Jump to content

  •  

CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.

Photo

A Tri-Bahtinov mask for SCT collimation and focusing

astrophotography imaging SCT
  • Please log in to reply
501 replies to this topic

#426 Psittacula

Psittacula

    Explorer 1

  • -----
  • Posts: 93
  • Joined: 27 May 2017
  • Loc: Tokyo

Posted 22 January 2021 - 01:52 PM

But also found that outW parameter doesn't apply when generating mask. You may want to check.

 

And I found that rounded B-mask generator doesn't go well with Fusion 360 on 'some' slits. I don't know what makes some slits are left unselected for extrusion.

 

The outW option seems to be working in my environment.

 

The problem with fusion360 may possibly be caused by the use of elliptical arc curve commands in the svg path for the rounded pattern. 

For example, I wonder if it would help to import the SVG file from the generator into Inkscape, export it as DXF, and then import the DXF file into Fusion360.

 

Satoru



#427 Jinux

Jinux

    Messenger

  • *****
  • Posts: 488
  • Joined: 25 Jun 2007
  • Loc: Bay Area, CA

Posted 22 January 2021 - 02:03 PM

The outW option seems to be working in my environment.

 

The problem with fusion360 may possibly be caused by the use of elliptical arc curve commands in the svg path for the rounded pattern. 

For example, I wonder if it would help to import the SVG file from the generator into Inkscape, export it as DXF, and then import the DXF file into Fusion360.

 

Satoru

I'll double check on outW. It's webapp, can't be different result, right? :)

 

I'm already using SVG -> Inkscape -> DXF -> Fusion360 as you found that it is the most accurate way to preserve geometry correctly.

Let me try SVG directly and see whether it makes difference. Scaling is manageable issue than non-selectable profile.

 

-Jinux


  • Psittacula likes this

#428 MarMax

MarMax

    Surveyor 1

  • *****
  • Posts: 1,747
  • Joined: 27 May 2020
  • Loc: Los Angeles

Posted 22 January 2021 - 03:54 PM

I'm already using SVG -> Inkscape -> DXF -> Fusion360 as you found that it is the most accurate way to preserve geometry correctly.

I can't get Inkscape to DXF, what am I doing wrong? I'm using Windows 7 and have AutoCad LT 2014.



#429 cytan299

cytan299

    Apollo

  • *****
  • topic starter
  • Posts: 1,097
  • Joined: 24 Nov 2014

Posted 22 January 2021 - 04:10 PM

Hi,

I've also added a rounded corners option to the original tri-bahtinov mask generator webApps.

https://satakagi.git...i-Bahtinov.html

It's a bit ugly, but probably not too inconvenient. It also works when there are boundary stems.

attachicon.gifTBM_rounding.png

 

Satoru

I think it's great!

 

cytan


  • Psittacula likes this

#430 Jinux

Jinux

    Messenger

  • *****
  • Posts: 488
  • Joined: 25 Jun 2007
  • Loc: Bay Area, CA

Posted 22 January 2021 - 04:53 PM

I can't get Inkscape to DXF, what am I doing wrong? I'm using Windows 7 and have AutoCad LT 2014.

I don't know how you do it, so I don't know what you're doing wrong. :)

Open it on Inkscape and use "Save As" and select DXF. Make sure you select "Base unit" as "mm"

Just in case you still have issue, I put three files in following link

 

https://app.box.com/...f4oqqqrust7rgau

 

STEP file for 2nd version mask I made targeted for F/6.3 but will work for native focal length too.

2nd version of mask generated and converted to DXF both of original and rounded

 

-Jinux

 

ps. Fusion 360 recognize SVG correctly on rounded corner. So culprit is Inkscape conversion to DXF.



