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

Refiguring a Dynamax 8" Schmidt Corrector

  • Please log in to reply
886 replies to this topic

#876 Geo31

Geo31

    Voyager 1

  • *****
  • Posts: 11208
  • Joined: 28 Jan 2013
  • Loc: Willis, TX

Posted 13 April 2019 - 11:16 AM

 

I'm a Big Fan of Dave's work, and especially on this DX-8.  I suspected that the mirrors in mine were excellent, and it was the lousy corrector that hobbled the scope.  I want to see if all Dave's hard work confirms my opinion.

It's probably something I wouldn't try for a number of reasons, especially since it's easy and relatively cheap to just buy a C8.  But if I had a D8 I'd probably figure "What have I got to lose?"  The flattening seemed straight-forward enough.  I'm really curious what it's going to take to fix the Schmidt curve.



#877 terraclarke

terraclarke

    Hubble

  • *****
  • Posts: 19957
  • Joined: 29 May 2012
  • Loc: Ohio Valley

Posted 13 April 2019 - 04:39 PM

if I had a D8 I'd probably figure "What have I got to lose?"  The flattening seemed straight-forward enough.  I'm really curious what it's going to take to fix the Schmidt curve.


It seems a lot more complex to me than grinding my Newtonian mirror and I thought that (parabola) to be pretty complex when I was making it 52 years ago.

Cross-section of a Schmidt corrector plate, skyward (flat) side up, (vertical profile exaggerated):

Attached Thumbnails

  • CE746E8D-22CF-4026-A81D-943143FF9C60.png

  • rcwolpert and Bomber Bob like this

#878 wfj

wfj

    Surveyor 1

  • *****
  • Posts: 1717
  • Joined: 10 Jan 2008
  • Loc: California, Santa Cruz County

Posted 13 April 2019 - 08:10 PM

I haven't found an underperforming SCT yet where it wasn't the corrector plate.

 

And I made one once using the vacuum method. While its under a specific negative pressure, you grind/polish a long FL sphere. Then when you're done, release the pressure and you have the curve.


  • Bomber Bob likes this

#879 rolo

rolo

    Cosmos

  • *****
  • Posts: 8793
  • Joined: 14 Jan 2007
  • Loc: GA

Posted 13 April 2019 - 08:14 PM

I haven't found an underperforming SCT yet where it wasn't the corrector plate.

 

And I made one once using the vacuum method. While its under a specific negative pressure, you grind/polish a long FL sphere. Then when you're done, release the pressure and you have the curve.

That's why back in the day Celestron advertised 1/10 wave mirrors but mentioned nothing about the image degrading correctorbombdrop.gif


  • Paul Hyndman, wfj and Bomber Bob like this

#880 DAVIDG

DAVIDG

    Cosmos

  • *****
  • topic starter
  • Posts: 8895
  • Joined: 02 Dec 2004
  • Loc: Hockessin, De

Posted 13 April 2019 - 09:23 PM

 My corrector has good areas and bad areas. The reason was because the glass was wavey to start. So when you pull it down against a Master surface those waves get transferred to the top surface along with the opposite Schmidt curve. Now you grind it flat and release it from the Master  and you have the Schmidt Curve but it also has the opposite shape of the waves. That is what I have now. The problem is the waves were big enough to cause errors that would take a long time to polish out. I found the same problem when I tried to polish back surface to be optically smooth. They were just too big and I had to grind the corrector against a flat surface to get it mechanically smooth enough to polish it optically smooth.

   So I tried pressing the Schmidt curve into a pitch to form a lap and then because of the shape of Schmidt curve one can only rotate the corrector to polish it.  I was hoping that since this surface was the one that ground and polished hat the error would be small enough to polish out.  That isn't the case so I'm back to having to grind them out. There are a number of articles on how to grind the Schmidt curve into a flat surface without the vacuum method.So I know this can be done and I'll ready have curved formed it just has some bad spots So my thought is to make a grinding tool like I did a pitch lap. I'll make the tool by pressing the corrector into Bondo instead of pitch. Bondo is hard enough so it will grind the glass with optical grit. I would mark the corrector were the good and bad areas are and  grind out the bad areas on the Bondo tool. Now I would rotate the corrector against this tool. So the bad areas on the corrector would grind against the areas on the tool that have the corrector shape. Hopefully the result will be an uniform Schmidt curve that then can be polished.  

 

 

                           - Dave 


  • tim53, andycknight, rcwolpert and 4 others like this

#881 wfj

wfj

    Surveyor 1

  • *****
  • Posts: 1717
  • Joined: 10 Jan 2008
  • Loc: California, Santa Cruz County

Posted 13 April 2019 - 11:36 PM

Off the top of my head - could you "map" the areas and laser mill (or chemically etch) them to remove the "wave" selectively, then fine grind to "feather" the changes, then use a petal lap to restore the figure?

 

Might not be so much work?



#882 tim53

tim53

    Voyager 1

  • *****
  • Posts: 14306
  • Joined: 17 Dec 2004
  • Loc: Highland Park, CA

Posted 14 April 2019 - 12:24 AM

I would be worried about the bondo shrinking, as it does quite a bit unless it's really thin.



#883 Geo31

Geo31

    Voyager 1

  • *****
  • Posts: 11208
  • Joined: 28 Jan 2013
  • Loc: Willis, TX

Posted 14 April 2019 - 08:37 AM

It seems a lot more complex to me than grinding my Newtonian mirror and I thought that (parabola) to be pretty complex when I was making it 52 years ago.

