Totally counterintuitive. I've done a fair bit of metalwork sanding and polishing, so I naturally assumed that optical polishing would be similar, only using something harder and flatter to grind the surface. It never occurred to me that the rotational grinding process would use something ductile, yet get better results. Thank you very much for the excellent explanation.
In metalworking terms, it would be very similar to using an aluminum, copper, or tin lap (as in watchmaker's "black polishing"). You want your lap to be softer than the material to be cut. Your lap becomes a matrix to hold the abrasive particles in place, and the cutting happens on the material that can't just grab and hold the abrasive. (Tin, by the way, gives absolutely amazing results when polishing steel. It's just _really_ stringy to machine when you're initially making the lap. Save it for your finest - sub-micron - grits.)
Very informative yet old video on lapping metal parts by rotation. Check this: ua-cam.com/video/fnoVV-RWIWY/v-deo.html It vey clearly explains "how it's made".
I feel there’s a general misconception when it comes to lapping a polishing that is a result of focusing on the lap material. As the first comment responder noted the lap only hold the cutting media. There are three fundamental rules for cutting to occur, though I only usually remember two. The important one here is that the cutting “tool” (in this case lapping or polishing compound) MUST be harder that the workpiece. The second is that there must be relative motion. The third escapes me. But in either case it is not the lap that cuts the work piece but the embedded abrasive. Apologies for the lecture comment but lapping and polishing seem to be no different fundamentally than any other metal removal process; hard removes soft. In the case of polishing silverware presumably there is some residual polishing compound on the cloth that is the effective mechanism for removing the oxide layer. Also, if I’m glaringly wrong please correct me. 👍 Edited for autocorrect errors.
Flat-out the best vid on making flat optical surfaces! Clearly thought out well, I couldn't pitch in any criticism. It's almost like we're on the same wavelength.
What a cheerfully bright comment! I for one found the video very illuminating, and it seemed to polish out all the rough spots in my dull and hazy knowledge. You could say that it expanded my bandwidth....
The true test of whether someone has mastered a field is their ability to explain it to the uninitiated in a clear and concise manner, you sir are a credit to the field.
How did I end up watching 12min of "how to polish something to the nm scale?" But you explained it so well that I understood it without any previous experience in polishing glass or anything for that matter
This is really good, so well detailed. It seems over the years there has been a lot of demonstrations of optical grinding/lapping, lots of "recipes" so to speak, without any detailed explanation as to why it works. I'm coming out of this with a deeper understanding of the process, only took 12 years, but better late than never!
I have been interested in optical engineering for a long time now and yours is the first channel I have seen that covers it well. So thank you for making such excellent videos.
I work in the optics industry for a couple of years now and I have to say that your channel is a real treasure! Your videos are both, highly educational and entertaining. Keep up the outstanding work!
The principle looks simple, its just a tar and turntable, but those who tried working with glass, especially polishing and making it precise, knows that its extremely hard, takes years of practice and patience. Great video.
I used a machine like this in the 90s to flatten hydraulic motor parts. Instead of pitch, the wheel surface was steel, and the surface was kept flat by adjusting three rings which also kept the parts in position on the wheel. We'd check it a couple times a day by washing the abrasive off and placing an optical flat on the wheel. But other than that it worked the same way.
@Andrew Crews I worked on Eaton and Sauer-Sundstrand axial piston pumps and motors. We would replace pistons and cylinder blocks and send them out to be resleeved/refinished.
I appreciate this comment very much after having done very similar work in the 90s. I used to test lapped and partially polished parts with an optical flat to infer the concavity of the part (convex or concave). Once the concavity was determined, I inferred the lapping wheel's (opposite) concavity and would adjust the rings' position to correct the wheel.
This is the most satisfying and informative presentation I've seen in ages. As a photographer I really appreciate the craftsmanship required to produce high quality optics. My hat off to you Sir! 🎩
If you think that is impressive the Laser Interferometer Gravitational-Wave Observatory uses interferometry to detect changes in length less than a 10 thousandth of the diameter of a proton.
I think there is a tradeoff. It would certainly help, but investing in climate control may not be worth it. Also, it seems heating the plate and then weighing it down removes enough deformity in a small amount of time. These techniques are fascinating.
I have been polishing precision optics for 12 years, I really enjoyed the video, thank you. I do the final polishing of the optics on a spindle in a zerodur plate with holes, I put planes with weights in them, according to a similar principle. pitch polishing pad for the night I turn over on a plate smeared with Regipol with good flatness
I've always thought there was something almost magical about being able to make incredibly accurate optical surfaces, flat or otherwise, with no precision tools whatsoever. I ground my 6" parabolic mirror by hand, and figured it to 1/20 wave accuracy on a pitch lap, with no references other than the Foucault test. Very nicely done. Subscribed. cheers from sunny Vienna, Scott
@@robertmccabe8632 Indeed. I use this principle to keep my sharpening stones flat. Using silicon carbide abrasive, I grind A against B, B against C, and C against A.
Very clear and detailed explanation, best I found so far (and no irritating background music). Cool how you built the turntable from a washing machine motor and rollerblade wheels!
Thanks for the video, it's very informative. So the polishing table is an aluminum disk with a motor stator attached to it, and you press the pitch lap by using a granite plate, but I was wondering if you could share what the pitch lap substrate is made of? It looks like a few inches thick disk. Is it also granite?
No actually in this particular case it is borosilicate, which has a thermal expansion coefficient that is about 3 times lower than granite (which is an advantage). You can however use granite without problems if you have good temperature control.
At work here in Germany they got mad at me for filing like this....that you work in nm tolerances and do so as well made me really happy. The understanding of why one would do it so they just couldn’t understand. Also an amazing video, very informative. Keep up the good work!
