1941 City of Los Angeles

[Jay Criswell]

.....and some, not so wise!

Lateral? I don't know, .010" to .020". Just enough so nothing binds.

Distance between gears is far more complex. It depends of the type of gears. There are specific formulas for making that determination. Tons of info on line if you really want to dig into it.

I work with a guy, Greg Scott, with American Gears that may be be able to help. Another company that I've dealt with is QTC or Quality Transmission Components. You might check their website; qtcgears.com. I have one of their catalogs/handbooks (they call it their Product Guide & Technical Manual). It probably has everything and more than you could ever want but this one is for metric gears so it may not apply.

Jim, to be honest, there are so many suppliers out there you just need to find one that can take your gears and, possibly, provide off the shelf replacements. I truly believe the builders of these models used gears that were readily available. I'll bet Boston Gear was a supplier to many of them. They're still in business too.

Jay

[ScaleCraft]

Axle play. I am surmising you want to know vertical and not horizontal.
The weight of the unit when on all wheels on track should eliminate any concerns about axles dropping...until you need one to drop for contact due to a horrible low spot in your rails (H0 stuff, if you read up..when expensive locomotives started into driver springing, mass-produced decided on vertical play on drivers to compensate, and that remains today).
The test if to see how much lash you have betwixt worm and worm gear with truck and wheels on rail and a comparable weight applied to the truck to simulate locomotive body and any frame.

[Jay Criswell]

Jim,

I'm not sure I totally agree with Dave. Don't get me wrong, he's a knowledgeable guy and I listen to what he has to say.

If you're referring to the clearance where the axles ride in the frames I would say .001 to .002"". The tighter the better (without causing a bind). A couple of reasons for my thinking. First of all, because you're not using a conventional gearbox where the gear mesh never changes and allows the axle to move up and down you need to keep what you have as close to ideal as possible. If your axle has too much play and flops around, so to speak, you'll eat up your gears prematurely. Gears don't like to have the alignment or mesh changed while in operation. The other issue is, even with the best tolerances achieved, materials, lubrication, and maintenance you'll experience wear over time. This alone will take your .001" and make it greater.

Just my thoughts, take them for whatever you think they're worth.

Jay

[R.K. Maroon]

Thanks, Dave and Jay, for that input. I think the action on this may be at the intersection of your talking points. We know that a rigid-frame truck will tilt going through a curve due to the coning of the wheels. If there were no side play in the axles, a three-axle truck could not keep all six wheels on the rails as it tilts over. Even with side play, it is not clear that all wheels make contact and roll without slip. I can't make a direct argument why this would, but this effect may explain why one sees the middle gear on these stripped but not the two outer, as with the one on the center-left here:



And that may be the reason why Adams at some point made their drives A-1-A. This one has has a sprung center axle, while others I have just have a lot of play in the journals:



So while I am reluctant to give up the extra straight-line capacity of the gears, it may be best to leave the center gears off, tighten up the outer journals (per Jay) and leave some play in the center axle (along Dave's line of thinking) so the truck will correctly roll in a curve.

Jim

[ScaleCraft]

I used to think gears had to be fully meshed at all times (and still do on separate gearboxes). While I try, one of the things you usually see is worm on top of worm gear. When the axles/bearing surfaces wear, the mesh decreases, hopefully to the same or similar rate that the gears themselves have worn.
What actually caused me to start thinking out of the box a bit was...winter of 1975/76, part of a large layoff in an avionics plant, went to work at a hobby shop doing repairs.
I was handed a GEM H0 0-4-0 with the valve gear off one side.
Fixed that, figured I'd break it in a bit (new loco...damaged in shipment).
Fitted it into my cradle upside down, and applied power. Had this funny noise, no drivers turning. Gee, thought I, I just ran it on the track and it was fine.

They had so much vertical play in the driven axle journal (all of them, if I remember right), that without the loco weight, the drivers just kicked out of the way and the gear teeth danced.

Put it on the track and it ran fine.

Best way to do that? No, but I started to think about mesh being maintained by weight.
Not too long ago I read about H0 sprung versus vertical slop drivers.

Is it ideal?
No.
With enough weight does it work?
Seems to.

Won't work with underslung worm or jackshaft.

Of course, you can always mill out those case halves and out NWSL boxes in there.

In #1 gauge, LGB made a Mogul starting in, oh, late 60's but most likely early 70's. Originally, center driver fixed bearing location that matched ends. All drivers, always, flanged.
They bucked a bit, didn't like sharp vertical changes to grade.
Next generation, looked the same, except the center driver axle was now in the super dooper plastic floating bearing, with spring wire holding it lightly down....problem solved. Not the same as worm mesh and weight, but floating center axle. Just stick a couple of chunks if, oh, 025 or smaller music wire across the center axle, one per side, to make sure it stays down.

[Jay Criswell]

Jim,

I don't concur with your explanation but I see no point in going any further. Do whatever you want.

LGB to Adams is apples and oranges. One is made to take a silly, non prototypical curve the other not.

Have fun, I'm not going to waste any more time on this.

Jay

[R.K. Maroon]

I'm not sure I agree with my own thinking either, Jay -- As I said "I can't make a direct argument".

So we'll leave it at that -- I just wish I had a solid explanation for why the center worm seems to wear more than the others, on average, and why Adams chose to eventually eliminate it.

Thanks for your input -- I do appreciate it.
Jim

[bob turner]

Of course the key is to support the worm shaft in between axles.

Boston gear was the go-to supplier. No more; the simplest part from them is $30. Multiply that by six axles and you get $360. That does not include anything but the gears.

