First a little introduction on the issue, this is copied from W&H Main Yards site
"When a train goes around a curve, the outside wheels have further to travel than the inside wheels. This is the same event that occurs in automobiles which requires a differential rather than a solid axle. In the case of railroads a solid axle is necessary for stregnth and simplicity, so another solution was applied.
The treads of a railroad wheel are not flat, but beveled into a hyperbole with the smaller diameter toward the outside and the curve leading into the flange. The distance between the inside faces of the railheads is slightly larger than the distance between the outside faces of the wheel flanges by about an inch. So as one wheel rides up its rail, the other wheel rides down its rail. The different diameters in contact with the rail have different circumfrances, which compensates for the different distances of travel.
Now here’s the question - how does NMRA RP-25 address this ? Does it at all ?
Depending on the answer we can now understand if it is worth the effort reducing friction in the axels. Because if RP-25 wheels on the curve are behaving as if they were flat then the friction there is a lot more than friction in the axels [oops]
The difference in profile may be a mute point when you consider that the profile on prototype wheels change with wear from rubbing brake shoes. Conversely, the crowned profile on a railhead can wear down to being perfectly flat from this braking action.
Philadelphia is one of the few cities that has always had trolleys. Driving down streets where there are still tracks in the streets shows this wear pattern. The rails in the middle of the blocks (between stops) shows very little wear, whereas the rails near the trolley stops are invariably worn down and flattened. No doubt this wear pattern is exasperated by the use of magnetic rubbing blocks, but this wear should also take place on other rail head to lessor extent.
The NMRA HO wheel profile should at least remain constant since nothing actually rubs against the full width of the tread.
The RP25 wheel profile is similar to the prototype. The main differences are the tread width, an the depth of the flange. As the above diagram illustrates, the tread is tapered, and there is a small ‘fillet’ between the flange and the tread. Both of these items help keep the wheels on the rails. Here is the link for the RP25 wheel contour: http://www.nmra.org/standards/rp25.html
I have been in the hobby for a number of years,and the RP25 wheels do roll better that the old S-3 wheels. I would suspect that any quality trucks have RP25 wheels sets in them. Metal wheel sets will roll better than plastic ones, and will not accumulate/spread dirt as bad either.
We been doing a little thinking here, dimastep!?! The subject is interesting to consider, but, likely has “little to do with the price of tea in China”. I think the RP-25 standard has more to do with keeping equipment on the rails and passing through switch points, frogs and guard rails then compensating for the difference in speed between the inner and outer wheels. However, never having read RP-25 I am likely talking out the side of my neck! I know the rail head on my Atlas code 100 track is flat. So, no matter what the wheel’s contour is, the wheel is riding on a pointed edge and not a rounded edge like the prototype. My un-educated, but, seemingly common sense opinion is the models would have more friction in turns then the prototype, proportionately! We’re out in theory land here, which is very close to La La Land
You see, I like to run long trains, and they are loaded. So I’ve been trying to reduce friction as much as possible. After moving to IM wheels and KD trucks I have pretty much reduced friction in the axle, but noticed I am still not getting good results. Which got me thinking…
Look at it this way - the rail wear on model trains is nonexistant. So goes for wheel wear. So if you think about it, if the rails and wheels were made prototypical so that (as described above in my first post) there is no friction on the curve - we’d get one step closer to realism. That and also reduced the number of locos required to pull a train.
Dimastep,Their is wear and tear on our track and wheels over a long period of time…
Now IF you find a solution for overcoming friction I am sure the real railroads will be most interested in your solution…
When building cars and locomotives (not diesels), I always put extra effort in checking the free rolling capabilities of the trucks and mechanisms. If I find binding in the mechanisms or the trucks do not roll as well as others I’ve checked, I re-work until they do. Beyond this, I don’t know what else can really be done. I think metal wheels are helpful. They don’t spread dirt and they roll better. Matching axle length to bearing surfaces in the trucks is important. Shorter axles then needed is worse than too long. The axles should meet the bearings on their points. Because the wheel treads are conical, some of the function you mention is met. However, I would guess it is far more important to make the trucks as free rolling as possible.
Our RP-25 wheels have a bevel on the surface and ride centered on the track, but considering that we only put feet on our cars instead of miles per year, how can we worry about tire wear. Look at how long it takes to wear the plating off the drivers on a brass engine.
