Western Railroad Discussion > Rail wear

In watching some train videos tonight, I started thinking about rail wear on curves and which rail would wear out faster. On one hand, the outer rail would have more force on it from the wheels and flanges. However I would think that the inner rail would get a lot of wear from the inner wheel likely to be the one that is doing the majority of the sliding from being on a solid axle with the outer wheel which is probably more heavily loaded and not as likely to slip. So, which rail typically wears out sooner?

Let's see if I can explain this the way an old division engineer told me.  The tread of the wheel is slightly tapered, that is, the wheel is slightly larger in diameter by the flange.
I believe freight car wheels have a 1 in 10 taper.  This is done so the whel will not slide on a curve.  It is just a bit shorter around the inside of a curve that the outside rail.  The curve may be super-elevated to force the wheel to ride on the outside of the tread on the outside of the curve, making up the slight difference in distance around the curve.
The trick is to elevate the curve just enough that a train moving a the most common speed will remain riding on the center of the rail.  A slower train probably ride to the low side, and a high speed train to the high side.  Consideration must also be given to the practical speed limits of the curve too.


Norm,

Norman Orfall
Helendale, CA
TIOGA PASS, a private railcar

Heres a picture I took, this was in Mullan Pass, MT..  You can see how the flanges dig into the outer rail in the curve. 
http://www.railpictures.net/viewphoto.php?id=565763&nseq=45 
 

In these days of unit trains, 143t cars and 12-20 mph speeds all super elevation is being removed from mountain grade curves.

Welded rail has pretty much made harmonic rocking a thing of the past allowing posted speeds in lower below 25mph range.

Posted from iPhone

There is significant wear on both rails, and they are different.
Norm is correct up to a point.  The goal of establishing superelevation is to have the net result for trains going regular speeds to have about 1+ inches of "underbalance" so there is somewhat more weight on the outside wheel.  This keeps it against the outside rail where it tracks around the curve and there is little danger of a stringline derailment (pulling the train off to the inside of the curve).
The result is that the outside rail wears against the gage face and top gage corner and the inside rail tends to flatten out.
Rail grinding can adjust the profiles of both rails to enhance the steering effect of the taper of the wheels and reduce the rail wear.
AAR interchange rules are very sloppy, wheels with what are now recognized as bad profiles are still legal for interchange.
Closed loop operations like the big iron ore haulers keep all their wheels to a good profile by turning on wheel lathes or milling machines as often as needed, which turns out to be not that often because the wheels steer well and do not experience heavy wear.
Rail grinding also removes micro cracking from rolling contact fatigue, if not done rail can die an early death not from wear but from fatigue fractures.
Come to the WRI conference and you will get four days of classes on this.  (Wheel Rail Seminar)

Originally, the wheel taper was 1 in 20, and the tie-plates are made accordingly (1 in 20) in order to tip the tails slightly "inward". For high speed passenger service on the Santa Fe, back in the 1940s, they discovered that the wheels "hunted" laterally way too much on the two axle EMD "F Unit" trucks, at speeds well over 60 MPH. Thus, the Santa Fe changed to cylindrical tread on all their 4-axle passenger locomotives. In the 1980s, some bean-counter at Amtrak inquired WHY they were "charged extra" by the Santa Fe for wheel truing, and WHY did the Amtrak units operating on Santa Fe require "different wheel truing cutters"? Well, to make a long story short, Amtrak told the Santa Fe to stop using the "special wheel truing cutters", i.e. for the cylindrical tread, on their Amtrak units.

With the loss of the cylindrical tread wheel tread profile, guess what? At speed over from about 65 MPH to over 100 MPH, the wheels began "hunting" laterally again. The EMD Engineering Dept. had to get involved, and after much research and testing, the "Wide Flange Uni-Point Contour" wheel profile with a 1 in 40 taper was developed, which pretty much cured the problem. However, later production GP40s, GP50s, GP60s, and F40PH units were subsequently delivered with, or re-equipped with, hydraulic dampers to prevent truck yaw at speeds over 60 MPH.

In conclusion, today the standard wheel profile taper is 1 in 40, and rails are still "tipped inward" at 1 in 20.

