Saw this and thought it might be interesting to some people. Its a kinky kind of rack-n-pinion, only different. They say it's Zero-Backlash! Check out Roller Pinion from Nexen (http://www.nexengroup.com/rp/). I may have to try one of these sometime just to see how well they work. Very unique!

Hey that thing is really cool! If you try it, be sure to report on it here. :) Any idea on the cost?

Arvid

They don't have their prices listed. Does anyone? I call these "Stealth Products". Like so many other motion control specialists they don't list prices and want you to play phone/email tag with their sales staff before you find anything out. They said in their brochure it costs more than a rack system but less than a ball screw. Whatever that means.

Here is a brochure they gave me:
Roller Pinion Brochure, PDF - Adobe Acrobat Format (http://home.comcast.net/~kr4495/temp/RollerPinion.pdf)

Revised:
The Linear Distance Per Revolution of the thing is 6.3" for the smallest model - RPS16. That is about twice the length of a regular pinion gear, assuming you chose the smallest regular pinion. This just means you have some more stepping-down to do.

Well, I got a reply from them on their smallest model. If you were going to make an axis that is around 60" long it would cost you about 1500 US after taxes and shipping. Around 3K for an 8 foot long x-axis (racks on both sides).
Is is definitely out of my league. But, it's still interesting even if I can't afford it.

:cheers:

Anybody else have any experience with this system? Seems pretty expensive.

This is a bit off-subject but Nexen has other interesting products as well. I have no connection with Nexen, I'm not their promoter.

Linear rail brakes, rotary brakes, the above roller gears among others.

Neat stuff but I haven't used any yet.

Dick Z

Is this new technology (as in last few years)? Seems like putting rollers on a pinion is something that engineers probably thought of decades ago. Does anyone else offer a roller pinion design?

Regardless, it is a very cool system. Ballscrews are impractical for large routers on low to medium budgets. Companies like Thermwood require support arms to prevent whip. Komo uses helical R&P. Bridge mills use massive ballscrews that probably cost more than high end routers.

Roller pinion... relatively inexpensive, easy to manufacture, lube-free. Pretty much the perfect linear drive system for woodworking IMO.

I know there is a very smiliar gear in use, since 1999, at a General motors plant which is in use to stamp quarter panels in a linear motion application. I'm not a liberty to cast judgement but Nexen's technology is very similar to what is used in the GM plant. Infrindgements on current patents? But to answer you question, yes there are others out there that have this technology.

Joseph Ives used roller pinions in clocks. I'm not sure if the nexen rollers rotate on a shaft like Ives pinions did or not.

It's the inverse of a chain and sprocket (imagine a short length of chain wrapped in a circle as the pinion and a linear 'sprocket')

It would be interesting to experiment with a fixed chain laid flat (or held rigid in a U shaped channel, or even upside-down in a suitable T slot) with a sprocket on the motor. It could give similar performance to the roller pinion at a fraction of the cost.

There were similar designs made a few years back b some guy named Leonardo DaVinci. He also mated the roller pinion against face gears. Clever fellow eh?

Dick Z

It's the inverse of a chain and sprocket (imagine a short length of chain wrapped in a circle as the pinion and a linear 'sprocket')

It would be interesting to experiment with a fixed chain laid flat (or held rigid in a U shaped channel, or even upside-down in a suitable T slot) with a sprocket on the motor. It could give similar performance to the roller pinion at a fraction of the cost.

Not exactly. If you use a chain sprocket curve then the line of motion of the of the drive pinion will not stay parallel to the line of motion of the rack. Say that three times fast. Here's what I mean:

http://e.imagehost.org/t/0424/rps.jpg (http://e.imagehost.org/view/0424/rps)

The sprocket profile looks more pointy than it really is. The export kind of lost something. The tooth profile is actually convex. Here's a better look:

http://e.imagehost.org/t/0973/rps3.jpg (http://e.imagehost.org/view/0973/rps3)

Note how the center line of the pinion follows a cycloidal path.

The pinion is in rotation about its center axis and in translation; also; the pinion path and the line of motion along the rack must be parallel. With those constraints I simulated the motion of the pin through rotation and translation. The actual rack profile should look like this:

http://e.imagehost.org/t/0061/rps2.jpg (http://e.imagehost.org/view/0061/rps2)

Both profiles above used the same pinion configuration: 3/8" rollers, 10 rollers per pinion, 2PI length of travel per rotation.

Compare the last profile with that used by Nexen:

http://img213.imageshack.us/img213/3096/nexenrp500ft2.th.jpg (http://img213.imageshack.us/my.php?image=nexenrp500ft2.jpg)

The Nexen profile is more triangular and less like a sprocket profile.

