Trains.com

Flanges, To big OR Is Wheel to Small!

11478 views
32 replies
1 rating 2 rating 3 rating 4 rating 5 rating
  • Member since
    March 2004
  • From: Over the Rainbow!
  • 760 posts
Flanges, To big OR Is Wheel to Small!
Posted by eZAK on Wednesday, March 24, 2004 9:40 PM
Now we all know that flanges are far larger than any proto type known.

AND

We also know that they were made that way for tight (0-27) turns
and rounded over rail.

Now what I am unsure of is what should be the scale diameter of the outside edge of the flange?

AND

What is the outside diameter of the wheel itself?

The point being is could the flange be the right size diameter and the actual inner wheel be too small?


On another note, Are large flanges really needed anymore?

Isn't it more of a weight issue?
Relax, Don't Worry, Have a Home Brew!</font id="size2"> Pat Zak</font id="size3">
  • Member since
    December 2003
  • From: St Paul, MN
  • 6,218 posts
Posted by Big_Boy_4005 on Wednesday, March 24, 2004 9:54 PM
Well Pat I can post very fast, I saw this within 4 minutes of your posting it. The truth is I DON'T KNOW the answer to this, and I think it's a very interesting set of questions.[:D][;)]
  • Member since
    February 2002
  • From: Mile High City
  • 296 posts
Posted by jkerklo on Thursday, March 25, 2004 8:20 AM
I have been wondering about flange depth and slanted wheels myself.

Noting that most switches and crossovers have some places where the wheel must ride on its outside rim to make it over gaps:

Has the depth of the flange changed over time? It appears to have been deeper for the prewar sheet metal wheels, but has it changed from postwar till now?

Is there a specification or standard? I have never seen any accepted wheel standards for O-gauge toy trains, yet wheel makers and switch makers have to cooperate or nothing would work. Are we still using standards developed so far in the past that they are a part of toy train history?

Whose idea was the slanted/tapered wheel? The only reason I can see is that wheel spacing on the axle can be sloppy.

I use some 711 switches (prewar, O72) on my layout. New cars "bump" more going through the switch than postwar cars. Yet, I can't see any difference with modern Lionel O72 switches. Has anyone done a serious comparison?

I would find any comments useful.

John Kerklo
TCA 94-38455
www.Three-Rail.com




  • Member since
    April 2003
  • 305,205 posts
Posted by Anonymous on Thursday, March 25, 2004 8:39 AM
There was a discussion on wheels & standards on the OGR Standards Forum. I'm no expert, but I'll repeat the little I know.

The current tapered wheels are called "fast angle wheels." They first appeared during the Lionel MPC era. That's when Lionel went from independent wheels in their trucks to two wheels on an axle. The forces of going through a curve are supposed to shift the wheels in such a way that the narrower diameter part of the inside wheel rides the rail, while the wider diameter part of the outside wheel rides the outer rail. This causes the inner wheel to spin faster, and the outer wheel to cover more ground, allowing the car to go through the curve easier & with less resistance.

If you push a car with fast-angle wheels down a straight into a curve & then do the same with an older PW car, you'll find that the newer car travels farther. Assuming you push with the same force, etc.

Having the train ride on the outer rim of the flange isn't strictly necessary from a theoretical standpoint, since the real things don't do this. I don't know why toy trains do, but they shouldn't have to. Maybe if you lay your own turnouts you can prevent that from happening.

Finally, I believe that 2 railers use wheels with a more prototypical cross section, but there's a very good chance I'm wrong about that.

Tony
  • Member since
    February 2002
  • From: Mile High City
  • 296 posts
Posted by jkerklo on Thursday, March 25, 2004 9:11 AM
Thanks Tony,

I had never realized that the wheels were fixed to the axles on the new cars. I have to think a bit to see if I like it or not. Postwar wheels do wobble, that would solve that.

Prototypical wheels are fixed on axles; curves do generate a lot of rail wear, which implies pulling friction. Interesting.

