Yesterday, while at one of my favorite trains watching spots, which includes a moderate curve, a long BNSF freight train passed. Most of the cars went through the curve without the flange hitting the rail, which I presume was the cause of the screech, but the wheel flange on several cars at various locations throughout the train were touching the rail as indicated by a loud screech.
Why would most of the cars get through the curve without the flange touching the rail but not all of them?
It has much less to do with the flanges than it does with the solid axle between the wheels.
When passing through a curve, one wheel will be sliding at least a little bit. Which one depends on a variety of factors.
When that sliding occurs, you’ll often hear the squeal. Tune in to the Desher rail cams for a regular dose.
Several things can affect the squeal. Dry rail is one - wet rail lubricates the contact and the sliding is relatively silent. Weight is another. An empty flat car is no where near as noisy as a loaded coil car.
Curves on the transfers at Deshler are 10 MPH. I haven’t really listened to trains on broader curves or at higher speeds.
If you can manage a close-up view of the roadbed on a sharp curve, you’ll likely see flakes of metal on the ballast from the sliding.
My house is 1/2 air mile from Sykesville’s station (operating as Baldwin’s Restaurant) on the CSX’s Old Main Line - the track follows a series of curves following the Patapsco River.
There are occasional notes of rail singing as the coal trains pass through town. The OML does have several ‘greaser’ locations along its length.
How freely a truck swivels on its center plate can affect the amount of flange contact on curves. The center plates are lubricated but can run dry over time.
I have seen metal flakes on the ties and ballast near the outer rail on sharp curves with none near the inner rail. I think the flakes come from the gage side of the railhead and the wheel flanges rather than the wheel tread. The gage side of the outer railhead was noticeably worn while the inner rail gage side was not worn at all.
(Center Bound trucks) … big concern especially on certain locomotives and articulated cars
pretty much the opposite of truck hunting
I would only be guessing, since I’m not in the business, but tires and rails that are both well along in their cycles of utility probably also have a bearing on noise. Tires eventually leave their profile of a truncated cone as they wear. I would guess that the contact profile changes as a result, and also the place where the contact takes place. That might increase both scrubbing and noise, along with radius.
I’m in a rail-side bar with friends in the west end of Toronto- The Blue Goose to be exact, a great old neighbourhood pub where the actual foot rest at the bar is a length of rail- the VIA and GO Transit yards are nearby. When the GO trains go by they don’t make any noise- freights made screeching noises. My question is, do the GO train coaches have a solid axle?
I thought the wheel taper was supposed to compensate for the difference in distance traveled as a train rounds a curve. The wheelset moves to the outside so its outer flange runs against the inside of the outer rail. That would be only the flange fillet, and not the entire flange wall. Then with the wheelset shifted off center like that, the outer wheel is riding higher on its taper than the inner wheel.
To a point. For most mainline curves that will work. On a tight, slow curve, not so much. We regularly get squeal on our 5.5 degree curves at 25 mph. I don’t know what the transfers/wyes at Deshler are.
Euclid -
That principle is valid only for a range of curves, from tangent up to some radius that I don’t know how to compute tonight (too many variables, constants, and formulas that I don’t have at hand). In principle, it would be the different diameters at the contact points - near the flange on the outer rail and hence a longer circumference, and further from the flange on the inner rail and hence a shorter circumference - the taper is 1:40, IIRC. The difference circumferences over the distance across the track between the two contact points forms a truncated cone. Solve for the distance from one of those contact points to the point of the cone will yield the radius. When the wheel on the outer rail is as tight as it can be against the rail, that’s the shortest radius / maximum curvature which the wheel can traverse without slipping, Any curve sharper than that will have a path that’s inconsistent with the circumference of one wheel or the other, and will create a slipping condition.
Since the Wye’s are 10 MPH, I would guess the curves that create them would be on the order of 10 or 12 degrees or more. Mudchicken or Paul North would know better than I would.
