Wear of Hardened Roller on Track

[bsmet95]

My company manufactures monorail type cranes and carriers using proprietary patented track. Some of our trolleys (carriers) have side guide rollers, See the attachment. We have a situation in which we are getting an unusual amount of wear on the guide rollers. two trolleys are tied together with a loadbar and the load is suspended from an eyebolt in the center. The trolleys in question have a 2500 lb. capacity per 2-wheels and are travelling around a large radius curve at about 75 FPM, pulled by a tractor. This system has been in use only a few months, with near constant travel in one direction only. The rollers contact the side of the rail but generally should not be seeing a load greater than 15% of the vertical load, on two rollers. Thus far the wear is approximately 0.001" deep.

Is there a way to calculate the approximate radial load which would cause the wear?

Thanks.

 

[SwinnyGG]

Which side of the curve are the worn rollers on, and which side is the tractor on? What's the radius of the curve?

If I'm envisioning this correctly, your guide roller load assumption of 15% of the vertical load does not account for any load applied by the tractor doing the towing.

Have you accounted for the fact that the tractor may be pulling consistently to one side, causing this wear? The tractor, even if it is small, is capable of applying MUCH larger loads than 2500 lb.

 

[MikeHalloran]

Is this a new design for an old company?

If so, the calculations you seek are probably on paper, in the drawer of an ancient desk.

... that was probably thrown away several managers ago, if you have MBAs at all.


Two things come to mind.

1. Lateral forces will tend to lift the inner vertical roller, allowing the rail to exert point contact, not line contact, with the guide roller, because there are no guide rollers contacting the bottom of the rail flange.

2. If the guide roller is a normal ball bearing, it's not intended for line contact on the outer race, which is normally supported by a bored hole in something. Using such a bearing in this way will flex the outer race too much, probably causing spalling of the contact area.
If the guide roller is a 'cam follower bearing', with extra thick outer race, then it should survive this service, so something else is going on.



Mike Halloran
Pembroke Pines, FL, USA

 

[bsmet95]

The rollers on the inside of the curve are wearing. Curve radius is 11'-0" on centerline. Rollers are custom-made for us, with a single row of ball bearings. I have layed out the trolleys, tractor, and curve and have calculated a side pull force. I neglected to mention that there is another 4-wheel trolley tied to the first one thru a loadbar, approximately 15 feet from the first one. As the trolleys are towed around the curve, this loadbar would cause more side pull due to chording across the curve.

Thanks.

 

[hendersdc]

While 75 ft/min is not very fast having the load attached by an eye bolt may allow the load to swing. Is it possible that the load suspended under the trolley is swinging outward as it goes around the curve? This could create the line contact Mike talks about.

I assume that you have looked at the track surface that the guide rollers contact and found it to be in good shape.

Are there any joints between rail sections where there may be a step or gap that could be taking small bites out of the rollers?

 

[desertfox]

Hi bsmet95

There is clearance between the rollers that are wearing and the track, so how do you know the rollers are actually rotating when and if they contact the track, is it possible that the rollers are sliding against the track instead of rotating. If the wear is even then clearly there is no sliding but I cant see any force applied to those rollers that ensures contact with the track.

“Do not worry about your problems with mathematics, I assure you mine are far greater.” Albert Einstein

 

[btrueblood]

Is the edge of the lower flange left in an as-rolled condition, or is it polished smooth? If left as-rolled I could see the oxides on the surface doing some abrasive wear on the rollers.

Got any pictures of the rail edges, and wear on the rollers?

 

[dhengr]

Bsmet95:
Where is the tractor w.r.t. the trolley you show, and is its angle of pull and loading constant? How is the tractor guided so its angle and pull load are constant or is it driven around the plant by a drunk? The 15' tow bar to the second (trailing) trolley will exert one hell of a side load on both trolleys on a 22' dia. circle of travel. That’s nearly a quarter of the way across/around the track circle. You might be better off to pull on the load, down near the ground, and let it exert whatever gravity and lateral loads to the two trolleys. You really need to describe this system better, we can’t see it from here, even if you can. Show some side views, end views and plans of the whole system and load and tractor, with some dimensions, etc. The way I’m visualizing this system at the moment, I would expect the wear to be on the outside guide rollers, or on diagonally opposite rollers when this is being pulled around the track. What kind of wear do you have on the trailing trolley?

