Gene
Reading through my copy of "Gears and Gear Cutting" by Ivan Law (#19 in the 
workshop practice series) and some more research suggests that since both the 
cup and the ring gear should have over 135 Teeth that the tooth shape 
approaches the diameter of a rack and therefore they really aren't involute 
anymore as much as just triangular.  That probably explains why the drawings 
I've seen suggest an ever so slight sine curve to the tooth.

In either case though, the paper talks about how as the gear bends it bends 
along the same neutral point that say a toothed drive belt bends.  That appears 
to cause less stress than bending them in the opposite direction.  

But I'm an amateur and a pretty poor mechanical designer since it's not my 
field or talent.

I bought some PETG yesterday.  I have some parts to make out of metal and still 
need those bearings so until all that arrives this project is on hold.
John


> -----Original Message-----
> From: Gene Heskett [mailto:ghesk...@shentel.net]
> Sent: December-14-21 2:09 AM
> To: emc-users@lists.sourceforge.net
> Subject: Re: [Emc-users] Harmonic Drive
> 
> On Tue, 14 Dec, 2021 at 12:31 AM, John Dammeyer <jo...@autoartisans.com> 
> wrote:
> 
> 
> To: 'enhanced machine controller (emc)'
> Todd,
> I'm not sure I 'get' how it works with the pinned gears.
> 
> I built a different kind of flex cup drive based on a paper on what is called 
> an HGD_EWG_Drive
> https://www.geartechnology.com/issues/1013x/harmonic_drive.pdf
> 
> Here's a really short video of the version of one I created.  I used that 
> earlier 3D print harmonic drive for the base tooth form
> although this type looks like you can use involute profile or even belt 
> profile as the flex teeth follow a circular rather than elliptical
> gear.  Something to play with now that I have some PETG.
> https://youtu.be/lb6shTgQ1wY
> 
> I really need to bring in a bigger stepper to play with this.  The surplus 
> unit with about 1A at 36V will stall if I even look at it cross
> eyed.
> 
> Here's the original I was playing with along side the new one.
> http://www.autoartisans.com/harmonicdrive/HarmonicDrive-35.jpg
> 
> I still don't have the right size bearings but did have these tiny ones I 
> bought 20 years ago at a model aircraft show in the UK.  So I
> created bearings that were 22mm diameter.  Big issue there they were really 
> only 21.3mm after printing and that caused all sorts of
> issues because I forgot that detail.   I then printed some oversize, created 
> an arbor for them on the lathe and turned them down to
> 22mm.  Now I get 100% connection and zero backlash.  This things spins so 
> smoothly just by placing a finger on the wave cup driver
> and pushing it around.  And you can't hold onto the output.
> 
> http://www.autoartisans.com/harmonicdrive/HarmonicDrive-29.jpg
> 
> And here's a diagram of most of the bits and pieces.
> http://www.autoartisans.com/harmonicdrive/HarmonicDrive-28.jpg
> 
> The main bearings are utility trailer 1" ID roller bearings.  I'm waiting for 
> 6101 and 6103 bearings to arrive.  Most  of the parts will be
> made from metal.
> 
> John Dammeyer
> 
> 
> 
> What happens to these designs if the teeth are triangles instead of involute, 
> with
> say 10% of the tips clipped. It seems to me that the needed flex to gain the 
> free space for the tooth to "hop" over the mating tooth
> would be reduced, prolonging the life even
> further. The pinned, 4 pinion model also looks interesting.
> 
> 
> The involute tooth profile used in most of the links shown here would need a 
> more
> rapid engage/disengage, which results in the overly large eccentricity and
> 
