Problems with a Walthers #8 curved turnout

At what spot does it derail? I have this turnout  with no problems.

Dante

Rick... Video?

Can you post a video of what's happening?

Check gauge

Common Problem

As soon as I saw the post title I could guess what you're experiencing.  Most Walthers code 83 #8 curved turnouts are wide in gauge through the points and all the way to the point end of the turnout.  Derailment problems caused by this manufacturing defect primarily manifest themselves when traveling through the diverging route, and only seem to affect some equipment (anything with semi-scale code 88 wheels especially, but other things too).

First, did you check the turnout with an NMRA gauge before installation?  If not, do so now, and pay special attention to the area from the end of the turnout by the points, and through the points with them thrown both directions.  I'll bet you find rails out of gauge.

If that's the case, the fix is actually quite simple.  Slice the plastic spikes from the inside of the diverging rail through the affected area.  Drill holes for new metal spikes (not track nails) to keep the rail in place and spike the rail (you only have to do this on the outside through the typically affected portion of the point/stock rail interface since there is no clearance for spikes on the inside).  Use flexible CA like PolyZap or MicroBond to hold the spikes so they never work loose.

Gauge problems are rampant on Shinohara turnouts, whether sold under their own brand or Walthers.

Check basics

Dear Rick,

Basic check:

- at what specific point does the loco derail?

- does it climb the outside rail?
(which, for the LH route of a LH turnout is the switchrail/diverging rail/frog)

- does it mis-take the frog, and try to "go the wrong route"?
(If YES, suggests that the combination of loco/axle gauging/axle-spacing + guardrail is not doing it's job.
Possibly the Guardrail flangeway gap is too big?)

- what is the gauging like at the point of derailment and +/- 1" of that point?
(break out the NMRA gauge)

- what is the gauging of the loco axlesets?
(side questions:
- does the loco derail when running thru the turnout Toe> Heel, Heel> Toe, or both?
- does the loco derail when running cab-first, hood first, or both?
- is it always the front truck, rear truck, or both trucks which derail?
- assuming a given truck, is it the inboard or outboard axle which is derailing?)

- based on the fact that the loco truck axle spacing is longer than most freight trucks, the axle yaw (angle of offset relative to "90-degree perpendicular to rail") will be greater when negotiating any form of curved trackage.

This will make the leading axle/outside wheel/flange grind against the inside face of the outer rail, and make it more prone to "pick" any imperfection/hump/gap/divot/notch/feeder-solder-lump/etc.

Run your finger along the top and inside face of the outer rail of the route which causes issues,
(which, if I understand correctly includes the frog),

any thing which catches your fingernail is a possibly culprit.

Rest assured it's find-able and fixable, just gotta diagnose the problem patiently and logically...

Happy Modelling,
Aim to Improve,
Prof Klyzlr

PS - Has this loco ever negotiated this turnout OK without derailing?
- If YES, what changed recently at the turnout or with the loco which might cause the fault condition?

PPS if you have a passenger car or similar with long axle-spaced trucks, does it also derail in the same/similar manner at the same location?

Sounds like Rob might have it pegged......

But I did have a somewhat similar problem with Peco code 75 curved turnout. In my case it turned out that the turnout was not level in the side to side plane. It was ever so slight....you could hardly see it with the naked eye but once I shimmed up the outside, the problem disappeared.  Rob seems to have experience with the Walthers curved turnouts so I'll defer to him but I just thought I'd throw my experience out there just in case.

Michael

Thanks, fellows! - it seems the problem is the loco

Bad Locomotive...

Go sit in the corner.

Far out, man?

How much out of gauge is "not much?"  Mine was about 1/16" wide for a couple inches.  That's more than enough to cause derailments at the points.  Bottom line - if the turnout's wide in gauge you will have problems.  Let it go and it will be harder to fix later once you have more finish work done on the layout.  Mine was rendered derailment free with less than 10 minutes of work.

Common Problem

"Gauge problems are rampant on Shinohara turnouts, whether sold under their own brand or Walthers."

Rob, I have 41 Walthers/Shinohara turnouts-many of them curved. I checked each with the NMRA gauge and found none that were out of gauge. Not exactly a rampant or common problem.

Dante

Problems with a Walthers #8 curved turnout

Rick,

A stab in the dark: is a coupler trip pin a little too low?

Dante

Loco identified, seek and distroy...

Dear Rick,

If a 6-axle monster can succesfully negotiate the turnout in question, I'd agree it starts pointing to the loco  as the main source of issue. (turnout gauging may be slightly suspect, but if only 1 item of rollingstock falls off when all else work fine, suspect the rollingstock...)

SO, things to check:

- all axles in gauge?

- all axles/wheels in a truck in alignment with each other?
A pair of wheels on a given axleset can be the right distance apart, but they may be biased to the Left or Right hand side of the truck (IE not symetrical on truck centreline)

Assume that one axle is "biased to the Left", and the other "biased to the right". The result is that the truck will "crab" down straight trackage, with the truck centreline on-an-angle relative to the track centreline.