 

[bsmet95]

The attachment is all I can provide. There are straight sections and curves on the monorail. The angle of pull varies as the carrier negotiates curves. The tractor has 2-wheel trolleys with the same guide rollers.

It seems as if the wear should be on the outside rollers due to side pull, but it's happening on the inside rollers.

Thanks.

 

[Nescius]

Shot in the dark: Vertical scrubbing action of the side rollers against the flange edge.

Perhaps the vertical, load-bearing rollers are out of round, damaged, or have debris stuck to them. Perhaps the track running surface is wavy, or fouled with debris. How about a poorly fit joint causing the vertical action?

 

[WKTaylor]

bsmet...

RE You roller bearing outer races...

Do the rolling surfaces have a dead-flat rolling surface or a crown-radiused rolling surface?

what is the hardness of the rolling surface?

Have You considered applying composite wear-strip material bonded to the track?

Have You considered a 'mohair sweeper' or squeegee running ahead of Your rollers in-contact with the track, to 'clean-off' accumulating dust/debris/crud?

Regards, Wil Taylor

o Trust - But Verify!
o We believe to be true what we prefer to be true. [Unknown]
o For those who believe, no proof is required; for those who cannot believe, no proof is possible. [variation,Stuart Chase]
o Unfortunately, in science what You 'believe' is irrelevant. ["Orion", Homebuiltairplanes.com forum]

 

[IRstuff]

Are the trolleys truly able to swivel? If the load bar was forcing the trolleys to stay in a straight line, there might be an expectation that the inside wheels would have more force on them than anticipated.

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert!

https://www.youtube.com/watch?v=BKorP55Aqvg

faq731-376 forum1529 Entire Forum list

http://www.eng-tips.com/forumlist.cfm

 

[desertfox]

Hi bsmet

How well tolerance is the hole centres on the load bar? Do you have a loci diagram of the load bar and trolleys moving round the track which shows the position and angles of the load bar as the tractor pulls them round. Like IR states that load bar might be forcing the trolleys apart.

“Do not worry about your problems with mathematics, I assure you mine are far greater.” Albert Einstein

 

[romke]

Normally a roller on a rail should be a type of line contact, when both roller and roller are sufficiently strong. However, when sufficient load is applied, the linecontact will temporarily be deformed to a elliptic contact - and some sliding will occur. In rollerbearings the negative effect of sliding on wear is counteracted by the use of a lubricant separating the two surfaces from another. Have you considered lubricating the contact between roller and track?

Letting the roller "rub" against a block of solid lubricant (somewhat like the brushes used in electric motors) like MoS2 or graphite so that the roller gets coated with a thin layer of solid lubricant that then also can be transfered by the roller to the track might well reduce wear to a more acceptable amount - starting out with not yet worn rollers that is.

 

[kiwitom235]

is the load swinging by any chance? the tractor speeding up and slowing down, or the tractor turning left and right would cause different speeds on the load... this would cause the centripedal force on the load to change and lead to lateral movement... this could cause the roller to hit against the rail...

Can you not video the roller as it moves along the rail?


Tom

 

[LittleInch]

bsmet 95

If I read this correctly the 4 wheel trolley thing has two sets of axle wheels but these appear to be fixed to each other??

If so then you will get wear on the inner rollers at either end because you're forcing this trolley round the curve.

At your allowance of drift, the rollers only just allow that radius. Imagine if the radius was half - the trolley wouldn't go round it if the two axles are rigidly connected to each other.

You either need to
increase the allowable sideways movement or allow the axles to rotate on the 4 wheel trolley thing to follow the curve.

In terms of sideways force the force will depend on how much play each set of rollers has and how much into the curve it takes before the trolleys tighten up and you need to start forcing it round.

