CNC for optics

Hi everyone, I'm new to this forum. I am based in the Netherlands and have a broad interest in CNC, mainly for the manufacture of optical components. In some forums it is customary to introduce yourself, I'm not sure if that is the case here. I guess the best way to introduce myself is just to refer to a recent video on a Loh CNC spherical radius grinder that I bought a few years ago on an auction and restored in its original state:

YouTube

Looking forward to reading all the interesting stuff on this forum!

@JeromeV That's a fantastic video you made (THANKS for making that !).

Holy smokes... A LOH sphereomatic CNC lens grinding machine , perfect condition picked up at auction for € 750 ???? (Seems the Dutch DOD takes good care of stuff ;-) ).

Less than a $1000 , Isn't that machine like at least $500K new ?

The government auctions I have been to (in the USA when I worked for 'gov ) rarely have intact / near pristine equipment like that; that's the find of century that you have there. At US gov auctions You'll find an electron microscope or something but its packed in crates in straw in 50 pieces.

I'm trying to get more into jig grinding (related) and there are some pretty amazing set ups that are possible with the Moore spin-table on a micro sin plate; I believe there is a set up where one can do aspherical / parabolic elements.

We are not doing glass but need to devise better surfaces in hard materials. So we use different autocollimators for optical alignments (even have an old Hilger watts, Davidson and others ). Really interesting how the high tolerance stuff goes hand in hand with the optical stuff. I'm messing around with more unusual spherical bearings for our applications, so spherical grinding is in general is a pretty miraculous process (in of itself, how it finds it's own surfaces (with a bit of luck and skill and extreme patience...) ).


[Neat trick you made for centering using the microscope on the edge of the optical blank !].

Anything grinding + CNC is hellatiously expensive so @Jerome you really lucked out man (Amazing !).

Some of the CNC's jig grinders like a Moore G18 that has been CNC'd can be more of a PITA than not "CNC'd" as your set ups can become much less flexible and the savings on 'Time" are questionable, and the new CNC jig grinders are cray cray expensive but the CNC mechatronics are generally not that more accurate than a GOOD Japanese CNC vertical mill. So @JermoeV your hydraulic actuation seems amazingly precise on the LOH (1990's machine.).

We are still muddling through with stuff.

Glad you decided to join this forum.

The abrasives section of the forum has some very knowledgeable folks that hang-out there too.

@JermoeV (aka ; "Huygens*" Optics ) do you mainly do mirrors or antique reproductions or do you get your lenses coated somewhere ? Or do you do that yourself ? Or IR imaging etc. ? [I have some amazingly accurate photogrammetric cameras with optics built by WILD (Heerbrugg) ~ Switzerland, when they still did that kind of thing) that use uncoated optics but can also image in the near infra red. Spherical lenses / Biogon type symmetric layout with 16 micron total radial distortion (pretty sweet, metrology grade + beautiful imagery for a very wide angle lens (system)/lens "cone". ). ].

____________________________________________________________________________________________________________


[Just as aside I grew up on the edge of London not far from William Hershel's home and museum and was always very inspired by his telescope building and metal mirrors that he made (literally in his back yard) as well as discovering the planet "George"** ;-) (Eventually to be called by the far more sensible name Uranus.). .].

** After King George the III (Georgium Sidus ~ Planet George .).

Hi Cameraman,

Thanks for your response and advice. I will check out the grinding section. I guess you are right, I had a bit of luck there. Somebody at the government warned me that an auction was upcoming. Unfortunately he did not know when. So in the end, I found out only minutes before it was closing, even had to retrieve a new password 3 minutes prior in order to make a bid (it was an online auction), so it was kind of a close call. But actually I was the only one that bid on that machine. And btw on another Loh machine too, a centering machine (see picture of a similar one, mine was missing the laser and measurement unit). That one was actually only 180 Euro's so under 200 dollars. I don't think a lot of people know what to do with them.



I don't know about the new price of the Spheromatic. I guess it has been expensive but I don't think 500k. Of course it came without any of the tooling. Actually 2 of the special diamond mills cost as much as the Loh Spheromatic.

I must admit I do not use it a lot. Manufacture of optics is a bulk industry and there is so much surplus on the market, that it is seldomly viable to manufacture a lens. Especially given the amount of time you have to spend on making it. If you look at Suplus Shed for example, you can get a variety of lenses for under 5 dollars, anti-reflection coated and all...

