Nostalgia & History > Are Flanges Necessary?

For over 60 years, since I was in Junior High (I think they call it Middle School, now), I've heard scientists tell me that railroad wheels stay on the track because of the conicity of the wheels treads.

Today, I received the August issue of Garden Railways magazine, and Vince Bass' column on page 55 restates this. He even includes a link to a video produced by Popular Mechanics magazine which attempts to illustrate the point. The video of a scientist explaining this is found at

popularmechanics.com/science/a25581/science-behind-train-tracks-wheels

But, I keep thinking that these people have never actually been around a railroad. Have never heard of stringlining or jacknifing. Have never felt truck hunting. Never heard the scream of flanges nor the groan of a wheelset being dragged around a sharp curve. Have never felt a steam engine trying to straighten out a 10 degree curve...

So, today I set out to convince myself that flanges really are not necessary, and that conicity alone keeps a train on top of the tracks.

Not having a full size wheelset nor lathe in my basement, I set out to test the theory that conicity alone can keep wheels on top of the rails of my model railroad.

First picture is a wheelset I turned down to remove the flanges, without removing the conicity.

First video shows a flanged wheelset placed at the top of a straight 15 foot long grade with an S-curve at the base. The wheelset stays on the track and follows the S-curve to a stop, showing that the flanges do indeed contact the rails on the curve.

Second video shows one of several attempts made at getting the unflanged wheels to do what the flanged wheels had done. Many attempts were made with results similar to that shown, the greatest difficulty being to place the wheelset nearly perfectly perpendicular to the track at the top of the grade.

Third video shows a well placed flangeless wheelset actually making it to the bottom of the straight track, but unable to negotiate the curve.

My conclusion: The taper of the wheels helps keep the wheelset on straight track, but is not reliable. Curves? Forget it!

Flanges ARE necessary.

-John



Edited 1 time(s). Last edit at 07/03/17 22:23 by LarryDoyle.

---

You must be a registered subscriber to watch videos. Join Today!

Third video,

Now, I wonder if I can convince the NSSR to let me try this out on Lemon Drop Hill.....

-John

You must be a registered subscriber to watch videos. Join Today!

Interesting.

You probably get a lot of vibration and minor surface irregularities at the model level that wouldn't exist in 1:1.

Maybe someone can try this with an old wheel, flanges ground off, on a private branch line to see how it works in actual RR size. With a forklift and powered flatcar for putting the wheel set back on track if it comes off.

I wonder how steep of a cone, or how long of a cone, would be necessary to go around the very sharpest tracks (16 degrees mainline, 40 degrees yard?) so that flanges really wouldn't be necessary.

Neat experiment,

Ted in OR

Didn't this issue get settled about 150 years ago?

Light weight model wheels can't be very stable. Did the wheels with flanges stay on the track every time? Are the flanges the same size proportionally to prototype wheels.

Are the tapers on you model wheels the same as on prototype wheels? What happens if you increase the taper? What happens if you do it with a truck (4 wheels, two axles)?

Back to weight? The weight of the wheels you used is much less proportionally that the weight of prototype wheels, and certainly with the of the weight of a car added. Small irregularities in the track and track gauge may be the cause of the derailment of light-weight model wheels.

Seems you have more experimenting to do before a conclusion can be drawn. See what you got yourself into?

Marty Bernard

On the prototype the taper on the wheel tread, or conicity, does most of the work, the evidence is that there is usually no gauge corner wear on tangent and mildly (e.g. up to 2 degrees or 3000 ft radius) curves.
The photo is a newly turned wheel on new rail. This is a special profile for a transit line but it shows the details of the tread and flange that are part of all wheel profiles
Wheel treads incorporate not only the tread and the flange, but also a gently curved transition between them.
Much of the curving effect on sharper curves and in turnouts is done by this transition area, the closer the flange gets to the rail the steeper the angle of contact and the greater rolling radius of that wheel.
Very seldom does the steepest part of the flange bear against the rail, that only happens in wild dynamic situations including entry into short switch points, low and out of line joints, and severe slack action.
One downside of conical wheels is that they can become unstable and "hunt", wandering from one side to the other, over-correcting when the wheel flange gets to near the rail on one side, then the other. This is usually suppressed by the friction in the snubbers and other devices in the trucks.
One exception to this has been Bay Area Rapid Transit (BART), who designed their system with cylindrical wheels (no taper or conicity).
Their flanges did all the work in keeping trains on the rail on both tangent and curves; the result of this was a lot of corrugation on the rail (and noise from the wheels passing over the corrugated rail) and short wheel life due to flange wear. They are in the process of transitioning to tapered wheels.

---

I couldn't get the Popular Mechanics link above to go live. But, it works if you cut and past it into your browsers search field

popularmechanics.com/science/a25581/science-behind-train-tracks-wheels



Edited 1 time(s). Last edit at 07/04/17 00:40 by LarryDoyle.

Thanks for the detailed explanation!

Question only tangentially (pun intended) related: I've noticed a lot of transit wheels with that particular set of machining marks. Can anyone tell me what sort of machining they're doing to produce that profile? Based on pondering it over my tea, I'm guessing it's a wide, spiraled mill cutter that is specifically cut to the desired wheel profile, and I'm further assuming the wheelset doesn't have to be dismounted from the truck, these profiles are cut in place. But the tool chatter is so regular and spaced longitudinally I'm very curious to know precisely how they're profiling these wheels!

