Advice on gear material choice

[randgust]

Back 'in the day' the Trix F-units and all diesels had all brass worms, a plastic worm gear idler, and all brass gears in the trucks.   Pretty much standard Trix design across all platforms.   Rapido GP's used a big brass worm gear on a brass worm, a plastic cup gear, brass in the trucks.   The Atlas/Rivarossi SW1500 was another brass geared wonder that when worn in and well lubricated, was a pretty good drive.

I lubricated the bejeesus out of them and on the brass gears, the vintage sticky gear lubricant of choice was vaseline applied with toothpick, with Labelle 108 on the shafts.    I never had trouble with the vaseline, but it did have an interesting habit of 'preserving' the particles of brass that would wear off the gears.   On the annual 'clean & lube' teardowns I used to do on those you could see what had come off the gears, and it was interesting as it was obvious there was gear wear going on.   Now, as those Trix dies aged, the gears got worse, tolerances got worse, and the new ones got noisier (model power era into Con-Cor).  I never had one actually wear out, but I was pretty disciplined about maintenance, too.  But I'll admit I had the brass gears pretty much swimming in stuff, too. 

In fairness, I've had a lot more trouble with some of the fine-toothed, poor-tolerance, cracked gear problems with plastic than I ever had with those brass gears.   The very worst were the 'red gear' Roco 70's ones, that was a soft plastic than a nylon, and the F/GP worm gears were more like little spikes on wheels.   I bought bags of those from Atlas as they'd wear out pretty much no matter what you would do.  My N Heisler is the only survivor of that drive, and I still replace the worm gear about every three years, it begins to misbehave on curves and bind up due to wear on that gear.

[peteski]

In fairness, I've had a lot more trouble with some of the fine-toothed, poor-tolerance, cracked gear problems with plastic than I ever had with those brass gears.   The very worst were the 'red gear' Roco 70's ones, that was a soft plastic than a nylon, and the F/GP worm gears were more like little spikes on wheels.   

Was wear of the plastic problem, or did  the gears just split over the metal axle (like the Bachmann Nylon gears do)?

[randgust]

It takes virtually nothing to get into fine-tooth gears to cause infinite grief, and if it does, the likelihood of damage by running with stuff in there is a lot higher. I've had to replace some Atlas shay gears due to damaged teeth resulting from debris pull-in, also had the 'atlas clicks' on the SD50 chassis become really unbearable to the point where I swapped out the entire chassis.    I really think that tricks like taping-off the bottom of newer Kato trucks is a good idea to prevent debris infiltration on those finer teeth.

I had a 'nearly never used' Bachmann doodlebug actually have the wheels fall off, due to the white gear material, and also have the gears crack on the same  chassis. 

[peteski]

It takes virtually nothing to get into fine-tooth gears to cause infinite grief, and if it does, the likelihood of damage by running with stuff in there is a lot higher. I've had to replace some Atlas shay gears due to damaged teeth resulting from debris pull-in, also had the 'atlas clicks' on the SD50 chassis become really unbearable to the point where I swapped out the entire chassis.    I really think that tricks like taping-off the bottom of newer Kato trucks is a good idea to prevent debris infiltration on those finer teeth.

I had a 'nearly never used' Bachmann doodlebug actually have the wheels fall off, due to the white gear material, and also have the gears crack on the same  chassis.

Some of the Kato's newer Diesels have gears with teeth finer than Roco, and while they are prone to clogging up, I  have never seen any broken teeth on any I serviced (yet). Yes, some years ago (after seeing the gunk between teeth) I came up with the idea taping the bottom of those Kato trucks.





Bachmann cracked gears can be successfully repaired too.  https://www.therailwire.net/forum/index.php?topic=49440



You must  have had lots of coarse loose ballast, and the Roco's red plastic must have been very soft!

[mmagliaro]

As far as I know, the Kato stuff is all Mod 0.3, except the GG1 and some SD40-2's, which are Mod 0.2.
Mod 0.2 are great for getting speed reduction in a small space, but I can see how their "fine-ness" might not
be a good choice for a commercial model.

[woodone]

Any one tried using UHMW material- there are several types that look like it would be good for worm wheels driven by brass worms?

[narrowminded]

Any one tried using UHMW material- there are several types that look like it would be good for worm wheels driven by brass worms?

