours ago, derrickf said:

See this page if you can take in any more without your head exploding :-)

Unless I missed something, you started this thread looking for solutions to imaging small scale DSOs - in that case you will not need to worry about using a large detector with any scope you choose since the DSO will typically only occupy a small part of the real estate.

FWIW I would go with an RC in your shoes - astigmatism is not an issue if you collimate properly.

 

Derrick

Take your point about chip size. In practice I would probably have a biggish chip though to double for wider field use - initalliy at least I'll use the Canon so I can check the effect on the periphery with the C9.25.

BTW did you mean to include a link?

[derrickf]

OOPS! Sorry, LINK here

 

Derrick

[ollypenrice]

22 hours ago, derrickf said:

The use of stellar FWHM as a yardstick has nothing to do with a desire to image the stars themselves but is simply a convenient way of evaluating the ability of an optical system to properly discriminate features (whether they be stellar or diffuse)   - this is analogous to specifying photographic lens resolution in lines/mm - few people actually photograph card with fine lines inscribed on them :-).

 

Derrick

I see what you mean. How good a yarstick the stars make, though, I don't know. They are are very different as intense near-point sources from fine extended detail in nebulae and galaxies. We're experimenting, here, with a Meade ACF and it gives pretty poor stars - very large - but nice fine galaxy detail. This is only based on one image but I have known the scope for a while and the previous owner felt the same about it.

Sorry, a bit off topic but I thought it was an interesting tangent.

Olly

[derrickf]

47 minutes ago, ollypenrice said:

it gives pretty poor stars - very large - but nice fine galaxy detail

I don't doubt that is what you are seeing but it is difficult to understand how that might be: the imaging system does not know the source of light impinging on a pixel and all light is subject to the same PSF (point spread function). Logically (and mathematically) - if stars are soft then all features in the image should also be soft.
Also if the stars are soft, how do you intend to focus the telescope?

 

An important reason why stellar FWHM is used as a yardstick of resolution is that we know their true geometry - they are a point source - knowing this and measuring the FWHM of stars in an image enables us to calculate the PSF which describes how the atmosphere and our optical system distort any incoming light. This is getting off topic and delving deeper will require resorting to fairly heavy maths and statistics so I'll leave it there.

Apropos your SCT it seems, from your posts in other topics, that you are wedded to the idea of binning images with the ACF - it might be worth trying a few images unbinned to see how the stars and diffuse features look.

 

Derrick

[ollypenrice]

36 minutes ago, derrickf said:

I don't doubt that is what you are seeing but it is difficult to understand how that might be: the imaging system does not know the source of light impinging on a pixel and all light is subject to the same PSF (point spread function). Logically (and mathematically) - if stars are soft then all features in the image should also be soft.
Also if the stars are soft, how do you intend to focus the telescope?

 

An important reason why stellar FWHM is used as a yardstick of resolution is that we know their true geometry - they are a point source - knowing this and measuring the FWHM of stars in an image enables us to calculate the PSF which describes how the atmosphere and our optical system distort any incoming light. This is getting off topic and delving deeper will require resorting to fairly heavy maths and statistics so I'll leave it there.

Apropos your SCT it seems, from your posts in other topics, that you are wedded to the idea of binning images with the ACF - it might be worth trying a few images unbinned to see how the stars and diffuse features look.

 

Derrick

At F10 focus is not all that critical. Initial tests were done with a B mask but the scope will be run robotically and will need to work in autofocus software. So far this has been a bit hit and and miss...

I don't know enough about optics to know why a scope might give large stars while also giving fine detail on other sources. It just seems that this is so.

Olly

[opticalpath]

Whilst I accept the critical sampling theory, it's interesting to see what the majority of amateur imagers actually do in practice.  DSO imagers typically under-sample, often significantly, compared to the FWHM/3.3 formula for critical sampling.  On the other hand, planetary imagers typically over-sample significantly.  Why doesn't everyone zero in on the critical sample plate scale if that is optimal? 

It seems to me that we all choose the particular factor that we want to optimise and move generously over to that side of critical sampling, even though it compromises other aspects of the image. So DSO imagers are concerned about noise in faint parts of their images, and about guiding/ tracking errors during long exposures, so will under-sample to maximise S/N in the wispy stuff and to mask the effects of small guiding errors, whilst accepting loss of the very finest detail resolution.   Planetary imagers are critically interested in the finest detail and so will 'err' on the side of over-sampling to ensure that no possible detail is missed in moments of excellent seeing (I know there are other theoretical reasons for over-sampling), whilst compromising on the additional exposure or gain that is required to compensate.

If critical sampling is indeed optimal, why isn't everyone operating at that scale?  Is it just that we're dealing with imperfect systems and making practical compromises?

Small DSOs like planetary nebulas are an interesting in-between case where we want small-scale detail and to optimise S/N in faint regions.  Perhaps for these objects critical sampling is indeed the ideal.

Adrian

 

 

[derrickf]

I certainly don't disagree with you: those imagers who care or understand about sampling rates decide on what configuration of scopes and cameras to invest in based on their personal imaging goals and the type of target that they are interested in; many others just copy what other people have that produces the kind of images they like. Both of these routes to deciding on which equipment to own/use are perfectly valid and reasonable.

Personally I am more interested in smaller DSOs so I want to capture the maximum detail permitted by the seeing conditions at my site.

 

The discussion started here by the OP was about the best gear for small DSOs and that, indubitably, is an imaging system that critically samples or slightly over-samples the seeing thus providing the maximum detail at the largest platescale.

 

Derrick