View Full Version : suggestions

06-06-2010, 08:58 AM
I have to make a wheel for an encoder. The wheel will count distance on sheets of steel. I was thinking an aluminum wheel with a rubber coating on the outside. I am just not sure what rubber i should use and how i should mount it. Any ideas ?

Rob :)

06-06-2010, 09:21 AM
Hot steel, cold steel, oily steel,wet steel, big pieces, little pieces? With ALL THE DETAILED INFORMATION provided, I would pick Grandma's old faithful. neoprene. Want to get fancy try silicone or nitrile.
By the way, how long is a piece of string?

Peter S
06-06-2010, 09:26 AM
How about an o-ring fitted into a groove? Or several o-rings side by side if you need width. A stretch fit and maybe a bit of super glue to make sure it doesn't slip.

You can also buy o-ring material and make your own lengths up using super glue.

Urethane is a good alternative to rubber when it comes to getting tyres made, a range of hardnesses available. A lot more expensive and difficult than an o-ring though!

Edit: I was referring to getting urethane tyres cast, but now I recall we used urethane cord to make tyres also, you can buy this cord in a range of diameters (larger than o ring material, e.g. 6mm, 8mm, 10mm and up), but you need to hot melt join it. It requires a special tool to do it properly, and it is difficult to get the join absolutely smooth. Suppliers usually make to order. As used for light driving bands etc.

06-06-2010, 09:54 AM
Well i gave you pretty much what i have. I do know it is a steel froming company. They convert sheets of steel into steel studs, in some machines, in others they do other stuff. Oil is used to lubircate the forming tools, but i was not told if the encoder is going on the in or the out of the machine.

How would you mount the neoprene?

Peter S

Cool idea. I did not think of that. I would be concerned that the o-rings would slip but i have used some glues, especially that o-ring glue, that really bonds. I like that idea. The customer could change them easily if they break.

Rob :)

06-06-2010, 11:44 AM
My first impulse would be to use a commercial encoder from BEI or similar. That makes life MUCH easier for the poor schlep who has to repair the machine ten years down the road.

Second to that I'd, make the disk out of black delrin.

06-06-2010, 12:48 PM
I would think a commercial wheel built to known dimensions (to ensure a source of replacement parts) would be a good idea. Skateboard wheels come to mind as a family of readily available wheel. You may require a larger scale but those are also available. Once you have a replaceble wheel of known specifications you need only adapt your encoder to the wheel.

06-06-2010, 02:28 PM
You would think the motor moving the sheets of steel for the steel studs would be the most likely candidate to receive the encoder wheel - and for that there is a wealth of ready made / off the shelf encoder wheels ready for your application

06-06-2010, 02:29 PM
Trouble with using a wheel of whatever diameter is what tolerance on length is acceptable for wear on whatever facing due to friction??Would have thought that an adjustable length stop was more acceptable.

Regards Ian.

06-06-2010, 02:34 PM
What's to be the diameter of the wheel- Maybe convert a hockey puck. That's some pretty hard rubber. Should wear well.

06-06-2010, 02:54 PM
Wear and slippage are problems here.

A 1mm error in diameter will be 1 x pi = 1 x 3.1416 = 3.1416mm in circumference and linear travel.

The smaller the wheel diameter the larger the number of turns of the wheel in a given length.

There is no allowance/s for wear and slippage.

Inertia may need to be considered as well.

The tolerance to length needs to be considered.

If the "studs" mentioned are wall studs, I'd guess that cropped/finish length tolerance/limits will be pretty tight as regards length per any individual pieces and for a whole batch. I'd imagine that a 10 foot stud would need to be in the order of 120 +/- 0.005" = 119.995/120.005".

As I understand it, an encoder will only measure angular step-to-step and this will be translated into the linear equivalent so the linear tolerance zone will need to be allowed for in the roller/wheel sizing and tolerancing and encoder step angular displacement.

I'd imagine that there will be some guaging and limit stops for setting and stopping as well.

doctor demo
06-06-2010, 03:31 PM
Well i gave you pretty much what i have. They convert sheets of steel into steel studs, in some machines, in others they do other stuff. i was not told if the encoder is going on the in or the out of the machine.
Rob :)
On the converting of sheet to studs, is the material coming off a coil and being cut to length or is it actually sheet steel?
The other question I have is, is the forming process a press brake or a roll mill?

