Can I put a Barlow in an OAG?

[SteveBz]

Hi Guys,

I have a 50.5 mm OAG n my Celestron C8N.  Can I image at longer focal lengths, say by putting a 31.7 mm Barlow through it, so that the guiding is unchanged, but the imaging is at a higher resolution?  Will I get focus?

Regards,

Steve.

[vlaiv]

Barlows tend to push focal plane further out so you'll just need some extension at camera side, alternatively, if you leave imaging camera as is - you might be unable to reach focus with guide camera - it will need additional "inward" travel.

[SteveBz]

7 hours ago, vlaiv said:

Barlows tend to push focal plane further out so you'll just need some extension at camera side, alternatively, if you leave imaging camera as is - you might be unable to reach focus with guide camera - it will need additional "inward" travel.

Mmm.... I don't really know how to fit it.  It would be an M42 male fitting from the TOAG, with somehow a Barlow inside it.  The Barlow would have it's own M42 for the DSLR T-ring.  Then I'd need an M42 - 31.7 mm stepdown extender.  Not sure how that would work.

Steve. 

[vlaiv]

17 minutes ago, SteveBz said:

Mmm.... I don't really know how to fit it.  It would be an M42 male fitting from the TOAG, with somehow a Barlow inside it.  The Barlow would have it's own M42 for the DSLR T-ring.  Then I'd need an M42 - 31.7 mm stepdown extender.  Not sure how that would work.

Steve. 

What sort of focal length are you hoping to achieve and what pixel size will you be using?

[SteveBz]

9 hours ago, vlaiv said:

What sort of focal length are you hoping to achieve and what pixel size will you be using?

So, if I could double the focal length that would be great.  It's currently an FL of 1m and a pixel size of about 5 microns.  I gout my box of adaptors out last night and there was no obvious way to do it.  Maybe I could unscrew the Barlow cell and use a step-up adaptor.   I'd still need some sort of extender.

Do I remember right Vlaiv, was it you who upgraded to a 203mm RC?  How's it going?  Are you still happy with it?  F/8, so I guess a focal length of 1.6 or something.

Steve.

[vlaiv]

That is F/5 scope?

With 1000mm and 5µm pixel size - you are already at 1"/px - that is highest sampling rate you should realistically go with. Using barlow will give you 0.5"/px - that is too much.

Yes you remember correctly - I have 8" F/8 RC and I use ASI1600 - which has 3.8µm pixels. That gives me 0.5"/px  - and I ended up binning x2 data every single time (sometimes even x3 because of poor seeing). Unfortunately - scope has not seen starlight this whole year, but that is hopefully about to change in 3-4 months - at that time I'll be moving into a new house and possibly have new obsy finished as well

Why do you want to barlow when you are already at max resolution? Going with higher resolution is not going to give you any more detail.

 

[SteveBz]

On 21/12/2020 at 11:28, vlaiv said:

That is F/5 scope?

With 1000mm and 5µm pixel size - you are already at 1"/px - that is highest sampling rate you should realistically go with. Using barlow will give you 0.5"/px - that is too much.

 

Hi Vlaiv,

I'm not really sure what this means.  The diffraction limited resolution on my scope is .57" and as you say my scale is about 1"/px.   I feel, for instance, that if I photograph M57, I should see the two stars within it, but I don't, I'm lucky if I can see the central white dwarf, but I can only imagine the second one.  Generally I feel my stars are a bit blobby.

In this photo, for example, the focus feels tight.  I don't see any coma.  The guiding looks OK, but I still feel I would like my stars to be smaller and more pin-like  A Barlow might not be the right answer, but I feel I could do better.  Do you have any other suggestions about what I could do?

Happy Christmas,

Steve.
 

[vlaiv]

Pixel per pixel, your image of M57 looks like this:

At this resolution, like you noticed, stars are no longer pin points but start to feel soft. This has nothing to do with telescope and it's resolving power - it has to do with atmosphere and tracking. Long exposure photography simply can't utilize resolutions below 1"/px and sometimes not even that - more like 1.5"/px.

Using barlow will just make things worse - you won't capture any additional detail as atmosphere is blurring it - and you'll just have lower SNR and you'll need more imaging time.

If you have very good SNR in your image - you could try sharpening - one like planetary imagers do. With even lucky imaging you need sharpening to achieve full resolution of telescope.

Even on this Jpeg image that is 8bit in nature - we can still get some more detail by using wavelets (although it amplifies the noise as well and Jpeg artifacts star to show - but at least stars are tighter and you can see both of them in nebula):

[SteveBz]

4 minutes ago, vlaiv said:

Pixel per pixel, your image of M57 looks like this:

At this resolution, like you noticed, stars are no longer pin points but start to feel soft. This has nothing to do with telescope and it's resolving power - it has to do with atmosphere and tracking. Long exposure photography simply can't utilize resolutions below 1"/px and sometimes not even that - more like 1.5"/px.

Using barlow will just make things worse - you won't capture any additional detail as atmosphere is blurring it - and you'll just have lower SNR and you'll need more imaging time.

If you have very good SNR in your image - you could try sharpening - one like planetary imagers do. With even lucky imaging you need sharpening to achieve full resolution of telescope.

Even on this Jpeg image that is 8bit in nature - we can still get some more detail by using wavelets (although it amplifies the noise as well and Jpeg artifacts star to show - but at least stars are tighter and you can see both of them in nebula):

Nice.  What software did you use?

[SteveBz]

4 minutes ago, vlaiv said:

Pixel per pixel, your image of M57 looks like this:

At this resolution, like you noticed, stars are no longer pin points but start to feel soft. This has nothing to do with telescope and it's resolving power - it has to do with atmosphere and tracking. Long exposure photography simply can't utilize resolutions below 1"/px and sometimes not even that - more like 1.5"/px.

