Building a Train Wheel

Machinery's Handbook does not differentiate between materials for interference (shrink or force) fits. Looks like this application would be appropriate for a Class FN3 fit. Table 6 on page 617 of my 24th edition says to use an H7 tolerance on the hole and t6 tolerance on the shaft. This would result, for example, in an interference of 2.6-4.9 thousandths of an inch on your 4.25" diameter fit.

Thanks for your input. Sounds good.

You have both an inside fit and an outside fit. In one instance you will have to reverse the heat and cold.

If your axles have bearings, you don't need a steel insert in the center of the wheel.

Will it matter if I work inside out or outside in with the shrink/heat? I was going to use the steel as the bearing on an alumimum axle. Rotational speed is not going to be very great. In fact, it will probably end up as a stationary display model, but capable of moving. The train these will go on will weigh around 100#; bearing on 4 axles (8 wheels).

I'd heat the wheel and insert the cold bushing first. Then cool the wheel and heat the rim for final assembly.

Be careful when you do the final machining. That transition from Al to steel can cause the cutter to grab, if you go too fast. You probably already know that different materials cut at different rates.

Good luck.

Based on this last statement, you dont need the meticulation. Machine the Al to size and undercut the steel by .002". The Al can be frozen for an hour and the wheel boiled for an hour. I believe this is all you need for a model.

Its the heating that one needs to be careful with.........and the heat should not be at too high a temperature or applied too applied too quickly or unevenly or distortion problems may occur.

For cooling, I think dry ice at -78⁰C should be OK, if not, liquid nitrogen would be the go.

..........and of course a good press.

I think (I know I shouldn't do that), it would pro ably be a plain bearing at an ID 1"

Actually, if you consider the differential thermal expansion of the materials, sunlight and ice cubes are all that would be required.

Let's not complicate this too much. _{(too late for that)}

I suppose if the dimensions turned out a little oversize/undersize you could always use Decon A or F dependent on the error..........I hope that isn't too complicated.

Sorry, only trying one on!!!!!

Based on the sizes involved and especially based on the likelihood that this will not be a dynamic assembly, I suggest that you just heat the outer ring (an oven set to 450F should be adequate) and drop the inner wheel into the hot ring that is sitting on a grating much like the rack from the oven. If you really want maximum clearance during assembly, put the aluminum core in the freezer (about 25-28F) as well but I believe that a room temperature core will provide adequate clearance if machined properly.

If the assembly was going to actually see serious mobility, I would suggest that you form a small inner lip on the aluminum core to minimize any shifting of the outer ring due to forces from the tracks during curves.

I wouldn't suggest more than polishing after assembly, you won't need it.

When cold the outer ring an d the AL interference fit should total 0.0025" - 0.004" with an 0.062 radius edge on the AL major diameter.

On a clear summer day set the outer ring in the sun on a lazy susan with Fresnel lens not clearly focused upon it at about 7 AM an put the AL in an ice chest with regular ice. At about 7:20 AM move lens an drop the AL into the external ring.

Do apply a UV protective coating to the AL or it will lose it's temper in the presence of sunlight and O2 rapidly.

"... Do apply a UV protective coating to the AL or it will lose it's temper in the presence of sunlight and O2 rapidly. ... "

Er....what? Aluminum will lose heat treating effects, if exposed to sunlight and O2?

I have never heard of such a thing. I cannot find any acedemic (or otherwise) papers on such a phenomena.

I'm certainly open to being wrong and learning something new, so if you know of any, please provide some links in support of the claim.

Bwire, has Mushroom season come around early this year?

I read all the posts, and I didn't see any mention about strength of materials!

Have you considered that steel is much stronger than aluminum, and that it may cause the aluminum to cold flow when it shrinks to tighten the joint? Also, if this is for a regular locomotive wheel (as opposed to a model train car), the loads can cause aluminum to deform -- it just merely doesn't have the strength to handle such loads.

Of course, I'm not talking about some alloys of aluminum which have higher yield strengths.

Look at stress-strain curves for aluminum and its alloys before you finalize your design. There are also references which give thermal expansions for various metals and alloys, such as Machinerys Handbook.