Not exactly. If you use a chain sprocket curve then the line of motion of the of the drive pinion will not stay parallel to the line of motion of the rack. Say that three times fast. ...

Yes you're quite right :) (although it had me scratching my head for a while to work out why).

I had initial only considered the motion of the circular roller and semi-circular part of the sprocket (gum?) but, if one considers the motion of the tooth, then clearly, in the case of the roller pinion the tooth follows a straight path while a the tooth of a sprocket follows a curve (as you pointed out a cycloidal path).

That then raises the questions;

Is it possible to design a conventional sprocket that would operate as required (i.e. backlash free and accurate) on a linear chain?

How many teeth would be required to maintain correct contact at all times? And, if this meant an unfeasibly large sprocket, would having two or more smaller sprockets suitably spaced along the chain, give you the desired control?

Yes you're quite right :) (although it had me scratching my head for a while to work out why).

I had initial only considered the motion of the circular roller and semi-circular part of the sprocket (gum?) but, if one considers the motion of the tooth, then clearly, in the case of the roller pinion the tooth follows a straight path while a the tooth of a sprocket follows a curve (as you pointed out a cycloidal path).

That then raises the questions;

Is it possible to design a conventional sprocket that would operate as required (i.e. backlash free and accurate) on a linear chain?

How many teeth would be required to maintain correct contact at all times? And, if this meant an unfeasibly large sprocket, would having two or more smaller sprockets suitably spaced along the chain, give you the desired control?

Yes, it is possible. The tooth pattern would have to be a roulette (not exactly cycloidal due to the point of concern changing as the point of contact moves) though and not an arc (a sprocket tooth is just two arcs). I'd say 3 would be the minimum number of teeth, or maybe 4 teeth.

You could use two sprockets on the same axil to take up the backlash if you can't get the profile perfect---similar to an anti-backlash gear setup.

Yes, it is possible. The tooth pattern would have to be a roulette (not exactly cycloidal due to the point of concern changing as the point of contact moves) though and not an arc (a sprocket tooth is just two arcs). I'd say 3 would be the minimum number of teeth, or maybe 4 teeth.

I was thinking about the number of teeth on the sprocket as a whole but I agree 3 or four teeth in contact should be plenty.

Yamaha_R6M - Thank you for that excellent explanation. I had not considered those details at all.

Does this also hold true for a belt drive, such as an AT 5 or AT10 profile ? In other words, is the motion cyclic ?

Thanks

HarryN

Yamaha_R6M - Thank you for that excellent explanation. I had not considered those details at all.

Does this also hold true for a belt drive, such as an AT 5 or AT10 profile ? In other words, is the motion cyclic ?

Thanks

HarryN

I'll go out on a limb and say yes this holds true for belts as well. The profile on a belt is designed to rotate a tooth around the axis of the pulley. The only places where the belt translates is between the pulleys and when the belt engages/disengages the pulley. When the belt engages/disengages the tooth of the belt doesn't stay parallel to the line of motion prior to engagement or after disengagement; however, this doesn't matter because the device that is being moved should be connected between the pulleys and the motion of the belt between the pulleys is always along the same path and the bulk movement "should" be the same between the pulleys for a given rotation. Now, belts sag in the middle and there is a difference in tension on either side of your device so the actual velocity of the device wont be linear for a constant pulley rotational velocity. The belt will sag in a catenary if the belt is allowed to sag. More accurate belt systems will slide the belt along a "level" surface (any material supported between two points will sag). Chain and belt systems are great if you want to transfer rotational motion between pulleys (like on a bicycle) or if you want to move something between the two pulleys or sprockets (your grocery store conveyor); however, the motion at the pulley/belt interface or the chain/sprocket interface is not smooth.

If you try to roll a pulley on a flat belt then the pulley axis will follow a cycloid. The tooth profile must be altered to accommodate the rotation and translation of the pulley or roller pin or whatever you are using.

wow im actually extremely disappointed i saw this product and im trying to design and put together a business plan for making my machine... but dang i cant afford the nexen like i truely wanted if its costs that much... and less then a ball screw is bull i was looking at around 800 fo r my whole machine at 7'x3' travel. Im extremely disappointed now...

But now the question comes up... would it be possible to make your own roller and rack... having a machinest do it could you get it any cheaper then 800... is this even reasonable?

wow im actually extremely disappointed i saw this product and im trying to design and put together a business plan for making my machine... but dang i cant afford the nexen like i truely wanted if its costs that much... and less then a ball screw is bull i was looking at around 800 fo r my whole machine at 7'x3' travel. Im extremely disappointed now...

