Derailing truck through #4 turnout - UPDATE!

[tstage]

Thanks for giving that update, Dave.

Tom

[tstage]

you tried with another new no. 4.  Was it installed in the same place?

Autobus,

No, the other #4 turnout is just down line from the other one.  Here's a diagram that I've created in the RTS 7.0 software to better visualize the yard trackage:

Click diagram to enlarge

Autobus, the two #4 turnouts that you are inquiring about are highlighted in red.  Turnout #1 is the one that first caused the derailing problems.  Turnout #2 is the other turnout I tried and was also able to replicate the problem on. 

You don't by any chance have a curve from the right side (looking from point end to frog end) leading into that switch, do you?

The Turnout #1 does have an 11" piece of flex-track curving into it from the right.  The unhighlighted turnout to the right of it is at an elevation of +1/4" then drops to 0" by the time it gets to the Turnout #1.  Turnout #2 does NOT have a curved track leading into it, yet I still had the derailing problem with it.

I hope that gives everyone a clearer picture of what is going on. 

Tom

[Autobus Prime]

T:

Yet another entry in this neverending discussion: you said, Tom, that you tried with another new no. 4.  Was it installed in the same place? You don't by any chance have a curve from the right side (looking from point end to frog end) leading into that switch, do you?

[tstage]

Don, 

No.  I tested the original tender and another identical tender, and neither one derails unattached to the locomotive.

As I mentioned in an earlier post, Don, the wiring harness connector on the tender seems slightly "short" because it appears to angle down slightly to make contact with the locomotive plug.  I'm guessing that that may be one part of the problem and why the added weight on the tender helps to rectify the derailing problem.  The tender does not bind rotationally when attached to the back of the locomotive.

Tom

[Don Gibson]

Tom:

Do the tender wheels still derail when pushed through the turnout when not draw barrr'd to the engine?

[tstage]

Tstage, you definatly are close to getting the Scientific Method down. Rename a couple things, (Hypothesis, Predictions, Variables, etc.) and you'd be a scientist! :)

But seriously, I think you might have it right.

Thanks, Mark.   Actually, I'm a research technician by trade and do this sort of observation and analysis on a frequent basis.  Yea, I guess I could use the fancier terms.  I'm a simple man so I just chose ones that were simpler to understand.

Tom

P.S. Mark, I actaully decided to go back and change a couple of the headings to other terms.  I thought "Scenario" and "Conclusion" made more sense than "Circumstances" and "Diagnosis".

[tstage]

Mike,

A very good and reasonable question.  I did check both trucks early on and they do rotate freely with no binding.

Tom

[mikebo]

One more thing to check. Does the axle rotate freely in the truck? If there is a slight bind in the truck such that the axle does rotate quite as freely as it should you will get derailments.  I just had a boxcar with that problem.

 Mike

[Hoople]

Hi everyone! 

I've got a slight problem and I'm wondering if you all can collaborate what I think may be happening.  I've been installing some new Atlas #4 turnouts in my yard this weekend.  This evening I was testing out the trackwork to make sure everything was working properly and discovered a consistent problem with one of my Proto 2000 0-8-0 switchers.

Circumstances 

Whenever I back through the frog, the axle of the tender truck nearest the cab rotates sharply CW and derails.  (See sequence below)

Click picture(s) to enlarge

Observations

• I've tried this with another new #4 turnout and it does the exact same thing
• I ran the locomotive through the turnout forwards - no derailment
• I pushed an identical Proto 2000 0-8-0 tender backwards through the same turnout - no derailment
• I backed a 2-8-2 BLI Mike through the turnout - no derailment (added)
  
• I pushed a 40' boxcar through the turnout - no derailment
• Trucks rotate freely and don't bind
  
• Only that particular axle on that tender derails

Diagnosis

My guess is that the flange on the axle closest to the cab might be slightly undersized; therefore it is not making contact soon enough with the frog and slips past it.  I haven't had a chance to measure the flanges on the axle with calipers to confirm this or not.

Am I way off base?  Or do you think it's an issue with the turnout(s) themselves?  I didn't have this problem with the Atlas Snap turnouts that have a sharper diverging angle.

Thanks for the help ahead of time.