Excellent explanation! I’ve been aware of over-arm polishers for lenses, but not aware of this method. You openly share a lot of which others would consider trade secrets. Even so, I have to believe that if I were to set up a similar polishing system, using everything you show here, I probably wouldn’t get results as good. Thank you!
Do you always use the same grit to polish? If not, do the grits find themselves trapped in the pitch? Do you have to use different pitch plates for each grit?
Polishing is the final stage of producing an optical quality surface and there is only one compound used on a given lap. Grinding the surface to a desired rudimentary curve is called grinding, not lapping. Pitch laps do not grind but only polish. The amount of material you'd remove by polishing is miniscule and nobody wants to waste time doing any more polishing than necessary. Grinding a surface starts as a rough cut to quickly give a desired curvature and to remove casting irregularities. Then a succession of finer grits is used to remove the deep scratches left by the previous grit. This is done with a hard tool on the workpiece and an abrasive slurry is run onto the surface. It looks a lot like lapping with a pitch lap, but because the tool is hard, it can be reliably cleaned off when it is time to change to the next finer grit size.
This is very interesting. I read somewhere that for lapping of silicium wafers they use some chemicals instead of abrasive particles. I can't imagine how precise they must polish wafers for 4nm technology.
It's generally a combination of both. If you use a chemical that etches your surface while you are polishing, you can use a very mild (or soft) abrasive agent, which results in a smoother result. By the way, when you use Cerium Oxide to polish glass, chemical interaction also helps speed up the polishing process. For wafers the actual flatness is less important that the smoothness, since modern wafer steppers make a heigh map of the wafer to correct for the total thickness variation when clamping a wafer to the chuck. Modern technology wafers are indeed incredibly smooth and flat (from the dimension of individual components to that of the full chip)
This video reminds me of how profound our technological advancement has been. Just think of how many different people had to cooperate and dedicate basically their entire lives to engineering better solutions in the relatively niche field of precision optics. Using a high viscosity fluid as a lapping surface? How the hell did anybody come up with that? Stuff like this just blows my mind.
No idea why this is in my recommended, but very interesting. I'm amazed this kind of thing can be DIY with the proper knowledge and materials. I would consider myself a maker more in the software area, it's always cool to see what other people are working on. Keep making!
Excellent. Very happy I found your channel. Honestly, as a nerd in training, a good friday night for me includes such terms as nanometer and interferometry. Looking forward to your next video! Edit: trainings -> training
When the pitch drapes off the sides, how do you get it back up on top? Do all the different grade abrasives get mixed into the tar? Do you then have to replace the tar? How do you setup for lens grinding as opposed to flats? How are multifocal optical lenses ground? eg for spectacled?
freeze then chip it off and melt and recast. the polishing compound is micon sized...not grit. used to use rouge, but there are better compounds not so messy.
Does this type of pitch polishing give you any assurances of the parts being square or parallel? I imagine you might often want an object with two opposite faces both very flat and very parallel to each other, but my totally uninformed intuition says this process might not be able to give you that?
That is correct. To create parallel surfaces, generally two-sided grinders and polishers are used that have two abrasive disks spinning in opposite direction while applying pressure.
The algorithm sends me to strange places sometimes, but it sure can be a fascinating journey. Today I learned something I did not know yesterday, thanks.
Last minute had the information I didn't know I needed. Blocking pitch does put pressure on glass held to a backing. All the old film of spectacles and camera lenses show blocking, but these are not expected to be accurate to fractions of a wavelength. OK now I will also avoid lots of messy cleaning up as well.
Very good stuff. I've done my share of glass pushing, long long hours of manual work. I tried to make 200mm flats but I constantly got into troubles in 1um (two rings) level... Perhaps I return to them sometimes, even though I hardly remember why I started making them LOL (ok it was some cassegrain telescope idea, and another for testing other flats)
When you go through the grits of silicon carbide or cerium oxide, dou you change pitch plates or clean it very well? How do you make sure some coarse grit doesnt ruin the later polishing steps?
The silicon carbide grinding is not done against pitch but against other glass or granite plates. You have to carefully clean and rinse between grains (not only the plates, the work space, but also your hands) in order to avoid contamination.
Sure, that is the way to store them over longer times. It also avoids the evaporation of the more volatile components from the pitch so it keeps the right viscosity. However, even then you have to put some effort in bringing them back in a usable state, because, they will loose shape on the sub-micron scale.
Like the other people have said, really great video and clear explanation. One question: What's the relation between the maximum size of flat you can grind vs. the minimum size of the large pitch plate below? I mean, how small can you make this machine vs. how big of a flat can then be produced. I am wondering about grinding flat mirrors for star diagonals for telescopes, not sure how big yet but maybe up to 90mm (?).
If you have a round mirror the size can be up to 35-40 percent of the pitch ring. However it is more difficult for rectangular objects like star diagonals. You need to contain them in a ring. For a 90mm diagonal you need at least a 350-400 mm pitch ring. Also you cannot attach this type of object at the angled surfaces to the holder, but only at surfaces perpendicular to the pitch surface. For example using hot glue. If you look carefully at the connection of the elliptical mirror at 6:58 in the video you see that it is connected at the short side with two dots of hot glue (on each side) to the aluminum holder.
@@HuygensOptics Could you just create the flat in whatever shape is most convenient for the machine then water jet cut it out or something like that to get your desired shape (ellipse, rectangle, etc.)?
It's possible, but it depends on the flatness you require. Generally the cutting will introduce stress/cracks in the material at the cutting interface, which can seriously deform the objects in the um scale range. If you require high quality optics it's not an option. Also the abbrasives in water jet cutting can very easily damage your optical surfaces, even with protective layers applied.
@@HuygensOptics hm... maybe you could use the larger blank of 'convenient shape' and cut out the desired part(s) about 99% of the way with a small diamond grinding bit of some kind, leaving a holding ring or maybe tabs, then do the flatness grinding/polishing, then just snap the parts out by breaking the thin retaining glass. Or maybe try EDM using one of the techniques for non-conductive materials. But, yeah, at that point there's probably not really an overall cost/effort savings. :)
That is a trick used with for example with the central hole in telescope mirrors: you drill almost completely through except fort the last 2mm or so. My personal experience with this method is not positive. Generally, if there is residual stress in the material, it will be partly released when cutting parts out, deforming the piece. I always bring the object in its final shape and then start grinding and polishing and focus my efforts on a good mounting method and the correct tool shape.
The closest thing I do is knife sharpening, but I always wondered about stuff like straightness, flatness, and smoothness. This videos was very interesting.
So-de-knetter... té gek dit! Leppen van metalen onderdelen doe ik met veel liefde. Leppen met pek is iets wat ik nog nooit gezien heb. Dus: fantastisch om nu met je neus zowat óp de lepschijf te zitten. Met commentaar van een kenner. Hartelijk dank!! Ik kijk erg uit naar de video over de vlakheidsmeting! Groeten - Nobby Assmann
10:00 So not necessarily, and not usually to that degree. Yes, surface plates do have stricter requirements for local flatness vs overall flatness. For instance a grade AA surface plate of 18x18 in, has an overall flatness requrement of 50uin, but a local flatness requirement of only 35uin, so that would definitely need to be checked for whatever specific surface plate you're using.
If pitch is a liquid, couldn't flatness be achieved without a reference flat (like the bruiser or surface plate)? Enclose the pitch in a container with walls higher than it and heat it up enough to reduce the viscosity so it flows fast enough for the surface to find its level in a reasonable amount of time, then let it cool down. Surface tension would mess with the flatness near the walls, but closer in should be fine. I suppose you lose the grooves though, darn.
You will have to wait for a long time before the surface reaches submicron flatness. Keep in mind that there is a volume contraction during cooling down. And of course it is not just about getting the tool flat, but keep it flat during processing.
good to see that physics works all engineers experience the same when going down to nanometers no matter what kind of engineering they are doing and that is that solid material is moving ( slowly )
Great video. Very very well explained! Brings back memories. I used to make achromats from blanks of optical class at home in high school for telescopes.
Love the videos! I have a couple of questions: - what pitch specifically do you use in Jerome’s Pitch Polisher? - when setting up the pitch polisher, how do you ensure identical angular rotation of object and pitch lap? Are dimensions important or does the design itself ‘handle’ it? Thanks!
10:07 - Well, in that image, the variation is actually 1μm over 200 mm (because it's between the centre and the edge), not 400, but I guess it won't always be such a continuous curve.
I'm an optical fiber telecommunications technician. We used to have to polish the end faces of our connectors when terminating them. 3 different ratings of polish paper and polishing in a " figure 8" motion. It was tedious. Faster speeds and the need for lower reflection at the connections has us using fusion splicing and factory terminated connections, now. No one misses " puck and polish" terminations.
i am a machinist by trade and when i was initially linked this video, and i only saw the title, i thought "oh great some charlatan is going to talk about taping sandpaper to a piece of float glass" but i was pleasantly surprised to see that was not the case and this is the real deal. subscribed. have you considered trying a cast iron master surface for the weight? as long as the room is at least loosely climate controlled, you could theoretically have a more economical option for a even flatter surface than the granite plate
In principle cast iron would work just as well as granite. The CTE is around 10-11 ppm, so compared with granite (8ppm) this is only slightly higher. And because of it's higher thermal conductivy, it will get to thermal equilibrium (and retain it's original shape) quicker than granite. So it would in fact be a better choice as a material. However, don't know how easy it is to make a cast iron disk of 400mm in diameter with a flatness of approx 1um. By the way, granite is not an expensive material.
@@HuygensOptics i know that in metrology cast iron is preferred for master surfaces, because it is more dimensionally stable and easier to make very (beyond AA grade) flat. for your specific use, it would also help as cast iron is denser so it would be heavier for the same size. that is why it came to mind.
@@SuperAWaC Actually in some super accurate machines, aluminium with internal cooling channels can be preferred too. This is because aluminium has way higher thermal conductivity (about 3 times higher than iron) and can be stabilized better, down to 0.1°C. Of course if you deal with big loads and need very high stiffness the cast iron or granite are preferred, but with low loads, aluminium counter intuitively can be even better. I can't find source and machines doing that right now, but I do remember seeing some year ago.
I am a Tool & Die machinist and horologist and it made me happy to see there are other intelligent machinists watching things like this. I would second the use of a cast iron master plate. There is a guy who comes once a year to certify the multitude of granite surface plates we use in my shop and the man that does this is one of the rare ones who actually can re-lap the granite plates back into spec for flatness using a Master cast iron plate using aluminum oxide abrasives of various grades. Cast iron master laps are always made in pairs of three to get true flat reference surfaces- I know they are still made but I am not sure who you go to to buy one but if you had one I am certain you would get better results since they are what are used to correct even the granite surfaces
@@movax20h I do know that many coordinate measuring machines have ways that are aluminum which is hard anodized and then lapped to extreme tolerances. It wouldn't surprise me if the very high end ones are also liquid cooled. I do know that they are kept in tightly temperature controlled environments with any heat sources removed (such as computers and humans) and put into different rooms, which is probably enough. But it would take longer for the whole machine to soak back to equilibrium if that environment were disturbed.
Dear Jeroen, Inspired by this video I settled out to try taking broken Nd:YAG crystals and polish them to be usable again. I have already hand-lapped then using 3µm abrasive to about 3 wavelength flatness. Currently, I am finishing building a pitch polisher following your example here to bring them to final flatness. However, since they are not glass I can't use cerium oxide. I found a paper suggesting using .3µm aluminium oxide. Do you have any recommendations for the concentration of abrasive in the slurry when using your type of polishing machine? and do you have a hint where I could source this polishing compound as a private person in small quantity for a reasonable price? (Using the keyword "Linde A" I found a lot of North American sellers but none of them seem to send to Europe where I am based) Thank you for your inspiring videos and in advance for any tips you can provide!
it's the blocking itself that introduces mechanical stress in the material. After deblocking it is automatically removed again, but then the shape of the objects change, generally in an unpredictable way.
@@HuygensOptics yes, that's exactly what I asked. after unblocking, the surface color changes unpredictably :-(( instagram.com/p/CE9lp0yqDlC/?igshid=1tatmxrin9m51
So, the only thing you can do here is keep blocking wax thickness as thin and uniform as possible. Also only block on flat surfaces. another option is to us pitch as a blocking substance, prefereably in a dot pattern. Next you give the pitch sufficient time (1-2 days) to relax, so that all stress flows out of the contact points between the two surfaces. This will reduce stress and deformation significantly.
Hi, to make a "Economic bruiser" can I use a large granite disk with small disks of Soda Lime glass glued on face? Or best, can use a disk of thin sodalime glass (same size in diameter) glued on face of these granite disk? I can not pay for zerodur or fused silica... I´m changing a large 1,10 meter CG to CP... and the bruiser is my biggest concern... I purchase the book recommended by you, but doesn't suggest alternatives! thank you a lot.
It is an option. Float glass and granite have approx. the same CTE (7-9 ppm/K) so that is good and I have actually done this in the past.The problem that you will encounter is changes in shape due to temperature changes and gradients. It will be very hard to get your process stable. So for example using a thick slice of Pyrex (CTE=3.3) has major advantages over granite. Pyrex blanks can sometimes be found for fairly low prices on Ebay.
@@HuygensOptics Amazing... I have a pyrex blank 24" diameter end 1,5" thick and I can use that as a bruiser... You save my day again! Thanks! in few days I can put a video about this machine on my channel and invite you to see.
growing up in a machine shop makes me fascinated at what high precision milling can do. I'd take parts home and use the machined metal parts to play with.
I've got absolutely no idea why this video was recommended but I'm so glad it was.
Fascinating. Well done.
Same
Same
Same
It's because your subliminally interested in light & optics.
It started when I clicked on a bartender showing how to make optically clear ice cubes for drinks, next day this was in my feed.
Totally counterintuitive. I've done a fair bit of metalwork sanding and polishing, so I naturally assumed that optical polishing would be similar, only using something harder and flatter to grind the surface. It never occurred to me that the rotational grinding process would use something ductile, yet get better results. Thank you very much for the excellent explanation.
In metalworking terms, it would be very similar to using an aluminum, copper, or tin lap (as in watchmaker's "black polishing"). You want your lap to be softer than the material to be cut. Your lap becomes a matrix to hold the abrasive particles in place, and the cutting happens on the material that can't just grab and hold the abrasive. (Tin, by the way, gives absolutely amazing results when polishing steel. It's just _really_ stringy to machine when you're initially making the lap. Save it for your finest - sub-micron - grits.)
You polish silverware with fine cloth which is softer than silver. If you use sandpaper the result would be terrible.
Very informative yet old video on lapping metal parts by rotation. Check this: ua-cam.com/video/fnoVV-RWIWY/v-deo.html It vey clearly explains "how it's made".
@@gvidas1338 This is great, thanks!
I feel there’s a general misconception when it comes to lapping a polishing that is a result of focusing on the lap material. As the first comment responder noted the lap only hold the cutting media. There are three fundamental rules for cutting to occur, though I only usually remember two. The important one here is that the cutting “tool” (in this case lapping or polishing compound) MUST be harder that the workpiece. The second is that there must be relative motion. The third escapes me.
But in either case it is not the lap that cuts the work piece but the embedded abrasive. Apologies for the lecture comment but lapping and polishing seem to be no different fundamentally than any other metal removal process; hard removes soft.
In the case of polishing silverware presumably there is some residual polishing compound on the cloth that is the effective mechanism for removing the oxide layer.
Also, if I’m glaringly wrong please correct me. 👍
Edited for autocorrect errors.
Flat-out the best vid on making flat optical surfaces! Clearly thought out well, I couldn't pitch in any criticism. It's almost like we're on the same wavelength.
What a cheerfully bright comment! I for one found the video very illuminating, and it seemed to polish out all the rough spots in my dull and hazy knowledge. You could say that it expanded my bandwidth....
@@digitalradiohacker makes me wanna leave my daily grind and do something else
Micronically inquisitive mind lapped up the precision explanations.
If y'all don't cease with immediacy I'm gonna jump into a woodchipper.
😂
The true test of whether someone has mastered a field is their ability to explain it to the uninitiated in a clear and concise manner, you sir are a credit to the field.
How did I end up watching 12min of "how to polish something to the nm scale?" But you explained it so well that I understood it without any previous experience in polishing glass or anything for that matter
It's rather hypnotic.
This is really good, so well detailed. It seems over the years there has been a lot of demonstrations of optical grinding/lapping, lots of "recipes" so to speak, without any detailed explanation as to why it works. I'm coming out of this with a deeper understanding of the process, only took 12 years, but better late than never!
