eccentrically cycloidal gearing

[svenakela]

Hi, much as I admire the ingenuity of the hypocycloid drive, I get the impression after reading the details that "it fits where it touches, when it touches".....that is, given enough time the sliding faces (they do slide, can not roll) will end up with more clearance than the worm drive which is being projected as a viable backlashless reduction drive replacement.....it can not exist without some backlash eventually ocurring, which is not adjustable, no matter what materials are used and no matter how tight the clearances initially are.

...

FYI, they do not slide, they roll. I use a bunch of hypos and the backslash is not noticable.

[handlewanker]

Hi Sven, if'n it rolls as you say, is there any reason why a rotary table drive of 60:1 ratio without backlash couldn't be made, or is made, and how would it compare to the conventional steel worm and bronze wormwheel set-up cost wise.

I can understand the EC set-up as it is easy to see the rolling motion from the patent notes that Bill Todd posted, but the one you showed in the video of the drive is difficult to envisage.....do they both follow the same principle of the eccentric gear with rounded teeth rolling round another internal gear also with rounded teeth, but one less than the pinion?
Ian.

[svenakela]

The biggest problem to overcome with hypo's are vibrations from the assymetrical position of the internal driver. It requires counter weight and/or reduction of material to rebalance the shaft. I have a fourth rotational axis with a hypo drive.

Google or search here at cnczone, they've been discussed before.

EDIT: Great summary here.

[handlewanker]

Hi, I don't think the vibration would be a problem if'n the drive was used in place of a rotary table's worm drive with bronze worm wheel to reduce/eliminate backlash....probably only if'n you were "fast forwarding" the table round to the next co-ordinate etc.

I can not see any way that any backlash can be taken out if'n it develops from wear.....not important in a reduction drive, but totally necessary in a rotary table/positioner application, and 4th axis work is totally reliant on a backlash free device.

What would the highest reduction be in this device?

IE, can you get as high as 1:2 reduction?....this would lend itself to the 1:2 reduction that is required in a 4 stroke engine to drive the camshaft at half engine speed, so making a very compact and simple crankshaft/camshaft geartrain as opposed to the two gears that otherwise puts the camshaft to one side of the engine due to the diam of the second driven gear.....the reducer would then just be a virtual collar that fits on the crankshaft centrally, with the rotors inside it.
Ian

[svenakela]

Yes, the vibrations are there if you have higher rev's, if you stick to lower revs it's not a problem but as I said a balancing is doable. You will have high rpms's pretty quick as the gearing ratio is low. 1:2 will not be doable, for any type of oscillating gear reduction.

Wear is not an issue, my hypo's are stuffed with grease and they will run forever before backlash will become a problem. They're made in a Cincinatti CNC, and we kept the tolerances really low. Also, the contact area is way larger than for example in a worm gear or rack/pinion. There are other guys out there who made hypo's in home made CNC's and they still perform very well.

[handlewanker]

So, if'n the rotors are made initially to a slack fit, and then hard chromed to inject longevity for wear and then ground to also control the clearance, I would think that the rotary table with one as the resolver would be practically indestructible, and maybe handed down from father to son like a family heirloom....LOL.

What is the max reduction you could get from a hypo device......four ring gear lobes and three rotor lobes?.....that would be 1:4 reduction...perhaps?.....that would suit my design needs as the 1:4 ratio could be utilised with two sets of cams at 180 deg to one another.

On second thoughts, for the engine timing gear arrangement, I think that a pinion with 20 teeth and involute tooth form, running inside a ring gear of 40 teeth would do the job of 1:2 reduction without having to go to the extremes of making a hypo drive for a relatively simple problem that does not care if some backlash was present.

I think the hypo solution would also suit the drive for a chain block which at present requires a set of planetary gears to make the reduction.
Ian.

[vegipete]

Try this thought experiment .

you are riding along on a nice flat road on horse & cart.

Unfortunately, the wainwright has built the wheels badly, the axle is not in the centre

You bump along, (in a cycloidal motion)

Then your luck changes. You come to a piece of road laid by the wainwright's brother.

The road surface is waving up & down in a way that cancels out the motion of the off-centre wheels.

No wheels sliding

I don't believe this. "No wheels sliding" can only occur if all 4 wheels are in phase. (I'm not sure, but I suspect the cart would actually get 'stuck' on the section of road made by the wainwright's brother if the wheels can't slide.) The EC-gearing has created essentially a single tooth pinion. All 'layers' of this pinion rotate at the same speed yet the radius varies. With varying radius comes varying tangential speed, yet the meshing gear obviously does not move with locally varying surface speeds.

Still, a really neat mechanism. Gotta think about fabricating one...

[BillTodd]

I don't believe this. "No wheels sliding" can only occur if all 4 wheels are in phase. (I'm not sure, but I suspect the cart would actually get 'stuck' on the section of road made by the wainwright's brother if the wheels can't slide.)

You're quite correct That's where my analogy breaks down. (But who in their right-mind would make a four wheel drive Horse'n'Cart :LOL

Novikov's modification is to change the road so only one wheel touches down at a time. when this is integrated into a 'worm' the result is a smooth rolling motion.

Bill

[handlewanker]

I think Bill's analogy applies to a cart with two wheels, that way the view looking at the side is for one wheel going up and down on the undulations, which simulates the EC design where one roller rolls up and down on the track as it moves forward.

Four wheels (two in line) would not work on a track like the EC type, as they would be working against each other to maintain contact with the track form, unless the track was in a straight line and the wheels were independently sprung.

Anyway, the wheel on the bumpy road is just an example of the EC motion, and the actual EC "wheel" is really a series of wheels pitched apart and across the track to form what appears to be a spiral, but is not a spiral in actuality.....more like a series of phased cams making contact with phased undulations on a track.
Ian.

[handlewanker]

Hi just had another look at the #1 post and the first video clip.....the end of the clip changes the two entities from a rounded lobe to a square form.......this would be "easy" to make using a 4th axis drive and an end mill.....amazing that a square sided rotor can rotate round a wheel with square sided teeth, it looks like a spiral gear drive but isn't......this has got to be doable with a CNC mill and 4th axis.
Ian.