#431 MarMax

MarMax

    Surveyor 1

  • *****
  • Posts: 1,747
  • Joined: 27 May 2020
  • Loc: Los Angeles

Posted 22 January 2021 - 05:40 PM

I don't know how you do it, so I don't know what you're doing wrong. smile.gif

Open it on Inkscape and use "Save As" and select DXF. Make sure you select "Base unit" as "mm"

Just in case you still have issue, I put three files in following link

 

https://app.box.com/...f4oqqqrust7rgau

 

STEP file for 2nd version mask I made targeted for F/6.3 but will work for native focal length too.

2nd version of mask generated and converted to DXF both of original and rounded

 

-Jinux

 

ps. Fusion 360 recognize SVG correctly on rounded corner. So culprit is Inkscape conversion to DXF.

Thanks for the links! I just noticed that the DXF export file options are "DesktopCuttingPlotter" so that could be whats wrong. There is no other DXF export option I see. And I have to admit that I don't know what you mean by select "Base unit" since I've not found where that is.



#432 Psittacula

Psittacula

    Explorer 1

  • -----
  • Posts: 93
  • Joined: 27 May 2017
  • Loc: Tokyo

Posted 25 January 2021 - 08:15 AM

Hi,
 

I've been experimenting with the rounded edge option of the Tri-bahtinov mask to see if it has any effect. The mask was created by cutting out a thin plastic sheet with cutting machine (Silhouette Cameo 3). 

roundedPic.jpg
 

The comparison is made with V2, which was created with all the same parameters except for the rounded option.

Condition: Aldebaran, Tri-bahtinov mask, C6, ASI224MC, 100ms, gain max.

 

pattern.png

Aldebaran TBM r2 rounded dif.jpg

 

I could hardly see any difference, so I stacked the images captured by SharpCap with AutoStakkert, just like the planets, and I could see some spuriousness.
The trend is definitely less spike-like spuriousness. However, it is not enough to bother me, and considering the fact that rounding reduces the aperture of the slit and makes the image a little dimmer, my personal impression is that I do not care either way.

Satoru


  • AstroNikko likes this

#433 cytan299

cytan299

    Apollo

  • *****
  • topic starter
  • Posts: 1,097
  • Joined: 24 Nov 2014

Posted 25 January 2021 - 09:56 AM

Hi,
 

I've been experimenting with the rounded edge option of the Tri-bahtinov mask to see if it has any effect. The mask was created by cutting out a thin plastic sheet with cutting machine (Silhouette Cameo 3). 

attachicon.gifroundedPic.jpg
 

The comparison is made with V2, which was created with all the same parameters except for the rounded option.

Condition: Aldebaran, Tri-bahtinov mask, C6, ASI224MC, 100ms, gain max.

 

attachicon.gifpattern.png

attachicon.gifAldebaran TBM r2 rounded dif.jpg

 

I could hardly see any difference, so I stacked the images captured by SharpCap with AutoStakkert, just like the planets, and I could see some spuriousness.
The trend is definitely less spike-like spuriousness. However, it is not enough to bother me, and considering the fact that rounding reduces the aperture of the slit and makes the image a little dimmer, my personal impression is that I do not care either way.

Satoru

I agree after looking at your results that rounding actually makes the mask *harder* to use because it's dimmer. So, I think your original design is superior :)

 

cytan


  • Psittacula likes this

#434 Psittacula

Psittacula

    Explorer 1

  • -----
  • Posts: 93
  • Joined: 27 May 2017
  • Loc: Tokyo

Posted 25 January 2021 - 10:21 AM

If we were going to round the edges of the slits, it seemed preferable to do so while at the same time reducing the pitch of the slits.

This would reduce the lack of aperture area due to the rounding, and would also spread the diffracted light strips outward.

 

Satoru



#435 cytan299

cytan299

    Apollo

  • *****
  • topic starter
  • Posts: 1,097
  • Joined: 24 Nov 2014

Posted 25 January 2021 - 12:31 PM

If we were going to round the edges of the slits, it seemed preferable to do so while at the same time reducing the pitch of the slits.

This would reduce the lack of aperture area due to the rounding, and would also spread the diffracted light strips outward.