Tom

Tstage, you definatly are close to getting the Scientific Method down. Rename a couple things, (Hypothesis, Predictions, Variables, etc.) and you'd be a scientist! :)

But seriously, I think you might have it right.

[tstage]

Terry,

Ahhhhhh...I understand your point now.  I'll have to try it and see how much more the opposing wheel raises up with a load behind it.

Tom 

[C&O Fan]

The next test will be to push a string of cars backwards thru the turn out

It helps determine how much weight to add to the tender

Terry,

FYI: Cars go through just fine.  I was pushing a string of loaded coal hoppers through that turnout, that's when I discovered the problem with my 0-8-0 tender derailing.

Tom

I guess i didn't say it right It's not the cars i'm worried about

The tendency for the tender to raise up increases exponentially by the number of cars it pushes

[tstage]

The next test will be to push a string of cars backwards thru the turn out

It helps determine how much weight to add to the tender

Terry,

FYI: Cars go through just fine.  I was pushing a string of loaded coal hoppers through that turnout, that's when I discovered the problem with my 0-8-0 tender derailing.

Tom

[Vail and Southwestern RR]

My personal opinion is that adding weight is a bandaid, at best.  All you are doing there is hiding a problem.  And sooner or later, some other car is going to have the same problem, if it is the switch.

 

[msowsun]

if I did decide to switch-out the wheel sets for different ones, does anyone know how you get the black plastic axle covers (that push the brass wheel pickups against the inside of the wheel) off?  Many thanks again for the help.

Tom

 

Those Proto 2000 tender wheels are very specialized. I doubt anything else will fit and still give you reliable pickup.

If you look closely, it looks like the brass pickups actually ride on the axles. The axles are split in the middle so that there is separate left and right pickup.

[Autobus Prime]

I used a pair of calipers this evening to check the flangeway gap on the diverging track of both the Snap and #4 turnouts.  Here's what I found out:

• Snap: 0.055"
• #4: 0.047"

I also checked the flange gap on the #4 turnout with the NMRA gauge to see if it was in gauge.  The tab wouldn't fit down into the flangeway gap so the flangeway is "in gauge".  With that being the case, it seems to me that weighting the tender is the quickest and cheapest solution to the derailing problem.

T:

Good work.  You should also use the NMRA gage to check the "check gauge" (that seems a bit awkward, there) between the inside of the wing rail and the outside of the guard rail.  If this is too small, your wheels can hit the frog point and go the wrong way.  Use the little tabs at the top that are marked "FLANGEWAYS".  I'm not sure if the flangeways measured above were the guardrail or frog flangeways.  If the guardrail flangeway and frog flangeway are both okay, but the check gauge is off, measure the track gauge across the running rails. 

To widen too-narrow check gauge and bring the wheel away from the frog point, you can add a thin shim to the side of the guardrail that faces the running rail, or carve a plastic guardrail entirely away and replace it with one bent from scrap rail, as I've done to fix badly worn ones.

http://www.nmra.org/standards/pdf/s-1.pdf

[Anonymous]

I would just check the gauge w/o removing them. Months ago my club had a major clean all the wheels on every thing before going back on the layout project. Some members had damaged the PUs on their Proto tenders and they are a bear to bend back and have work properly.

Just checking gauge may not catch the problem.  The wheel may be off center, or skewed on the axle.  Flangeand/or wheel tread may not be concentric.

  What's probably going on here is the condition know as "tolerance stack".  This describes what happens when a number of tolerances are at or near the limit.  Any one by itself won't cause problems, but combined they cause a malfunction.  This kind of defect can be particularly frustrating to track down because most pieces check out ok by themselves, but combined they throw things off enough to cause trouble.  The fix is to eliminate enough of the problem to get things to work reliably.

  In the case at hand we have a known offender, the Atlas switch, which however usually works ok with other combinations, a couple of probables: wheel set, tender to engine wire, tender under reccomended NMRA weight, and a couple of possibles;  tender trucks too tight etc.

  If adding weight brings things back in a combination that allows reliable operation, congragulations, you have successfully solved the problems.  If you don't feel comfortable changing the wheel sets and everything works ok, let it go at that. Alternatively, write the manufacturer/importer and ask them to replace the defective part.