You are mostly right about prototypical wheels not riding on flanges, but for an exception see the Rochelle webcam on this website.

There is a double track crossover and there has to be gaps in the track to clear the flanges of the opposing train.

Must be a maintenance nightmare. I have frequently observed trucks parked by the crossing, with welding flashes coming from the rails area.

John Kerklo
TCA 94-38455
www.Three-Rail.com
  • Member since
    December 2003
  • From: St Paul, MN
  • 6,218 posts
Posted by Big_Boy_4005 on Thursday, March 25, 2004 10:49 AM
Now that I have taken some time to think about this, one of the reasons the flanges are so large has to be the fact that the head of the rail on traditional track is ROUND, as eZAK originally noted. While other more realistic track styles have been around for a long time, O and 027 styles will probably NEVER go away, and neither will the large flanges. The large flanges allow the trains to continue to stay on the rails long after scale type wheels would have derailed, making them more forgiving on rough track.

Wheels on N and HO scales as well as 2 rail O also have oversized flanges, though theirs are not nearly as grotesque as those of 3 rail O. The NMRA has been working on wheel contours for years, and as track quality has improved, flange size and wheel shape have moved closer to prototype.

There is a movement in both HO and O for what is called FINE SCALE though this is not the standard that manufacturers produce to. Replacement wheel sets are made and used by a growing number of serious modelers. This would be an acceptable option for 3 rail, except for fact that when cars change owners, they may need their wheels changed back. Beside, there is no easy way to change locomotive wheels.

Real railroad wheels actually have very small flanges, as they don't have to worry about BOUNCE, the way models do since they are so heavy. Real railroad wheels have come in a variety of sizes over the years, though they are pretty much standardized now. I'm not sure of the actual dimension.

Finally, these trains were originally intended for children, and the large flanges helped kids get the trains on the track. Even American Flyer S gauge has large flanges and it too is considered a "TOY" type train.

Do we really want to change this distinctive feature, this tradition that has been with us for so long? Personally I have come to love, accept, and even appreciate the large flanges over the years, and have no desire to change. I can get trouble free operation from less than perfect track as a result of this detail, and consider it part of what makes 3 rail different and fun. The bottom line is, if anything needs to be fixed to make things run better, it should be the track first, not the wheels. (Sorry if I read more into this question than you intended Pat.)

Remember, A horse, designed by committee, is a camel.[swg]
  • Member since
    August 2003
  • 6,434 posts
Posted by FJ and G on Thursday, March 25, 2004 10:56 AM
I believe that in real trains, the flange is not much deeper than 1 inch. That's it folks, only one inch of flange on hundreds of wheels of rocking, rolling train that keeps it from plunging off the bridge into the great chasm.

Why then, with our deep flanges, do our trains derail much more frequently than the real ones?

---------------------

This discussion has become more prevelent in HO and 2-rail O, because the flanges in those scales are also oversized. Proto48 and Proto87 has become a thriving subculture of modelers who increase the fidelity of their trains, including the flange size. They are very careful with their trackwork.
  • Member since
    February 2002
  • From: Mile High City
  • 296 posts
Posted by jkerklo on Thursday, March 25, 2004 11:00 AM
Were the wheel flanges on logging railroads larger. Those I have seen didn't stand out as oversized.

John Kerklo
TCA 94-38455
www.Three-Rail.com
  • Member since
    August 2003
  • 6,434 posts
Posted by FJ and G on Thursday, March 25, 2004 11:14 AM
John,

No. It's a wonder they stayed on the unballasted, primitive tie tracks.

I recall reading memoirs of a Shay fireman (or engineer?) who declared that you could always tell when the train derailed because the ride suddenly felt smoother.
  • Member since
    December 2001
  • From: Austin, TX
  • 10,096 posts
Posted by lionelsoni on Thursday, March 25, 2004 12:40 PM
The thing at Rochelle is a crossing. A crossover is something else.