But now the question comes up... would it be possible to make your own roller and rack... having a machinest do it could you get it any cheaper then 800... is this even reasonable?

Would it be possible --- yes. For $800 --- probably not.

You "might" be able to find someone to CNC something for you for less; however, you'd be hard pressed to do so. I recently had a 4500psi manifold machined out of 4340 and the cost for the one off prototype was just over $3500. My company paid that; I never would have. I've found shops locally that are slow and negotiated pretty good deals with the owners to have work done for me. I had a wells link plate, rods for said wells link, lower and upper control arms and new perches machined for my car last year for the cost of the material+the CNC operators hourly+10% to the owner. I found a shop that was slow and worked the deal out. All the above cam in less than $1000. $800? You'd be hard pressed IMO. It also depends on your area too.

i was saying all that and now i remember my best friends father owns a business with laser cnc machines... and how much do you think material would cost... because i know he is really slow right now but i also know the machines cost alot of money to run lol so idk how that would work out but its a thought. i was looking at a rack and pinion setup with this one company and honestly it was rediculous what they quoted me... $750 for each split pinion for antibacklash i need for and then like $150 for every 1000mm of rack... is that not ridiculous?

This is the idea I was thinking about:

The chain is fixed straight and rigid in an extrusion. The axis is moved with a sprocket the shape of which is modified to suit the straight motion of the chain.

This is the idea I was thinking about:

The chain is fixed straight and rigid in an extrusion. The axis is moved with a sprocket the shape of which is modified to suit the straight motion of the chain.


thats a good idea... i think it would actually work but i think you can make it even better. by skipping the chain and maybe using bolts and metal rods with holes down the center and bolting them down a channel like that. theoreticallyt he same idea but the chain would move even if under tension and yea it would have to be adjusted and what not.

how do I delete this post? don't see any delete button, can only edit
thankx

Hello folks, nice animated gif Bill. (Fellow drafter?) I am constructing a table currently and found a 100' roll of rollerchain for $100 and am also thinking about this setup instead of $300 belts. With a chain there tends to be a little play between the roller and the pin however. Not sure how big of a problem it would be. If you were to tack weld the chain to a backing strip you might not need the enclosure and it also would eliminate the chance of the chain sliding around in there. Also, a guide bearing on the bottom would be a good idea (I think) to keep the sprocket from jumping up on a quick direction change. Kyle's idea in theory might work. Metal suppliers sometimes have pre-punched 1"x1/2" channel for building railings with 1/2" holes every 4" on center. If you had some of that punched with holes the entire length, and could somehow make a drive sprocket with conical rubber cleats (instead of bolts), it just might work although the cleats might get chewed up by the edges of the holes unless you were to spend alot of time deburring/softening the holes.


here are some links to a "negative" version of this idea:
https://sdp-si.com/eStore/Direct.asp?GroupID=1024
https://sdp-si.com/eStore/Direct.asp?GroupID=1025
not too much of a savings over the regular endless timing belts

here is a material that is very affordable per foot but might be too wimpy:
https://sdp-si.com/eStore/Direct.asp?GroupID=769
https://sdp-si.com/eStore/Direct.asp?GroupID=171

dreaming the impossible dream, bamwa ( prolly going belts, so far....)

ok this is the concept i was thinking of http://i2.photobucket.com/albums/y38/shelbygt500buddy/concept.jpg

also i found a rack that the grooves are round im sure making a roller pinion wouldnt be so hard.

I have been looking at threads concerning the Nexen RPS units. Assuming I'm willing to pay the price. How would be the best way to use the system? It's going to need some sort of reduction drive system to use economical stepper or servo drive motors. Thats going to introduces cogged belts or gears into the mix. Now you have potential backlash in a system that's sole benifit is lack of just that. So how could I direct drive this at lower the IPM that I typically use. Seems like I'd need to find the drive system first.
Anyone though this out?

I have been looking at threads concerning the Nexen RPS units. Assuming I'm willing to pay the price. How would be the best way to use the system? It's going to need some sort of reduction drive system to use economical stepper or servo drive motors. Thats going to introduces cogged belts or gears into the mix. Now you have potential backlash in a system that's sole benifit is lack of just that. So how could I direct drive this at lower the IPM that I typically use. Seems like I'd need to find the drive system first.
Anyone though this out?

The notched or timing belt drives actually work quite well. I have (3) 4:1 belt reductions on my burning table. I did use 3/4" wide belts & tentioned them only snug enough there isn't any whip in the belt. I can detect no backlash at all in the system. At least visually. Nothing shows up in the work as backlash either. Gears is another story as any standard gear must have some backlash to allow for heat expansion & lubrication.