The Co. that I do work for from time to time has three roll mill lines and a press brake. On the roll lines the material comes off a coil then goes through a straightener . The material length is measured as it comes off the straightener and is then cut off with a flying cut off that is in front of station one of the roll former.
I added a disc brake to the last station of the straightener on one of the lines recently and also worked on the clutch for the flying cut off, but did not pay any attention to how the length measuring is performed. I don't recall any encoder wheels on the material though.


06-06-2010, 03:34 PM
Comes to mind that you could fashion a two sided wheel on which to mount an o-ring. Side one would have the hub and the main groove to mount the o-ring, plus whatever encoder pattern, and the second side would tighten up against the side of the o-ring. You'd assemble with a certain amount of tightening to begin with, then surface grind the o-ring to a specified diameter so it can roll out a distance accurately. As it wears, tighten the thing a tad and check it against a known length from time to time.

06-06-2010, 10:31 PM
thanks guys i have been reading the responses.

I will answer a whole bunch of questions with one simple answer. This is not my design. The elctrician i get alot of work from has asked me to build a wheel for his Omron E6C2-C encoder.

I want to do the best job for my customer with the limitations they have put on me.

I am hoping to find something at the local supplier i can buy or easily modify. I wanted feedback from the board so i have a better idea of what to look for.

Thanks for all the replies. I have even considered a knurled wheel. I just don't want it to damage the metal going in or coming out.

it is a roll forming machine, with the metal coming off of a roll. If it is the machine i think it is.

Again thanks for the feedback. any more suggestions i am all ears

Rob :)

J Tiers
06-06-2010, 11:06 PM
I'd be concerned about oil.....

it makes the wheel slip, and it eats some rubber compositions, and swells up others. Either problem affects accuracy.

Since all you need is to make the wheel, find a material that is the least sensitive to oil, and preferably get it in a standard form, like an O-ring, etc.

A toothed whee;l would slip less if spring-loaded etc.... I doubt it would damage the material free-running, but you might ask. It would need to be hardened within an inch of its life though, as it is harder to replace, and wear would be bad.

06-07-2010, 12:27 AM
Something else comes to mind- if there's going to be a problem with the point contact of the wheel slipping, you could always use two wheels- cogs, actually- and run a toothed belt around them. The belt is what rides on the steel you are measuring. Of course, this method is prone to debris getting in the teeth, and it may not be easy to get it calibrated.

J Tiers
06-07-2010, 12:36 AM
If, as might be expected, the wheel is one of a pair that "pinch" the metal, the pressure may change the effective radius of the wheel..... if there is a rubber component involved.

I'm starting to like the sharp-knurled wheel a lot better, but it may not fit the existing device setup.

Did you get a particular OD and tolerance?

06-07-2010, 01:13 AM
An entirely diffrent method might be desireable here if accuracy is critical
For example, a common digitial caliper/height gauge or digitial TDI with the output port could be used, Just attach it to a 10' (or however long a standard stud is) jig with a stop at the end. Have the stud hit the stop and have the jaw 'come in' to measure.

Or the entire jig could drop down and just push against the beam as it glides along, using pnumatics to get a constant pressure without caring much about actual location.

06-07-2010, 01:35 AM
Use a magnetic wheel. It will eliminate slip.

Something like this:

The large wheel is the stock contact wheel. The small wheel is a swarf collector. It has stronger magnets than the large wheel and picks off swarf.


06-07-2010, 03:16 AM
Use a magnetic wheel. It will eliminate slip.

Something like this:

The large wheel is the stock contact wheel. The small wheel is a swarf collector. It has stronger magnets than the large wheel and picks off swarf.


so how does that work then :confused:

06-07-2010, 04:44 AM
Worked for a company that had a ten station roll former and the cut to length problem was an on going candidate to "Electronickise". Slippage and repeatability always came into the equasion and the solution we kept coming back to was a guillotine mounted on a sliding carriage triggered via an adjustable length stop. When material hits stop, the guillotine carriage slides in sync with the material flow, crops , length ejected sideways and carriage returns to original position as material is still feeding.