Curve oilers were used where their was a sharp curve with a high drag (read heavy cars and lots of them) and the flanges rub against the inside of the rail. The bad thing about curve oilers was it cut down on the traction for the loco because it oiled the top of the rail.
Railroads have been know to “open” the gauge a little on the curves but not very much as the cars would fall into the center of the track. One of the problems for the Pennsy [the standard (?) of the world] with the T-1, was the long wheelbase and it couldn’t go all over the line. It would look, to me, like as long as they’d been designing locos they would have known better, but I guess it was their halfhimers.
I haven’t looked at my track through a magnifier but I doubt that the rail heads are square as that is another problem for the manufacturers, curved would be easier to maintain.
o - Metal wheels(Atlas, Jaybee, P2K, NWSL, Intermountain/Reboxx, Kadee)
o - Metal needlepoint axles(NWSL, Jaybee??, Intermountain/Reboxx)
o - Matched axle length(Reboxx)
Other than using the Intermountain ‘roller bearings’, a correct length needlepoint
metal axle riding in a conical shaped delrin(or engineering plastic) sideframe will
give the best ‘low friction’ rolling trucks. That is a lot of work and expense to do!
I use for the most part, P2K metal wheels in standard plastic(delrin) trucks on all
of my rolling stock. I use a Reboxx ‘tool’ to core out the journal area on the trucks
so I have a nice conical depression for the needlepoint axle to roll in. The P2K
wheel sets have some kind of engineering plastic axles, but they seem to roll in
the truck sideframes quite well. Even the Athearn wheels/trucks will roll pretty good
if you clean up the flash on the inside of the truck sideframes, and core out the journals.
The problem is the plastic wheels - they just attract dirt/gunk, and spread it around
the layout. Some of the worst rolling wheels are my Kadee wheels. The Kadee axles
are also made out of some kind of engineering plastic, though it appears to be much
‘softer’ that the P2K wheels sets. They seem to roll in plastic truck sideframes OK, but
start to roll worse in metal sideframes as time goes on. My $28 Kadee boxcars start
to become the worst rolling cars on the layout! After inspection/cleaning, I have found
that the metal journals are full of the rounded axle ends. The metal trucks are grinding
away the ends of the axle! I still have a lot of Kadee wheel sets, so I clean up the trucks,
core out the journals with my Reboxx ‘tool’, and replace the wheel sets. One of these
days I am going to take a micrometer to some new Kadee axles, then order some
exact match Reboxx wheel sets and
I think everyone got a bit off-topic here. [:)]
The original question isn’t about friction in the axels, it is about friction on the curve because of different distance inside and outside wheels have to travel. This is why we have differentials in our cars, but rail cars DON’T have them. So they made a solution by shaping wheels properly. NMRA wheels also shaped specially, they are not flat, they form a slight cone. I wonder why ?
dimastep,
The tapered NMRA wheels should perform similarly as the real wheels around curves. I am sure there is some math that went into the hyperbolic shape of the real thing, but the typical taper of the NMRA wheels at their small scale should be considered “close enough”. Again, you can see the shape of an RP 25 wheelset here: http://www.nmra.org/standards/rp25.html. Reading the text, it says a taper is not required, but is allowed. Again, for molding or casing purposes, the wheels will likely be tapered.
If not from a lack of taper on the wheels, what else would cause the rolling friction to be higher around curves? Most model railroad axles typically have a conical point on the end that rides in a conical hole in the truck side frames. When a car goes around a curve, there will be a side load on the car, causing those points to be pressed harder into the conical holes, thus increasing friction. The same basic concept applies to the real things too. No matter what bearing surface between the axles and the trucks, a curve will put a side load on those bearings and thus increase the rolling friction. Of course, the better the bearings, the lower the friction, and the less this effect is felt. However, friction is just a fact of life. You can work to minimize it, but you can not get rid of it.
Well, I think we have made this mole hill into a large enough mountain, now! Sometimes I think we are so enamored with our perceived wisdom, we forget this is really only a simple hobby; the miniature locomotive pulls a string of cars around the layout! And as the Crusher used to say “And dat’s all da peoples needs to know”!
I’ve seen a hellava lot bigger mountain out of a smaller molehill. It just depends on how much idle time people have - like I’m waiting for the game to begin and don’t have time to go trainning before hand.