Yes, I can see how the taper compensation could only work in a certain range of speed and sharpness of curve. I have seen slow tight curves in yards and junctions where the trains are pulled hard and want to stringline, so the flanges on the inner rail are up tight against the inner rail. Then apparently, the inner wheels turn slower than the passage of rail would call for under outer rail. This badly mushrooms the rail head of the outer rail, creating a huge burr off of the rail head, nearly doubling its width.
Usually your wye curves are a function of what real estate you have to work with…have seen plenty of the “or-more” side of the knuckle buster 20+ degree curves where bypassed knuckles are the rule and locomotives will never couple. (I imagine PDN had it worse than me on older railroad R/W’s)
Ideally, you work towards the same degree of curve as the equivillent curve in the turnout, but that never happens. [X-)]
…and who is maintaining the wheel taper (maybe when new?) [:-,]
Logic says that in a stringline situation, the flanges would be tight against the inner rail. A train moving with enough speed might force the flanges against the outer rail, due to centrifugal force.
AFAIK, the rails at Deshler haven’t been swapped for several years, and I haven’t noticed any indication of the mushrooming you mention during my visits there.
Which wheel would slip would be a function of the center of gravity of the car. If the curve is canted, and the speed is slow, that will be the inner rail. The outer wheel will slip.
There are probably 15 trains each on the SE and SW transfers per day at Deshler.
If there’s enough speed to throw the center of gravity toward the outside of the curve, then the inner wheel would likely slip.
28 stone rail grinders[swg] - only really have been in play for the last 25-30 years.
Euclid… the proper terms are shelling and spaulling.
As have 286K and 315K weights, and various kinds of head-hardened rail, that make repeated ‘maintenance’ grinding workable and ‘cost-justified’. Euclid might enjoy reading up on the theory of ‘magic wear rate’ and the assumptions involved in it.
The ‘flip side’ of rail grinding is the practical use of underfloor lathes for reprofiling wheels on a regular basis, which is much more of a ‘thing’ than it used to be.
He will probably find more references if he spells the latter ‘spalling’ as in discussions of gear teeth or bearings. The mushrooming requires axle load high enough that the rail steel under the work-hardened surface ‘cold flows’ under the combination of vertical and lateral load (this is a more common problem than it ‘should’ be in assuring ‘hard coating’ integrity in actual service) and as MC indicates it is better addressed with periodic grinding than ‘lighter loads’ or different operation, although it could be argued that lower weight on contact patches – which BTW should be characteristic of commuter equipment not fully loaded, which would often be the case to observers – would produce perhaps dramatically lower wear or distortion.
There are further considerations regarding braking forces, and the whole phenomenon of gauge-corner cracking that would have such implications at Hatfield in the UK.
[/quote]
As Larry pointed out, I had inner and outer reversed in my above post, so I have edited that. In any case, I don’t think this is shelling or spalling. It appears that a lot of the rail head has been cold worked into a massive burr that hangs off rail head to the outside of the curve. It does not look like wheelslip rail burn, since it is quite consistent throughout the curve. The burr overhangs the side of the rail head by an inch or more, and effectively widens the rail head top surface to maybe 3.5”-4”. Although, in the burr area, the surface is rather organic in form.
Thereby explaining why, in the days of “stick” rail, many roads swapped north and south rails to extend rail life.
I can imagine a giant machine that would do this with welded rail but I doubt it would be practical.
On CSX it is done with the Curve Patch rail gangs. The high rail used at one location may have been the low rail at another location. Rail is a valuable capital asset - carriers will use it as necessary to get their full economic value out of it.
Remember as the carriers removed stick rail from the main tracks, they loaded it up on rail trains and moved it to their rail plants where it was cropped and welded and sent back out for installation at some other location.
Only when rail is totally worn out will it be considered ‘scrap’.
I thought the tapered wheel served as enough of a differential.