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

hogging two identical 12" f6 mirrors?

  • Please log in to reply
15 replies to this topic

#1 savarin

savarin

    Lift Off

  • -----
  • topic starter
  • Posts: 16
  • Joined: 16 Nov 2013

Posted 20 October 2019 - 10:26 PM

I am in the process (still) of building a 12" f6 binocular telescope.
The tubes are finished so now its time to grind the mirrors on a mirrormatic grinding/polishing machine.
These mirrors will be flexed to final shape.
The back face of the mirrors will have a radius of curvature 2.5 inches smaller than the front at 144 inches.
I have two methods to rough grind in mind.
Method 1
using a swinging pendulum with a diamond grinding disk at the point, swung over the mirror to rough out the curve as its turning on the grinding machine. Both front surfaces can be done this way.
To rough the rear surfaces a plaster tool can be hollowed the same way and lined with steel dumps, re ground to the correct roc and the rear curves can be ground using this tool.
(I have done this and it works very well)
All polishing is carried out in the usual manner with suitable sized pitch laps.
To support the curved rear face of the mirror for polishing I sit the front surface downwards on a smooth flat surface apply a 10mm square ring of rubber around and use an expanding cake ring as a collar then pour in the dental stone.
When set the rubber and cake ring are removed leaving the mirror 10mm proud of the plaster that is now 20mm wider in dia than the mirror and the mirror now supported by the plaster cast is set on the machine and polished in the usual fashion.
Again I have done this with no problems except I accidentally dropped and broke the mirror but early testing showed it was behaving as expected.
The good points I can see are its easy to ensure both mirrors have the same RoC.
The only downside I can see with this method is the the very high tripod required (12 feet)and the inevitable flex in its legs. My experience to date has been on very short focal lengths where its easily managed.

The tripod legs have been braced since this shot.

 

tripod.jpg

 

Method 2
Spin grind mirror A on top of mirror B.
this will produce the correct curvature in mirror B with a convex bottom surface on mirror A.

now flip them and use the flat bottom of mirror B as the tot tool to grind mirror A front surface which now has a curved back surface so must be supported in plaster and producing the curved back surface to mirror B
I believe it will be possible to make both mirrors this way but ensuring they are the exact same RoC when finished could be harder.
I'm also unsure if the radius of curvature of the back surfaces with be the same ROC as the front.
I guess its not a difficult problem to solve as I can make a concave tool to then grind them to the correct 141.5" back face ROC.

 

grinding-idea.jpg

 

Polishing and testing will be via Foucault and Ronchi as I only want a sphere.

Can anyone see any gotchas that I may not have thought of?

Thanks

 


  • retroman2, robertasumendi and Lumix.guy like this

#2 TOMDEY

TOMDEY

    Aurora

  • *****
  • Posts: 4938
  • Joined: 10 Feb 2014
  • Loc: Springwater, NY

Posted 20 October 2019 - 11:50 PM

Hi, Savarin, Cool!

 

Well... the "flexed to final shape" part seems unnecessarily risky... regarding achieving decent, stable, reliable paraboloids with no degrading  other aberrations manifesting. Applying forces to do that comes out nifty on the computer... but is near-impossible to reliably achieve in practice. I worked on that stuff at work, long ago... with interesting results. It's intriguing... but otherwise an exercise in windmill-tilting.

 

you see:

 

> astigmatism "potato chip" is easy to bend in --- takes very little force

> power  "pucker" is harder to bend in --- takes way more force

> spherical "correction" is VERY hard to bend in --- takes WAY more force

 

Consequently, when you try to do the third one, you will most certainly/inadvertently... introduce the other two (and many others) ... by random large amounts.

 

So, why not just polish/figure the mirrors to paraboloids, in the traditional way... and just not have to worry over it in the field?

 

The additional challenges of turning the OTAs into functioning binoculars will be plenty challenging enough to occupy your creativity, time and effort!    Tom

Attached Thumbnails

  • 155.1 don quixote tilting windmills.jpg

  • Dave O and retroman2 like this

#3 savarin

savarin

    Lift Off

  • -----
  • topic starter
  • Posts: 16
  • Joined: 16 Nov 2013

Posted 21 October 2019 - 02:56 AM

Hi Tom,

I've flexed an 8" f3.6 or 7 that was a commercial mirror but so far under corrected as to be nearly spherical.

It had a superb surface which we didnt want to destroy so I made a plaster tool with steel dumps and ground the back to the required RoC (I forget what it was) and it flexed into a an excellent optic pin point stars right out to the edge.

You do have a point with just carrying out the figuring in the normal way and I really am torn between the two, the flex method from my experience is just leading the way at the moment.


  • retroman2 likes this

#4 m. allan noah

m. allan noah

    Apollo

  • *****
  • Posts: 1144
  • Joined: 14 Aug 2009
  • Loc: Virginia, USA

Posted 21 October 2019 - 06:46 AM

The important part of your question is the order of operations, and how to produce the same ROC. I don't have nearly as much experience with glass as folks here, but I've done lots of repeat operations in metal and wood. I would guess that when you use mirror B as a tool, it will flex more than mirror A did when used as a tool, because it has had one side ground already, and it is not blocked in plaster. I think you will get different ROCs. I would use a third tool, and alternate mirrors frequently. I might even build a second turntable, so I could leave both mirrors mounted, and switch the tool back and forth between them.



#5 stargazer193857

stargazer193857

    Cosmos

  • *****
  • Posts: 7766
  • Joined: 01 Dec 2013
  • Loc: Southern Idaho

Posted 22 October 2019 - 09:29 PM

Why not just hog close and then use a fixed post on both with spherometer until they match?

 


Edited by stargazer193857, 22 October 2019 - 10:04 PM.