That stuff might be great but when you machine it it leaves burrs and they are impossible to de-burr.  A flame might do it but I actually redesigned a part to be sliced in a guillotine type shear instead of machine facing and then trying to de-burr.  When you machine it, file it, sand it, it leaves "hair" hanging off the edge.  You remove some with a file or sanding but as it's being removed, more is generated.  You could sit there and repeat that until there's nothing left.   And when you flame de-burr it can look OK cosmetically but if it's a machined detail exit that matters, the flame doesn't cut or burn away the material but basically melts the small "hair" and because it does, that can leave the burr material melted at the edge which then has the effect of leaving a "rolled" edge, screwing with the entry and exit dimensions.  It might be made to work but be prepared for some serious screwing around dialing it in.  This is why Delrin (Acetal) is such a good choice.  It has much of the positive properties with almost no downside. 

Now, if you need an abrasion resistant surface, like in pneumatic or slurry conveying of sand, fly ash, etc, then UHMWPE will outperform hardened steel and just about everything else you can think of.  It's amazing stuff.  But I don't think the worst ballast disaster imaginable would warrant that.

[mmagliaro]

Soooo.... a replacement Arnold brass gear is in Lee's engine.  After a little running, there's brass dust, even though the gears have oil on them.   There's no way the wear should be severe enough to see brass dust after just a little bit of running.   I wanted to open this up to the community for suggestions.

My first thought is that perhaps the idler has a few rough or sharp spots - maybe only on one tooth, and that is chewing up the worm.  Maybe that's how the worm got destroyed in the first place.  These Arnold 4-6-2's abound and they lasted decades, so although brass-on-brass may not be a good idea, there is no way the wear could ordinarily be this bad or none of them would have lasted a month out of the store.

I'm thinking run some strips of 400 grit and then 1000 grit sandpaper through each idler tooth just to take off any sharp edge and see how that goes.

Opinions?

[peteski]

Are you sure that it is a tooth on the gear that is "chewing up the worm", and not the other way around (worm grinding down the teeth)? 

I'm not a metallurgist, and I don't play one on TRW, but I'm pondering whether there are different types of brass alloy, and some are softer than others?  If that is the case here, it is possible that whichever part is made from the harder alloy is chewing up the softer part. 

When I serviced Cody's brass PRR H-10, it the (lubricated) gearbox contained a brass worm and worm gear.  The worm was not too worn out, but the worm gear had its teeth heavily worn down (the surfaces of the teeth were worn away so much that the tops of the teeth were sharp points.

[woodone]

There are a bunch of different brass types . Some will be harder than others. One thing used in a brass make up is lead, which makes the brass machine very well, I bet it may be softer than sone others.

[narrowminded]

There are a bunch of different brass types . Some will be harder than others. One thing used in a brass make up is lead, which makes the brass machine very well, I bet it may be softer than sone others.

Yes, there are.  And dissimilar hardness would be an appropriate pursuit if looking for improved life.  This is a standard approach in any bearing application, two metals mated to each other, but I'm not one to chase this into the weeds as I don't think there is enough of a hardness difference in brasses to make a huge difference in service life.  When there are long known issues with similar, non-hardened materials (brass, mild steel, aluminum, etc) making lousy bearing mates and good mating materials identified that are the results of proper research by the manufacturer of the product(s), I see no point in starting a new design that proves again that it's a bad idea.   

[narrowminded]

Are you sure that it is a tooth on the gear that is "chewing up the worm", and not the other way around (worm grinding down the teeth)? 

I'm not a metallurgist, and I don't play one on TRW, but I'm pondering whether there are different types of brass alloy, and some are softer than others?  If that is the case here, it is possible that whichever part is made from the harder alloy is chewing up the softer part. 

When I serviced Cody's brass PRR H-10, it the (lubricated) gearbox contained a brass worm and worm gear.  The worm was not too worn out, but the worm gear had its teeth heavily worn down (the surfaces of the teeth were worn away so much that the tops of the teeth were sharp points.

It's the logical result of the gear action.  Every tooth and every square inch of surface area on the worm and worm gear sees the same engagement in the ratio of the gear set.  So if it's a 14:1 gear, the worm's surface area sees at least 14x the surface engagement with the like material.  This is even worse in practice due to the fact that a minimum sized worm gear will see at least three teeth engaged per revolution of the worm and aggravated by the involute form of engagement (reduced surface area), but for logical deducing, trying to get a grasp on the picture without using decimals the basic idea holds true.  This is why you would usually use the harder material in a set and in the case of this worm gear set, with little load, the worm should be at least mild steel. 