If you are sure the spacing is the same as All Nation, then it is time to contact Bill Pope's kids.

I am with Jay - get those axle bearings tight. Solder is a good bearing material, and it is renewable. I am using it for tailwheel bearings on my airplanes - works great!

[ScaleCraft]

WHOLE lot of difference between a precision machined axle hole and a single tailwheel.
For that matter, how about a wooden tailwheel for a Helio Courier?
I'd like to know what a new axle fits like. We may not have that big a problem.
Thought about shouldered axles? Pin honing the axle holes?
The sides of the boxes are flat, so getting a 90 degree bore shouldn't be that much of an issue.
But what is the c/l to c/l of worm shaft to axle shaft?
Have to find a new block to make sure you have it right.
Otherwise it's an effort in futility.
14T gears.....have to find someone to make them for you?
Like I said...mill out the cases and put NWSL boxes in.

[John Webster]

A couple of opinions here:

Get a good book on gear design and you will find formulas for the separating forces generated by transmitting power through gears. As Bob mentioned, those forces are probably large enough that they are bending the worm shaft away from the center axle gear. Take the center axle out and push on the center worm to get an idea of how much deflection just a couple of pounds will produce.

The worm shaft should have as close to zero end play as you can get. Too much end play in the worm shaft manifests as bucking, especially going down grade while retarding throttle. It might be a good idea to check that the worms aren't moving on the shaft too.

Centering the axle gears on the axle shafts and limiting axle shaft end play is also important. If the axle gears are too far off center the edge of the gear tooth will be scraping the face of the worm, removing any lubricants. In a multi gear setup like your diesel trucks an off center axle gear has its contact patch farther from the worm shaft centerline than the others which means that it is trying to run at a different surface contact speed thus concentrating wear on that gear.

I'm enjoying your dive into the mechanical intricacies of this restoration.

[bob turner]

When you are lining worn bores with solder, you do not have to be a precision machinist. Just run an adjustable ream through with the blocks bolted together.

How do I know this works? Tailwheels are not the only thing I fix in my shop.

Also, for that worm shaft - I can envision a simple half-bearing that could have some adjustment, either with shims or just locking the machine screws with Loctite.

Most worm shafts need ball bearings, but I have found that these bronze blocks do not seem to need them. They do need continuous lubrication.

I tried ball bearngs on locomotive axles. I like solid better for slow moving parts like axles.

Personal preference.

[webenda]

Solder is a good bearing material, and it is renewable. I am using it for tailwheel bearings on my airplanes - works great!

Thank you for the hint Bob. I am wondering how Maroon is going to repair his axle bearings. I thought of Babbitt (mostly tin or mostly lead depending on application), but I imagine 60/40 to 50/50 Pb/Sn would work well for model trains. I am following this with great interest to see what choices he makes in solving the many problems in restoring these models.

[aww]

Bob

Yes great idea, solder. I hadn't thought of that.

Adams drives, I had some experience with them, limited, but I noticed a lot off gear failures. Never could figure out how to get new gears.

I suspect the distance between the axle and the worm shift is too close for optimum geometry of a worm and a wormgear. The two want to push apart from each other and in the middle gear they succeed.

Allan

[Mitch]

Jim

Just a few observations by a guy who knows nothing of the "stops" but I do know a little about gears, gears meshing, and gear wear. An example is setting up the rear differential in a car or truck. We are allowed .004 to .007 backlash between ring and pinion, and I can guarantee you that your vehicle's differential gets much hotter than the model train. I usually try to hit .005, as the factory states the closer to the minimum spec, the better off you are.
1st off, (and we'll use the second from the left in the pic), are these gear cases brass or bronze? Either way, I would think that the steel axle shaft is harder than the case, so the case is gonna wear first.
2nd, the cases have no axle bearings other than the surface of the case itself, and it sure don't look like it's ever been lubed at the bearing surface.
3rd, it appears we have press fit "limiters" between the wheels and the case, and they show no rubs where the axle could have been shifting from side to side. This seems strange to me, 'cause the engine would have certainly gone around a curve or two in its lifetime. Again, no evidence of lube on those surfaces, either.
4th, the "driven" gear closest to the motor connection seems to be in very good condition, while the center gear is hugely concave and the forward gear slightly concave. The question may be: Where does most of the weight of the engine rest on each unit?
The design of the gear case, IMHO, leaves a lot to be desired. I like the worm and sector arrangement, but there needs to be a different support system for the axles. Even if you could use plumber's solder, (or even better, poured melted wheel weight lead, (high antimony content), and re-drill and hone for a .001/.002 clearance, you would still need to lube in there, most likely every run. That makes it highly maintenance intensive.
I assume the objective here is to maintain the original design and make it work, but if were my money, (and I don't have any ) I'd go with a version of Jay's design and line bore the case to accept sealed bearings and have the axle gears fitted to give .002 to .005 clearance between the worm and the driven gear. Even then, I'd probably use Lubriplate or one of the newer synthetics, and forget about it for a while.

[bob turner]

That is true - you simply cannot run a steel/brass interface dry. It will start to gall and chatter.

But these gearboxes are built to hold a lot of grease - I just pack them with Lubriplate. As puny as these gears are, they seemed to last a very long time.

Brass is sort of ok for a bearing surface, but horrible as a gear material. You need bronze, or plastic.

And the theory of dissimilar metal bearings is that things imbed in the softer material and wear at the harder material - I have no reference for that, so

Opinion.

My Max Gray Mogul got the solder treatment for worn bearings - it has a fixed worm location - really straightened it out and eliminated a lot of noise.