Cross-section of a Schmidt corrector plate, skyward (flat) side up, (vertical profile exaggerated):

Yep.  Know what a Schimdt corrector curve looks like.  That's why I'm curious what it's going to take to fix it.  Again, I certainly wouldn't go looking for a D8 (why look for problems?), but if I had one, I don't think I'd have much to lose.  I think that's Dave's point.


  • Bomber Bob likes this

#884 DAVIDG

DAVIDG

    Cosmos

  • *****
  • topic starter
  • Posts: 8895
  • Joined: 02 Dec 2004
  • Loc: Hockessin, De

Posted 14 April 2019 - 11:47 AM

Off the top of my head - could you "map" the areas and laser mill (or chemically etch) them to remove the "wave" selectively, then fine grind to "feather" the changes, then use a petal lap to restore the figure?

 

Might not be so much work?

  I have already mapped the bad areas. It is easy to do by just by placing the schmidt side against my optical flat. There are a couple of pictures in the thread in the ATM forum that shows this. Here is  a picture as well. The areas from 10 to 11 and 1 to 2 o'clock in the image are what the curve should look like so those are a section of a "good" area.  There are 4 high areas on my plate that are around 90 degrees apart. 

   So If I cast a tool in which I press the curve side of plate into, just like I did with a pitch lap I have the curve I need to grind the bad areas away. I just cut away the bad areas leave the areas with the correct curve and grind the corrector against that tool. The grinding motion is just rotation and no side to side. Remember I'm  trying to develop a procedure that use fairly common techniques so others might be able to follow what I have done.Laser milling and the unlike it too exotic for what I'm try to accomplish

   Like I said there are a number of articles on how you make a Schmidt corrector starting off with flat piece of glass and by first grinding it against a tool that uses peddle shaped areas then polishing it against a pitch lap that is also peddle shaped. So I already have 85% of the curve I just need to clean up the bad areas. So the process will be similar to what is outlined in those articles.

 

              - Dave 

 

 

schmidtcorrector2hrs.jpg     


  • Geo31 and Bomber Bob like this

#885 wfj

wfj

    Surveyor 1

  • *****
  • Posts: 1717
  • Joined: 10 Jan 2008
  • Loc: California, Santa Cruz County

Posted 14 April 2019 - 01:38 PM

My comment was in consideration of other optical techniques to achieve same (have been experimenting with nonlinear optics and additive manufacturing processes to understand scope of what they can do, mostly for evaluating new business opportunities at nano scale). I've been wondering if we'll always grinding/polishing optics, or, if like with "3D printing", other means of fabrication might displace them. Currently semiconductor processing have made them "accessible".

 

 

  I have already mapped the bad areas. It is easy to do by just by placing the schmidt side against my optical flat. There are a couple of pictures in the thread in the ATM forum that shows this. Here is  a picture as well. The areas from 10 to 11 and 1 to 2 o'clock in the image are what the curve should look like so those are a section of a "good" area.  There are 4 high areas on my plate that are around 90 degrees apart.

You're showing the "waves"/"ripples" at the contact zone with the plate (and perhaps the center). By pressing one can "see" other zones when they are brought into contact. The overall pattern, not just the given point contact was what I was referring to.

 

As a 2D map of all of these, one could selectively etch/mill the "lands" that stick out. This would leave in place the existing Schmidt curve largely intact. The residual artifacts of the process would be over/under adjacent "sleeks" that would polish out quickly with the petal lap that would refine/enhance the desired curve.

 

Same technique might make optical refinement a more common, cheaper, quicker, deterministic means to fix poor optics. Per rolo's "image degrading corrector", it might give customers a stronger case to insist vendors "fix their broken optics", where they could automate, recoat, and return to customer as warranty repair.



#886 DAVIDG

DAVIDG

    Cosmos

  • *****
  • topic starter
  • Posts: 8895
  • Joined: 02 Dec 2004
  • Loc: Hockessin, De

Posted 14 April 2019 - 08:56 PM

 It would be difficult to repair a corrector economically as a commercial process.It is cheaper to make one from scratch and you still have the problem of then touching up the secondary to null out the secondary .  So the labor cost would be more  than the a new scope. That is why Celestron and Questar who match optics will only replace the complete set of optics. Is just cheaper to do it that way. 

             Technically the way to fix these fairly quickly is to use a CNC lathe with diamond turning. So the profile can be programmed in and the curve cut. It is done all the time but you need about $100K in tooling. So you'll have to make a lot of correctors to make your money back. 

   As I have stated thou the purpose of this project is to develop a process that uses common tooling and material so a ATM that  has a made a mirror or two would have good chance of success. 

  

 

                    - Dave 


  • Bomber Bob and Augustus like this

#887 wfj

wfj

    Surveyor 1

  • *****
  • Posts: 1717
  • Joined: 10 Jan 2008
  • Loc: California, Santa Cruz County

Posted 14 April 2019 - 10:25 PM

Wasn't suggesting this as a commercial service to fix old scopes. Did suggest means to "push back" on vendors of new scopes, which will be using new fabrication technology. How soon is hard to guess.

 

Optical fabrication is changing. We're already using additive manufacturing to replace "classic" parts. I'm fabricating using it right now. It's cheaper/faster than traditional.

 

I'm just exploring next steps as it applies to this community. Perhaps its outside of the "scope" of this topic, where "classic" techniques are all one should use on classic scopes. My bad.




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






Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics