Wow! Those spindles found a good home! I’m subscribed now so I’ll be eagerly watching for updates. Very nice meeting you at ASPE, and I hope to see you again next year. Hopefully you’ll get entered in the student challenge. #ASPE
Nice project :) I was wondering since you put a magnetic rotor on the spindle - if the magnetic field propagates through the air bearing spindle, did you end up with magnetic front? Did you check the alignment of the motor rotor and motor stator axis - if they are not concentric, you can get additional radial force and preload the spindle?
Hey Cyrus, I Just watched your video from 5 months ago on investigating form error from clamping. Glad to see the results. I was wondering how that drawbar system would work out…. Much better than I expected! I think the springback in your work holding is unavoidable - even with a more complaint flexure. Would you consider magnets embedded in the notches of the coupling? This will keep the preload in-line with the contacts so there’s no (or just less?) bending out of plane. I could see this bringing up balance issues by location error of the magnets and each being of slightly different mass. I could imagine that drawbar having some shaft whirl as well - though I reckon the spindle speed is slow enough to not create an issue Looking forward to the next update as always!
How insanely impressive! OMG! That has got to be the best of the best right there. I would absolutely DIE to own a lathe set up with that incorporated into it🙏🙂👍
Incredibly cool and beyond kind of PI to help you guys out! Interested to see how the Hale connection works out. Do you have a link to a paper/thesis where he goes into the details of it?
patents.google.com/patent/US6065898A/en digital.library.unt.edu/ark:/67531/metadc784803/m2/1/high_res_d/8431.pdf they are only briefly mentioned in his phd thesis but i still have linked it because its an awesome read
Not sure if it makes a difference on repeatability but if you add a hex to the brass knob you can use a torque wrench to tighten the draw bar. You could also number the cut outs on the “chuck” and aluminum plate so that it is going on the same way every time.
Too much clamping force (or any at all, really) will distort the workpiece. It doesn't sound like much, but this can easily be beyond the tolerance of optical components such as mirrors. I'm not sure how much distortion the hand tightening imparts, but you really want the absolute minimum necessary for stable machining/finishing.
Great work, thanks for sharing. Is the Hale connection very susceptible to drawbar torque? Just a guess but i would have thought any slight misalignment of the machined mating surfaces would cause a bit of elastic throw to it? Just a guess though as ive never made one (or heard of them!)
The way in which they are machined, there is a very small chance for misalignment, which is why they’re so repeatable. There certainly could be some deflection if you over tighten the draw bar, but you really only need to snug it lightly. As long as there is a decently centered seating force, those things aren’t going to move anywhere.
@@cylosgarage ok thanks. I had a look at the phd and it doesn't mention manufacturing method, just that it needs precisely straight lines. Did you do it on a rotary table? Do you think 3d surfacing would be precise enough?
@@cylosgarage ah cool, that's good to know. I do a fair amount of 3d surfacing on my CNC, so I'll definitely give the design a go. I think i will extrude the surface where the bolt goes through so there isn't an air gap. Thanks again for sharing
@@lawmate Nice, that's a good idea, that way you get a repeatable preload force. Be wary of the parallelism of the boss that the thread goes in though. If that surface and the one it mates with aren't perfectly aligned, it'll ruin the precision.
Very cool. In the thesis you have in the description one of the two surfaces of the coupling is rounded so you have no flat on flat but Zylinder on flat contact. Do you see less precision with a flat on flat design? Seems to be way easier to make.
Yes, the design we're using is a flat on flat variation of the original design. This has been successfully demonstrated by a certain precision focused university in the past. While it technically can't get as good of a result as the round on flat design, is it a good choice for this application because it provides higher stiffness while still being extremely repeatable. It well satisfies the requirements. Again, keep in mind these rough milled prototypes are already repeating to under 20 microinches. ground and lapped ones will be even better
@@cylosgarage awesome! What is the repeatability like when you rotate it 120 or 240 degree? As there is no compliance in the system like it is on erowa or 3r pallets you must make them real precise, no? Looking forward to see more of it!
@@sandeeps7979 15000 RPM max i think. Try checking the title or watching the video and see if you can figure out the company by yourself (hint: it’s in the title of the video 😱)
Have you tried multiple coupling connections? It seems like the hale coupling would only be repeatable for a given pair of coupling halves. Are you going to make a master coupling gauge?
Wow! Those spindles found a good home! I’m subscribed now so I’ll be eagerly watching for updates. Very nice meeting you at ASPE, and I hope to see you again next year. Hopefully you’ll get entered in the student challenge.
#ASPE
Very generous of PI to give those spindles. I love those guys even more now!
That’s beyond satisfying to see the aluminum disk go back on with absolute zero runout
there's actually 2 thou of runout shhhhhhhhh
Nice project :) I was wondering since you put a magnetic rotor on the spindle - if the magnetic field propagates through the air bearing spindle, did you end up with magnetic front?
Did you check the alignment of the motor rotor and motor stator axis - if they are not concentric, you can get additional radial force and preload the spindle?
Hey Cyrus,
I Just watched your video from 5 months ago on investigating form error from clamping. Glad to see the results. I was wondering how that drawbar system would work out…. Much better than I expected!
I think the springback in your work holding is unavoidable - even with a more complaint flexure.