If problems persist you can consider what other remedies you want to take, but I would expect that anything that is off is not off by much.  It's simply the combination of flaws that are causing the problems.

JBB

[C&O Fan]

The next test will be to push a string of cars backwards thru the turn out

It helps determine how much weight to add to the tender

[tstage]

I used a pair of calipers this evening to check the flangeway gap on the diverging track of both the Snap and #4 turnouts.  Here's what I found out:

• Snap: 0.055"
• #4: 0.047"

I also checked the flange gap on the #4 turnout with the NMRA gauge to see if it was in gauge.  The tab wouldn't fit down into the flangeway gap so the flangeway is "in gauge".  With that being the case, it seems to me that weighting the tender is the quickest and cheapest solution to the derailing problem.

Even so, if I did decide to switch-out the wheel sets for different ones, does anyone know how you get the black plastic axle covers (that push the brass wheel pickups against the inside of the wheel) off?  Many thanks again for the help.

Tom

[bogp40]

Once again, many thanks all who have contributed to this thread so far.  As I was contemplating again this morning the problem with my 0-8-0 switcher tender, I remembered something that I did but had forgotten to include in my update above.

After I discovered the problem of the opposing wheel on my tender lifting up on the guard(?) rail of the diverging tract, I manually pushed a boxcar through the offending turnout to observe what it would do.  As the wheels on the right side of the boxcar went through the guard rail, NONE of them raised up or derailed - either going forward or backward through the turnout.  Since the boxcar (Accurail 40') is weighted, that seems to confirm Bob and Dave's idea of added weight to eliminate the problem.  The small chunk of 4" long 1 x 2 pine on top of the tender (that I mentioned earlier) did allow the tender to back through the turnout without issue.

I wouldn't be surprised if the stiffness of the wire harness connector between the tender and locomotive is the main culprit and the lightness of the tender is the response to that.  I think I'll go the added weight route, as that appears to be the easiest fix.  Before I do that, I will check the wheel sets and turnout (Thanks, Terry & Crandell! ) for gauge.

SIDE NOTE: As far as switching out wheel sets on the Proto 2000 0-8-0 tender: How does one remove the black plastic sheaths (that press the pick up bars against the inside of the wheels) without destroying the wheels?  Is this something that it would be wiser to order already assembled from Walthers?

Tom

Tom,

I would just check the gauge w/o removing them. Months ago my club had a major clean all the wheels on every thing before going back on the layout project. Some members had damaged the PUs on their Proto tenders and they are a bear to bend back and have work properly.

[BRAKIE]

Please do not take this response as facetious but I thought that Jeff Wimberly's analysis of the troubles was reserved for N Scale because I have experienced just this trouble on numerous occasions. I don't ever remember encountering these off center wheels in my HO Scale days.

Jeffery is correct..Some wheels are dog legged from the factory for some reason..I had this problem with 3 wheels/axles of P2K 33" wheels and 1 set of Atlas wheels/axle..Not sure what causes the problem.

[tstage]

Terry & Crandell,

I have seen that very shimming technique pictured in one of the MR help books as a fix to a troublesome turnout.  I'm pretty sure I have that somewhere in my library.  Thanks for bringing that one up.

Tom

[tstage]

Once again, many thanks all who have contributed to this thread so far.  As I was contemplating again this morning the problem with my 0-8-0 switcher tender, I remembered something that I did but had forgotten to include in my update above.

After I discovered the problem of the opposing wheel on my tender lifting up on the guard(?) rail of the diverging tract, I manually pushed a boxcar through the offending turnout to observe what it would do.  As the wheels on the right side of the boxcar went through the guard rail, NONE of them raised up or derailed - either going forward or backward through the turnout.  Since the boxcar (Accurail 40') is weighted, that seems to confirm Bob and Dave's idea of added weight to eliminate the problem.  The small chunk of 4" long 1 x 2 pine on top of the tender (that I mentioned earlier) did allow the tender to back through the turnout without issue.

I wouldn't be surprised if the stiffness of the wire harness connector between the tender and locomotive is the main culprit and the lightness of the tender is the response to that.  I think I'll go the added weight route, as that appears to be the easiest fix.  Before I do that, I will check the wheel sets and turnout (Thanks, Terry & Crandell! ) for gauge.

