Heat and cold speed restrictions

Amtrak Unlimited Discussion Forum

Help Support Amtrak Unlimited Discussion Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
Short answer. Yes
When Ribbon rail is laid it is heated and expanded to a nuetral temp. Depending on area anywhere between 95-115 degrees I believe. This gives room for typical expansion and contraction.

when conventional rail was laid there was a cardboard fiber strip of varying thicknesses placed between the rail ends to allow expansion.
 
Last edited:
No cardboard filler strip that I have ever seen. Only think I can think of is the about 1/4 inch thick chunk of insulating material used between rail ends at insulated joints.

The rail length does NOT change with temperature in continuously welded rail. The usual intent is to stretch or heat the rail the rail so that it is stress free at a relatively high temperature. The internal stress in the rail steel goes up and down depending upon the difference between the actual temperature of the steel and the stress free temperature of the steel. The 95 to 115 (Fahrenheit) temperature is the usual range for zero stress temperature in the US. The point is to keep the rail in tension most of the time. There are several reasons for this. First, push a rubber band, then stretch it. That should make it obvious why compression is undesirable. The tensile force you get in the rail even with a 100 degree difference between air temperature and rail temperature is still way below the tensile strength of the rail. When you have it in compression, it will buckle, that is kink if the stress gets too high, and too high is dependent more on the stability and strength of the ties and ballast than the rail's strength. Vibration tends to loosen things, so this kink may occur under a train as it goes over it, hence derailment potential. In signaled track a buckle in the track does not affect the signals. A pull-apart with a rail break does cause the signal to go red. Hence if you are going to have a temperature related failure, one that causes a rail break is the better one. The length of the string of rail has nothing to o with the internal stress due to temperature. Use of a good angular ballast stone, well compacted and with a wide, usually 12 inch preferred width beyond end of tie is important in controlling buckling due to compression in the rails, but really does nothing for you in tension in straight track. High tensile forces can tend to pull the whole curve to the inside, but usually not by much. The total force is related to rail size, thus the condition of ballast and ties is more important to prevent buckling with a large rail section, but the unit force within the rail depends solely on temperature difference, completely independent of the size of the rail section.
 
Oh yeah, back to the question of speed restrictions: Many systems do have hot weather speed restrictions, but usually not cold weather restrictions except in the more northerly areas, and that commonly due to frozen ballast as much as rail issues. The defining "hot" temperature is usually experience based and related to the zero stress temperature, so "hot" will be a higher temperature in Arizona than it will be in Montana, for example. Sometimes a large change in temperature in a short period of time will also be used to define certain speed restrictions.
 
Back to the maintenance question: Use a large well compacted ballast section, shoulders about 12 inches wide beyond the tie ends. Some of the European systems use a heaped up ballast shoulder, but I consider that near pointless. In wood ties, "box anchor" every other tie to reduce potential rail movement. If concrete ties, use a design with scalloped sides to improve their "bite" into the ballast. Use glued insulated joints to eliminated the effect of a discontinuity. Stretch the rail so that it has the highest zero stress temperature you can feel comfortable with. In AREMA the recommended zero stress temperature is defined by (2H+L)/3 + C, where "H" is the highest anticipated RAIL temperature, which could easily be up to 30 degrees higher than the highest air temperature, "L" is low anticipated rail temperature, which usually is considered to be the same as the lowest air temperature, "C" is an addition to the temperature based on experience and a certain amount of just feeling, and is usually in the range of 20 to 30 degrees. This is what gets you to the 90 to 115 degree usual selected value. Since the force transferred across insulated joints is carried by the glue, a large bond area is important, hence the usual 36 inch long joint bar. There is some use of 48 inch bars to reduce the unit force in the glue. My opinion is that this is a good idea.
 
Durango and Silverton had a heat kink in their jointed rail. MBTA had to re stress all the rail from BOS to Worcester as CSX had not done it properly and as well poor documentation.

The PRR type variable tension CAT has problems as well. Amtrak has to place heat slows due to CAT sagging before track slow. That is one case where MARC diesel trains can actually go faster than MARC electrics for a narrow temp band.

Then COLD PRR CAT can get too tight and might snap if electric trains go faster than some speed. Suspect it is different at different locations. These problems just one more reason to have constant tension CAT .
 
Then COLD PRR CAT can get too tight and might snap if electric trains go faster than some speed. Suspect it is different at different locations. These problems just one more reason to have constant tension CAT .
Unfortunately the world's richest country apparently cannot find the money to convert its most premier rail corridor to constant tension catenary in this century, as it would appear. 🤷‍♂️
 
Unfortunately the world's richest country apparently cannot find the money to convert its most premier rail corridor to constant tension catenary in this century, as it would appear. 🤷‍♂️
Never understood that one, either. In the overall picture, this is relatively cheap, and should pay for itself in fairly short order.