  • savarin likes this

#6 savarin

savarin

    Lift Off

  • -----
  • topic starter
  • Posts: 16
  • Joined: 16 Nov 2013

Posted 23 October 2019 - 01:19 AM

Why not just hog close and then use a fixed post on both with spherometer until they match?

 

I have been procrastinating and thinking and I think this is the best way to go.

I will post some pics once I start.

Too many years ago I had a simple program to work out the sag is there still one available?


  • Dave O likes this

#7 JohnH

JohnH

    Gemini

  • ****-
  • Posts: 3109
  • Joined: 04 Oct 2005
  • Loc: Squamish BC Moved!!!!!

Posted 23 October 2019 - 02:33 AM

My best idea to grind out to identical mirrors is to make two matching tile tools start grinding, but oh, and this is very important, start swapping out the tools and the mirror with the mirror pairs and keep doing so as you grind finer and finer. Obviously if you grind one too many times more than the other, the curves will no longer match she would have to pay very careful attention to your number Strokes, and the past the tool takes in relation to the Mirror Etc

#8 Dave O

Dave O

    Viking 1

  • *****
  • Posts: 623
  • Joined: 21 Dec 2011
  • Loc: Sri Lanka

Posted 23 October 2019 - 06:21 AM

I really think that you are "overthinking" this ...


  • John Lightholder and stargazer193857 like this

#9 JohnH

JohnH

    Gemini

  • ****-
  • Posts: 3109
  • Joined: 04 Oct 2005
  • Loc: Squamish BC Moved!!!!!

Posted 23 October 2019 - 07:31 AM

I had considered doing a project like this, even to the point of getting a few selections of pairs of blank so I could use that were identical, and even finding some very oversized roof prisons properly orient the images.


The method I consider to get to mirrors to the nearly exact focal length each was to grind each in specific manner and then with short intervals in between, swap the tools between the mirrors. With care and forethought, you could easily generate two mirrors that had the same focal length, especially if you checked periodically with a spherometer and adjusted your method as necessary.

Edited by JohnH, 23 October 2019 - 07:32 AM.


#10 MitchAlsup

MitchAlsup

    Skylab

  • -----
  • Posts: 4002
  • Joined: 31 Aug 2009

Posted 23 October 2019 - 07:11 PM

I did a pair of 12" blanks by using a single tool and a pair of blanks, alternating the blanks every other wet. When they were done grinding, the single tool would start sticking at 9 micron at about 2 minutes into the wet, each.


Edited by MitchAlsup, 23 October 2019 - 07:13 PM.

  • savarin and stargazer193857 like this

#11 stargazer193857

stargazer193857

    Cosmos

  • *****
  • Posts: 7766
  • Joined: 01 Dec 2013
  • Loc: Southern Idaho

Posted 23 October 2019 - 08:38 PM

I have been procrastinating and thinking and I think this is the best way to go.
I will post some pics once I start.
Too many years ago I had a simple program to work out the sag is there still one available?


Pathagorean theorem. 12" f6 is 144" radius on axis and 6" radius on mirror. 144-Sqrt(144*144-6*6) = 0.12505, 1/8".
The simple formula you refer to is an approximation for a square root that is accurate for slow scopes. I think what I did is simple enough.
  • savarin likes this

#12 savarin

savarin

    Lift Off

  • -----
  • topic starter
  • Posts: 16
  • Joined: 16 Nov 2013

Posted 24 October 2019 - 06:37 AM

sounds good to me, thanks



#13 savarin

savarin

    Lift Off

  • -----
  • topic starter
  • Posts: 16
  • Joined: 16 Nov 2013

Posted 24 October 2019 - 06:46 AM

The story so far.

Just finishing off two turnbuckles for the IPD adjustment.

The glass disk at the bottom was just for a reference mark.

bino1.jpg


  • Pierre Lemay and brave_ulysses like this

#14 stargazer193857

stargazer193857

    Cosmos

  • *****
  • Posts: 7766
  • Joined: 01 Dec 2013
  • Loc: Southern Idaho

Posted 24 October 2019 - 10:53 PM

sounds good to me, thanks

If you tell me how far apart the feet are on your spherometer, I can tell you what depth on it corresponds to 1/8" for the 12".

The triangle is cut into six 30-60-90 triangles, and the root 3 ratio is used to get the distance from the center to on leg. You can also just measure the distance as 2/3 the height of the triangle.

Say it is 2". Then 12/2=6, and 1/8" / 6*6 = 0.00347", three and a half thousandths of an inch. The reason for squaring the 6 is it is a parabola that the sphere approximated closely enough.

Edited by stargazer193857, 24 October 2019 - 10:59 PM.


#15 Pierre Lemay

Pierre Lemay

    Apollo

  • *****
  • Posts: 1368
  • Joined: 30 Jan 2008
  • Loc: MontrĂ©al, Canada

Posted 25 October 2019 - 05:20 AM

The story so far.

Just finishing off two turnbuckles for the IPD adjustment.

The glass disk at the bottom was just for a reference mark.

attachicon.gif bino1.jpg

Please show more detailed pictures of your build. Especially the IPD adjustment. Thanks. 



#16 savarin

savarin

    Lift Off

  • -----
  • topic starter
  • Posts: 16
  • Joined: 16 Nov 2013

Posted 25 October 2019 - 09:53 AM

Hi Pierre, the full story of the build can be found here

https://www.hobby-ma...inocular.55688/

along with the stuffups and solutions as they were found.

The IPD stuff is on page 4

The turn buckles are to take the place of the original straight lengths of ally I originally used.

The trouble is the supplier sent 4 right hand threaded ones instead of two left and two right hand ones so I'm awaiting their replacement.




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