Would you consider magnets embedded in the notches of the coupling? This will keep the preload in-line with the contacts so there’s no (or just less?) bending out of plane.
I could see this bringing up balance issues by location error of the magnets and each being of slightly different mass.
I could imagine that drawbar having some shaft whirl as well - though I reckon the spindle speed is slow enough to not create an issue
Looking forward to the next update as always!
So Awesome!! Glad to see your project getting the love and support it deserves. Bed ways next?
yes! working on them right now. switching to hydrostatic!
Dude!! I visited the PI facility when I was at the conference too. Nicest blokes ever!!
Yess! Are you the same Kumar we met in the precision machine design class?
@@cylosgarage That's correct!! Small world
How insanely impressive! OMG! That has got to be the best of the best right there. I would absolutely DIE to own a lathe set up with that incorporated into it🙏🙂👍
Incredibly cool and beyond kind of PI to help you guys out! Interested to see how the Hale connection works out. Do you have a link to a paper/thesis where he goes into the details of it?
patents.google.com/patent/US6065898A/en
digital.library.unt.edu/ark:/67531/metadc784803/m2/1/high_res_d/8431.pdf
they are only briefly mentioned in his phd thesis but i still have linked it because its an awesome read
@@cylosgarage Thanks!
This is Absolutely crazy! You seem to always get lucky with something.
Not sure if it makes a difference on repeatability but if you add a hex to the brass knob you can use a torque wrench to tighten the draw bar. You could also number the cut outs on the “chuck” and aluminum plate so that it is going on the same way every time.
Too much clamping force (or any at all, really) will distort the workpiece. It doesn't sound like much, but this can easily be beyond the tolerance of optical components such as mirrors. I'm not sure how much distortion the hand tightening imparts, but you really want the absolute minimum necessary for stable machining/finishing.
@@denniszhang9278 I was thinking a low torque around 15-20 in lbs would similar to hand tight but more repeatable.
thanks for sharing, looking forward to the first chips ;)
The idea of tuning linear motors for a diamond lathe makes my stomach turn :D
Great work, thanks for sharing. Is the Hale connection very susceptible to drawbar torque? Just a guess but i would have thought any slight misalignment of the machined mating surfaces would cause a bit of elastic throw to it? Just a guess though as ive never made one (or heard of them!)
The way in which they are machined, there is a very small chance for misalignment, which is why they’re so repeatable. There certainly could be some deflection if you over tighten the draw bar, but you really only need to snug it lightly. As long as there is a decently centered seating force, those things aren’t going to move anywhere.
@@cylosgarage ok thanks. I had a look at the phd and it doesn't mention manufacturing method, just that it needs precisely straight lines. Did you do it on a rotary table? Do you think 3d surfacing would be precise enough?
@@lawmate a very good rotary table would be the ideal way to do it yes. However, the most accurate way we have of making them is with the cnc.
@@cylosgarage ah cool, that's good to know. I do a fair amount of 3d surfacing on my CNC, so I'll definitely give the design a go. I think i will extrude the surface where the bolt goes through so there isn't an air gap. Thanks again for sharing
@@lawmate Nice, that's a good idea, that way you get a repeatable preload force. Be wary of the parallelism of the boss that the thread goes in though. If that surface and the one it mates with aren't perfectly aligned, it'll ruin the precision.
Very cool. In the thesis you have in the description one of the two surfaces of the coupling is rounded so you have no flat on flat but Zylinder on flat contact.
Do you see less precision with a flat on flat design? Seems to be way easier to make.
Yes, the design we're using is a flat on flat variation of the original design. This has been successfully demonstrated by a certain precision focused university in the past. While it technically can't get as good of a result as the round on flat design, is it a good choice for this application because it provides higher stiffness while still being extremely repeatable. It well satisfies the requirements. Again, keep in mind these rough milled prototypes are already repeating to under 20 microinches. ground and lapped ones will be even better
@@cylosgarage awesome! What is the repeatability like when you rotate it 120 or 240 degree? As there is no compliance in the system like it is on erowa or 3r pallets you must make them real precise, no?
Looking forward to see more of it!
@@hampfi747 I suspect that the main compliance mode is the theeth toppling outwards slightly in Cylo's design.
Hi sir, what’s the maximum rpm the spindles can handle. And which company bearings was used sir.
@@sandeeps7979 15000 RPM max i think. Try checking the title or watching the video and see if you can figure out the company by yourself (hint: it’s in the title of the video 😱)
Have you tried multiple coupling connections? It seems like the hale coupling would only be repeatable for a given pair of coupling halves. Are you going to make a master coupling gauge?
A Hirth coupling
Good work so far! I make ultra-precision diamond tooling, I would love to help you out with tooling!
Thanks! We’re actually in the process of looking for tooling right now so we’d be very very interested. Please email cyruslloyd4@gmail.com
Is that the Blockhead 3R or 4R?
It’s a custom opposed cone spindle they built a while ago but never sold (read: “garbage” they had lying around). ~3 inch though
I would love to ask you questions but it would literally be like a caveman talking to Einstein , Hawking etc
@@marcromanello3770 that’s very nice but also not true. I would love to answer your questions
@@cylosgarage you probably wouldn’t understand me I have a pre k understanding of lathe operations and I can’t read a mic or dial indicator