A recent discussion on a different list has prompted me to pose this question to those on this list who are qualified to answer. Do superelevated curves pose a higher risk of derailment to trains travelling slowly through them and why? My belief is YES, they do, although others argue that there is no increased risk. How do the laws of physics and gravity apply to my theory? Thanks, in advance.
…{I’m not qualified to give the professonal answer}, but just a comment. Rounding a high elevated curve with a train such as the Triple Crown at a slow pace, certainly by the law of physics should show the situation changes the center of gravity and we’ve had that train derail {stringline}, twice here on a wide sweeping curve.
Yes there is a very slight increase in the risk of a derailment to the inside for freight, of there is a corresponding decrease in the risk of freight derailing to the outside at normal speeds. That is why there is a limit on maximum superelevation. No freight car properly loaded should have a problem, but of course not every car is properly loaded.
‘Higher risk’ compared to what ? The same curve - with less or no SE - but the same train, speed, and in-train ‘draft’ = pulling forces ? I think the latter causes the problem more than the SE - it’s just that right at the ‘tipping point’, the SE is easy to point to and blame, because there’s usually no way to look at the in-train forces at that point. Sure, the locomotive ‘tapes’ will disclose what notch it was in and whether the sand was on, etc., but that can be quite a few cars ahead, so it doesn’t help us see what was happening back at the curve. Also, keep in mind that the most tractive effort = pull occurs at slow speeds - mainly because the engineer is likely trying to keep the train up to speed, or get it moving faster. At higher speeds, that pull isn’t needed, because the train is going faster - and the locomotives can’t do it anyway, because of the declining nature of the Tractive Effort vs. Speed curve - see a much longer separate discussion for that.
Good explanation.
Only minor quibbles would be that the resultant force line only goes through the center line of track when the car goes around the curve at “equilibrium” speed, where there is no lateral force placed on either the high or low rail. Since railroads typically don’t elevate curves for equilibrium speed (that would be a very high SE) they run with what is called “unbalance” which is a theoretical amount added to the actual elevation to get to the equilibrium elevation. Most railroads run at 3" unbalance and can get a waiver to go to 4", based on the ability of the equipment to negotiate it (usually mostly passenger equipment).
The max Velocity equation is the square root (can’t write it out mathematically so have to spell it) of Actual Elevation + unbalance (3 or 4) divided by .00007 times the degree of curve. If you have designed your curves so that V ends up being in the low 80’s for example, you probably would not want to set your max operating speed for pasengers at more than 75. That way, if you loose a small amount of elevation or a sharp spot develops in the curve you don’t drop below your max authorized speed.
You may be able to add a bit more elevation to off set this, but that depends on how much you have already versus what your standard maximum allowable is.
Many lines that had had good superelevation for passenger service had this reduced when they lost the passenger service with the advent of Amtrak. I don’t think it was so much because of the risk of derailment as it was that the lower rail would get more wear on it if the trains were moving too slowly for the superelevation. I suppose there would be a risk of the rail turning under the increased forces placed on it, but I think the wear would be more critical. Tie plates on curves may have more cant to them, which actually might bring the rail to a nearly vertical orientation. I’m hoping our resident curve-layer and Mudchicken will put his distinctive slant on this discussion.
Okay, thanks. So does that mean there IS an increased risk, or NO, there is no increased risk? That’s what I really want to know. [sigh]
As do I, CarI. Not taking anything away from the responses provided thus far, as I really do value solid input. However, I clearly stated that I was soliciting responses from only those “qualified” to do so. Gabe, where are you when I need you? [swg]
Well from exsperance I must say NO. Reason we have a super elevated curve ( had) its been cut back some but been in emergency on that curve and it looked like you coulde just tip the cars over and nothing happens, but even a freight hangs in there on that curve
The short answer is no. Slower speeds increase the load on the low rail, for the most part, which leads to accelerated wear, hence the lowering of curves after passenger trains are removed from the mix, as noted elsewhere.
The greater risk comes from over speed for the actual elevation in place. In the example I gave, if the curve looses super, the risk of running at the higher MAS goes up since now you are going outside the limits of the physics involved. Think of going around a curve in a car at a speed much higher than the posted limit. Which way does the car want to go?
Now think of how big the wheel flange is that is trying to resist going over the top of the rail, which is worn and may be lubricated, and the tie plate moves a little bit and is cut down into the tie on the outside a little, all leading to opening the guage so it may fall in before is actually climbs over.
An interesting read is the NTSB report on the derailments of Amtrak on BNSF in Washington due to rail seat abrasion on the outside of the curve in concrete tie track. (RAB-06/03, issued 10-18-06)
What’s wrong with Paul & Steve’s analysis.??? I don’t see any.
I’m gonna be more concerned with surface defects or reverse elevation.
Curves w/ no problem at equilibrium speed usually get 1 inch of super just to avoid reverse elevation and the chaos that comes with lateral force induced wheel lift.
Excess elevation is more of a detriment to the rail (low rail) if it is forced to handle tonnage at low speed and high speed passenger trains all at the same time.