I was going to use 6061-T6 as it is easily available and not as expensive as the 7000 and 8000 alloys which may be a bit stronger. It will be for a train car. That's why I was asking about the shrink tolerances; steel being stronger than aluminum. I believe the COE of aluminum to be greater than for steel. Am I correct? I need 8 wheels for now, but if they work ok, I will be making more.

CTE of Al is 12.3 (10^-6 in/in ^oF) and steel is 7.3(10^-6 in/in ^oF).

I'd be more concerned about breaking the outer steel band than deforming the Al.

It won't take much interference to lock things in place.

Yes, the COE of Al is greater than steel, but I'm still concerned about the mixing of the two metals in such an application. I have seen railroad car "hot boxes (called that for a reason)" which are on fire! Sometimes, the lubricating grease runs out, sits in the box long enough to get semi-hard and doesn't properly lubricate the bearing, etc.

With such a potential for temperature differences, I would be skeptical of the use of AL and steel/iron.

Is your design to reduce weight? I can't imagine that it would be less expensive to manufacture, and I can't get my head around saving enough fuel by reducing weight to pay for the greater manufacturing costs.

Oops! My eyesight isn't as good as it once was . . . I didn't see the dimensions in the drawing this morning!

Should work OK. I can't imagine that this would ever get hot or weigh enough to cause problems.

Especially since post #6 states, "it will probably end up as a stationary display model, but capable of moving".

One thing no one has mentioned, is iron oxide (rust) + aluminum = thermite (google it). Thermite is set off with a magnesium fuse, and is used to weld railroad track together, magnesium is used to get the temperature high enough for chemical reaction to proceed at a rapid rate. At lower temperatures, it happens justs a lot slower. What this means is once you get your parts together, you may not be able to get them separated, just something to consider when using dissimilar metals.

" What this means is once you get your parts together, you may not be able to get them separated,"

Hell man, that's the whole point of this discussion! Where have you been?

As usual, we have complicated this discussion beyond necessity. Take a look at the coolest temperature your wheel assembly will be exposed to. If you have a slight interference at this temperature(.0005) between the ring and wheel that's all you need. It won't get looser as it gets hotter.

There's another issue here. That is the accuracy of the equipment you have, and your proficiency running it. Can you work to the tight tolerances we are discussing here?

Machining that outer ring from steel may be a challenge, unless you have a tube the right size and are simply cutting sections off the end.

I'd say try one. If it falls apart, you can always bond it together with epoxy.

This is a neat idea, but as the brakes are applied the steel O.D. of the wheel will heat

up first and could break away, and or slip off the aluminum core.

Just a thought to consider.

Ken.

Back to the drawing board Ron - The ID of your composite wheel is 0.375. It is steel, the shaft you propose is ALUMINUM? This combination of materials will not work, and there is no room allowing for a bearing. If you want to put a bought off the shelf hardened steel sleeve over the aluminum shaft (Why aluminum?) and insert a inexpensive bought off the shelf needle bearing as a press fit on the wheel. That will work. If you are doing small Q's and want something inexpensive and serviceable you could consider an actual composite. A steel rim with a bearing ID. and a epoxy wheel. Reinforce the epoxy as needed if plain old out of the can won't due.

Why not just make it a nice transition fit, and use some Loctite sleeve retainer, it's amazing stuff. I use it all the time for serious engineering applications with great success. If the wheel gets hot in service then the aluminium hub will expand to give an even tighter fit.

Steer clear of red grease, hard simoniz wax is preferable

Lyn has said it all.

This has gotten way too complicated for what it is.

Anyone out there ever fitted a ring-gear to an auto flywheel?

Forget about the ice cubes and the lah-de-da.

Machine the components to a 5-8thou interference fit and heat the tyre with a soft gas flame ( oxy-acet, LPG, etc), evenly, for about a minute and slip it onto the suitably supported, room temp., wheel ( so as not to have the tyre fall of the other side, Eh?) wearing leather gloves. It doesn't need to be that hot.

Let it cool and job done.

The bush can be simply pressed in ( or drawn-in using some 'all thread', a couple of nuts, and two hefty washers) if it's correctly machined to a 'light press fit' aided by some Loctite bearing mount, if you've got the fit a bit loose.

Easy.

Stu.

I suppose I too subtle in post #9 then

Leave the rim set in the morning sun 20-30 minutes under curved glass on a black back ground. No need to waste gas and distort the rim eh it doesn't need to get that hot. It would be simple to use a nylon inner hub unless you don't have nylon but tubing would be well suited. Galling too often and so easily occurs when AL is mated to steel as in a bushed condition. Nylon tube and a little wax or soap goes a long way as dry lubricant.

A six inch steel C-clamp would suffice as a press to fit these pieces together.

I say bwire, what a clever little bass............sorry I had not seen that before.

Thanks,

There are other clever critters in that area too. One day I wasn't getting a bite when I saw a cottonmouth with a frog in it's mouth an I thought to myself frogs are good bait. So I quick snatched that snake just behind the head an got the frog but then how was I to let go without getting bit. Genius! I grabbed up the bottle of Jim Beam and swilled some into it's mouth, the snake went limp so I tossed it.

A while later I felt something bump against my shoe an that snake was back with two frogs in it's mouth, life's good

Couple of months back we won a job to reco' half a dozen BIG reduction gearmotor sets. Big? The size of a CAT D8 bullgear.

Didn't have a bearing heater. Office got quotes, to hire and to buy.

I settled for using the workshop BBQ. Placing the bearings on the open griddle.

Worked a treat.

Cost? Zero.

Stu

WOW! Ronseto, thats a lot of work for a model! In the Live Steam hobby, we use solid blanks or castings for the entire wheel. We don't shrink on tires. As far as shrinking the tire on, how about making a scale fire ring like the railroads used.

Cost is a big factor here. Material cost vs labor cost. I'm retired so labor is not an issue. I could get blanks cast up, but by the time that's done and the cost of machining, wheels can cost $30+ each. Material and my time an bring that down to $5 each.

I understand, one of the factors when building equipment is, time or money. To make all of the parts out of "scrap" costs time. To shorten the build time, use castings, but that costs money. I have built a 50Ton, 1.6"Sc Shay using scrap, except for the plumbing. Cost in scrap was about $750 including getting the wheels cast. However, since it is my wifes locomotive, she wanted Superscale's fittings, valves, water injector and so on, cost $4k. The water injector, I put in a jewelry box and gave her that as a christmas present.

Large scale locomotives can be costly. I'm trying to approach the live steam build from a low cost direction. I don't model steamers. I model diesels and electrics. I built a model of a Swedish class D, 1-C-1 electric locomotive in 1/8 scale. It is around 5' longx14" tallx12" wide. It runs with a DC motor from a scooter powered by a 12V storage battery. It weighs around 100 pounds. Total cost was around $70 including some metals, end mills, taps and drill bits. Most materials came from my wood scrap bin. The wheels were made of Baltic plywood which have held up very well. I would like to make the wheels from metal for my next model.

I was thinking; how about if I were to use a cast resin in place of the aluminum? It would be easier to set up a jig. Final machining would be after the resin had set. Do you have a recommendation for a resin? Epoxy, Polyester urethane? It shouldn't be flexible.

Nix the resin. I suggest leaded bronze a.k.a. Naval Brass or Cast Iron

This is an update on my scale model locomotive building. I cast some spoked driving wheels using an aluminum filled epoxy resin and they came out great. They are strong enough for running and they would be fine as an exhibition model run by remote control, but not for hauling passengers. I have other model making projects and cast epoxy is the way to go.

Hello, I would like to know more about your project

I have come to the conclusion that the best way to solve my situation is to just do it based on basic engineering and see how it turns out. In other words using the tried and true "trial and error" method. Most knowledge that we subscribe to was at one time derived from trial and error. The actual engineering numbers came after and confirmed the results of trial and error.

The rule-of-thumb for steel tyres/tires on a steel or cast iron wheel is to machine them 1/1000 too small, i.e. 40thou undersize for a 40in wheel (imperial measure), or 1.0mm undersize for a 1.0m wheel (metric measure). Railway/railroad repair shops will be able to confirm this.

It is rare for the inner wheel to be cooled as the outer tyre/tire is heated; usually heating the tyre/tire is sufficient, and then the wheel is lowered into the tyre/tire before it cools; once it has then it won't come off.