This video reminds me of the optician who fell into a lens grinder and made a spectacle of himself.
That's not as bad as the glass blower who accidentally inhaled and now has a pane in his chest.
@@MikeWiggins1235711 Still not as bad as that chef who, while cooking some some chicken broth, fell into the pot and made a laughing stock of himself.
Clearly, I didn't see that coming.
@@Cynthia_Cantrell Did you hear about the guy who wondered why the baseball kept getting bigger and bigger, then it hit him.
Reminds me of the lab technician who spilled some acid on himself. Really left him fuming.
One of the most clearly explained process ever seen on UA-cam. Beautiful. I feel like building one of these now.
I have been interested in optical engineering for a long time now and yours is the first channel I have seen that covers it well. So thank you for making such excellent videos.
I work in the optics industry for a couple of years now and I have to say that your channel is a real treasure! Your videos are both, highly educational and entertaining. Keep up the outstanding work!
Hats off to your narration... I simply was thrilled. I am a retired engineer, 73 yrs.
The principle looks simple, its just a tar and turntable, but those who tried working with glass, especially polishing and making it precise, knows that its extremely hard, takes years of practice and patience. Great video.
jeez why didnt youtube algorithm recommend this to me earlier??? Its fascinating!!
I used a machine like this in the 90s to flatten hydraulic motor parts. Instead of pitch, the wheel surface was steel, and the surface was kept flat by adjusting three rings which also kept the parts in position on the wheel. We'd check it a couple times a day by washing the abrasive off and placing an optical flat on the wheel. But other than that it worked the same way.
its not the same as true level
@@hindugoat2302 Reality is poison! I can't live like this!
@Andrew Crews I worked on Eaton and Sauer-Sundstrand axial piston pumps and motors. We would replace pistons and cylinder blocks and send them out to be resleeved/refinished.
I appreciate this comment very much after having done very similar work in the 90s. I used to test lapped and partially polished parts with an optical flat to infer the concavity of the part (convex or concave). Once the concavity was determined, I inferred the lapping wheel's (opposite) concavity and would adjust the rings' position to correct the wheel.
Pl.explain the process in detail
This is the most satisfying and informative presentation I've seen in ages. As a photographer I really appreciate the craftsmanship required to produce high quality optics. My hat off to you Sir! 🎩
I've been polishing for a year now. This was super informative. Putting images to techniques helps me understand more of what I do all day long lol
Using interferometric fringes to test optics, what an ingenious setup!
If you think that is impressive the Laser Interferometer Gravitational-Wave Observatory uses interferometry to detect changes in length less than a 10 thousandth of the diameter of a proton.
@@kellymoses8566 luckily we don't need *quite* such precision for optics :)
When you shut the machine down overnight, is it necessary to maintain the lab at a cool temperature to minimize pitch movement?
Nah...you just use auto tune. Everybody does these days. Hehehe
@@shannonpincombe8485 That's how those T-pain sunglasses are made.
I think there is a tradeoff. It would certainly help, but investing in climate control may not be worth it. Also, it seems heating the plate and then weighing it down removes enough deformity in a small amount of time.
These techniques are fascinating.
@@shannonpincombe8485 I respect the pun
They run 24hrs
Excellent vector drawing demonstrating the constant angular velocity.
I have been polishing precision optics for 12 years, I really enjoyed the video, thank you. I do the final polishing of the optics on a spindle in a zerodur plate with holes, I put planes with weights in them, according to a similar principle. pitch polishing pad for the night I turn over on a plate smeared with Regipol with good flatness
I've always thought there was something almost magical about being able to make incredibly accurate optical surfaces, flat or otherwise, with no precision tools whatsoever. I ground my 6" parabolic mirror by hand, and figured it to 1/20 wave accuracy on a pitch lap, with no references other than the Foucault test.
Very nicely done. Subscribed.
cheers from sunny Vienna, Scott
if you like that; then the three plates to make a surface plate is another example of this principle (though wil hardish surfaces)
@@robertmccabe8632 Indeed. I use this principle to keep my sharpening stones flat. Using silicon carbide abrasive, I grind A against B, B against C, and C against A.
Very clear and detailed explanation, best I found so far (and no irritating background music).
Cool how you built the turntable from a washing machine motor and rollerblade wheels!
Thanks for the video, it's very informative. So the polishing table is an aluminum disk with a motor stator attached to it, and you press the pitch lap by using a granite plate, but I was wondering if you could share what the pitch lap substrate is made of? It looks like a few inches thick disk. Is it also granite?
No actually in this particular case it is borosilicate, which has a thermal expansion coefficient that is about 3 times lower than granite (which is an advantage). You can however use granite without problems if you have good temperature control.
Am I the only one that loves this type of stuff, but at the same time, completely understands how boring most people probably would find it?
Very fascinating seeing someone so specialized and advanced in their field, thank you for sharing.
At work here in Germany they got mad at me for filing like this....that you work in nm tolerances and do so as well made me really happy. The understanding of why one would do it so they just couldn’t understand. Also an amazing video, very informative. Keep up the good work!
Excellent explanation! I’ve been aware of over-arm polishers for lenses, but not aware of this method. You openly share a lot of which others would consider trade secrets. Even so, I have to believe that if I were to set up a similar polishing system, using everything you show here, I probably wouldn’t get results as good. Thank you!
Thank you, i always wanted to know how it was done. Very clear description.
I have always wanted to see Mach 3 used in a unique way like this. BRAVO !