 

Satoru

If you reduce the pitch, I think plastic between slits will also be reduced, thus making the mask more fragile. It's always a compromise. IMO, rounding of the edges only gives you a small improvement and may not be worth doing.

 

cytan


  • Psittacula likes this

#436 MarMax

MarMax

    Surveyor 1

  • *****
  • Posts: 1,747
  • Joined: 27 May 2020
  • Loc: Los Angeles

Posted 25 January 2021 - 03:19 PM

Hi,
 

I've been experimenting with the rounded edge option of the Tri-bahtinov mask to see if it has any effect. The mask was created by cutting out a thin plastic sheet with cutting machine (Silhouette Cameo 3). 

 

The comparison is made with V2, which was created with all the same parameters except for the rounded option.

Condition: Aldebaran, Tri-bahtinov mask, C6, ASI224MC, 100ms, gain max.

 

I could hardly see any difference, so I stacked the images captured by SharpCap with AutoStakkert, just like the planets, and I could see some spuriousness.
The trend is definitely less spike-like spuriousness. However, it is not enough to bother me, and considering the fact that rounding reduces the aperture of the slit and makes the image a little dimmer, my personal impression is that I do not care either way.

Satoru

Could you put up the star test images of all three types of masks (original, V2 and V2 rounded) next to each other. I'm still trying to see and understand the differences. I know there are all here in the thread but I'm not sure which is the original. Thank you.



#437 Psittacula

Psittacula

    Explorer 1

  • -----
  • Posts: 93
  • Joined: 27 May 2017
  • Loc: Tokyo

Posted 25 January 2021 - 05:55 PM

Could you put up the star test images of all three types of masks (original, V2 and V2 rounded) next to each other. I'm still trying to see and understand the differences. I know there are all here in the thread but I'm not sure which is the original. Thank you.

There is no side-by-side comparison of the three masks, but I do have a comparison between the original and V2 on the page here. There is a trade-off between these as well.

 

https://satakagi.git...ensitivity.html

 

Regards,

Satoru



#438 MarMax

MarMax

    Surveyor 1

  • *****
  • Posts: 1,747
  • Joined: 27 May 2020
  • Loc: Los Angeles

Posted 25 January 2021 - 08:57 PM

There is no side-by-side comparison of the three masks, but I do have a comparison between the original and V2 on the page here. There is a trade-off between these as well.

 

https://satakagi.git...ensitivity.html

 

Regards,

Satoru

Great, thanks for the link. I like the original pattern, seems more distinct. Thinner and brighter spikes seem to stand out better IMO. I was trying to understand what the benefit of the V2 was and IMO the original looks more distinct. Maybe it's just more light passing through the original.



#439 Whichwayisnorth

Whichwayisnorth

    Gemini

  • *****
  • Posts: 3,349
  • Joined: 04 Jul 2011
  • Loc: Southern California

Posted 27 January 2021 - 04:11 PM

Sorry if this has already been discussed. 18 pages of this thread  I just skimmed.

I wanted to print one out using my 3d printer.  I used the site to generate an .svg file and then imported it using tinkercad but when I do so it makes the mask expand out to something like 650mm and cuts it apart.  So I selected import as "art" and then set the bed dimensions to 300/300 which is what I have.  Then it puts it exactly on the bed 300/300.    In other words, it doesn't seem to carry over the actual dimensions in mm into a file that tinkercad can read.

 

Is there another way to convert .svg to .stl while maintaining the correct dimensions so it can be printed on a 3d printer?

 

Thanks



#440 Jinux

Jinux

    Messenger

  • *****
  • Posts: 488
  • Joined: 25 Jun 2007
  • Loc: Bay Area, CA

Posted 27 January 2021 - 04:59 PM

Sorry if this has already been discussed. 18 pages of this thread  I just skimmed.