To answer your question, I did manufacture large area telescopic mirrors in the past (up to 500mm) and I used a home made diamond radius mill for that. With these mirrors you have to remove 2kg of glass or more to get them in the right curvature before you can start grinding. Actually I stopped that activity about 4 years ago. Now I mainly do CAD design, the optics is mainly a hobby now. Cheers!

JeromeV said:

Hi Cameraman,

Thanks for your response and advice. I will check out the grinding section. I guess you are right, I had a bit of luck there. Somebody at the government warned me that an auction was upcoming. Unfortunately he did not know when. So in the end, I found out only minutes before it was closing, even had to retrieve a new password 3 minutes prior in order to make a bid (it was an online auction), so it was kind of a close call. But actually I was the only one that bid on that machine. And btw on another Loh machine too, a centering machine (see picture of a similar one, mine was missing the laser and measurement unit). That one was actually only 180 Euro's so under 200 dollars. I don't think a lot of people know what to do with them.
View attachment 257818


I don't know about the new price of the Spheromatic. I guess it has been expensive but I don't think 500k. Of course it came without any of the tooling. Actually 2 of the special diamond mills cost as much as the Loh Spheromatic.

I must admit I do not use it a lot. Manufacture of optics is a bulk industry and there is so much surplus on the market, that it is seldomly viable to manufacture a lens. Especially given the amount of time you have to spend on making it. If you look at Suplus Shed for example, you can get a variety of lenses for under 5 dollars, anti-reflection coated and all...

To answer your question, I did manufacture large area telescopic mirrors in the past (up to 500mm) and I used a home made diamond radius mill for that. With these mirrors you have to remove 2kg of glass or more to get them in the right curvature before you can start grinding. Actually I stopped that activity about 4 years ago. Now I mainly do CAD design, the optics is mainly a hobby now. Cheers!

Click to expand...

That's super cool !

OMG a centering machine for 180 Euro (face palm).


I used to have a good friend in the Netherlands (when I lived in the UK) that I used to buy a lot of X- military / defense stuff from , mainly surveying equipment and some very rare and unusual aerial reconnaissance stuff and various older metrology stuff for pretty decent price (HIGH) nothing like the steals you got there amazing.


1/2 meter telescope mirror that's HUGE ! Wow, a real labor of love.

____________________________________________________________________________________

You have a number of different cups for your machine ? (or you have to make a few ?).

I think the hydraulic aspect of your machine is also for very fine controlled movement, I'm guessing that massive forces are not really required ? (shrugging shoulders.).


BTW I think your video is really excellent very didactic … I might comment there under a different handle / follow / subscribe . Look forward to your new videos !

________________________________________


Without giving too much away we tend to work with off the shelf optics / lenses (Mainly Schneider digitar and Linhoff some Zeiss but we completely take them apart and rebuild , integrate and then calibrate the hell out of them for principal distance, radial and tangential distortion and some other stuff I can't talk about. To commission a similar performing lense assembly cost nearly $200K just for a pair of lenses... So we have figured out how make things better and more affordable for our applications that (we integrate into a larger system ) and ultimately our customers. I.e. So we can afford to do what we do so our broader customers (eventually) can do the same appreciate the same price / performance / quality (when we formally launch / roll out.).

@JeromeV - great video!

Question - OK, that machine generates spherical lenses. What does the machine that generates aspherical lenses look like? (A pointer to some web site or book would be great.)

I believe you can make an aspherical lens with that machine (although can't say for sure, my optics experience is on the semiconductor and fiber side). The same tooling is used, the only difference is the lens radius of curvature isn't constant. Edmunds Optics has a video on it, you can even see the same tooling:

Yesterday I tried answering the above questions but somehow technology made my post disappear. I'll try again.

@Cameraman: I bought a few milling cups from a company here in the Netherlands. Unfortunately they went out of business. I have a large collection of diamond drills. For small radius of curvature on small objects, you can generally also use a glass diamond drill. I made a adapter so that the machine also accepts these drills. By the way, we still have these military surplus auctions in the Netherlands, sometimes you can still buy nice equipment for very low prices.

@brian_machine: The difference in surface shape between spherical and a-spherical optics is generally not that large, a few microns. (unless of course you have an extremely a-spherical surface). A few microns can easily be polished away using the traditional polishing agents. So you first make a spherical surface, including the polishing stage. Then, after the spherical surface is completely finished, you will make the correction. This can be done by deterministic polishing (small tool, local area see Pariels video), or by using specific strokes and larger tools where the amount of contact time between tool and glass surface is varied. The problem with small tools is that they can easily lead to higher surface roughness (> 50 nanometers), which is undesirable in optics. Gordon Waite's channel on Youtube has a few very nice video's that illustrate how that works with large mirrors. He demonstrates this by hand, but of course you can also use a CNC machine for this. I use a combination. Most of the time required for producing aspherical optics goes into the iterative process of measuring the properties and correcting the surface based on the measurements.