Thanks for all the very useful information.  A followup question - if rails and wheels are profiled such that sliding around curves is minimized, why are flange greasers still in use (and more being used all the time)?  Are flange greasers trying to solve another problem, or are the various wheel/rail profiles not enough by themselves to eliminate wheel slide?

milwrdfan Wrote:
-------------------------------------------------------
> Thanks for all the very useful information.  A
> followup question - if rails and wheels are
> profiled such that sliding around curves is
> minimized, why are flange greasers still in use
> (and more being used all the time)?  Are flange
> greasers trying to solve another problem, or are
> the various wheel/rail profiles not enough by
> themselves to eliminate wheel slide?

Well, the wheels don't really "slide". That's why the tread is tapered. The flange greasers serve a two-fold use: 1) help to reduce flange wear as the flange is forced against the side of the rail head in the sharper curves, and 2) help to reduce rolling resistance on grades.

> In these days of unit trains, 143t cars and 12-20
> mph speeds all super elevation is being removed
> from mountain grade curves.

What is the usual on 10--deg curves now-- 2 inches?

HotWater Wrote:
-------------------------------------------------------
> Originally, the wheel taper was 1 in 20, and the
> tie-plates are made accordingly (1 in 20) in order
> to tip the tails slightly "inward". For high speed
> passenger service on the Santa Fe, back in the
> 1940s, they discovered that the wheels "hunted"
> laterally way too much on the two axle EMD "F
> Unit" trucks, at speeds well over 60 MPH. Thus,
> the Santa Fe changed to cylindrical tread on all
> their 4-axle passenger locomotives. In the 1980s,
> some bean-counter at Amtrak inquired WHY they were
> "charged extra" by the Santa Fe for wheel truing,
> and WHY did the Amtrak units operating on Santa Fe
> require "different wheel truing cutters"? Well, to
> make a long story short, Amtrak told the Santa Fe
> to stop using the "special wheel truing cutters",
> i.e. for the cylindrical tread, on their Amtrak
> units.
>
> With the loss of the cylindrical tread wheel tread
> profile, guess what? At speed over from about 65
> MPH to over 100 MPH, the wheels began "hunting"
> laterally again. The EMD Engineering Dept. had to
> get involved, and after much research and testing,
> the "Wide Flange Uni-Point Contour" wheel profile
> with a 1 in 40 taper was developed, which pretty
> much cured the problem. However, later production
> GP40s, GP50s, GP60s, and F40PH units were
> subsequently delivered with, or re-equipped with,
> hydraulic dampers to prevent truck yaw at speeds
> over 60 MPH.
>
> In conclusion, today the standard wheel profile
> taper is 1 in 40, and rails are still "tipped
> inward" at 1 in 20.

Let me clarify this a bit. Today's standard wheel profile has a 1:20 taper and a flange profile known as AAR1-B which is nearly the same as the Unipoint EMD developed in the 1960's to reduce flange wear, particularly on 3 axle trucks. This profile has a compound curve transitioning from the tread to the flange that effectively increases the taper as the wheel moves toward the rail. The AAR1-B is the profile used on most freight car wheels and was derived by an AAR study of worn wheel profiles in the early 1980's..

Taper is good for curving but the more taper on the wheels, the more the tendency for instability, generally known as hunting, to occur. As Jack noted, the cylindrical tread was used by the Santa Fe to control truck hunting at higher speeds, as the EMD GP swing hanger truck is notorious for hunting above about 60 mph with worn 1:20 taper wheels. A cylindrical profile won't generate steering forces so it wears the flanges rapidly and tends to track with one wheel or the other on each axle up against the rail head continuously. EMD developed the 1:40 taper as a compromise between the cylindrical and the 1:20 to try to balance flange wear and stability. The 1:40 is used by most passenger operators today but it has a simple 11/16" radius transitioning tread to flange rather than the compound radius like the Unipoint or AAR1-B. It's known as the APTA 140 profile - APTA has a nice standard for wheel profiles that provides a lot of background on wheel profiles and a good explanation of flange angle and why it's important - you can it find by searching for: APTA-PR-M-S-015-06.pdf

Dave