The recent cold snap in the American Southwest caused New Mexico’s Rail Runner service to be curtailed. According to reports, the single digit temperatures (Fahrenheit) caused the rails to contract, making for an unsafe running condition.
I am curious about what specifically was or is the problem. Other rail lines in the western states apparently did not shut down because of the cold weather.
More generally, how are tracks constructed and maintained to permit safe running from below zero temperatures to above 100 degree temperatures?
When the rail is laid, it may be heated to a specified temperature. This becomes its neutral temperature; the point at which it is neither trying to expand nor contract. Thereafter the rail is attempting to kink or pull apart depending on the relationship of actual rail temperature to the neutral temperature. Presumably rail laid in the desert southwest will have a higher neutral temperature than that in the Midwest, even when owned by the same railroad.
Building off these two posts, up here in Massachusetts, the MBTA runs with restricted speeds whenever the temperature gets into the upper 70s. The neutral temperature is colder, so we can run at normal speed most of the time in the winter.
The rail, as already stated, is under a great deal of tension when the temperature drops well below the neutral temperature. Especially if the rail is new and recently laid, the first really cold snap can find any trivial imperfection in the rail or welds. It’s a miserable job for the track forces, who get called out, often in the middle of the coldest night of the year, to get the railroad running again.
A common starting point is to lay a rope soaked in kerosene or diesel oil next to the rail and light it. The heat causes the rail to expand back its appropriate length.
In very cold weather, I assume the rail contraction is localized. Otherwise, track workers would have to lay miles and miles of burning rope. Is the contraction problem most prevalent around switches?
Another question: In very hot weather, how long does it take for kinks to form? Is kinking a gradual process, or does stress build to the point that the rail just kinks in a matter of seconds or minutes? If one rail kinks, does the other follow?
With heat kinks, since both tracks are tied together through the wooden or concrete tie, they do go together when a heat kink occurs. Heat kinks are much rarer when there is sufficient ballast to hold the tracks in place, but if there is insufficient ballast, worn-down ballast, no ballast at all (freshly laid track), or severe heating, like in desert conditions, heat kinks are common. Usually, it is a sudden thing that happens, as the sideways forces from rail ends pushed against each other finally overcome the hold of the ballast, and a part of the track then jets sideways to relieve the pressure. This is why with continuous rail, there are still occasional joints-it’s to give the rails some room to expand without pushing sideways. About all that can be done is preperation for the environment a track will be laid (setting the right neutral temp at time of laying), and patrols during hot summer months.
In case you’re wondering, derailments have happened from joints not having enough room between the rails, and one rail then climbs higher than the one in front of/behind it during expansion. As for the sideways kinks, anything can still usually make it through at slow speed with advance warning, though usually maintenance is called to deal with them instead.
That’s it, I suspect. Cold temperatures that are attained over a period or a couple days aren’t usually as much of a problem - esp. when there are heavy freight trains running over the line to shake things up from time to time and sometimes prevent concentrations of stress at a point (and sometimes aggravate that effect !).
Also, if some portion of the track is fairly new, and this was its first encounter with temperatures that cold, there might have been a good reason for an abundance of caution about that condition serving as a real-life testing procedure for finding the ‘weak links’ in the chain of rails, until it became clear that all of the rail and joints survived intact. And/ or if the officials had some reason to doubt the integrity or quality of their track system, esp. the rail’s neutral temperature and the rail fastenings, etc. . . . [:-^] some
It will depend on how large the gap is, which in turn is affected by how well the rail anchors were applied and maintained.
If the gap is small, heating it as described may be enough but you might need to warm the rails to as much as 120F above air temperature in the middle of a howling blizzard. First though you drill holes in both ends so a joint bar can be applied once the gap closes. It will be advisable to adjust the rail anchors immediately afterward, re-applying any that might have been removed to help the rail grow.
Another method is to cut out a piece and insert a short piece of rail, perhaps 20 feet long with joint bars at each end. The shape of the break can be a factor in deciding which method to use.
Either way, there is more work to be done in the spring before summer heat becomes the concern. The joints need to be field welded and the rail restressed to the proper neutral temperature. If a piece of rail has been inserted you need to keep track of how big the gap was in the broken rail since the replacement rail is now too long by that amount.
For an in-depth discussion of the topic of “neutral temperature” check this link out for the NTSB report on the 2002 Autotrain sun kink derailment. The really interesting stuff starts at about page 43 of the report.
One interesting point is that over time, the neutral temperature of a length of CWR can change due to various things, which is probably a phenomenon that give folks like our muddy feathered friend a lot of work.
Thanks Chuck, for that link. From the bottom of page 29 of that NTSB Auto-Train accident report, under “CWR Temperature Control” [emphasis added - PDN]:
"When the longitudinal mechanical and thermal compressive forces in CWR buildto a point at which they exceed the track structure’s ability to restrain them, a track buckle will occur. According to AREMA, the magnitude of force that may be developed within CWR, independent of its length, is the product of the difference in temperature between the neutral rail temperature [61] and the current rail temperature, multiplied by the cross-sectional area of the rail and the factor of 195 pounds per square inch (psi). [62] In the daytime, rail temperature is commonly about 30° F above the ambient temperature.
Thanks Chuck, for that link. From the bottom of page 29 of that NTSB Auto-Train accident report, under “CWR Temperature Control” [emphasis added - PDN]:
"When the longitudinal mechanical and thermal compressive forces in CWR buildto a point at which they exceed the track structure’s ability to restrain them, a track buckle will occur. According to AREMA, the magnitude of force that may be developed within CWR, independent of its length, is the product of the difference in temperature between the neutral rail temperature [61] and the current rail temperature, multiplied by the cross-sectional area of the rail and the factor of 195 pounds per square inch (psi). [62] In the daytime, rail temperature is commonly about 30° F above the ambient temperature.
Now if someone could just design telescoping track that problem would be solved. I’d like a tape measure with no tape but instead a holographic tape. A holographic tape would never bend and fall to the ground after being extended out 12 feet, it would just extend forever.
I have seen slip joints used on transit lines, usually near changes in track structure or bridges I think. One side is like a switch point, and the other rail can slide against it. So telescoping track has already been designed. I suspect slip joints may also have been used very occasionally on freight roads at critical places such as a movable bridge.