Does c80 Atlas flex go out of gauge?

[NtheBasement]

I have a helix, close to 20 years old, and a newer locomotive. The helix has a turnout in it and the loco on downhill runs keeps stalling at the turnout. Turns out the front truck derails on a diagonal, so that the leading wheels are outside the curve and the trailing wheels are inside. And it turns out that it does that long before the turnout, and then stops when the leading wheel creates a short at the turnout.

I looked for all kinds of things that might cause this including trip pin height and fouled track. When I got out my Micro trains track gauge I found out that most of the helix was on the narrow side, and the part where the wheels first jump is more on the narrow side than the rest.

I don't know if that is what is causing the derail. AFAIK the track hasn't changed since I laid it. Has anyone else run into this issue with old Atlas track going out-of-guage?

I'm hoping I don't have to tear it out and lay down new flex, any alternatives?

[Allen H.]

I've never heard of it going out of gauge, but the plastic could be weakening I suppose?
I know this next question has been discussed over the years but I've always been told to put the loose rail on the outside of the track as it has a little bit of side to side movement because it needs to be loose in order to "flex".

By putting it on the outside, the rail will lean away from the tight rail, causing the gauge to widen which is better than too narrow, which is what sounds like is happening in your case. The rails are to narrow causing the wheels to pop up and over the rails.

[rrjim1]

I just went down and checked my 35 year old Atlas track that I have in one yard, even the two curved tracks all measured perfect with my NMRA gauge.

[el Gato Gordo]

Granted that the track is a bit narrow. Do any of your other loks or cars have trouble on this stretch? And have you checked the wheels on the new lok against your gauge?

Cheers!
Gordon

[NtheBasement]

Fat cat, I checked the wheels. Good idea but negative, they fit right at the inside of the flange slots in my Microtrains coupler height gauge. Out of curiosity I checked some other locos including my ancient Atlas/Kato GP35s which ran the helix for 15 years when the system was DC. All the Kato wheels are identical. Strangely when I checked a recent Atlas DCC GP 30, all wheels are on the narrow side by a good millimeter. I wish it had the pulling power to take the thru freight thru the loop.
@allen H, the loose rail is on the inside of the curve. I'm thinking of cutting a gap (on a straight) and seeing if I can pull the ends together and re-solder. That should shorten the inner rail which will widen the gauge.

[Allen H.]

Well I don't know if that would take care of it.
They are not laying over because they are under strain from expansion.
It's because the webbing is wider than the rail head. So when you bend it, the rail head will have the tendency to bend before the wider web on the bottom and thus causes the rail to layover bend towards the outer rail making the gauge narrower. The spikes on the loose rail does not hold it tightly so that it bends like the tight rail.

Does this make sense?

[MetroRedLn]

Unless your layout is subject to extreme expansion/contraction temperature conditions, you should be fine. I use Atlas/ME Code 55 track and don't even have rail gauge problems.

[NtheBasement]

Thanks all for the suggestions.
@allen H, if you mean the rail will tip then why will it tip toward the inside?

The rest of the layout is ME code 55; no problems there. I had that c80 flex left over from the previous layout. I checked some other Atlas c80 (I still have pieces not installed on the layout) and they gauge OK, wider than the helix. To be honest I'm not even sure I am focusing on the right thing - just can't think of any other reason for the derail.

I ordered an Atlas rerailer track. I'll install it in the helix before the turnout and see if it solves the problem. The helix is an oval and the straights are > 5" long so it should fit OK. Not happy with this approach as it covers up the problem (if it works) instead of going to whatever the root cause is.

[Rodsup9000]

I think @allen H is right on what is happening. I was taught over 40 years ago when laying track with one loose rail, such as Atlas code 80, always lay the loose rail to the outside of any curves. The way the track is made, it has a tendency to for the gauge of the rails to go narrow side of the gauge when laid to the inside. When laying with the loose rail to the outside, it'll go just a hair wide on the gauge. I'm guessing why it hasn't given you any problems so far is that for the most part, locos are shipped with their wheels slightly narrower than the NMRA gauge. When I add new loco's to my layout, I take them apart and add conductalube to all axle points, clean and lube motors, trucks, and gears, add decoders and then gauge the wheels. 95% of the time, the wheels are too narrow to go into the NMRA gauge.

The NMRA gauge is the standard that I (most model railroaders use it all scales) use for reliability on my layout and if you want to run trains without derailments, I can only suggest you should too.

[rrjim1]

My layout is in my basement, temperatures range from 65 degrees in the summer to 72 degrees in the winter, (heated). You would not believe how much the track expands and contracts. Without expansion joints my track would have all kinds of problems. My guess is your track contracted and that's why it narrow. If you add a bunch of expansion joints I bet your track will be fine.

[NtheBasement]

Thanks guys for the tips. I always laid the Atlas with the loose rail on the inside. Otherwise you have to cut off some ties to even the ends; I went for closer tie spacing figuring it would be more robust. I googled around a bit for flex track tips and saw two recommendations to put the loose track on the inside of the curve.