It's hard for me to believe but, I lapped and polished for 20 years (Gator Diamond, Inc) and didn't know half of this info. Thanks, Bill
Thank you for nice video, will you please let me know how to make single 40 to 80mm discs flate up to 1/6 wave or 50 nanometre regards.
Fascinating videos, with plenty of details. Thank you for all this useful info and diagrams!
Do you always use the same grit to polish? If not, do the grits find themselves trapped in the pitch? Do you have to use different pitch plates for each grit?
Polishing is the final stage of producing an optical quality surface and there is only one compound used on a given lap.
Grinding the surface to a desired rudimentary curve is called grinding, not lapping. Pitch laps do not grind but only polish. The amount of material you'd remove by polishing is miniscule and nobody wants to waste time doing any more polishing than necessary.
Grinding a surface starts as a rough cut to quickly give a desired curvature and to remove casting irregularities. Then a succession of finer grits is used to remove the deep scratches left by the previous grit. This is done with a hard tool on the workpiece and an abrasive slurry is run onto the surface. It looks a lot like lapping with a pitch lap, but because the tool is hard, it can be reliably cleaned off when it is time to change to the next finer grit size.
This is very interesting. I read somewhere that for lapping of silicium wafers they use some chemicals instead of abrasive particles. I can't imagine how precise they must polish wafers for 4nm technology.
It's generally a combination of both. If you use a chemical that etches your surface while you are polishing, you can use a very mild (or soft) abrasive agent, which results in a smoother result. By the way, when you use Cerium Oxide to polish glass, chemical interaction also helps speed up the polishing process. For wafers the actual flatness is less important that the smoothness, since modern wafer steppers make a heigh map of the wafer to correct for the total thickness variation when clamping a wafer to the chuck. Modern technology wafers are indeed incredibly smooth and flat (from the dimension of individual components to that of the full chip)
I wasn't aware how much I like this content
Prachtig werk! Ik heb me altijd al afgevraagd hoe die dingen zo vlak konden worden gemaakt.
The explanations you give is perfect. Even a brick can understand this.
this filled in some gaps in my knowledge, really great stuff.
Using a Fisher and Paykel motor driven by a VFD is genius! So much simpler than the old belt and pulley reduction.
This video reminds me of how profound our technological advancement has been. Just think of how many different people had to cooperate and dedicate basically their entire lives to engineering better solutions in the relatively niche field of precision optics. Using a high viscosity fluid as a lapping surface? How the hell did anybody come up with that? Stuff like this just blows my mind.
I would have never guessed that you'd use a malleable disc to do the polishing. Really fascinating to see how this is performed, thanks for the video!
Hello youtube algorithm...thank you for suggesting something I didn't know i needed to know
No idea why this is in my recommended, but very interesting. I'm amazed this kind of thing can be DIY with the proper knowledge and materials.
I would consider myself a maker more in the software area, it's always cool to see what other people are working on. Keep making!
"proper knowledge and materials"--including an old washing machine motor and rollerblade wheels.
Your channel deserves more subscribers
Excellent. Very happy I found your channel. Honestly, as a nerd in training, a good friday night for me includes such terms as nanometer and interferometry. Looking forward to your next video!
Edit: trainings -> training
Thank you kind algorithm for bringing me here. This was very interesting.
When the pitch drapes off the sides, how do you get it back up on top? Do all the different grade abrasives get mixed into the tar? Do you then have to replace the tar? How do you setup for lens grinding as opposed to flats? How are multifocal optical lenses ground? eg for spectacled?
freeze then chip it off and melt and recast. the polishing compound is micon sized...not grit. used to use rouge, but there are better compounds not so messy.
This is an excellent way to do low speed control with a VFD. Thank you for the good idea.
Does this type of pitch polishing give you any assurances of the parts being square or parallel? I imagine you might often want an object with two opposite faces both very flat and very parallel to each other, but my totally uninformed intuition says this process might not be able to give you that?
That is correct. To create parallel surfaces, generally two-sided grinders and polishers are used that have two abrasive disks spinning in opposite direction while applying pressure.
I can't shake the feeling that I'm learning knowledge that I will never need to use!
Agreed. Like watching This Old Tony.
physicist here, every single video of yours is mesmerizing, I have no words!
Your videos constantly inspire me to want to build my own lenses for different optical projects. Thank you for sharing your designs and knowledge!
What a fascinating microscopic world lies behind these seemingly primitive (to the layman only, of course) machines!
well explained, no questions, sir! great video
i have no idea why UA-cam recommend me this video. What more fascinating
is i watched it until end, and still have no idea what is that.
The algorithm sends me to strange places sometimes, but it sure can be a fascinating journey. Today I learned something I did not know yesterday, thanks.
So fascinating the science and math that goes into allowing us to achieve this. I would have never guessed it was pitch
Last minute had the information I didn't know I needed. Blocking pitch does put pressure on glass held to a backing. All the old film of spectacles and camera lenses show blocking, but these are not expected to be accurate to fractions of a wavelength. OK now I will also avoid lots of messy cleaning up as well.
Very good stuff. I've done my share of glass pushing, long long hours of manual work. I tried to make 200mm flats but I constantly got into troubles in 1um (two rings) level... Perhaps I return to them sometimes, even though I hardly remember why I started making them LOL (ok it was some cassegrain telescope idea, and another for testing other flats)
When you go through the grits of silicon carbide or cerium oxide, dou you change pitch plates or clean it very well? How do you make sure some coarse grit doesnt ruin the later polishing steps?
The silicon carbide grinding is not done against pitch but against other glass or granite plates. You have to carefully clean and rinse between grains (not only the plates, the work space, but also your hands) in order to avoid contamination.
Very good explanations with great schematics. Great video overall.