Another possibility would be to investigate bronzes.  That may be what is done in some of these pairings that folks say have worked adequately for them.  Brass and bronze look pretty similar from a galloping horse but are in fact two very different, up to and including very VERY different, materials.  Look up some of the specs for bronzes and then go another step and look up aluminum bronze or some of the "Ampco" bronzes if you want to see this.  "Self healing" is a term that can surface when describing their properties.  Then try machining them.   And the first time your drill becomes pressed/ seized into the hole you're drilling you'll never see them as similar again.

In my automation business we regularly machined double start acme threaded nuts in aluminum bronze that was mating to SS threaded stems with thread engagements at 4x to as much as 8x the stem diameter.  (In a power screw the thread becomes a bearing, not a nut.)  Looks like brass but I assure you, if you were in doubt, one light pass with a cutter is all it will take to know it is NOT brass.

[woodone]

7675
That stuff might be great but when you machine it it leaves burrs and they are impossible to de-burr.  A flame might do it but I actually redesigned a part to be sliced in a guillotine type shear instead of machine facing and then trying to de-burr.  When you machine it, file it, sand it, it leaves "hair" hanging off the edge.  You remove some with a file or sanding but as it's being removed, more is generated.  You could sit there and repeat that until there's nothing left.   And when you flame de-burr it can look OK cosmetically but if it's a machined detail exit that matters, the flame doesn't cut or burn away the material but basically melts the small "hair" and because it does, that can leave the burr material melted at the edge which then has the effect of leaving a "rolled" edge, screwing with the entry and exit dimensions.  It might be made to work but be prepared for some serious screwing around dialing it in.  This is why Delrin (Acetal) is such a good choice.  It has much of the positive properties with almost no downside. 

Now, if you need an abrasion resistant surface, like in pneumatic or slurry conveying of sand, fly ash, etc, then UHMWPE will outperform hardened steel and just about everything else you can think of.  It's amazing stuff.  But I don't think the worst ballast disaster imaginable would warrant thatM.

There seems to be a science to tool sharpping when marching UHMW PLUS RPM & feed rates, but like you said why bother when yo can get Delrin.

[narrowminded]

There seems to be a science to tool sharpping when marching UHMW PLUS RPM & feed rates, but like you said why bother when yo can get Delrin.

Actually, there is an art to tool sharpening for all materials, extremely sharp being one of the details for most plastics, but that's not the crux of the issue here.  The same characteristic of the material that makes it so uniquely capable of abrasion resistance is what causes it to be so miserable at deburring time.  Fool around with a piece and you'll see the problem in a hurry.  Make sure you de-burr all of the edges, pockets, drilled holes, thread crests, etc. 

Better than taking one guy's experience with the material, do a search for "deburring UHMWPE" and pick one.  Line after line of machinists sharing their frustrations with the stuff and some suggestions of how to go at it.  Not how to make it easy but how to force your way through if you absolutely must do it.   They include the flame I mentioned and then get into cryogenic solutions including entire operations that offer this as a service.  Easy edges would be one thing but parts with various features (like gears) don't always present the solvable issues, like specific tool sharpening.

This link was the very first one that came up when I did that.  It pretty well outlines the issues and describes pretty much exactly what I did, too.  The flame is the most likely to be useful for many jobs on short run parts but if it's a critical detail edge, like the crest on a worm, it might tend to make the part bad when it melts the "hair" at the edge.  But a change to an equally or totally adequate material eliminates all of the difficulty described and is why UHMWPE doesn't even make my list for a machined gear material.  Great stuff for some jobs, just not this one.  Now, if you did get a part out of it (it can be molded) I can almost guarantee it would work just fine.

Enjoy: https://www.practicalmachinist.com/vb/general/deburrig-uhmw-315184/

[mmagliaro]

I can get a steel mod 0.4 worm with a 1.8mm  bore that should fit well in the Arnold steam loco.  My only concern is not understanding WHY there is so much wear on the brass worm in the first place. 

In the back of my mind, the nagging worry is that the steel worm may not wear out, but it might start wearing out the worm gear instead.

Regardless, putting a steel worm in there will probably fix the problem.   I'm heavily leaning in that direction because I'd like to fix this thing with a sure-file "kill" and not have it become a science project.