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

Twyman effect has ruined my flat

This topic has been archived. This means that you cannot reply to this topic.
23 replies to this topic

#1 ccaissie

ccaissie

    Vanguard

  • *****
  • topic starter
  • Posts: 2,319
  • Joined: 13 Sep 2010

Posted 20 November 2016 - 11:03 AM

I cored a 12.5 flat with coarse grit, and now that I am comparing Foucault and autocollimation 10" f/10 for a Chief, I've determined that the coring raised the center of the flat.  It's not been quantified, but I suspect it's around 1/8 wave high, and extends from the 1.7" hole about 1/2".  I ignored it when making previous deep mirrors, but this time I have ZERO obstruction to hide any center defect.  

 

I recall a discussion that said this effect from coring can be alleviated by subsequent fine grinding or polishing the i.d.

 

Can someone point to articles on this?  Now would be perfect time to attempt and verify the remedy.



#2 DAVIDG

DAVIDG

    Voyager 1

  • *****
  • Posts: 10,426
  • Joined: 02 Dec 2004

Posted 20 November 2016 - 11:20 AM

 I have personally cored a number of flats and have never seen a problem so  I would double check that the defect is really in the flat and not the mirror. Double pass is very sensitive and zones that are missed in a Foucault test are easily visible in double pass. 

   If you have a smaller uncoated flat you can check the larger flat near  the hole even thou the larger flat is coated. If you take a piece of non-woven lens tissue and separate a single layer of fibers, then place it between the flats. The single layer of fibers will reduce the reflectivity of the coated flat enough so the fringes can be seen. 

 

               - Dave 



#3 ccaissie

ccaissie

    Vanguard

  • *****
  • topic starter
  • Posts: 2,319
  • Joined: 13 Sep 2010

Posted 20 November 2016 - 12:26 PM

Cool trick on the coated vs uncoated flat check. 



#4 Pinbout

Pinbout

    ISS

  • *****
  • Posts: 25,850
  • Joined: 22 Feb 2010

Posted 20 November 2016 - 02:08 PM

I check out coated flats all the time without the tissue

 

https://youtu.be/BG5tzwAxroc

 

 

Here's a coated /uncoated

 

https://youtu.be/O1QRjIQSFIA



#5 mark cowan

mark cowan

    Vendor (Veritas Optics)

  • *****
  • Vendors
  • Posts: 9,896
  • Joined: 03 Jun 2005

Posted 20 November 2016 - 05:13 PM

I cored a 12.5 flat with coarse grit, and now that I am comparing Foucault and autocollimation 10" f/10 for a Chief, I've determined that the coring raised the center of the flat.  It's not been quantified, but I suspect it's around 1/8 wave high, and extends from the 1.7" hole about 1/2".  I ignored it when making previous deep mirrors, but this time I have ZERO obstruction to hide any center defect.  

 

I recall a discussion that said this effect from coring can be alleviated by subsequent fine grinding or polishing the i.d.

 

Can someone point to articles on this?  Now would be perfect time to attempt and verify the remedy.

Just take the grind down to 500 or better if you want to reduce the induced error, or better, do that and then flash polish it.  The distortion is all mechanical in origin.  This assumes the flat hasn't got significant stress baked in though, as releasing that will be a one-way street for distortion.



#6 Eikonal

Eikonal

    Vostok 1

  • *****
  • Posts: 179
  • Joined: 11 May 2011

Posted 20 November 2016 - 09:30 PM

I'd say the same thing as Mark:

fine grind, and preferably polish the hole to relieve the stresses introduced by coring.


Edited by Eikonal, 20 November 2016 - 09:32 PM.


#7 ds-in-wa

ds-in-wa

    Explorer 1

  • -----
  • Posts: 63
  • Joined: 08 Feb 2012

Posted 24 November 2016 - 12:15 PM

I am working on three 8" flats (full thickness Pyrex) and plan to core one of them.  I had planned to core from the back to within 1/8" of the front after fine grinding, and to finish coring from the front after figuring.  However, the discussion above leads me to think that it would be better to finish polishing and figuring before I do any coring, and then to core and fine grind and polish the hole afterwards.  Comments?


Edited by ds-in-wa, 24 November 2016 - 12:17 PM.