This gave me a moment of connectivity to my Grandad. He ground rifle scope glass for a few years. Now that’s not flat but more of polishing
Today this was recommended to me and I watched till the end
I'm wondering if you can put the pitch lap in the freezer when not in use for weeks or months?
Sure, that is the way to store them over longer times. It also avoids the evaporation of the more volatile components from the pitch so it keeps the right viscosity. However, even then you have to put some effort in bringing them back in a usable state, because, they will loose shape on the sub-micron scale.
Like the other people have said, really great video and clear explanation. One question: What's the relation between the maximum size of flat you can grind vs. the minimum size of the large pitch plate below? I mean, how small can you make this machine vs. how big of a flat can then be produced. I am wondering about grinding flat mirrors for star diagonals for telescopes, not sure how big yet but maybe up to 90mm (?).
If you have a round mirror the size can be up to 35-40 percent of the pitch ring. However it is more difficult for rectangular objects like star diagonals. You need to contain them in a ring. For a 90mm diagonal you need at least a 350-400 mm pitch ring. Also you cannot attach this type of object at the angled surfaces to the holder, but only at surfaces perpendicular to the pitch surface. For example using hot glue. If you look carefully at the connection of the elliptical mirror at 6:58 in the video you see that it is connected at the short side with two dots of hot glue (on each side) to the aluminum holder.
@@HuygensOptics Could you just create the flat in whatever shape is most convenient for the machine then water jet cut it out or something like that to get your desired shape (ellipse, rectangle, etc.)?
It's possible, but it depends on the flatness you require. Generally the cutting will introduce stress/cracks in the material at the cutting interface, which can seriously deform the objects in the um scale range. If you require high quality optics it's not an option. Also the abbrasives in water jet cutting can very easily damage your optical surfaces, even with protective layers applied.
@@HuygensOptics hm... maybe you could use the larger blank of 'convenient shape' and cut out the desired part(s) about 99% of the way with a small diamond grinding bit of some kind, leaving a holding ring or maybe tabs, then do the flatness grinding/polishing, then just snap the parts out by breaking the thin retaining glass. Or maybe try EDM using one of the techniques for non-conductive materials. But, yeah, at that point there's probably not really an overall cost/effort savings. :)
That is a trick used with for example with the central hole in telescope mirrors: you drill almost completely through except fort the last 2mm or so. My personal experience with this method is not positive. Generally, if there is residual stress in the material, it will be partly released when cutting parts out, deforming the piece. I always bring the object in its final shape and then start grinding and polishing and focus my efforts on a good mounting method and the correct tool shape.
is this similar to how they make the large telescope mirrors?
I had no idea about any of this. Fascinating.
The closest thing I do is knife sharpening, but I always wondered about stuff like straightness, flatness, and smoothness. This videos was very interesting.
Thanks you, Sir. Very good explanation of optical polisher.
So-de-knetter... té gek dit! Leppen van metalen onderdelen doe ik met veel liefde. Leppen met pek is iets wat ik nog nooit gezien heb. Dus: fantastisch om nu met je neus zowat óp de lepschijf te zitten. Met commentaar van een kenner. Hartelijk dank!! Ik kijk erg uit naar de video over de vlakheidsmeting! Groeten - Nobby Assmann
10:00 So not necessarily, and not usually to that degree. Yes, surface plates do have stricter requirements for local flatness vs overall flatness.
For instance a grade AA surface plate of 18x18 in, has an overall flatness requrement of 50uin, but a local flatness requirement of only 35uin, so that would definitely need to be checked for whatever specific surface plate you're using.
If pitch is a liquid, couldn't flatness be achieved without a reference flat (like the bruiser or surface plate)?
Enclose the pitch in a container with walls higher than it and heat it up enough to reduce the viscosity so it flows fast enough for the surface to find its level in a reasonable amount of time, then let it cool down. Surface tension would mess with the flatness near the walls, but closer in should be fine.
I suppose you lose the grooves though, darn.
You will have to wait for a long time before the surface reaches submicron flatness. Keep in mind that there is a volume contraction during cooling down. And of course it is not just about getting the tool flat, but keep it flat during processing.
good to see that physics works
all engineers experience the same when going down to nanometers
no matter what kind of engineering they are doing
and that is that solid material is moving ( slowly )
Great video. Very very well explained! Brings back memories. I used to make achromats from blanks of optical class at home in high school for telescopes.
Can a copper disc be very flat polished?
Thank you for making and sharing this video! Great content and very informative!
Love the videos! I have a couple of questions:
- what pitch specifically do you use in Jerome’s Pitch Polisher?
- when setting up the pitch polisher, how do you ensure identical angular rotation of object and pitch lap? Are dimensions important or does the design itself ‘handle’ it?
Thanks!
Great video, Christiaan would be proud if he was here today!
Awesome work! I am going to try this...
10:07 - Well, in that image, the variation is actually 1μm over 200 mm (because it's between the centre and the edge), not 400, but I guess it won't always be such a continuous curve.
Good day sir...great video...just curious if you can lap any material on it..for example carbon and silicon and tungsten like seal face materials?
Can glass be used as a sureface plate ?
Really well made video! Great job.