Using barlow will just make things worse - you won't capture any additional detail as atmosphere is blurring it - and you'll just have lower SNR and you'll need more imaging time.

If you have very good SNR in your image - you could try sharpening - one like planetary imagers do. With even lucky imaging you need sharpening to achieve full resolution of telescope.

Even on this Jpeg image that is 8bit in nature - we can still get some more detail by using wavelets (although it amplifies the noise as well and Jpeg artifacts star to show - but at least stars are tighter and you can see both of them in nebula):

Secondly, unless I go full scale observatory with lasers and active optics, there's not much I can do.  However, I do see better examples of M57 with both stars showing from people with larger scopes.  How have they managed that?  Should I take hundreds of shorter exposures, like 15 secs, and take the top ten percent?

Regards

Steve.

[SteveBz]

This photo here, from memory, is about 2 or 3 hours of 120 sec exposures. 

[vlaiv]

4 minutes ago, SteveBz said:

Nice.  What software did you use?

I used Registax 6 as it has wavelet sharpening often used in planetary imaging. Good results can be obtained with Gold-Meinel sharpening in Gimp 2.10 / gmic filter collection:

They work in different ways but both like good 16/32 bit data.

5 minutes ago, SteveBz said:

How have they managed that?  Should I take hundreds of shorter exposures, like 15 secs, and take the top ten percent?

I think it is down to processing. Could you post 16 bit tiff of crop around the nebula and I'll give it a round of sharpening. You can have better results in better seeing or with narrowband imaging (NB is often sharper).

You could try lucky imaging approach - go for a lot of short exposures and then only stack those with good FWHM values. Then apply good sharpening to result.

[SteveBz]

58 minutes ago, vlaiv said:

Could you post 16 bit tiff of crop around the nebula and I'll give it a round of sharpening. You can have better results in better seeing or with narrowband imaging (NB is often sharper).

Here's a fit.  Is that OK?

M57-400-120s1DegMPCC3C8-N-LP-stacked_v3.fit

[vlaiv]

I could not get much, but here it is after some deconvolution work to try to isolate the stars:

Issue of course is the noise that gets quite noticeable and forms this sort of foam pattern when deconvolution is applied - but it does shrink the stars and now both stars can be seen.

Maybe simple unsharp mask will be enough to get nice looking image:

No, that is not detail in the nebula - those are artifacts from sharpening - but it does look "Hubble" like

 

[SteveBz]

10 hours ago, vlaiv said:

I could not get much, but here it is after some deconvolution work to try to isolate the stars:

Issue of course is the noise that gets quite noticeable and forms this sort of foam pattern when deconvolution is applied - but it does shrink the stars and now both stars can be seen.

Maybe simple unsharp mask will be enough to get nice looking image:

No, that is not detail in the nebula - those are artifacts from sharpening - but it does look "Hubble" like

 

Very nice.  Did you use Registax or Gimp to do this?  I might try StarTools again and see what I can do,

[vlaiv]

40 minutes ago, SteveBz said:

Very nice.  Did you use Registax or Gimp to do this?  I might try StarTools again and see what I can do,

First one was done with AstraImage (demo version, I could not save but took a screen shot), and second one was done in Gimp

[almcl]

 

1 hour ago, SteveBz said:

 I might try StarTools again and see what I can do,

Since you mentioned StarTools, I ran your .fits file through version 1.7.455 and (binning to 35% and cropping pretty heavily) got this:

 

I am sure that with a bit less haste and more attention to detail a much better result would be possible.

[bottletopburly]

Run through Startools 1.7.456 looks nice to me 

 

[SteveBz]

5 hours ago, vlaiv said:

First one was done with AstraImage (demo version, I could not save but took a screen shot), and second one was done in Gimp

I think I prefer the Gimp one.

[SteveBz]

4 hours ago, almcl said:

 

Since you mentioned StarTools, I ran your .fits file through version 1.7.455 and (binning to 35% and cropping pretty heavily) got this:

 

I am sure that with a bit less haste and more attention to detail a much better result would be possible.

Very nice.  OK. I'd better have another go.

[SteveBz]

2 hours ago, bottletopburly said:

Run through Startools 1.7.456 looks nice to me 

 

Spectacular. Is that really the same photo!!  I'm going to have a go and re-post.

[bottletopburly]

2 minutes ago, SteveBz said:

Spectacular. Is that really the same photo!! 

Certainly is ,Startools really starting to shine now 

[SteveBz]

26 minutes ago, bottletopburly said:

Certainly is ,Startools really starting to shine now 

Well yours is certainly better than mine.  This is what I did:  Autodev with ROI, Bin (no crop), Wipe, (tried HDR, decon & Contrast), Colour with Artistic detail aware with Cannon 450D.  Then FilmDev - restretch.

[bottletopburly]

i have attached my data work flow ,im using 1.7.455 in the shrink module it has an unglow feature which is handy , create a mask containing nebula and use unglow  that works well any questions just ask ,looking through the work flow will give you an idea of work flow, ignore blocks of code better to use auto dev create a ROI covering just the nebula.

sgl user data.txt

Edited December 26, 2020 by bottletopburly

[SteveBz]

37 minutes ago, bottletopburly said:

i have attached my data work flow ,im using 1.7.455 in the shrink module it has an unglow feature which is handy , create a mask containing nebula and use unglow  that works well any questions just ask ,looking through the work flow will give you an idea of work flow, ignore blocks of code better to use auto dev create a ROI covering just the nebula.

sgl user data.txt 57.7 kB · 0 downloads

Very nice - I'll just keep practising: