Making a wheelholder/cutter for N Scale

[mmagliaro]

There is a tool for holding locomotive wheels so that they can have flanges cut or a traction tire groove cut in them.
It is made by Fohrmann tools.

It looks like this:

https://www.fohrmann.com/en/wheel-holding-fixture.html

One look at these photos and I think you can see how it works.  It's genius!  It will center the wheel over the slightly-tapered shaft, and then the two large nuts, one from each side, tighten up against the wheel to hold it firmly.
Then, the big shaft can be chucked up in a mill or lathe, or drill press, and you can turn flanges or TT grooves.






SO....
They only make it for HO and larger scales.  I decided to try to make one myself.   I turned the whole thing out of steel.

The problem is that thin threaded "snout".  In N Scale, it has to be thin enough to go through our wheel centers.
I turned it down to .047 and then used a 00-90 die to cut 00-90 threads on it.

The whole thing works... EXCEPT... that snout is nowhere near strong enough to withstand any cutting forces.
I put a piece of brass in it (about like a brass N Scale flywheel size), and once the cutter presses on it, that
little snout snaps right off.

I think I see why they only offer this in HO and larger scales.

I tried again, and I heated the steel piece red hot and quenched it in water, hoping that if I hardened it, it would be more durable, but no... it is just too thin to hold up.

Any ideas?

I was just using a regular soft steel bolt to make the basic body of this thing.  If I used, say, a Grade 8 bolt,
would that get me a lot more strength?   My fear there is that a grade 8 bolt might be so hard it would be impossible to cut in the mill, or to use that 00-90 die to cut the threads on the snout.

Another thought I had was that maybe the whole thing only works, even in HO, if you ONLY cut on a thin wheel that
is locked down tight against the tapered cone.  In other words, I was testing on something much wider, something like a 10mm long flywheel, so maybe the only problem is that the cutter cannot be used very far down that thin snout.

[aikorob]

some type of live center/ support to hold the threaded end?
that should reduce side pressure on your "snout"

[robert3985]

Max @mmagliaro ,

Hmmmm....took me a couple of minutes to see how this works. 

Let me 'splain and tell me if I've got it right.

(1) Mount and center the threaded steel shaft in lathe chuck with threaded brass collar moved back from the exposed end

(2) Slide the wheel you have onto the tapered/threaded central shaft until it stops (stopping by interference on central taper thru axle hole)

(3) Turn threaded brass collar on steel shaft until it "kisses" the backside of the wheel

(4) Slide the slotted intermediate dual-diameter bushing onto central threaded shaft, making sure the slot is over any rod-attachment hole on a driver (also choosing which of two diameters does not impinge on driver weight)  Exact centering on this bushing is not required and due to its design would be difficult to do anyway.

(5) Mount threaded end-cap onto central threaded shaft and turn until mounted model wheel is held semi-steady by pressure generated between the central steel shaft threaded collar and the slotted brass bushing held in place by the threaded end-cap

(6) Near perfect centering of model wheel is achieved by having its axle hole "kiss" the central steel tapered/threaded shaft AND "kiss" the surface of the threaded brass collar on the central steel shaft...being pressed against these two surfaces by the pressure applied by the intermediate slotted brass bushing and the end cap which is threaded to match the steel tapered/threaded central shaft attached to the threaded steel shaft held in the lathe chuck.

(7) A bit of fiddling would be necessary to get both surfaces (on axle hole corners and backside of driver) to simultaneously register driver tire to be exactly centered.

Some assumptions are being made for this to work.  The first assumption is that the driver's tire is concentric with the axle hole, and secondly, that the inner edge of the axle hole is perfect (has not been dressed by hand to remove burrs).  If either of these assumptions is incorrect, it will be nearly impossible to get the tire centered on the lathe. Sure, it can wobble by a couple of mils and still be operable, but the goal is for it to be centered by about half a mil or less.

As you have discovered, getting enough pressure against the wheel to hold it securely while machining is the main weakness of this design (if the above two assumptions are true) as the central tapered/threaded shaft will almost always be mechanically too small in diameter (too weak) to apply proper pressure  and overcome the force applied to outside of the tire by the lathe tool...even in HO and O with their advantages of having a larger axle hole to work with over N-scale.

As I see it, there are a couple of ways to get this to work well in N-scale.  I'll explore the easiest possible solution, which involves eliminating the threaded part of the central steel tapered/threaded shaft.

Lathes have a tailstock, to keep long or fragile pieces centered while turning by using a tapered "center"...either "dead" or "live"...the "live" center using ball bearings to allow the center to turn.  A live center is essential for this solution.

This solution involves fashioning the central steel shaft and brass threaded collar just like the photos indicate, including a central/ precisely centered taper for the wheel's axle hole, but no threaded portion to that central, tapered shaft.

The intermediate brass, slotted bushing would also be the same

The threaded end-cap would be the same also, without the threads, but with the central hole drilled to allow a tight clearance fit over the short straight section of the central tapered shaft.

Setup would be the same as I described above until you get to step (5)  New steps continuing from new step five would be to:

(5) Slide end cap over the straight section of the central tapered/threaded shaft

(6) Slide lathe tailstock with live center up to wheel-turning fixture, inserting tip of live center into central hole of end cap and clamp tailstock to lathe ways

(7) Slowly turn adjustment wheel on tailstock until model wheel in wheel-turning fixture is barely immobile

(8 ) A bit of fiddling will be necessary to get both surfaces (on axle hole corners and backside of driver) to register driver tire to be exactly centered.

(9) Tighten the model wheel between tailstock live center and steel shaft/brass threaded collar untill it is firmly held tight enough to apply cutting tool to tire.

This will do the trick as holding pressure on the model wheel is not dependent on any small, threaded rod running through the axle hole.

The weak point to this fixture is another assumption, and that assumption is that the wheel tire is held tightly enough on the wheel body to withstand the considerable forces and heat generated by turning it against the lathe tool.  Some wheels have an insulative layer between the tire and wheel body, and others have an adhesive holding them on securely, both of which may, or may not be sufficient to withstand the force and heat of turning the tire.

(edit: After looking at the design of both the commercial version and the modified version, I can see that it's imperative to ensure that excessive pressure NOT be put on the taper stuck in the axle hole.  Too much pressure is going to deform the "edge" of this hole, requiring it to be reamed until it fits back on the axle shaft...which opens up a whole new can-'O-worms.  Also, I can see that it would also be easy to push the rim slightly parallel to the central wheel's tire mounting surface, so making sure that the edge of the axle hole bearing on the central tapered shaft, and the surface of the tire bearing on the adjustable threaded brass collar on the central mounting shaft are EXACTLY aligned is absolutely essential.  Caution should also be used when applying pressure using the lathe's tailstock adjustment wheel.  Hmmmm....this is a bad design IMO)

Alternative Design:
I believe it would be fairly easy to design a holding fixture that is registered by the wheel axle, but presses against the tire's front and back surfaces, with enough clearance to allow access to flange and/or rim (to allow cutting traction tire slot), which would completely eliminate cutting forces on the wheel body, insulative layer and any adhesive between tire and wheel body.  These metal pieces would also act as huge heat sinks to suck most of the heat away from the insulative layer and/or rim adhesive too.

(edit: Here's a quick, rough drawing of the tire-holding-fixture protocol I'm writing about:
)

The only "problem" with the tire-holding design would be that both the brass threaded collar on the central threaded steel shaft and the cap that fits over the straight section on the central tapered centering shaft would have to be specifically designed for different driver tire diameters, especially if turning flanges.  The end bushing which is pressed against the tire by the lathe's tailstock/live center would certainly need to be matched to each tire's diameter...especially O.D. and also I.D. depending on the design of the wheel body.

Anyway, this is what I would do if I needed to turn dismounted driver tires.

Cheero!
Bob Gilmore

[rodsup9000]

 I have some old RR bigboys that I want to turn the flanges down on and this is what I'm going to try.
I do have a lathe and that helps. Over 10 years ago when I started installing ball bearing to my 20.3 rolling stock, I made this holder to hold the wheel sets so I could turn the axle stubs down to fit the ID of the bearings. I plan on making something similar (but a lot smaller) to hold the drivers so I can turn the flanges down.  Using this same design but the center would be milled out for clearance for the gear.





 

[mmagliaro]

Gentlemen,
Thank you for such detailed analysis and photographs.  I wonder how well the HO version of this thing works.
But I bet it does work as long as you are careful when you set it up.

Yes, the fact that the whole thing is centered based on a taper going into the axle hole means you want to
"just" nudge that taper in there, "just" spin that brass collar assembly down to the front of the wheel,
and then tighten up the nut against the back of the wheel.  And hope that you didn't overdo it and deform the axle hole.

rodsup9000 - I like that fixture, but how does it end up centered around the axle?  Is it bored to just
exactly match the axle diameter, so that when you tighten the set screws, it closes that gap in the fixture completely
"just" as it tightens?

But anything that grips only the axle could be big trouble.  The wheel will slip around the axle for sure.
I liked the Fohrmann design because it could work with any size wheel, any size axle hole, and it gripped the wheel itself, not the axle.  Ideally, there would be a set of those brass spacers (the things that go over the thin "snout"
and press on the wheel).  They could be a set of brass rings, in a whole set of diameters, so you could choose the one that just about exactly lines up with the wheel rim and not the center.  Once it is tightened down, it would stay put, and it you are clamping directly onto the rim, you wouldn't risk spinning, tearing out, or otherwise messing up
the wheel center.

[robert3985]

...rodsup9000 - I like that fixture, but how does it end up centered around the axle?  Is it bored to just
exactly match the axle diameter, so that when you tighten the set screws, it closes that gap in the fixture completely
"just" as it tightens?...

I'm not rodsup900 but I'm going to bet the diameter of the central hole is a tight, barely interference fit with the wheelset's axle.  The fixture was probably machined in one piece, then the screw holes were drilled & threaded, with a clearance fit on one side, threaded on the other...then, it was slit down the middle with a mill.

The near exact-sized diameter hole automatically centers the fixture in the lathe's self-centering chuck, but could also be perfectly centered using an independent-jaw chuck and dial indicator.

...But anything that grips only the axle could be big trouble.  The wheel will slip around the axle for sure...

Max, I agree with this, however this fixture was made to turn and modify the axles on rodsup900's engines.  I don't believe it will work for actually turning down a flange or cutting a traction tire slot using a cutting tool in a lathe without the wheel spinning on the axle.  It would probably work with a grinding tool and light filing/sanding/polishing using really light pressure and taking your time for flange reduction, but making a sharply angled slot wouldn't work.

...I liked the Fohrmann design because it could work with any size wheel, any size axle hole, and it gripped the wheel itself, not the axle.  Ideally, there would be a set of those brass spacers (the things that go over the thin "snout"
and press on the wheel).  They could be a set of brass rings, in a whole set of diameters, so you could choose the one that just about exactly lines up with the wheel rim and not the center.  Once it is tightened down, it would stay put, and it you are clamping directly onto the rim, you wouldn't risk spinning, tearing out, or otherwise messing up
the wheel center.

Unfortunately, the Fohrmann design doesn't work on "any size axle hole" as your N-scale effort graphically shows.  It's the major weak point in the design.  Also, gripping the "wheel" is also a design flaw because of the uncertainty of how strongly the tire is attached to the "wheel" which may cause failure due to torque or heating problems...or maybe not.  It's an uncertainty.  Also, the Fohrmann design requires a brass, or hard metal casting as the main wheel body since the fixture attaches to that, so turning drivers with tires pressed on to plastic wheel bodies is out of the question.

Yes, the Fohrmann design takes into account different axle diameters and tire diameters, but...it doesn't work in N-scale because it REQUIRES a threaded rod that goes through the axle hole to attach the wheel to the turning fixture...and that threaded rod isn't strong enough to resist the resulting torque generated by machining the tire stuck waaaay out there in relation to the center of the thin, threaded bolt...bad machining setup. 

Where the fixture needs to grip the wheel is at the tire front and back surfaces, and the tapered pin in the center acts only as a centering device, not an attachment device. in my design, only two components would have to be replicated with different diameters...the threaded brass collar which fits on the steel central shaft, and the pressure bushing, both of which are pretty easy to make.

I don't know if it's a major drawback to both the Fohrmann design and mine but they both require dismounting the wheel from the axle.  I've got some ideas percolating that might work for being able to chuck up a wheelset without dismounting the wheels from the axle, allowing turning and slotting of the tire and not losing the quartering.  Hmmmm....

Cheerio!
Bob Gilmore

[rodsup9000]

I'm not rodsup900 but I'm going to bet the diameter of the central hole is a tight, barely interference fit with the wheelset's axle.  The fixture was probably machined in one piece, then the screw holes were drilled & threaded, with a clearance fit on one side, threaded on the other...then, it was slit down the middle with a mill.

The near exact-sized diameter hole automatically centers the fixture in the lathe's self-centering chuck, but could also be perfectly centered using an independent-jaw chuck and dial indicator.

Max, I agree with this, however this fixture was made to turn and modify the axles on rodsup900's engines.  I don't believe it will work for actually turning down a flange or cutting a traction tire slot using a cutting tool in a lathe without the wheel spinning on the axle.  It would probably work with a grinding tool and light filing/sanding/polishing using really light pressure and taking your time for flange reduction, but making a sharply angled slot wouldn't work.


Cheerio!
Bob Gilmore

It started as 2 pieces bolted together, milled axle slot with ball endmill to have a slight interference fit. Installed a rod the same size as axle and indicated it in and turned the OD.

 You both are right in that trying to turn a driver, it would slip on the axle.
 
 I will have to make a live center adapter for the tailstock, (like the pressure bushing in Bob's drawing) that would put pressure on the front face of the driver to press the back against the fixture.

 

[narrowminded]

You DO manage to take on some impossible tasks, Max. 

The problem is that the 00-90 screw is only .032" at the thread root.  Is this for a 1.2mm hole?  If larger, like 1.5mm you could go up to a #0-80 and the area afforded from the increased diameter would go up exponentially.  A 1.5mm bore is within the low end of the specs for a #0-80 which, in these little dimensions is percentage wise much stronger.  In fact, a quick check on the back of an envelope says that it's only .013" diameter difference but it's twice the area at the root.  That's a lot.  So, if you wanted to pursue that concept, even if you had a screw that was on the upper end of the tolerance and wouldn't go through, sand it a little and you're in.  Still in spec, too.  In that vein, you could also go up to a #1, same concept, and take the diameter down at the crest to just fit the 1.5.  Even though you have removed a lot of the thread you're still just above the pitch line and can get the strength back by doubling or even tripling the nut engagement.  These are weird and application specific, mickey mouse, but would be much stronger than the 00-90 screw.  Strong enough for a big overhang?  I doubt it but that's also dependent on how much of a cut you take and how much of an overhang.  Might coax something in but...

Then, final, would be to not use a screw at all for the minor diameter.  Put a solid piano wire through, which is already MUCH stronger than any material that you're likely to be successful tapping or die threading plus it's left at full diameter.  Instead of the threaded portion projecting beyond the taper, drill that end of the arbor for the wire.  Then, on that same bigger arbor that goes into the chuck, cross drill for a set screw well back on the meat of the thing.  Then on the wire projecting, mount a threaded rod of appropriate size to accomodate a through drill.  Several inches engagement might do OK with red loctite but the wire could also just be bent over on the far end.  Set screw clamps the rod at the coarse setting, the threaded rod section is your small clamp.  Still mousy but...

And finally, if you just wanted to add traction tires, not dress the flange, just make a collet at the flange diameter and depth and split it back an inch or so, cross drill it for a pinch bolt right behind the flange bore/ shoulder, pinch the collet, chuck it up and go.  To do the flanges you might do the reverse but you'll be on the tread taper which I would do but requires a better fit and gentle on the cut. 

Meanwhile, if you've got the equipment to pull this off well, at some point it might be easier to remake the wheels. Not really.

[Lemosteam]

Max, one question, when the threaded shaft separated form the taper was it twisted off or did it look like a bend fracture?  In ask this beaches I'df it looks like torque then any sort of process that relies on axial clamp load only likely will nit work as the rod going through the wheel must withstand the torque applied when the cutter meets the wheel.