I'm an optical fiber telecommunications technician. We used to have to polish the end faces of our connectors when terminating them. 3 different ratings of polish paper and polishing in a " figure 8" motion. It was tedious. Faster speeds and the need for lower reflection at the connections has us using fusion splicing and factory terminated connections, now. No one misses " puck and polish" terminations.
i am a machinist by trade and when i was initially linked this video, and i only saw the title, i thought "oh great some charlatan is going to talk about taping sandpaper to a piece of float glass" but i was pleasantly surprised to see that was not the case and this is the real deal. subscribed.
have you considered trying a cast iron master surface for the weight? as long as the room is at least loosely climate controlled, you could theoretically have a more economical option for a even flatter surface than the granite plate
In principle cast iron would work just as well as granite. The CTE is around 10-11 ppm, so compared with granite (8ppm) this is only slightly higher. And because of it's higher thermal conductivy, it will get to thermal equilibrium (and retain it's original shape) quicker than granite. So it would in fact be a better choice as a material. However, don't know how easy it is to make a cast iron disk of 400mm in diameter with a flatness of approx 1um. By the way, granite is not an expensive material.
@@HuygensOptics i know that in metrology cast iron is preferred for master surfaces, because it is more dimensionally stable and easier to make very (beyond AA grade) flat. for your specific use, it would also help as cast iron is denser so it would be heavier for the same size. that is why it came to mind.
@@SuperAWaC Actually in some super accurate machines, aluminium with internal cooling channels can be preferred too. This is because aluminium has way higher thermal conductivity (about 3 times higher than iron) and can be stabilized better, down to 0.1°C. Of course if you deal with big loads and need very high stiffness the cast iron or granite are preferred, but with low loads, aluminium counter intuitively can be even better. I can't find source and machines doing that right now, but I do remember seeing some year ago.
I am a Tool & Die machinist and horologist and it made me happy to see there are other intelligent machinists watching things like this.
I would second the use of a cast iron master plate. There is a guy who comes once a year to certify the multitude of granite surface plates we use in my shop and the man that does this is one of the rare ones who actually can re-lap the granite plates back into spec for flatness using a Master cast iron plate using aluminum oxide abrasives of various grades.
Cast iron master laps are always made in pairs of three to get true flat reference surfaces- I know they are still made but I am not sure who you go to to buy one but if you had one I am certain you would get better results since they are what are used to correct even the granite surfaces
@@movax20h I do know that many coordinate measuring machines have ways that are aluminum which is hard anodized and then lapped to extreme tolerances. It wouldn't surprise me if the very high end ones are also liquid cooled. I do know that they are kept in tightly temperature controlled environments with any heat sources removed (such as computers and humans) and put into different rooms, which is probably enough. But it would take longer for the whole machine to soak back to equilibrium if that environment were disturbed.
how does this method compare to the three plate method - scraping three surfaces, two at a time ensures flatness with good tolerances.
Dear Jeroen,
Inspired by this video I settled out to try taking broken Nd:YAG crystals and polish them to be usable again. I have already hand-lapped then using 3µm abrasive to about 3 wavelength flatness. Currently, I am finishing building a pitch polisher following your example here to bring them to final flatness. However, since they are not glass I can't use cerium oxide. I found a paper suggesting using .3µm aluminium oxide. Do you have any recommendations for the concentration of abrasive in the slurry when using your type of polishing machine? and do you have a hint where I could source this polishing compound as a private person in small quantity for a reasonable price? (Using the keyword "Linde A" I found a lot of North American sellers but none of them seem to send to Europe where I am based)
Thank you for your inspiring videos and in advance for any tips you can provide!
Hi. I have the same your laser rod that need polishing aganst. Did you succeed?
great!
how to remove stress on the lens? large lens. after removing it from the block.
it's the blocking itself that introduces mechanical stress in the material. After deblocking it is automatically removed again, but then the shape of the objects change, generally in an unpredictable way.
@@HuygensOptics yes, that's exactly what I asked. after unblocking, the surface color changes unpredictably :-((
instagram.com/p/CE9lp0yqDlC/?igshid=1tatmxrin9m51
So, the only thing you can do here is keep blocking wax thickness as thin and uniform as possible. Also only block on flat surfaces. another option is to us pitch as a blocking substance, prefereably in a dot pattern. Next you give the pitch sufficient time (1-2 days) to relax, so that all stress flows out of the contact points between the two surfaces. This will reduce stress and deformation significantly.
Today is the day i understood why tires have grooves
its amazing how a bumpy surface can grind down to the nanometer range
It's only bumpy on the macro scale.
Hi. Огромное спасибо за ваш вклад в образование !
Fascinating, and very well explained! Thanks a lot!
This video made me happy.
Great Video!!! Is there a way to make flat and parellel surfaces ?
How does the bruiser plate get so flat? Also how can you assume that the bruiser plate tolerance is a shaped in a perfect arc?
Hi, to make a "Economic bruiser" can I use a large granite disk with small disks of Soda Lime glass glued on face? Or best, can use a disk of thin sodalime glass (same size in diameter) glued on face of these granite disk? I can not pay for zerodur or fused silica... I´m changing a large 1,10 meter CG to CP... and the bruiser is my biggest concern... I purchase the book recommended by you, but doesn't suggest alternatives! thank you a lot.
It is an option. Float glass and granite have approx. the same CTE (7-9 ppm/K) so that is good and I have actually done this in the past.The problem that you will encounter is changes in shape due to temperature changes and gradients. It will be very hard to get your process stable. So for example using a thick slice of Pyrex (CTE=3.3) has major advantages over granite. Pyrex blanks can sometimes be found for fairly low prices on Ebay.
@@HuygensOptics Amazing... I have a pyrex blank 24" diameter end 1,5" thick and I can use that as a bruiser... You save my day again! Thanks! in few days I can put a video about this machine on my channel and invite you to see.
@@sandroac34 looking forward to that!
growing up in a machine shop makes me fascinated at what high precision milling can do. I'd take parts home and use the machined metal parts to play with.
Yes, I have a question. How on earth you do these sorceries?
Any links to making that motor (from washing machine and skate wheels)?
hats down, top grade quality video, so rare