Great video Jerome, I look forward to more!

I tried to answer some of the questions above a few days ago. However for some reason my answers need to pass moderation first. This post is to test if the moderation requirements has been lifted.

So I tried again and failed there is something in my reply that makes the automated forum system suspicious. And the moderators don't lift this suspicion. I'm afraid I'm not able to post anything substantial at this moment.

Someone else here that really enjoyed your video. Well made, excellent presentation of a complex process. Looking forward to many more of your videos.

Thanks Jerome. Great video. I particularly like the method to fix and align the lens. Not to derail your thread with a question about manual polishing of glass, but do you (or anyone else) have any suggestions about simple equipment to regrind/polish flat glass? We sometimes get parts from our manufacturer that need rework (looks like pits under a microscope) and its a pain to send one or two of them back. I thought about using grinding equipment like you do for metallographic preparation. I don't want to have a room full of grinding pads. I guess that using only one of these machines with different grades would be difficult with contamination as you change for finer grits. Perhaps the following could work? https://www.amazon.com/gp/product/B07BDP5K61/ref=ox_sc_saved_title_6?smid=A3KE0122UDBPIY&psc=1

JeromeV said:

I tried to answer some of the questions above a few days ago. However for some reason my answers need to pass moderation first. This post is to test if the moderation requirements has been lifted.

Click to expand...

Certain types of links get jammed … I believe there is a filter that blocks certain types of strings in a URL ?

No worries will probably sort itself out in a few days …

andehe said:

Thanks Jerome. Great video. I particularly like the method to fix and align the lens. Not to derail your thread with a question about manual polishing of glass, but do you (or anyone else) have any suggestions about simple equipment to regrind/polish flat glass? We sometimes get parts from our manufacturer that need rework (looks like pits under a microscope) and its a pain to send one or two of them back. I thought about using grinding equipment like you do for metallographic preparation. I don't want to have a room full of grinding pads. I guess that using only one of these machines with different grades would be difficult with contamination as you change for finer grits. Perhaps the following could work? https://www.amazon.com/gp/product/B07BDP5K61/ref=ox_sc_saved_title_6?smid=A3KE0122UDBPIY&psc=1

Click to expand...

Is there a particular optical flatness that you have to adhere to ? IS it an optically active surface ? I assume uncoated but do you need to coat afterwards ?

Out of interest are you allowed to say what diameter and or application ? (Type of glass ?).

____________________________________________________________________________________________________

Overall I have to admit I'm wondering about (in general) how well CNC machines (for machining) "Mix" with an environment of grinding and polishing and lapping type machines and equipment. For a smaller shop is that relatively controllable OR after a critical messy tipping point with grinding machines do you eventually knacker your newer comparatively shinny CNC machines (i.e. for machining/ milling/ turning (not CNC grinding) ) ? I don't have enough experience of that as it seems for larger operations all of that is normally kept very separate ? How easy is that all to control (how much of a threat is such "contamination" ? Does one need to be ultra strict with parts washing between processes (I'm thinking more metals here process wise going in the "wrong " direction back to cnc)). [It's not just a "Cultural" divide between grinding/ grind hand / tool maker (mind set) vs. CNC lol ;-) ].

Some of the smaller / specialized knife makers manage to make it work (it seems).

__________________________________________


* metallographic preparation Stuff we used (like a million years ago) with strips of emery paper of various progressive fineness (from left to right), the metallographic samples set in resin (epoxy or polyester ) and then hand rub in each lane and then turn 90 degrees move to the next progressively finer lane. The rig is set up on a slant so water goes down hill and is collected... Some filtration (not like using emery paper on a flat old (granite) inspection plate). Final samples clean with acetone, then hit with "Nital" for ferrous stuff (to etch the surface and reveal grain structure) and then to microscope. The grits used in metallographic "Prep" are probabaly not fine enough ((@andehe to fix your glass components) so some sort of optical compounds and diamond based slurry will be needed... Wondering if some kind of air tool for hand polishing might be more the ticket if you don't need an "Optical" flat ? 1/2 arsed orbital polishing by hand - quick and dirty ? I'm sure JeromeV has some good tips for this... As soon as he's allowed to post properly.

andehe: As Cameraman mentioned it really depends on the application of your workpieces. Basically you have two kinds of polising: 1) Cosmetic and 2) Optical. With the cosmetic polishing it is just about getting rid of nasty pits and scratches or a damaged coating. Since you don't really care about the optical properties of the surface it is just about having it look OK. For this purpose, the amazon product you refer to could be a solution. By the way, this particular product is only suitable for very small objects!

