The Valley River Model Railroad Club is building an N-scale layout inside an old Southern caboose. For the moment, we do NOT have climate control…and the heat inside that metal car can be intense (102 degrees yesterday afternoon). We are using Atlas Nickle Silver flextrack over a 24 foot straightaway. We assume the heat will cause some expansion. How does one compensate for it without having bad bumps on the mainline???
Have gaps in the rail every so often, don’t solder all the rail joints. Cut or adjust the gaps when the rail is cool.
Actually you will have more problems with changes in humidity and the wooden benchwork than you will with the expansion of the rail.
Dave H.
People have done tests on track expansion, and the results indicate that the track itself is not likely to be a problem in most cases. The track is much more temperature and humidity stable than the real problem - plywood or particle-board sub-roadbed. Also, the issue of temperature is far less serious than humidity.
Still, I would recommend a small gap between sections of track, and don’t solder all the rail joiners. This will allow a small amount of expansion without compromising performance. Of course, with non-soldered rails, you will need to be pretty meticulous about providing feeders to your track.
My advice for layouts with poor temperature control is to use foam, not wood, for your layout base. It’s immune to humidity.
Question about that-- if you’re using, say, foam over wood L-joists supported by wood legs & bracing, will the wood be a factor then? (I doubt it would be a factor in my case anyway since I’m building indoors in a dry, climate-controlled environment). But I’m just curious for my own edification 
I grew up in a fairly warm part of Australia and had a heavy degree of expansion of track laid in winter (August here) buckling in January (summer here) 100 degree heat outside - much warmer in shed I would think - when I was a teen. The shed my layout was in took the extremes (not on the lower end as you guys experience) and I had a buckle of about 1/2 inch. The track was not glued and ballasted at the time, let alone soldered.
I wound up “relieving” the kink by cutting the track to fit and relaying it and it did not “revert” as such and equaling the expansion gaps. The best I could offer advice is to lay track mid autumn and spring to “equalize” movement. Even so it does go with the territory.
Hope this helps!
Trevor www.xdford.digitalzones.com FYI
It doesn’t take very much track expansion to cause a problem, which will appear much bigger than it is. you can figure out the geometry of it if you are so inclined. But, the good thing is that it also doesn’t take a very big gap to take care of. And unsoldered joint, with a gap about the width of the thickness of a business card, every ten feet of so, and you are good, trackwise. As has been stated, the benchwork and roadbed can cause trouble too, so allowing for a little flexibility here and there so that things can move is a good idea,
Use gaps, metal benchwork and foam instead of wood.
I maybe wrong here…but if you are installing the track when the temperature is at or near its highest mark & the track has been allowed to adjust to such temperature before hand…then it should be contraction that you need to be concerned with
dbduck, you are correct. Either way, if you don’t allow for some mechanism to compensate, sort of like a buffer does, you end up with either large gaps or buckled track.
Code 100 NS rail will expand a mere 1/4" over 100’ in contiguous length over a 30 degree rise in temps…not much at all. So temps are about 10-15% of the problem for the vast, vast, majority of us. It is, instead, the expansion and contraction across and along grains in organic/wood products in the supporting structures on a layout, from the frame up to the roadbed. It is humidity that does this. So, if you intend to build in a non-climate controlled building, leave gaps in several places along a length, solder the rails on curves, and ballast it all. Remember that the gaps can be actual gaps with no joiners, or can be joined rail ends where the joiner is needed/desired for alignment…but really the ballast can do it all.
Any rail length not joined will need a feeder, and even the odd joiner, or most, will eventually fail to power the rail.
I want to thank you all for your quick response…we’ll take your expertise anytime. We are building on a foam base (1" blueboard) over an L-girder construction benchwork. The blueboard DOES have a particle-board underlayment, however, so we’ll be cognizant of the possible problems with that.
A million years ago, when I took science in high school, I learned that laying the track in the heat was the optimum approach. Then, the gaps will widen in the cold as the metal contracts. It’s never going to get THAT cold in our caboose…we hope…but the advice is welcomed nonetheless.
The N-scale community will be a great asset to our club as we go along.
Again…thanks to all!
About the only way I know is to allow gaps so that when the track expands, it has some room to “breath” and not compress together causing kinks.
ALL the above, plus
Add some slight ‘S’ bends in long (3 or more flextrack) straights. The curves will also add interest running. Engines, just running, will heat metal rail .
You will be wanting to add some heat & air conditioning just to operate.
Make sure you have enough Electrical ‘Service’ to inlude heat & air conditioning. (Installing is a one-time expense). Upgrading is not cheap.
Expect local electrical codes and inspections to apply. It’s what Fire Marshall’s do.