I took an old piece of unused Atlas flex, bent it both ways and checked the gauge. No difference. Perhaps over time the outer spikes loosen up and allow the rail top to tip toward the other rail.

I'm not looking forward to re-laying the helix; hopefully the rerailer will do the job. If not, I'm not sure if I'll go with Atlas laid loose-rail-out, or ME.

[randgust]

Either the roadbed expanding and contracting to humidity, or the rail expanding and contracting due to temperature can trigger this.

Or, it could have been there since you first laid it and you finally have a piece of equipment that has discovered it.

Either way, I've found the 'loose rail' Atlas flex to be a PITA and I don't use it because I've seen some pretty irritating issues on curve gauge like this. I've gone over to Peco C55 on all wide curves and Atlas sectional (soldered up) on curves 11" and under. The Peco is really stiff and rugged, you'll still need expansion joints at places.

It really takes very little temperature variation to stress out flex on curves. When it starts popping track out of ballast you'll believe.

[NtheBasement]

@randgust, 1st thing I checked was expansion - popped the suspect end of the oval off the formica and let it "hang loose". The gauge didn't change and the loco derailed like always.

I've never had any track movement issues on this layout but its in a temp-stable room. Although the helix is track contact-cemented on formica, the rest of the layout is contact-cemented on cork which is Elmer'd to plywood. Back in the day I used a fiberboard instead of plywood at the recommendation of a newgroup faq; what a nightmare! When the humidity went up even plaster would split. Have to ask if you have track popping out of ballast, what are you laying the track on?

[randgust]

Our firm has a sceniced, multi-train 5x9 HO layout in a large commercial display (street level) window for kids to run from the street. Behind that much glass, with a southern exposure, it is subjected to intense temperature and humidity variation - from freezing up to 120 degrees. It kinks track to the point where I simply soldered up curves and let them float on the cork roadbed over plywood. Rail is sort of secured to cork, and it broke loose the ballast glue in a couple places. It would kink curves at rail joiners.

It's extreme, but I've seen it happen - everything from kinking to moving ballast to narrowing gauge at flex. Ugly stuff.

I had a friend in N run into the same issues, but all humidity based, not temperature - he laid track on Homasote and it pretty well expanded enough to pop rails at joiners in a couple spots in the summer humidity. As a result I pretty much banned 'loose rail' flex track in general, and put in expansion joints every 4' using modified sectional track if the expansion joint had to be on a curve.

Current layout is 8' x 5'6", and is 30+ years, temperature and humidity controlled, no track problems although the Masonite backdrops show the most expansion and contraction issues with humidity changes.

[NtheBasement]

Thanks everyone for the responses. I have an update...

This thread has turned into "the elements of track laying" so starting there, the basement layout room has minimal temperate changes but is more humid in the summer. The helix is flex contact-cemented to formica. The helix is race-track style oval, the geometry being 13 inch radius "u-turn"s and 6 inch straights. It is 100% soldered, no expansion joints. The first thing I did was pop the problematic u-turn off the formica to let it float. Nothing visible happened when I did this and the ties stayed aligned to the contact cement grooves. The gauge remained too narrow. My conclusion is there is zero stress on the rails, the ties, or the formica that contributed to the issue.

I should add that I used the evil homasote in a previous layout at the recommendation of folks on rec.models.railroad "because it takes nails easily". Never again. Huge mistake, and yet I see people still use it. I suspect a lot of the problems people report about "track expansion due to temperature" are actually "substrate expansion due to humidity". The rest of my layout is 5/8 plywood on 1x4 stringers and I know that in the summer my entrance aisle, which is spanned by a duck-under bridge, is 1/8 inch narrower than in the winter. This is wood expansion guys, not track expansion.

We now return to our regular thread...

The turnout is on a u-turn. I replaced the track "upstream" with new (actually really old) Atlas flex up to the first straight, which I replaced with an Atlas re-railer. This was pretty painful due to the next helix level being 2 inches above the track. I then ran the train and the effort seemed successful. But... after an hour or so (maybe 6 circuits of the track and the helix) the problem recurred. So now I am looking at replacing all of the helix track. Ouch!

One final note, the flex track that I replaced, now loose, remains out-of-gauge until I straighten it at which point it is in gauge. I tried flexing other pieces (I have spares from the 90's) and they do not do this. At this point I am thinking of using Peco for the replacements.

[Rodsup9000]

I should add that I used the evil homasote in a previous layout at the recommendation of folks on rec.models.railroad "because it takes nails easily". Never again. Huge mistake, and yet I see people still use it. I suspect a lot of the problems people report about "track expansion due to temperature" are actually "substrate expansion due to humidity".

Did you seal it at all before you laid your track???? I used it in my staging yard and helix, but I put 2 coats of paint on all sides before I installed it permanently and I've never have had a problem. On the rest of my layout, I used Micore.

[NtheBasement]

Did you seal it at all before you laid your track????

No but it should help if you seal it after cutting and paint all the sides and edges. I don't hand-lay track and gave up nailing the Atlas because I couldn't find nails that were tiny enough to fit the holes in the ties even back when I had the homasote layout. To me gluing is way easier.