The normal flangeway depth is about 2 inches. You can often see wear on the bottom of the flangeway of a crossing frog. Most of the welding at crossings involves building up the rail at the inside corners of the frogs, where it gets beaten down by wheels going over the gap formed by the flangeway of the other track.

Prototype wheel treads are also conical, but not nearly so much as fast-angle wheels. It is a popular misconception that this is done to keep the wheelset centered on the track. Actually, it promotes a hunting behavior in which the wheels yaw back and forth. This does keep the flanges out of contact with the rails for most of the time, which is a good thing.

The real reason for conical treads is to provide more metal close to the flanges to allow for the greater wear there. When a wheel wears more-or-less flat, it is condemned. If you look closely, you can see that the rails are canted very slightly inward by the tie plates, by half the slope of a new wheel's tread, so that the rail wear is distributed evenly between the inside and outside edges by the mix of new and old wheels.

Most prototype wheels are 33 inches in diameter. Those on heavier cars are 36 inches. Those on low-slung cars are often 28 inches. I don't have a model here to measure. Anyone?

Bob Nelson

  • Member since
    April 2003
  • 305,205 posts
Posted by Anonymous on Thursday, March 25, 2004 12:56 PM
Bob:

Thanks for that post. Very informative.

Elliot:

I'm with you on this one. I don't want to change the wheels. Maybe the couplers, but not the wheels.

Tony
  • Member since
    April 2003
  • 305,205 posts
Posted by Anonymous on Thursday, March 25, 2004 1:48 PM
i feel pretty sure that if you made your tracks as level and smooth as the prototype does AND if you made your engines and rolling stock weigh as much (in scale of course) as the prototype does AND you make your turns as wide as the prototype does...

the choo-choos would stay on the tracks with them skinny little prototype flanges.

BUT...[:p]
our tracks are too rough and uneven, out turns are too tight and our rolling stock is too light and we need the big flanges to keep them on the tracks.

look closely at any scale model wheels, if you made them life size they would ALL look silly compared to the prototype. i own almost entirely postwar lionel and they all look silly next to the real thing. take a picture of your favorite toy train and superimpose it over a picture of the real thing in your photo editing software. whew!
  • Member since
    February 2002
  • From: Mile High City
  • 296 posts
Posted by jkerklo on Thursday, March 25, 2004 3:13 PM
Thanks Bob.

Crossing, of course.

I never knew rails and wheels were so complicated. Next time I can see a train and track up close, I am going look for the details you mentioned.

I have spent time up on Tennessee pass with a drink beer/take pictures of trains gang. The rails on a curve there seemed to be worn on the outside and I couldn't figure it out. Until one of the gang who had worked on the railroad explained that on a curve, the worn inside rail is moved to the outside of the curve to get more life out of the rail. Indeed, there was rail discarded at the side that was worn on both sides, and must have been the previous outer rail, that had once been the previous inner rail.

This makes sense once one understands the "pulling a train around a curve" concept.

John Kerklo
TCA 94-38455
www.Three-Rail.com
  • Member since
    August 2003
  • 6,434 posts
Posted by FJ and G on Thursday, March 25, 2004 3:31 PM
John,

I often dream of being rich enough to purchase T Pass. I'd run a tourist line or at least have handcar races there. Would be a wild ride going downhill, esp. after a few beers.
  • Member since
    February 2002
  • From: Mile High City
  • 296 posts
Posted by jkerklo on Thursday, March 25, 2004 4:00 PM
Dave,

There is an excursion/dinner train that runs year 'round on the Tennessee Pass tracks through Royal Gorge.

http://www.royalgorgeroute.com

I know a guy that knows the guy running the excursion. If you would like to use the track for handcar races, I might be able to make a connection for you. Sounds like fun. When would you start?

John Kerklo
TCA 94-38455
www.Three-Rail.com

  • Member since
    December 2003
  • From: St Paul, MN
  • 6,218 posts
Posted by Big_Boy_4005 on Thursday, March 25, 2004 4:19 PM
EVERYONE, the term for the pieces of track at Rochelle is DIAMOND, using it will help avoid confusion, and I believe that that IS the term real railroads use. I have nicknamed Rochelle the "QUAD DIAMONDS" because of the track there.

Flangeways on diamonds or turnouts have to be there of course, but the flange itself should never be called on to support the car, real or model. there will always be some noise and unsmoothness when the wheels have to cross a rail, but the goal is always to minimize that. Less noise, less bounce, fewer derailments at frogs.

Frogs can take a real pounding, and get worn out in a hurry. It happens all the time on the real railroads. A couple of summers ago, my daughter and I watched as a track repair crew welded a new chunk of steel into place on the point of a damaged frog. This was a temporary fix and has probably been redone since, but this spring they will replace the entire turnout.

Frog damage can even happen on models, though most people don't run their trains that much. It did happen at the mall display on a few frogs.

As for rough and uneven track, I would think that it wolud not as big of a problen on tabletop type permanent layouts, but there will always be floor runners and of course KIDS.[:D]
  • Member since
    February 2002
  • From: Mile High City
  • 296 posts
Posted by jkerklo on Thursday, March 25, 2004 5:31 PM
Elliot,

Not sure about wheel flanges not ever supporting the wheel.

On toy trains.
If I look carefully at a 45 degree DIAMOND, the plastic part in the center has a raised area adjacent to the outside rails. This area has clear marks, worn grooves even, from wheel flanges.

As I look at a well used O22 switch, there is a strip of metal by the gap. The metal has a clear marking of grease.

My question about wheel flange height was motivated from an observation that some cars went over switches smoothly, and some bounced.

For real trains, I haven't looked at a diamond closely, but I don't see how wheels can go over a gap without either using the flange or dropping down a bit. Dropping down has to be a more serious problem, since the wheel would then have to be pulled up again.

John Kerklo
TCA 94-38455
www.Three-Rail.com
  • Member since
    December 2001
  • From: Austin, TX
  • 10,096 posts
Posted by lionelsoni on Thursday, March 25, 2004 5:57 PM
You can call it a crossing, or a diamond, or a diamond crossing. See http://www.sdrm.org/faqs/defs.html

Bob Nelson

  • Member since
    April 2003
  • 305,205 posts
Posted by Anonymous on Thursday, March 25, 2004 6:49 PM
call it what you will... i'm gonna call it the "four-way-crossover-switcheroo-thingy"!

most toy train switch frogs have a rail shoulder for the flanges to ride up on, some dont.

the prototype doesn't ever ride on the flange, unless derailed.
  • Member since
    December 2003
  • From: St Paul, MN
  • 6,218 posts
Posted by Big_Boy_4005 on Thursday, March 25, 2004 11:41 PM
Yes Bob, but crossing can be confused with GRADE CROSSING, where a road crosses the track.

So to recap we have crossover, crossing and grade crossing. Before I get crossed up or cross, I think I'll stick to calling the intersection of two tracks a diamond. [%-)][swg]

Beside, have you ever tried to give the woman in your life a crossing? I think she'd perfer a diamond, don't forget to ask her if she wants O or 027, and what degree.[:P]Just watch out for flying trains, if you ever cross her.[swg]
  • Member since
    April 2003
  • 305,205 posts
Posted by Anonymous on Friday, March 26, 2004 8:38 AM
I believe I read somewhere that the real railroads have tolerances for the difference in height between two pieces of rail that we simply can't reproduce on a layout. The logic goes like this:

Suppose you work really hard & don't have any two sections of track that have a height difference greater than 0.1 inch. That seems pretty close, right? Well, if you scale that up to the real thing, thats a height difference that's closer to 4.8 inches. No railroad would allow two adjacent pieces of rail to be that far out of whack.