I think maybe you've got lucky that the radius and the sideways gap on the side rollers just matches. Any tighter and I think your trolley wouldn't run around the rail or would start to jump upwards.

As said this is all premised on the 4 wheel trolleys being rigidly attached to the two wheel bogies.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[bsmet95]

The individual 2-wheel trolleys are articulating. In theory both rollers on each trolley would contact the side of the rail. Sorry, my layout doesn't specifically show that.

 

[MikeHalloran]

Have you, personally, been to the installation and witnessed it in operation?

Is the 'tractor' really pulling, or is it pushing?



Mike Halloran
Pembroke Pines, FL, USA

 

[btrueblood]

Ok, thought experiment: without the side rollers, what happens to a single bogey trying to negotiate the curve? Because your load wheels don't have a conical shape (that I could see from your pictures/drawings), they will try and roll in a straight line until they bump/scrub the beam web. The side rollers prevent this from happening.

I'd argue that the rollers do all of the steering work for the system, and that this steering work is sliding friction (think skid steering like a tracked vehicle) of the load wheels against the rail.

I.e. the side rollers see some distribution of mu*load, where mu is the coefficient of static friction of what appears to be bare steel on bare steel, and this (from various sources) could be anywhere from 0.5 to 1.2, but is usually quoted at around 0.8.

You could reduce the effect somewhat by putting a conical taper on the load wheels, so that lateral shift of the bogeys allows them to self-steer somewhat, like a railroad wheel set. Or add lubrication on the horizontal web/wheel interface.

 

[dhengr]

Bsmet95:
I still can’t see it from here, we need much more meaningful info. The photo would be much better if it showed everything below the bottom of the current photo on down to the load. The way I see your underslung loadbar and eyebolt arrangement, as little as you’ve shown, you are probably applying that nasty lateral pulling load, from the long loadbar dragging the trailing trolleys along, about 12"+ below the top surface of your patented bot. flg. It is about 4" from that top surface on the bot. flg. to the top of the wear surface on your 3" wide ‘U’ bar from the load wheels. Then you have some structure, “eyebolt connection,” tying two, 2 wheel trolleys to make up a 4 wheel trolley, another 3-4" lower. Then you have another structure, “loadbar for 8-wheel trolley,” another 4-5" lower. Then you finally have the long loadbar, another how many inches lower. Pulling that 8 wheel trailing trolley around the circle is going to take a serious load. You have the regular rolling friction of the 8 wheels, plus a lateral rolling frictional component (beam flg. to wheel), plus a frictional resistance component due to torsional tipping of that whole trailing trolley system. The sum of these is the long loadbar axial load, and it is applied 12"+ below the top surface of your patented bot. flg. This rolls the whole 8 wheel front trolley on the main beam and imparts a torsional loading in the main beams too, to roll the beams a bit.

Your max. gravity loading is potentially 20kips = (8)(2.5k). I think that the lateral loading and its twisting (torsional) action on the 8 wheel trolley systems, tips that system enough to put much more loading on the 4 outer wheels. This also torsionally twists the main beams a little. Then, the outer wheels track in a way that pulls the whole 8 wheel trolley to the outside of the curve, which brings the inner guide wheels into almost constant contact with your patented bot. flg. I understand that your patented bot. flg. is hardened, one of its selling points. The hardness of the guide roller should match your patented bot. flg. But, do you see any wear on its upper, inner flg. corner or inner edge, where it is wearing on your guide roller? You should not have a 15' chord length (long loadbar) on that system on a 22' dia. curve if you want it to work well. A slightly longer loadbar would almost equal the circle dia., then you wouldn’t have any trailing trolley pulling component from the long loadbar. Can you design the long loadbar so there is a 3' canti. at each end and a 9' middle span? This might pull and track better and would cause less lateral loading on the trolleys. But, again, we can’t see any of this part of the whole system, so only you know how it works; except, apparently not very well. You might also consider just putting a second tractor on the trailing set of trolleys, and run them in unison off the same controller.