This does remind me of something that might work though.  If you are familiar with a quick release on a bicycle axle, the design presented two mechanisms that assist in clamping the hub to the fork Clovis, a nut and a cam lever.  The more you turn the nut onto the rod, the harder it is to apply the cam lever, but the additional axial clamping force is applied across the joint.  So if you forget the taper, and find a correctly sized piano wire, and fashion a cam lock at one  end, with a notched pressure plate to clear the rod boss, and at the other end a free spinning thumb wheel on the other side of a brass cylinder, you could simulate the quick release hub.  Tourque would again try to overcome the clamp load,M but with infinite cam load, it might stand a chance.  Another way to absorb the torque would be to add a dog on the sandwich plate over to the cylinder, thus when the cutting load hits the wheel, it is transferred to the boss, and then the notch to the dog back to the cylinder in the chuck overcoming the torque and allowing you to cut.  I will draw a pic tomorrow if it will help understand this rambling... 

[mmagliaro]

Answers:

0-80 is too big.  A 00-90 screw will just snugly fit through a Kato Mikado driver hole.  0-80 is out.
(In fact, the 00-90 is so perfect a fit, that to do a Mikado driver you could use a steel 00-90 screw
with washers and a nut, just chuck the long end of the screw (that sticks out through the nut) in a
mill or lathe chuck and turn away (to do flanges, but probably could not withstand the force to cut a TT groove).  I've done this with Trix K4 drivers too.  Those fit on a 00-90 screw perfectly enough to do the flanges. 

John, I sort of follow what you are saying with the piano wire.   But wouldn't I then have to make a piece with piano wire sized to every different wheel hole I need to work with?   Although, granted, there are probably only 5 or so different sizes in N Scale, I bet.  1.0mm, 1.2mm, 1.5mm, and then some of the brass and other steam has some big sizes that are probably up around 2, 2.5mm.

Actually, you might be able to take the correct size machined rod, cut some threads on it with a die
(even if you have to turn the outer tip down to a good size for the die).   Like, if you needed a 2.0mm rod,
you just turn down 1/2" or so the end so it matches the 0-80 or 1-72 thread, and you can put a nut on there.  Now, you press the rod through the wheel, grip the other end (the smooth end) in the mill or lathe chuck, and push the wheel back right up against the chuck.  Then you tighten the nut down.  The wheel would be sandwiched between the but and the chuck and at least on my chuck, the nose of that thing is machined really square.

Oh... another thing, while it's a little scary gripping only the wheel center, it's not as bad as it seems.
I made another Fohrmann-type fixture, and instead of beating it up trying to turn a big brass cylinder in it, I just put a Mikado wheel on there and cut a TT groove in it.
It worked great, just held by the wheel center.  Those outer tires are not that hard.  A sharp cutter
cuts them off with very very light pressure.  It worked great.

And I learned that what we all suspected about Mikado drivers is true.  The wheel tire is too thin to allow cutting a TT groove.  I wanted to go .007" in, and it just started to skin through the tire and expose the plastic wheel center.  But that's beside the point.  Gripping by the wheel center really worked.

And one other other thing.... We do not really have to be within 1/2 a thou on this.
I have run Mikado drivers in my mill with a dial indicator running on them.  They are good to within .002" or .003".  That's pretty darn good for a hobby product.  But they are not down to .0005".  No way.
(And when I chuck a plain brass rod in that same chuck, and turn it true first, and check it with the same dial indicator, it varies by only about .0005", so I know that my setup is better than the driver).

[robert3985]

Max...Haha!!  All that thinkin' and drawing...PFFFT!!  Not necessary.

If it works for ya, that's all that counts!

Cheerio!
Bob Gilmore

[mmagliaro]

Max...Haha!!  All that thinkin' and drawing...PFFFT!!  Not necessary.

If it works for ya, that's all that counts!

Cheerio!
Bob Gilmore

Ha ha ha !   Yes, I have a bunch of guinea pig Mikado drivers lying around, so there was nothing to lose just to see if it would work.  It was pretty cobbled up.  I used a stack of washers as the spacer fixture (that nice thing with the notches in it that Fohrmann has to clear the crank boss and counterweight).  I'm going to get rid of that and machine a nice spacer with some notches out of some heavy-wall brass tube, so my fixture doesn't look like a spinning junk yard when it runs in the mill.  But I think I can make it work.  Centering the wheel via a tapered metal shaft actually works darn well.  You just have to press the wheel snugly up against the cone by hand without being too heavy handed, and then snug up the nuts from either side. 

I'm going to cut a real wheel, and then test the TT groove with the wheel back on an axle, spun in the mill with a dial indicator to really see how good it is.  I will post pics of all this after I do it.