Jim Barrett emailed me his procedures for getting smooth track. He uses cork roadbed & planes them so that the transitions between pieces are smooth. Maybeyou can get the same effect with a sander, I don't know. But getting smooth track work takes effort and patience, it seems.

Tony
  • Member since
    August 2003
  • 6,434 posts
Posted by FJ and G on Friday, March 26, 2004 8:48 AM
John,

That's good news indeed. I had imagined them tearing out the tracks soon. There are some speeder clubs around the country who might enjoy tripping there.

Tony,

Some folks have planed homosote but that creates a dust storm. Other folks have laid the track and then tested and retested the tracks for months, making minor adjustments to the L-girder risers and to the cant of the tracks using shims.

In my experience I find O scale much more forgiving than HO.

Dave V
  • Member since
    August 2003
  • 6,434 posts
Posted by FJ and G on Friday, March 26, 2004 8:54 AM
Here's a neat website relating to Proto48 and tiny flanges:

http://www.proto48.org/p48_art_05.htm
  • Member since
    February 2002
  • From: Mile High City
  • 296 posts
Posted by jkerklo on Friday, March 26, 2004 8:24 PM
Here's a shot of a steam engine driver, showing the flange. Apologies, but I didn't take a ruler to hold up. Next time.



The flange was about an inch, maybe a little more. Definately not 2 inches.

The wheel/rail surface is not square and flat. It was slanted, curved a little, and higher next to the flange, as Bob described. The rails it is sitting on are rounded on the edges, and the top is a gentle curve, not flat.

I couldn't verify the "toe-in" aspect of the rails, but since this was yard rail (in a building actually) it wouldn't necessarily have answered the question.

By-the-way, Elliot, this is a shot of a driver on your namesake.

Hey, as I signed on, I saw that Big_Girl_4005 was also on. Any relation?

John Kerklo
TCA 94-38455
  • Member since
    December 2003
  • From: St Paul, MN
  • 6,218 posts
Posted by Big_Boy_4005 on Friday, March 26, 2004 9:13 PM
Yup John, she shares the "roundhouse" with me these days.[;)][:X][:D]

Thanks for the cool photo. Makes me want to visit all the more.

By the way, it is true that some of our toy train's flanges do srtike parts of different track elements. While there may be some inconsistancy in wheel size and flange size the bigger problem may be the amount of slop around frogs and flangeways. If the rigid wheelsets found on modern equipment were gauged properly and uniformly, a lot of the track tolerances could be tightened.

Some of the worst offending equipment has to be post war with it's wheels that spin on the axles. I made the decision a long time ago to not run my post war trains any more, they just cause too much trouble.
  • Member since
    February 2002
  • From: Mile High City
  • 296 posts
Posted by jkerklo on Friday, March 26, 2004 9:36 PM
I run mostly postwar, and your right about the problems. It is not so much that the wheels spin on the axles, but that they wobble and shift, sometimes a lot. Cars that derail get removed, but I am going to have to address the problem someday, or keep buying rolling stock, forever.

I tried to come up with an easy way to add some sleeves to wheels, but the shear volume is overwhelming. The machinist in me is still tempted.

What about real train diamonds? The area that the Forney museum in located in is rather industrial, and there is a lot of train track around, but I couldn't find a diamond. No real need for one, I guess.

Someone is going to have to go to Rochelle and take a look. Who is closest?

I have some more pictures for you. I will post them in a couple of days. 4005 has a good home.

John Kerklo
TCA 94-38455
www.Three-Rail.com

  • Member since
    March 2004
  • From: Jelloway Creek, OH - Elv. 1100
  • 7,578 posts
Posted by Buckeye Riveter on Friday, March 26, 2004 10:08 PM
If you can keep your track gauge tolerance the same as the prototype railroads and then reduced to scale you are on your way to reducing the size of the flanges on your model trains. From some rough calculations, based upon a 1:50 ratio and a tolerance of gauge at + 1/2 inch for the prototypes, Big Boy better keep that Gargraves track tolerance (slop) to + 1/100 of an inch or be fined by the FRA. If he can meet or exceed the 1/100 tolerance, he is on his way to reducing his flange size proportionally to the real flanges.