SIDE NOTE: As far as switching out wheel sets on the Proto 2000 0-8-0 tender: How does one remove the black plastic sheaths (that press the pick up bars against the inside of the wheels) without destroying the wheels?  Is this something that it would be wiser to order already assembled from Walthers?

Tom

[selector]

It appears to me that the space between the guard rail on the #4 is greater than the space between the guard rail and the stock rail on the snap turnout.  If that is truly the case, then perhaps you are experiencing the very problem that Joe Fugate pointed out in his recent thread on constructing handlaid turnouts.  The commercial turnouts are notorious for being out of gauge for the distance between the inner wheel face and the flange of the opposite wheel.  The #4 guardrail is lifting the flange, or at least deflecting it upward.  It should not be the spikes at all if the wheelsets conform to RP-25.

That's my best guess.

[C&O Fan]

Tom

It appears from the 2nd photo you have some dirt in the gap between the point and the guard rail

Sure enough.  Good eye, Terry!   However, I don't have the problem with the Snap turnout.  It's the #4 turnouts where the derailing occurs.

Tom

I just knew it couldn't be that easy a fix !

A friend of mine in Austin shims all his Atlas Turnout's flange ways with 0.10 styrene

he says this helps prevent derailments because the flangeways are too wide

you may want to take a guage and check them

Adding Weight to my C&O Mike's tender solved my problem

 

[tstage]

Tom

It appears from the 2nd photo you have some dirt in the gap between the point and the guard rail

Sure enough.  Good eye, Terry!   However, I don't have the problem with the Snap turnout.  It's the #4 turnouts where the derailing occurs.

Tom

[Anonymous]

Not to beat a dead horse here, but when I was learning how to be a gunsmith, I was taught to work on the cheapest, most easily fixed piece that might be causing the problem.  If that didn't work, move on to next cheapest easily fixed suspect part.

  In the case at hand, the track doesn't seem to the problem, since other equipment, including an engine of the same type run thru it fine (not to say there can't be problems with Atlas switches, and they don't have tolerance problems, we've all been there)

If it was me, I'd replace both wheelsets in the suspect truck (easy and cheap,less than a buck a wheel set)  Adding weight to the tender also seems like a good move, because in additon to helping tracking, it will also help with electrical pickup.  Loosening the wire between the tender and engine also sounds like a good move, once again, easy and cheap.

I'd be inclined to mess with the track only if the problem persists.  Otherwise you risk introducing another variable which may have nothing to do with the problem.

JBB

 

[Phoebe Vet]

LOL. I picked up some weight at my local Tire Shop yesterday, the sticky back type that they use on custom wheels.  

I use those peel and stick tire weights, too.

My local tire store gives them too me for free.  I must buy too many tires .... lol.

[Autobus Prime]

The only direct cure is to change turnouts. Find one that doesn't have such a ridiculously deep frog gap.

j_mz:

But he didn't have the problem with Snap-switches, which have an even larger frog gap. Even if the wheels do drop, the guardrail should be keeping it from drifting to the other side.  If one wheel drops on a rigid four-wheel truck, it's the kitty-corner wheel that lifts off, not the opposite one.

Hmm...maybe when the first truck wheel drops, lifting the kitty-corner wheel slightly, the combination of that and the loco's side-swinging rear overhang is enough to force the flange over the guardrail, causing the climbing the OP observed.

tst:

Do you have an NMRA gauge to check the wheelsets? If you don't, you could try shuffling the wheelsets to see if the derailment goes away.  I don't think your problem is with the spikes, but you could check that by removing the truck and rolling it through the switch frog.  If it bumps spike heads, you'll feel it.  Then take the removed truck and push it through the frog manually, applying some down pressure and side pressure with your finger.  Try to make it derail and see where the trouble happens.

Another thing you could try, if wheel-drop is the major problem, is to tighten up the front tender truck so that it can swivel but not rock. 

Is there a lot of offset in the loco-tender coupling at the point where the problems occur?

[C&O Fan]

 

Atlas Snap turnout

 

 Do you see any detrimental effects that I might be overlooking? 

 

Anyhow, thanks again for your insight and thoughts.

Tom

Tom

It appears from the 2nd photo you have some dirt in the gap between the point and the guard rail