I wanted to print one out using my 3d printer.  I used the site to generate an .svg file and then imported it using tinkercad but when I do so it makes the mask expand out to something like 650mm and cuts it apart.  So I selected import as "art" and then set the bed dimensions to 300/300 which is what I have.  Then it puts it exactly on the bed 300/300.    In other words, it doesn't seem to carry over the actual dimensions in mm into a file that tinkercad can read.

 

Is there another way to convert .svg to .stl while maintaining the correct dimensions so it can be printed on a 3d printer?

 

Thanks

It's described in Satoru's webapp link too.

Download Inkscape and import SVG there and export DXF. Make sure you set mm as unit. All scaling will carry over exactly, or at least to Fusion 360.



#441 MarMax

MarMax

    Surveyor 1

  • *****
  • Posts: 1,747
  • Joined: 27 May 2020
  • Loc: Los Angeles

Posted 31 January 2021 - 03:30 PM

I was finally able to test my "original" type Tri-B mask. It was cut out of 3/16" (4.5mm) gator board with a razor knife. It has been cloudy and rainy here so last night was the first opportunity for a quick test and it works as advertised.

 

I was not able to get a picture so I've edited Satoru's "original" picture to show what I saw. Below is a crude hand drawn approximation.

 

gallery_332504_14251_237363.jpg

 

Visually the center is not blown out (overexposed). I'm sure the picture was taken this way intentionally to show the spikes so the center had to be overexposed. What I saw was clear lines pointed to a center which looked like intersecting lines. But most importantly, it looks like my collimation is spot on, as in I don't think it's possible to get it any better.



#442 cytan299

cytan299

    Apollo

  • *****
  • topic starter
  • Posts: 1,097
  • Joined: 24 Nov 2014

Posted 31 January 2021 - 06:04 PM

I was finally able to test my "original" type Tri-B mask. It was cut out of 3/16" (4.5mm) gator board with a razor knife. It has been cloudy and rainy here so last night was the first opportunity for a quick test and it works as advertised.

 

I was not able to get a picture so I've edited Satoru's "original" picture to show what I saw. Below is a crude hand drawn approximation.

 

gallery_332504_14251_237363.jpg

 

Visually the center is not blown out (overexposed). I'm sure the picture was taken this way intentionally to show the spikes so the center had to be overexposed. What I saw was clear lines pointed to a center which looked like intersecting lines. But most importantly, it looks like my collimation is spot on, as in I don't think it's possible to get it any better.

As long as this was done at the centre of the scope image, then the scope is in collimation.

 

You can double check this by going into focus and out of focus to check that in fact it is. Here's an example when I checked it with my ETX125:

 

https://www.cloudyni...-goodbad-is-it/

 

 

cytan

 

cytan


Edited by cytan299, 31 January 2021 - 06:04 PM.


#443 MarMax

MarMax

    Surveyor 1

  • *****
  • Posts: 1,747
  • Joined: 27 May 2020
  • Loc: Los Angeles

Posted 01 February 2021 - 12:21 AM

As long as this was done at the centre of the scope image, then the scope is in collimation.

 

You can double check this by going into focus and out of focus to check that in fact it is. Here's an example when I checked it with my ETX125:

 

https://www.cloudyni...-goodbad-is-it/

 

 

cytan

 

cytan

You are way beyond my comprehension with all the things in that post . . lol

 

And I was definitely in the center of the scope image with the illustration. There was a touch of turbulence but the center spike in each set would drift equally on each side with the average being in the center. Next time the conditions allow I'll make sure to get a decent picture or video.



#444 archer1960

archer1960

    Viking 1

  • *****
  • Posts: 909
  • Joined: 26 Jul 2011
  • Loc: Southern New England

Posted 07 February 2021 - 10:52 AM

Just got my Tri-Bahtinov yesterday, and tried it on my AT-10RC. I think this shows that my collimation is decent, but not perfect. Agree?