SRK

railstiesballast Wrote:
-------------------------------------------------------
> On the prototype the taper on the wheel tread, or
> conicity, does most of the work, the evidence is
> that there is usually no gauge corner wear on
> tangent and mildly (e.g. up to 2 degrees or 3000
> ft radius) curves.
> The photo is a newly turned wheel on new rail.
> This is a special profile for a transit line but
> it shows the details of the tread and flange that
> are part of all wheel profiles
> Wheel treads incorporate not only the tread and
> the flange, but also a gently curved transition
> between them.
> Much of the curving effect on sharper curves and
> in turnouts is done by this transition area, the
> closer the flange gets to the rail the steeper the
> angle of contact and the greater rolling radius of
> that wheel.
> Very seldom does the steepest part of the flange
> bear against the rail, that only happens in wild
> dynamic situations including entry into short
> switch points, low and out of line joints, and
> severe slack action.
> One downside of conical wheels is that they can
> become unstable and "hunt", wandering from one
> side to the other, over-correcting when the wheel
> flange gets to near the rail on one side, then the
> other. This is usually suppressed by the friction
> in the snubbers and other devices in the trucks.
> One exception to this has been Bay Area Rapid
> Transit (BART), who designed their system with
> cylindrical wheels (no taper or conicity).
> Their flanges did all the work in keeping trains
> on the rail on both tangent and curves; the result
> of this was a lot of corrugation on the rail (and
> noise from the wheels passing over the corrugated
> rail) and short wheel life due to flange wear.
> They are in the process of transitioning to
> tapered wheels.

Steve Krause
Chillicothe, IL

railstiesballast Wrote:
-------------------------------------------------------

> One downside of conical wheels is that they can
> become unstable and "hunt", wandering from one
> side to the other, over-correcting when the wheel
> flange gets to near the rail on one side, then the
> other. . . . One exception to this has been Bay Area Rapid
> Transit (BART), who designed their system with
> cylindrical wheels (no taper or conicity).

I've read that Santa Fe used a cylindrical wheel profile on passenger diesels to reduce hunting at high speeds.

CPR_4000 Wrote:
-------------------------------------------------------
> railstiesballast Wrote:
> --------------------------------------------------
> -----
>
> > One downside of conical wheels is that they can
> > become unstable and "hunt", wandering from one
> > side to the other, over-correcting when the
> wheel
> > flange gets to near the rail on one side, then
> the
> > other. . . . One exception to this has been Bay
> Area Rapid
> > Transit (BART), who designed their system with
> > cylindrical wheels (no taper or conicity).
>
> I've read that Santa Fe used a cylindrical wheel
> profile on passenger diesels to reduce hunting at
> high speeds.

That is indeed correct about the Santa Fe using cylindrical profile wheels. They surprisingly learned about such wheels profiles, and how smooth riding they were, from the Chicago North Shore & Milwaukee electrified railroad. Both EMD and the Santa Fe were struggling with the EMD Hyatt roller bearing boxes pumping themselves almost dry, on the passenger assigned FT & F3 units, between Chicago and LA. The constant "hunting" action of the wheels at speeds well over 80 MPH, was causing the end of the axle to act like a piston pump inside the roller bearing journal boxes, thus slowly pumping all the lubricating oil out of the boxes. Once the Santa Fe began using cylindrical profile wheels on the 2-axle F unit passenger power, the problems disappeared. Both EMD and Santa Fe discovered, through extensive testing, that the "old standard" 1 in 20 taper wheel profile produced wheel-axle hunting within the 2-axle truck, at speeds in excess of 70 MPH, and the hunting became even more pronounced at speeds over 100 MPH.

The Santa Fe continued using the cylindrical profile well into the Amtrak era, until some beancounter at Amtrak discovered that the Santa Fe was charging Amtrak slightly extra for turning/truing wheels on the Amtrak units due to the "special profile". Upon investigation, the folks at Amtrak no longer realized what the cylindrical wheel profile was all about, so that told the Santa Fe Mechanical Dept. to discontinue its use. Imagine what then began to happen? As a result, the EMD Engineering Dept. had to get involved, and eventually developed the "Wide flange, Uni-point Contour, 1 in 40 taper" wheel profile, which is universally used to this day.

One might also add at this time in the history of experience, flanges would not be there if they were not necessary.

That wheel, and presumably others you have seen, is shaped by a rotating milling head like you see in this pair of pictures.
There are two ways to re-profile wheels: milling and turning on a lathe.
Both milling machines and lathes, when mounted in the shop floor like this, can re-profile wheels on the truck, on the whole vehicle, or loose (and tied down) like this.
As I understand it the lathe is more expensive to purchase but has the advantage of being able to cut any profile that is set up in the controls. It also produces a finish that has mild cut marks in a spiral from the rotation of the wheel while the cutting tool moves across the tread.
A milling machine will cut only the profile that is set up in the cutting head.

---

---

Another factor - strong crosswinds.
...Lorenzo

Five years ago, we began experiencing truck hunting on TIOGA PASS, a private car. We finally had to jack the car up and roll out the trucks for a complete inspection. We measured the wheel profiles, wheel diameters, checked the gauge of all wheelsets, and carefully examined all the truck parts. After not really finding any significant problem, we replaced the Gadtke polymer lubricating pads between the truck bolster and the body bolster. These pads are 24 inches in diameter about 2 inches thick, are made of a polymer/lubricant material, with a 6 inch hole in the center to accommodate the center pin that holds the car body and truck together. The truck hunting has ceased, and the guests that ride with me comment on how well the car rides.

Passenger car trucks are a very complex system of parts to accomplish certain ride-control purposes. The more I work with them, the more amazed I am how they work so well.

Norm

Norman Orfall
Helendale, CA
TIOGA PASS, a private railcar