> clearance gaps seen in the original paper. The smaller pinions in the 4 pinion
> model shown here would make the involute gear shape more desirable.
> 
> 
> 
> FWIW, the one I built, the one that's 100% printed from PETG except for the 
> crosman
> BB's as loose balls for the 3 bearings, has now been running at about 500 
> revs of
> its input shaft, for about 60 days now, has not failed other than I ripped 
> the output
> pulley out of the output disk trying to find a tooth count that resulted in a 
> tight,
> zero backlash belt to the A axle of my 6040 mill. One more tooth was too tight
> with the available belt.
> 
> 
> But the drive itself is still running. Driven by a 1NM 3 phase stepper. Its a 
> 30/1
> ratio.  Could be made higher, but with additional gear down in the belt to 
> the axle
> of about 4/1, its a bit slow. OTOH, the ability to drive while something in 
> the chuck
> is being gnawed on by an eighth inch tool doing 20k certainly opens up lots 
> of possibilities. :o)
> 
> 
> 
> > -----Original Message-----
> > From: Todd Zuercher 
> > [mailto:to...@pgrahamdunn.com<mailto:to...@pgrahamdunn.com>]
> > Sent: December-03-21 8:37 AM
> > To: Enhanced Machine Controller (EMC)
> > Subject: Re: [Emc-users] Harmonic Drive
> >
> > It was easier to model in Freecad than I expected.  I'll try to attach a 
> > jpg of it and see if it goes through.
> >
> > And if it doesn't go through a link to it on my Google drive.
> > https://drive.google.com/file/d/1_eSu-olALMsevTnOpzQ1o1r_rPjfexwU/view?usp=sharing
> >
> > Here is a link to the Freecad file (Freecad v0.18).
> > https://drive.google.com/file/d/1GGwG9nzmYowKeuMAWMvOExeGrvSzZCnY/view?usp=sharing
> >
> >
> > Todd Zuercher
> > P. Graham Dunn Inc.
> > 630 Henry Street
> > Dalton, Ohio 44618
> > Phone:  (330)828-2105ext. 2031
> >
> > -----Original Message-----
> > From: Andy Pugh <bodge...@gmail.com<mailto:bodge...@gmail.com>>
> > Sent: Thursday, December 02, 2021 6:40 PM
> > To: Enhanced Machine Controller (EMC) 
> > <emc-users@lists.sourceforge.net<mailto:emc-users@lists.sourceforge.net>>
> > Subject: Re: [Emc-users] Harmonic Drive
> >
> > [EXTERNAL EMAIL] Be sure links are safe.
> >
> > > On 2 Dec 2021, at 21:14, Todd Zuercher 
> > > <to...@pgrahamdunn.com<mailto:to...@pgrahamdunn.com>> wrote:
> > >
> > >  What if you used two outer ring gears of the same diameter one with one 
> > > fewer teeth.  Then have 4 sets of planetaries  stacked
> in
> > pairs.  Each pair would be pinned together
> >
> > You can just use pinions twice as long as the ring gears. I think I have 
> > seen this done
> >
> > It would work with all-external gears too.
> >
> >
> >
> >
> > _______________________________________________
> > Emc-users mailing list
> > Emc-users@lists.sourceforge.net<mailto:Emc-users@lists.sourceforge.net>
> > https://nam04.safelinks.protection.outlook.com/?url=https%3A%2F%2Flists.sourceforge.net%2Flists%2Flistinfo%2Femc-
> >
> users&amp;data=04%7C01%7Ctoddz%40pgrahamdunn.com%7Ce891b8fadd884297a79608d9b5ed4669%7C5758544c573f47cebee9
> >
> 6c3e0806fb43%7C0%7C0%7C637740852975732935%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJ
> >
> BTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000&amp;sdata=16s%2BM61vPdzTAtr3Qg6ttYnRU6%2Fg4vkOsrYcbPRxmuY%3D&amp;reserv
> > ed=0
> 
> 
> 
> _______________________________________________
> Emc-users mailing list
> Emc-users@lists.sourceforge.net<mailto:Emc-users@lists.sourceforge.net>
> https://lists.sourceforge.net/lists/listinfo/emc-users
> .
> _______________________________________________
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users



_______________________________________________
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Reply via email to