What are turnout tangent track recommendations for the head end?

[mobilman44]

Hi,

I've got a simple question and hope I communicate it correctly....... 

I'm building an HO layout, and as I've been working with HO (with success) since the '60s, I am not a newbie in this area and have always considered my trackwork "bulletproof".

However, one of a number of Atlas code 100 # 6 turnouts on a lower level staging track is just not reliable 100%.  When I let a car "freefall" down the 2 % grade to the various staging tracks, one turnout occasionally causes a derailment.  At slow speeds there is no problem however.  I have replaced the turnout, filed points, checked gauge and clearances, and still have the occasional derailment.

The turnout in question and the track leading to it and the tracks leading from it are all on level, flat well supported "ground".  As the derailment happens at the points, the problem is at the head end of the turnout.  The track leading into the turnout is the easement end of a 26 inch radius curve.  So this leads me to suspect that the problem may be that there is no length of tangent (straight) track of significance at the head end.  Note, all trackage is Atlas flextrack, mounted on cork roadbed over plywood.

My question is:  Is there a recommendation to have any tangent track at the head end of a turnout? 

I confess I'm kind of anal about trackwork, especially that which will be hard to get to once the main level benchwork is installed.

Thanks,

Mobilman44

[grizlump9]

 hi. i had a similar problem on my old layout years ago and it turned out to be track geometry.   if you are sure that your points and trackwork are right, then go back and double check the first thing you did. like filing and smoothing the points.  often we assume that we have done it right the first time only to find that it isn't quite good enough.  

you could do what i did and install a short guard rail just ahead of the points on the inside rail of the curve and that will keep the wheels from picking the point by holding them away from the outer rail.  i have seen several examples of this on prototype track especially on tight curves in industrial areas.

good luck.

graizlump

[cuyama]

If the easement leading into the point end of the turnout does not come fully to tanget or does not meet the turnout squarely, you can see problems like this. Especially since you said you have already tried replacing the tunout, that might be a place to look.

It shouldn't be necessary to have a length of tangent track ahead of the points, but it is easier to determine that things meet up squarely when it is tangent.

Byron
Model RR Blog 
Layout Design Gallery

[wjstix]

I had a similar problem on my previous layout, a 32"R curve leading into a no. 6 turnout on the mainline often caused derailements. I reduced the curve down to 30"R and that gave me room to add about 1 to 1.5" of straight track ahead of the points. That solved the problem.

If you can't adjust the track easily due to it's location, I would try adding a guardrail. Years ago on an old layout I did that and it worked. If the place you're having trouble isn't on the sceniced part of the layout it doesn't have to be fancy, even just a piece of stripwood of the right height glued in place would probably work.

[selector]

I wonder how you would find the exact cause, or the precise location, where this happens if it only happens at a relatively high speed.  I would be inclined to take a video of it and play the video in slow mo so that you can see where the event occurs.

I would suspect points that stick out, or are too blunt, or the gauge is a bit tight, maybe at the very point tip or along the closure rails, or the frog is in need of work.  It could be the cars, themselves, for all we know.  But the guards near the frog may require an adjustment of the gaps for the flanges.

-Crandell

[tomikawaTT]

To paraphrase an advertising slogan, "Is it the point, or is it the rail joint just before the point?  Since there's only about an inch between the two, only your wheel flanges know for sure."

I hand-lay my turnouts, and always avoid that joint just short of the points that's a (non)feature of commercial turnouts.  Even so, I have installed pre-point guard rails on a few to-be-hidden turnouts on sharp curves, just to avoid the situation you describe.

One thing that will minimize joint problems is to knock off the sharp corner at the top inside of the railhead with a stroke or two of a file.  The little facet should be barely visible - about 1/16 long and only a few thousandths of an inch measured across the railhead, with a very gentle angle.  If you can slide a fingertip along the inside of the rail and NOT feel it pick, you've got it right.  (I treat every rail-end at every joint to this little touch, and derailments are an almost unknown phenomenon on my layout.)

Chuck (Modeling Central Japan in September, 1964

[mobilman44]

Thank you all!

  I've decided to move the turnout a few inches down into the siding, thus allowing a few inches of tangent track before the turnout which may well solve the problem.  If not, I'll install guardrails (sounds like a terrific idea) and that just has to be the "final answer"......................

You guys are good!

Mobilman44   

[mobilman44]

Good Morning!

  After absorbing your responses to my problem, I again tackled the derailment problem last evening.  I applied a straight edge to the point end of the turnout and lead in track and found it to be out of tangent by close to 3/32.  I had previously "eyeballed" the area for smooth flow of the track, but apparently that was not a good enough check.

I loosened up the track joint, pulled the track nails in the area, and got the track in alignment and got everything re-secured.  So far, testing has been 100% successful - and I am one happy guy.

This problem actually turned out to be a good thing in one way.  It re-enforced my thoughts that testing trackage (& wiring) cannot be overstated - even for those of us that have a lot of experience.

Thanks again,

Mobilman44 

[tangerine-jack]

I have pondered your problem at great length and I am glad you found a solution, but I will share with you what I came up with anyway. You seem to care about getting track right, and I have a lot of information on that topic at my fingertips, but it is far too complex and detailed to even begin to scratch the surface of it here. Your issue evidently was with the transition curve from the tangent track off the switch. According to my information a transition curve must be placed between a straight section and a circular curve and between adjacent circular curves with different radii (compound curve) on main routes and direct lines and other main lines in general. If at least one of the adjacent track elements has a superelevation, a ramp must be inserted additionally to handle the different superelevations.

Here are the formulas for future planning.

The biggest area ignored by modelers is the transition curve (as well as the issues associated with superelevations) and the effects it has on how the trains run and quality of trackage. Again, I have a vast amount of information on this topic, but it's far to involved to post here.

 

[4merroad4man]

General rule of thumb to keep stuff like this from occuring is the same as an "S" curve issue:

one length of tangent track equal to the length of the longest car or locomotive wheelbase you will run, at the point end of the turnout.