For those that do not understand what we are talking about here, the original Pennsylvania Railroad Catenary (overhead wire) has fixed terminal points at intervals along the line so, like continuous welded rail, it cannot change in length with change in temperature. Unlike rail, there is nothing to constrain it when the temperature is above the zero stress temperature, so it is slack and can bounce and wobble around. When the temperature is very low, the tensile force gets high to the point that there is danger of the wire breaking. This overhead is now in the realm of 80 to 110 years old.

For the last 50 plus years, the concept is to have the wire held at a constant tension. This is achieved by have wire segments with a more or less centered fixed point and weights free to move up and down attached to ends of these segments so that there will be a constant tension in the wire regardless of temperature. Consider this a much simplified explanation. The extension of the catenary New Haven to Boston is constant tension.
 
Unfortunately the world's richest country apparently cannot find the money to convert its most premier rail corridor to constant tension catenary in this century, as it would appear. 🤷‍♂️
I didn't realize that Luxembourg was having such trouble upgrading it's 275 route-kilometers of track. :)

(The richest country in the world is apparently Luxembourg. The Richest Countries in the World (2017-2022) (focus-economics.com)
It's also a little humbling to see that Ireland is Number two, and the United States is only number 6, after Switzerland, Norway, and Denmark.)
 
I didn't realize that Luxembourg was having such trouble upgrading it's 275 route-kilometers of track. :)

(The richest country in the world is apparently Luxembourg. The Richest Countries in the World (2017-2022) (focus-economics.com)
It's also a little humbling to see that Ireland is Number two, and the United States is only number 6, after Switzerland, Norway, and Denmark.)
The confusion in your mind is of your own making, so have fun with it, :D

I was talking of total GDP, not per capita GDP. I know that US is not the richest in terms of per capita GDP. But at the end of the day that is mostly a side issue since it is not clear what per capita GDP has to do with the ability to fund a relatively modest project, and has little effect on the fact that somehow we are unable to fund simple things like bringing arguably the most important rail corridor upto current level of technology.
 
"...American engineers have considered compensation devices not worth the complication, a belief well-justified by the behavior of the lines which are in service without those devices." - page 580 of the 1925 Electric Railway Handbook after pages and pages on wire sag and stress.

Perhaps a new edition is due.
 
The confusion in your mind is of your own making, so have fun with it, :D

I was talking of total GDP, not per capita GDP. I know that US is not the richest in terms of per capita GDP. But at the end of the day that is mostly a side issue since it is not clear what per capita GDP has to do with the ability to fund a relatively modest project, and has little effect on the fact that somehow we are unable to fund simple things like bringing arguably the most important rail corridor upto current level of technology.
Actually, I agree with your main point, of course, but per-capital GDP isn't exactly an irrelevant measure of wealth. If there are more people that need to share that large pot of total GDP and thus the per-capita is less, then the people in the country with the lower per-capital GDP won't feel as "rich" and might feel that the costs of the project aren't as "modest" as some think. This is also complicated by the fact that the per-capital GDP might not be distributed very evenly, so the majority of voters may even think that they're poor (and may really even be poor), despite the county's great total (and per-capita) wealth.

Anyway, it's amusing to think how those "USA-Number one in everything!" folks would react to the fact that we're not the richest nation in the world, and we're even beat by Ireland, of all places.
 
I am sorry that a careless mention of "richest" by me has given an opportunity to divert this thread wildly off course. I hope we can get back to speed restrictions and constant tension catenary. 🤷‍♂️
"...American engineers have considered compensation devices not worth the complication, a belief well-justified by the behavior of the lines which are in service without those devices." - page 580 of the 1925 Electric Railway Handbook after pages and pages on wire sag and stress.

Perhaps a new edition is due.
I was wondering what were the compensation devices that were in vogue back then. The simple device of using gravity for providing constant tension had not quite come into vogue yet I suppose. And of course since then the technology for catenary has developed much more rapidly outside the US than within. In any case, at present the NEC is the one glaring large stand out within the US.

AFAICT all new electrification of all colors tend to be constant tension, and the Connecticut portion of the NEC has been converted to constant tension by MNRR. I found this interesting discussion thread at the Trains Magazine site regarding the unique catenary suspension systems that were used on the NH segment that has now all bee converted to constant tension by MNRR:

http://cs.trains.com/trn/f/111/p/261670/2946166.aspx
Interestingly NJT after trying constant tension (Matawan - Long Branch), reverted back from it (Montclair Bay Strret - Great Notch), and then decided to fiddle around with dual mode instead of electrifying the remainder that are yet to be electrified in its system. But then again it is also taking them more than a decade to build some 15 miles of new track setting a new record for slow track building.