#8 MKV

MKV

    Cosmos

  • *****
  • Posts: 9,691
  • Joined: 20 Jan 2011

Posted 24 November 2016 - 02:04 PM

I am working on three 8" flats (full thickness Pyrex) and plan to core one of them.  I had planned to core from the back to within 1/8" of the front after fine grinding, and to finish coring from the front after figuring.  However, the discussion above leads me to think that it would be better to finish polishing and figuring before I do any coring, and then to core and fine grind and polish the hole afterwards.  Comments?

Traditional wisdom and advice was to core from the back and stop 1/8 inch short of the opposite surface, then core form the front after the flat has been ground, polished and figured. I did mine after the fact and was lucky, but if I were to make a new set I would go with the traditional approach. It assures that hardly any stress will be released that hasn't already been released. In the past, also some would core all the way through and then replace the core and seal with beeswax. The problem I see with that method is that abrasive particles will inevitably get stuck into the beeswax and very likely cause serious scratches.  

 

Mladen

 

three flats_LR.jpg


Edited by MKV, 24 November 2016 - 02:06 PM.


#9 ds-in-wa

ds-in-wa

    Explorer 1

  • -----
  • Posts: 63
  • Joined: 08 Feb 2012

Posted 24 November 2016 - 07:44 PM

Thanks for the reply Mladen.  What you have described is pretty much what I was thinking before the above discussion on the Twyman effect.  However, if I understand the discussion correctly, the Twyman effect is caused by the introduction of stresses by coring with coarse grit, and that fine grinding and polishing the ground edges relieves these mechanical stresses.  You seem to be saying that coring with coarse grit releases stresses in the glass which is exactly the opposite of what Mark and Eikonal described above.

 

My conjecture is that the traditional approach you described, coring all but the front 1/8th inch and then finishing off the hole after figuring, relies on the stresses introduced by coring being stable, and since most of those stresses would be introduced before figuring and before coring the final 1/8th inch, the effect on the final figure of coring the final 1/8th inch would be minimal.  Also, chamfering the edge of the cored hole with fine grit would likely relieve much of the stress introduced by the last 1/8th inch of coring. 

 

I am not sure whether I'm making a mountain out of a mole-hill or not.  After all, David says he has cored a number of flats without a problem. Which leads me to think that doing all the coring after the flat is figured is not a terrible idea, particularly if a Twyman-effect raised edge is unlikely, and if it does show up, it could probably be remedied by fine grinding and polishing the sides of the hole.

 

Douglas



#10 brucesdad13

brucesdad13

    Apollo

  • *****
  • Posts: 1,034
  • Joined: 25 Jan 2015

Posted 24 November 2016 - 11:49 PM

You can't introduce new strain to the glass--it's baked in during insufficient annealing. I mean you can temporarily strain the glass with external mechanical force but it won't stick. That's my layman's understanding...

 

Edit: I envision it as an internal tug of war and when you remove some glass it's like taking a few people off one end of the rope... suddenly a balanced strain is imbalanced and everything shifts...


Edited by brucesdad13, 24 November 2016 - 11:51 PM.


#11 MKV

MKV

    Cosmos

  • *****
  • Posts: 9,691
  • Joined: 20 Jan 2011

Posted 25 November 2016 - 04:32 AM

...if I understand the discussion correctly, the Twyman effect is caused by the introduction of stresses by coring with coarse grit, and that fine grinding and polishing the ground edges relieves these mechanical stresses.  You seem to be saying that coring with coarse grit releases stresses in the glass which is exactly the opposite of what Mark and Eikonal described above.

Hi Douglas, it depends on how you look at it. The ground side definitely becomes weaker. The surface tension is released and cracks appear continuously. Polishing restores the surface tension.

 

https://www.ncbi.nlm...pubmed/21127579

 

If you take a polished glass plate and grind one side the polished ground side will tend to become convex, due to released surface tension (release of stress). The other sides experiences compression -- increased stress. 

 

Also, I don't recall specifying coarse grit. I used a fine diamond hole saw, lots of coolant and slow feed rate.

 

Malden


Edited by MKV, 25 November 2016 - 10:30 AM.


#12 mark cowan

mark cowan

    Vendor (Veritas Optics)

  • *****
  • Vendors
  • Posts: 9,896
  • Joined: 03 Jun 2005

Posted 25 November 2016 - 08:16 AM


The Twyman effect is caused by the introduction of stresses by coring with coarse grit, and that fine grinding and polishing the ground edges relieves these mechanical stresses.  You seem to be saying that coring with coarse grit releases stresses in the glass which is exactly the opposite of what Mark and Eikonal described above.