If the objects you refer to are part of an optical system, it is quite unsuitable. You need to do pitch polishing which generally requires a lot of experience. The required surface accuracy is in the order of 100 nanometer or smaller. If a surface of a lens deviates more than this value from the target shape than it will negatively influence the optical performance. 100nm material can be removed within seconds using polishing when some pressure is applied.The tool that you referred to on Amazon will also generally not yield a flat surface but a silghtly convex one, with "dropped down" edges of several microns.

I actually build a CNC-polisher for large optics a few years ago and am planning to post a video about it in the near furure. If you are interested, I can post a link once it is published.

So yesterday evening I finally got a message from the moderator saying that the NSA has approved my post ;-). Apparently it has to do with certain characters or sometimes links. I found out that it is very likely the "at"-symbol that prohibited by posts from being posted. Cameraman and brian_machine, please see post Nr 7 of this thread for answers to your previous questions.

andehe said:

We sometimes get parts from our manufacturer that need rework (looks like pits under a microscope) and its a pain to send one or two of them back. I thought about using grinding equipment like you do for metallographic preparation.

Click to expand...

That Amazon product doesn't allow for rotation of the workpiece, which means you either stand there and do it yourself, or get uneven material removal.

Why is returning items to the vendor a pain? I'd say they either have to send extra parts so you're covered if some are bad, or better still, they need to improve their QC so they're not sending out bad products. I can't fathom thinking it's better to reprocess their bad work at your company's time and cost.

JeromeV said:

<snip>

I actually build a CNC-polisher for large optics a few years ago and am planning to post a video about it in the near furure. If you are interested, I can post a link once it is published.

Click to expand...

Definitely be interested in that (big time)^^^ .

Good to know about the military surplus auctions in the Netherlands, I should check that out. Probably need boots on the ground from the U.K. and maybe ship lighter stuff to the USA.

Funny the Dutch military bought a lot of British built optical / metrology equipment (that is still really good today), but the British companies don't exist anymore, same with some of the Swiss equipment too. I don't know if that was a general NATO "Thing" / deal.


____________________________________________________________________________________________________________________


* I used to live about 4 miles away from NATO head quarters [In the U.K. ](HMS Warrior) so there used to be a lot of a mix of US and Dutch and UK service personnel at all ranks in Northwood (NW edge of London), and then the NATO "Tornado" swing wing fighter bomber elsewhere creating more of a European "Mix" + Italians and some German. We figured in the 70's and 80's if the Russians decided to launch ICBM's we'd be vapourized in seconds anyway being so close to NATO HQ. You forget as a kid the question of "What are you going to do in your last 5 minutes ?" was not an uncommon one; (5 minute advanced warning of a nuclear attack .). Probably explains why putting super glue in your hair and listening the Sex-Pistols seemed like a sensible option at the time (considering). Meanwhile highly decorated friends/ coronels etc. in the military would be several miles down in an underground complex at HMS warrior devising the means to wage nuclear combat.

Milland said:

That Amazon product doesn't allow for rotation of the workpiece, which means you either stand there and do it yourself, or get uneven material removal.

Why is returning items to the vendor a pain? I'd say they either have to send extra parts so you're covered if some are bad, or better still, they need to improve their QC so they're not sending out bad products. I can't fathom thinking it's better to reprocess their bad work at your company's time and cost.

Click to expand...

I agree that "Thing" is not even worth thinking about...


Lapping Polishing Machines Compounds and Consumable Products | Lapmaster Wolters

^^^ These three ring "Desk top" lapping machines are super interesting how they achieve the results they do (at least for metal components.).

I was wondering if there was some other process that was causing pitting for Andehe ?

Or if the glass is an "Observation window" for something ?

And for today's small weird world moment - I looked at Gordon Waite's youtube channel and yup, it's the guy I worked for, and with, in various ways, in the early 1980s....
(Long out of touch - I'm really bad at keeping in touch with people. But he was always a really smart guy and into telescopes, so his youtube topic is not a surprize.)