Interestingly, the new constant tension catenary being installed in the UK is tensioned using some sort of a compensating spring device, and not using weights suspended from pulleys at the end of each wire segment.
1627834962309.png

All of this is described in Network Rail - A Guide to Overhead Electrification (PDF).
 
Last edited:
Interestingly, the new constant tension catenary being installed in the UK is tensioned using some sort of a compensating spring device, and not using weights suspended from pulleys at the end of each wire segment.
View attachment 23814

All of this is described in Network Rail - A Guide to Overhead Electrification (PDF).
Good grief!! This is what I would call complicating the simple. Maybe this was the sort of stuff they were thinking about in the 1925 Electric Railway Handbook.
 
Good grief!! This is what I would call complicating the simple.
The Brits do have a penchant for complicating things and then moaning about how expensive it all is :D

They have been blowing both their budgets and schedules on all electrification projects in England, while mysteriously things seem to come out on or under budget and on schedule in Scotland! Go figure! Oddly there are some noticeable similarities of the sorts of things that go wrong between Network Rail in the UK and Amtrak in the US when it comes to electrification or upgrade of electrification. One of them surprisingly is the apparent inability to dig vertical holes into the ground consistently. :D
 
Last edited:
By the way, a spring does NOT give you constant tension. The tension will vary linearly with the amount of extension you have in the spring, and if the temperature gets high enough, will still result in zero tension in the wire. All it really does for you is give you a wider range of temperature that can be tolerated by the system.
 
When it comes to Dallas's light rail and speed restrictions due to heat, please tell me, is that restriction due to what? An issue with the rails, or the train wheels, or the connection to the electric wires??

 
Potentially both. When metal gets hot, it expands. When a long skinny piece of metal (such as a wire or a rail) that expansion makes it longer. If a rail gets longer but doesn't have gaps to fill, then you get a sun kink, and the rails are no longer parallel. That's a bad thing when you try and run a train over them.

Similarly with overhead wire. Take two poles 10 feet apart and put a 10' piece of rope between them. Now replace that piece of rope with an 11' piece, and you'll see that it hangs lower. Eventually the overhead wire and the pantograph collecting electricity affixed to the train get entangled, and wires get pulled down and trains stop.

You can do a little bit to help out with the first thing by heating the rail as you install it. Once everything is welded together it cools off and the rail is actually under tension under "normal" temperatures. But there is a limit to that because then they'll shrink when it gets cold out and the rail will break (which is a better failure method because the broken rail will be visible to the signal system and trains can stop). Similar with the second - you can vary the tension that the catenary is installed at, but physics is physics and there are limits (the superior workaround there is the constant tension catenary that isn't fixed at the ends, but runs over a pulley with a weight so that as the catenary lengthens, the weights get closer to the ground, but the tension in the line remains (as the name suggests)... constant.

That's an incredibly gross oversimplification, but hopefully close enough for us interested observers to understand.
 
Dallas Light Rail uses Constant Tension Catenary, so the effect of temperature change should be minimal. The speed restrictions are mostly for the possibility of getting rail buckling due to expansion that cannot be absorbed in compression alone.
 
Do the lower speed restrictions guarantee no buckling, or do the lower speeds allow for survivability if a derailment occurs from buckling?

How do rail companies accommodate this if the rail is in very hot desert regions?
 
Do the lower speed restrictions guarantee no buckling, or do the lower speeds allow for survivability if a derailment occurs from buckling?

How do rail companies accommodate this if the rail is in very hot desert regions?
The heat restrictions are basically so a train can stop before a dramatic heat kink if one is sighted and slower speeds will significantly reduce the possibility of derailment is the track is somewhat deformed but passable. Speed of trains likely have little or nothing to do with causing heat kinks. Heat kinks are caused by high outside temperatures.

The continuous welded rail is heated when laid to be stress free for the ambient temperature for the region it is laid in. The process was described by @George Harris in posts 3, 4, and 5. So CWR is heated to a higher temperature while being laid in southern Arizona than in northern Minnesota, since the ambient temperature is higher there.
 
Last edited:
I'm on the Acela right now just outside of Philadelphia. It's been hot as h*ll all the way from Boston, but we had no problem doing 150 mph is the sections in MA and RI. On the other hand, since we left New York, we've never gone faster than 100 mph. The temperature at Philadelphia International Airport was posted at 97 F. Perhaps the difference is due to the constant tension catenary in MA and RI vs. the old PRR style catenary south of New York.
 
I'm on the Acela right now just outside of Philadelphia. It's been hot as h*ll all the way from Boston, but we had no problem doing 150 mph is the sections in MA and RI. On the other hand, since we left New York, we've never gone faster than 100 mph. The temperature at Philadelphia International Airport was posted at 97 F. Perhaps the difference is due to the constant tension catenary in MA and RI vs. the old PRR style catenary south of New York.
Between Midway and Clark is constant tension in NJ, the portion where 150mph is allowed.
 
Back
Top