Although the model is a scale or semi-scale representation of the real engine or car, I don't know of any model that has exact scale operating characteristics such as wheel friction, axel loading, spring coefficients, deformation, etc, etc.

http://www.ble.org/pr/archive/headline033100a.html

Celebrating 18 years on the CTT Forum. Smile, Wink & Grin

Buckeye Riveter......... OTTS Charter Member, a Roseyville Raider and a member of the CTT Forum since 2004..

Jelloway Creek, OH - ELV 1,100 - Home of the Baltimore, Ohio & Wabash RR

TCA 09-64284

  • Member since
    March 2016
  • 1,447 posts
Posted by Eriediamond on Saturday, March 27, 2004 7:16 AM
Gentleman-Ladies, There seems to be some debate here for the reasons of tin-plate wheel design. Some answers and information I agree with and some I don't. One reply stated that it's a misconception that conical wheels are designed to keep the wheels centered between the rails. They are designed, infact for that exact purpose, the yawing is infact, caused by the wheels doing that and the flanges are there only to help, as he stated that they touched the rail occasinaly. If everything was perfect, no friction in the bolster, both rails in perfect alignment, both rails being the exact same height, no centrifical forces on curves and no cross wind condtions, then the wheels would keep themselves centered and no flanges would actually be needed. Now, for our tin-plate Lionel, American Flyer, Marx and others. The main reason for those high angle conical wheels with oversized flanges has been hit upon here but two of the biggest reasons seem to have been overlooked. First, thse trains were designed to be set up on the floor and played with and not operated on scale like curves or speed or perfect track. Second and the one most important reason for the wheel design is that all these trains had truck mounted couplers so these trains could negotiate the sharp curves, both forward and in reverse as well. How many times have you scale modellers answered to people with derailment problems to body mount their couplers. The tin-plate manufacturers answer to these problems was the cone wheels with deep flanges. We sometimes forget that these trains were designed to be played with by kids and made as inexpensively as possible and not for fine scale modellers. Anyway, this is just my believe, and anyone is welcome to disagree with me.
  • Member since
    March 2004
  • From: Over the Rainbow!
  • 760 posts
Posted by eZAK on Saturday, March 27, 2004 11:45 AM
Given what I've read so far, even though no one has actually answered the underline Q,
a 1/48 scale wheel would be aprox. 3/4" O.D.

That would make the O.D. of flanges on our toys closer to scale than the wheel it self.

In other words the flange is NOT over sized! It is the wheel that is under sized [:D]
Relax, Don't Worry, Have a Home Brew!</font id="size2"> Pat Zak</font id="size3">
  • Member since
    December 2001
  • From: Austin, TX
  • 10,096 posts
Posted by lionelsoni on Monday, March 29, 2004 8:12 AM
Ken, because of the wheel conicity, the yaw rate is proportional to the (negative of the) lateral displacement. The lateral-displacement rate is proportional to the yaw. Thus the second derivative of lateral displacement is proportional to the negative of lateral displacement. The solution of this differential equation is sinusoidal. That is, once disturbed, the truck oscillates left and right, as I have had the opportunity to observe, riding in a car with holes in the floor.

John, if you can get your hands on a tie plate, you can easily measure, or even see, the difference in thickness from one side of the bed to the other. The rail itself is symmetrical--the tie plate is what gives it the cant.

There is a diamond crossing nearby. I would be happy to examine it for you. What would you like to know about it?

Bob Nelson

Join our Community!

Our community is FREE to join. To participate you must either login or register for an account.

Search the Community

FREE EMAIL NEWSLETTER

Get the Classic Toy Trains newsletter delivered to your inbox twice a month