Technically, grinding a polished surface induces stress, it doesn't "release" it.  Conversely, polishing a ground surface produces strain of the opposite sign and direction.  So a piece of float glass that is ground on one side bends towards the polished side, as the surface area of the ground side is higher than the polished side, basically, but it won't quite fit into the same space anymore.  Polishing relieves this stress.  It has nothing to do with anneal.
 
See for instance, https://www.ncbi.nlm...pubmed/21127579
  
 

 

In the Twyman effect (1905), when one side of a thin plate with both sides polished is ground, the plate bends: The ground side becomes convex and is in a state of compressive residual stress, described in terms of force per unit length (Newtons per meter) induced by grinding, the stress (Newtons per square meter) induced by grinding, and the depth of the compressive layer (micrometers).

I see Mladen cited the same paper - but it's not due to surface tension - that's the result of a liquid and ionic bonds.
 

 


If you take a polished glass plate and grind one side the polished side will tend to become convex, due to released surface tension (release of stress).

 
 
Anytime you induce a crack (or pit) in the surface of a piece of glass it causes stress, because the VOLUME of the material after it cracks now has to include the volume of the air space as well - and so the new volume only fits into the old space under compressive tension.  Also, the polished side becomes concave, not convex...
 
 

 

My conjecture is that the traditional approach you described, coring all but the front 1/8th inch and then finishing off the hole after figuring, relies on the stresses introduced by coring being stable, and since most of those stresses would be introduced before figuring and before coring the final 1/8th inch, the effect on the final figure of coring the final 1/8th inch would be minimal.  Also, chamfering the edge of the cored hole with fine grit would likely relieve much of the stress introduced by the last 1/8th inch of coring.

Right, but if you THEN polished the entire core edge you'd remove that compensated strain and end up worse off.


Edited by mark cowan, 25 November 2016 - 08:30 AM.


#13 MKV

MKV

    Cosmos

  • *****
  • Posts: 9,691
  • Joined: 20 Jan 2011

Posted 25 November 2016 - 10:46 AM

I see Mladen cited the same paper - but it's not due to surface tension - that's the result of a liquid and ionic bonds.

Also, the polished side becomes concave, not convex...

Yes, you're right, that's what the link states. It was way past my bedtime.  I went back and made a correction this AM.

 

Another way of looking at it is that If something is under tension then tearing or cracking eases or releases the tension. The question is -- is the glass disk under tension or not and why? I think well annealed disks don't crack or warp after coring. Any change in the figure or cracks in the substrate as a result of coring to me indicate tension/stress. 

 

The way I see it, if something is under tension, then any force of cutting, drilling, sawing etc. will release the tension. Forcing the object back into its previous shape by sewing, gluing, screwing, or clamping a cracked object will increase tension again. Aren't crystalline materials under high tension by their very nature which makes them hard and dense but also brittle? 

 

Mladen 

 

Mladen



#14 mark cowan

mark cowan

    Vendor (Veritas Optics)

  • *****
  • Vendors
  • Posts: 9,896
  • Joined: 03 Jun 2005

Posted 25 November 2016 - 04:27 PM

The glass disk polished on both sides (and the edge!) and well annealed is NOT under tension.  It's what people have called "happy glass" - it responds most predictably to figuring.  Quartz (fused silica) has no internal strain due to how it's made but if you grind it or polish it after-the-face you change the overall shape, at least as far as fine figuring is concerned...  

 

Surface tension only occurs with liquids where the attraction of the bulk of the material exerts an unequal pull on surface atoms, so material under surface tension wants to minimize its surface area - hence a sphere.  Liquid glass has surface tension.  Annealing relieves this and any internal stress through bond breaking and relaxation during the time at temps below liquid and above the "freezing" point.

 

If you put a conch chip in a piece of glass (through sudden impact) it clearly shows strain, and attempting to grind out the chip will reveal exactly how far it went.  But once you grind it out and repolish it there is no strain left.  It can be done incorrectly with torches - here's a pic of the strain in a quartz blank that had a chip that was treated with a torch - it's a non-starter for mirrors:

 

strain.jpg


Edited by mark cowan, 25 November 2016 - 04:31 PM.


#15 ds-in-wa

ds-in-wa

    Explorer 1

  • -----
  • Posts: 63
  • Joined: 08 Feb 2012

Posted 25 November 2016 - 10:06 PM

 

My conjecture is that the traditional approach you described, coring all but the front 1/8th inch and then finishing off the hole after figuring, relies on the stresses introduced by coring being stable, and since most of those stresses would be introduced before figuring and before coring the final 1/8th inch, the effect on the final figure of coring the final 1/8th inch would be minimal.  Also, chamfering the edge of the cored hole with fine grit would likely relieve much of the stress introduced by the last 1/8th inch of coring.