Regards Ian.

Peter S
06-07-2010, 06:41 AM
I would be concerned that the o-rings would slip but i have used some glues, especially that o-ring glue, that really bonds. I like that idea. The customer could change them easily if they break.

Now that I think about this, it seems unlikely an o-ring will slip on the wheel, it will be gripping over quite a large area on the circumference. It is probably more likely to be slip between o-ring and steel sheet? Perhaps some spring pressure or weight to get extra traction if required (though you would have to consider what the encoder bearings can handle)?

06-07-2010, 08:02 AM
Seems some form of non-contact encoder is going to be required to eliminate the issues of oil, side loading, slippage etc, do a search for non-contact encoders.

Here is a couple


J Tiers
06-07-2010, 08:40 AM
Circlip has the general industry solution....

The sliding cutoff is as much to prevent material piling up (crumpling) while the jaws are closed, as for anything else, but the length measurement is fairly foolproof, does not depend on speed, friction, or whatever.... Just 'see" the material get to the right place, and whack it.

In this case, the design is NOT the OP's, however..... it's a "Milacron minute"...... "just take it from me that it HAS to be done this way... now what's the best VERSION of this way?"

That's what the OP is asking.

Smart or stupid, it doesn't matter.... that's what his commission is. A wheel.

06-07-2010, 11:58 AM
I am not a fan of encoders. They limit the operators, and on site machanics in small shops from being able to adjust and set up their machines. Programers keep coming back to modify the program as materials and demands on the machines change. YES they have their use, but being high tech, because you can be...is not always the best option. I would be using a light, prox switch, or something like that. The electrician could be using the enocder as a reference. AS in, when we have 80% of the stud made, slow down to 60% so the cut will be better. Honestly i have NO CLUE

I did not think of the oil slippage problem. Wow what a catch 22 that is. I have no clue where the electrician is going to put his wheel, and seeing as he accounts for 50% of my revenue, i figure i would sit back and watch. I might learn something. I did ask for a diameter, and said not to worry about it. He would adjust the program for the diameter.

So here is what i Orignally thought.

1. Aluminum base with a rubber rim for contact and avoid slippage (i did not think of the oil). I like the O-ring idea. Easy for the in house mechanics to change.

2. Plastic wheel. It would be lighter and easier on the encoder bearings but i was worried about longevity. I could even have one printed out on one of those 3D printers. Hehe that i like

3. Caster wheel machined to accept a hub that fit on the encoder. Again worried about weight.

4. Fiber glass. Make a hub and then pour a fiber glass mould. Machine it down. It would be pretty light and strong for its weight

5. Purchase something...but they will want me to come up with the material/diameter. I figure bout 4" AS the encoder is 2.5" in Dia and you will have to have a mounting brakcet.

6. Aluminum wheel with a knurling on the outside. I am not sure how long it will last though.

What do you think of my ideas?


Rob :)

P.S. Thanks for all the replies.

06-07-2010, 12:21 PM
J Tiers,

Thanks for the back-up. Several where questioning the idea. Not my idea, just my problem! haha


I beleive there are slididng guilatines already on the unit. I don't know which one it is. I have only been to the shop twice. To work on a brake for one of the rollers. Neat to watch everything work...and loud!!


Thanks for those links. Cool. i will definatley forward them to the electrician.


Neat but i think that is a little pricier then they want to pay :)

Rob :)

Thanks for the feed back :)

06-07-2010, 02:19 PM
OK Rob, they don't want to/can't afford to (Cheapskates) do it properly. Two wheels with an "O" ring as the tyre, sprung loaded together with the sheet "Nipped" between them.??

Regards Ian.

J Tiers
06-07-2010, 10:50 PM
if you knurl, I'd NOT do aluminum......

hard tool steel is the ticket...... and sharp knurls. You want it to bite a bit into the zinc coating on the studs. They'll never notice or be concerned about the marks, they won't be deep because there should be minimal drag on the wheel.

I bet their feed wheels make bigger marks, I think I have seen them on similar products.

backup wheel opposite it, I hope.......... just for consistent pressure