Result? the truck will "prefer" to turn in one direction,
(or more appropriately, have the rails curve underneath it),

but will be crazy-picky about any rail imperfection when turning the other way.

- all axles flat relative to each other?
Place each truck on a pane of glass, and ensure that all 4 wheels (for a 2 axle truck) are touching the glass. If one of the wheels is "lifted", then one of the axles is not sitting "flat", and this is a prime candiate for oddball derailments.

- if 1/2 axle systems, both halves solidly anchored into the centre "gear" section?
if not, one or both the 1/2 axle/wheel parts may be eccentric, meaning that as the axleset rotates, the bottom of the wheels (where they touch the rail) will appear to alternately go in-out-in-out of gauge.
(only suspect this if the loco randomly derails at a given point in the track. If it derails every time then look elsewhere first...)

Happy Loco Bug Hunting!
Aim to Improve,
Prof Klyzlr

Shinohara

Rob mentioned in a long ago post how Shinohara turnouts are commonly out of gauge. Freaked me out since I have a whole box full of them in wait. I unboxed a few at random and checked them with my gauge. I too found none of the units I measured to be out of gauge. At least, less than 1/16" if there was difference. That reassured me a bit but you can bet your bottom dollar I will measure each and every one of them before laying. Rob is so rarely wrong.

For the moment my loco is red tagged

Trip to the Maintenance shop

Prototypical reason to have it for an overhaul in the maintenance shop.  Even model the hood off or something.  I always thought that would be a neat foreground scene with a dummy loco you could slip an engine assembly into.  It wouldn't need to be super-detailed in the shop, just enough to make it appear to be "real" through the window or door of the shop.

Rob, what was involved in the

Rob, what was involved in the 10 minute repair of the Walthers turnout?

Stuff

I can relate experiences from my own current layout, as well as others I work on.  I picked up 11 Shinohara #6 code 70 turnouts.  EVERY SINGLE ONE was wide in check gauge through the frog (check gauge, measured with the square prongs atop the NMRA standards gauge, is important wherever guard rails are involved because a wide check gauge can cause flanges to bind or lift).  I acquired six Walthers/Shinohara code 83 #8 straight turnouts.  EVERY SINGLE ONE was wide in gauge at the end of all three routes.  I have one Walthers/Shinohara code 83 #8 curved turnout, which was badly out of gauge through the points.

I have several Walthers/Shinohara code 83 #6 turnouts, one code 70 curvable turnout, and three Walthers/Shinohara code 83 #7 turnouts.  All of these have been in gauge.  Code 83 #8s manufactured before the current DCC friendly design also tend to be in gauge.  I have many other turnouts from other brands, all of which have their own quirks.

In addition to working on turnouts on my own layout, I help other local modelers with their track.  Correcting gauge problems on Shinohara turnouts is a regular part of trackwork tuneup.  One guy special ordered something like 40 code 70 #6, and again EVERY ONE was wide in check gauge at the frog.  Fixing the code 83 #8 curved DCC friendly versions has proven to be important because of how far out of gauge they can be, and so far all of the ones my work group has used have required correction.  I've yet to encounter one where the issue could not be corrected so the turnout performed satisfactorily, but to suggest this problem isn't common ignores a fair amount of hard evidence.  Maybe all the defective examples end up at my local shops.

From my earlier reply to Rick on the first page:

Slice the plastic spikes from the inside of the diverging rail through the affected area.  Drill holes for new metal spikes (not track nails) to keep the rail in place and spike the rail (you only have to do this on the outside through the typically affected portion of the point/stock rail interface since there is no clearance for spikes on the inside).  Use flexible CA like PolyZap or MicroBond to hold the spikes so they never work loose.

This really did work and was quick to do.

With apologies to Rick for

With apologies to Rick for the hi-jack, I am responding to Dave's comment on a loco in for service.  This is an HO Blue Box Athearn SD9 refitted with a Walthers prime mover and some extra parts for the radiator rack, piping and aux generators and connections to the electrical cabinet (which you can't see from here).  It was inspired by a photo of a real one inside the Missabe's loco shop with the engine compartment cowl removed, number boards out (not sure why), etc.

Because the Proctor shops are just cardboard cutouts right now, I posed this in front of the Hull-Rust crusher, a long way from where it should be in this condition.

Stuff

Rob,

I certainly cannot refute your experience, but neither can I deny mine. My turnouts are all Code 83. 29 are the newer DCC friendly and 11 not (correction: I have 39, not 41). 13 are curved of all sizes, including 2-#8s. Gauges were checked before installation, and they generate no derailments in operation (except when I foolishly run against the point position).

Your correction method for those that are out-of-gauge makes good sense, and I will keep it in mind if I encounter the problem.

I should also mention that my motive power ranges from a classic original 0-4-0 Varney Docksider to 4 & 6 axle diesels and includes a Mike, a brass Pacific and a brass gas-electric that requires 22" radius curves (it does derail on 18" curves that are part of the 24/18 curved turnouts-no, they are not the advertised 24/20). 

Dante

Exactly what I was thinking, Milt

Very nice.  One day I will likely do the same.