Right, but if you THEN polished the entire core edge you'd remove that compensated strain and end up worse off.

 

 

Thank you Mark.  Your explanations helped me visualize how the Twyman effect stress is introduced, i.e., by the fracturing we've all seen in illustrations of how coarse grinding fractures the glass to a certain depth that must necessarily be ground out by successively finer grits. 

 

However, it seems to me that one key assumption behind the "traditional approach" is that the stresses introduced by coring with coarse grits are stable enough not to change over time or over the range of conditions they will be subjected to.  Although fractures do tend to propagate, I suspect that the assumption is reasonable.  ??? 



#16 mark cowan

mark cowan

    Vendor (Veritas Optics)

  • *****
  • Vendors
  • Posts: 9,896
  • Joined: 03 Jun 2005

Posted 25 November 2016 - 11:26 PM

Grinding produces shallow enough fractures - especially coring with a diamond bit - that this is true.



#17 MKV

MKV

    Cosmos

  • *****
  • Posts: 9,691
  • Joined: 20 Jan 2011

Posted 26 November 2016 - 05:13 AM

Thanks, Mark.



#18 steveastrouk

steveastrouk

    Apollo

  • *****
  • Vendors
  • Posts: 1,247
  • Joined: 01 Aug 2013

Posted 27 November 2016 - 11:22 PM

Grinding produces shallow enough fractures - especially coring with a diamond bit - that this is true.

Hi Mark,

 

What would abrasive waterjetting do ? I have a 16" optical window I want to bore a 2" hole instead of the 1/2" it has now.

 

Steve



#19 mark cowan

mark cowan

    Vendor (Veritas Optics)

  • *****
  • Vendors
  • Posts: 9,896
  • Joined: 03 Jun 2005

Posted 28 November 2016 - 12:06 AM

It's all the same SFAIK, it depends on the roughness of the resultant surface.  Waterjetting creates ridges along with the cut itself but I have no experience with what that does to the surface particularly.  Diamond coring is a lot smoother overall, when I've done that commercially I finish up the surface with fine diamond cloth glued to a sanding drum to smooth it down to quite fine finish.



#20 brucesdad13

brucesdad13

    Apollo

  • *****
  • Posts: 1,034
  • Joined: 25 Jan 2015

Posted 28 November 2016 - 02:20 AM

Mladen once mentioned using a ball trailer hitch to polish the opening.



#21 MKV

MKV

    Cosmos

  • *****
  • Posts: 9,691
  • Joined: 20 Jan 2011

Posted 28 November 2016 - 02:52 AM

Mladen once mentioned using a ball trailer hitch to polish the opening.

To chamfer the edges down to 5 microns, not the inside walls. The Delmarva class uses beer bottles for the edges.-- much  better and cheaper, imo. :o)

 

Mladen


Edited by MKV, 28 November 2016 - 02:54 AM.


#22 Pinbout

Pinbout

    ISS

  • *****
  • Posts: 25,850
  • Joined: 22 Feb 2010

Posted 28 November 2016 - 07:30 AM

And more fun...



#23 brucesdad13

brucesdad13

    Apollo

  • *****
  • Posts: 1,034
  • Joined: 25 Jan 2015

Posted 29 November 2016 - 12:43 AM

lol

 

 

Mladen once mentioned using a ball trailer hitch to polish the opening.

To chamfer the edges down to 5 microns, not the inside walls. The Delmarva class uses beer bottles for the edges.-- much  better and cheaper, imo. :o)

 

Mladen

 



#24 DAVIDG

DAVIDG

    Voyager 1

  • *****
  • Posts: 10,426
  • Joined: 02 Dec 2004

Posted 29 November 2016 - 09:44 AM

 

lol

 

 

Mladen once mentioned using a ball trailer hitch to polish the opening.

To chamfer the edges down to 5 microns, not the inside walls. The Delmarva class uses beer bottles for the edges.-- much  better and cheaper, imo. :o)

 

Mladen

 

 

Here is a  link to  John Downie's Gallery showing him  chamfering his cass. blank at the Delmarva class using a glass bottle  http://www.cloudynig...242-dmchamfers/

 

                  - Dave 




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