Attached Thumbnails

  • LIGHT_5s_1600iso_20210206_19h58m07s961_+18c.JPG


#445 cytan299

cytan299

    Apollo

  • *****
  • topic starter
  • Posts: 1,097
  • Joined: 24 Nov 2014

Posted 07 February 2021 - 11:18 AM

Just got my Tri-Bahtinov yesterday, and tried it on my AT-10RC. I think this shows that my collimation is decent, but not perfect. Agree?

Not perfect. Clearly, the spikes at around 10-11 o’clock and 3-4 o’clock are not symmetric. A bit of tweaking will get you spot on.

 

cytan


  • archer1960 and Psittacula like this

#446 meegja

meegja

    Viking 1

  • -----
  • Posts: 514
  • Joined: 02 Jul 2020
  • Loc: The Netherlands

Posted 17 March 2021 - 04:07 AM

Hi everybody

Wondering if the TriBahtinov Grabber software is still in development?
Last mention of it was a year ago by Jinux, see https://www.cloudyni...ing/?p=10044031



#447 mlord

mlord

    Surveyor 1

  • *****
  • Posts: 1,826
  • Joined: 25 Oct 2020
  • Loc: Ottawa, Canada

Posted 22 March 2021 - 09:57 PM

I used the Modified Tri-Bahtinov Mask Drawings Generator https://satakagi.git..._symmetric.html with the default settings of 2069.973 / 225.425 / 73.025 to generate a mask suitable for my (new-ish) Evolution 8.

 

A friend found a 3D printer large enough to make one (most home printers seem to be about 5-10mm too small for that size mask) and I got the mask back today.  I asked for a 2mm thick print, with a 3mm thick outside border for extra durability.  I haven't eyed it for collimation yet, but the feedback for extremely accurate focus is rather good!

 

Here is the .stl file we used for the mask as printed: http://rtr.ca/astron...ov_for_evo8.stl


  • cytan299 and Psittacula like this

#448 Kaydubbed

Kaydubbed

    Mariner 2

  • *****
  • Posts: 288
  • Joined: 29 Sep 2018
  • Loc: Kansas City, Missouri

Posted 01 April 2021 - 08:57 AM

I used the Modified Tri-Bahtinov Mask Drawings Generator https://satakagi.git..._symmetric.html with the default settings of 2069.973 / 225.425 / 73.025 to generate a mask suitable for my (new-ish) Evolution 8.

 

A friend found a 3D printer large enough to make one (most home printers seem to be about 5-10mm too small for that size mask) and I got the mask back today.  I asked for a 2mm thick print, with a 3mm thick outside border for extra durability.  I haven't eyed it for collimation yet, but the feedback for extremely accurate focus is rather good!

 

Here is the .stl file we used for the mask as printed: http://rtr.ca/astron...ov_for_evo8.stl

I'll buy one for my Edge 11 SCT if your friend wants to print one!

 

I just don't have the resources [or energy] to hunt down someone local with a 3D printer and go through that trial and error if they aren't familiar with these masks. 



#449 mlord

mlord

    Surveyor 1

  • *****
  • Posts: 1,826
  • Joined: 25 Oct 2020
  • Loc: Ottawa, Canada

Posted 01 April 2021 - 09:32 AM

There are lots of 3D-print companies in the USA who would print/ship one for you.  Have a look for them.

 

Cheers



#450 MarMax

MarMax

    Surveyor 1

  • *****
  • Posts: 1,747
  • Joined: 27 May 2020
  • Loc: Los Angeles

Posted 01 April 2021 - 10:42 AM

I'll buy one for my Edge 11 SCT if your friend wants to print one!

 

I just don't have the resources [or energy] to hunt down someone local with a 3D printer and go through that trial and error if they aren't familiar with these masks. 

Buckeyestargazer is one of the CN vendors 3D printing the Tri-B masks. I bought one for my CPC 1100 and it's very good.




CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.


Recent Topics





Also tagged with one or more of these keywords: astrophotography, imaging, SCT



Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics