Kato FEF and Peco code 80 - oh dear.

[mmagliaro]

Pardon me whilst I steal your photo and annotate!  Perhaps this photo, and some more
detail, can help get to the bottom of this.

So in the photo below, the engine is backing through the turnout, from left to right?
And the wheels progress along the indicated "Closure Rail", following along the noted
locations 1, 2, 3, 4 ?  Is that right?  And about where (say, between 2 and 3? ) does the wheel climb up
and over the rail?

Perhaps the wheel starts to lift just as it rolls from the point rail onto the closure rail (at the hinge joint?)

This is indeed amazing and perplexing.

[Dave V]

For reasons I don't understand, my Peco code 80 switches...going on 9 years old...have a tendency to change gauge over time.

[peteski]

Another thing I was going to suggest is to remove the tender (via the drawbar) and loco's trailing truck (they both simply snap off). That way we might be able to better view the derailing event.  Well, the trailing truck is strictly cosmetic but the lack of tender being attached might slightly change the dynamics of the problem.  Still, it would be interesting to see.

[bobthebear]

So in the photo below, the engine is backing through the turnout, from left to right?
And the wheels progress along the indicated "Closure Rail", following along the noted
locations 1, 2, 3, 4 ?  Is that right?  And about where (say, between 2 and 3? ) does the wheel climb up
and over the rail?

From left to right. And between 2 & 3 is about correct. I seriously think there is not enough downwards pressure from the springs onto the rear drivers. Maybe just a fraction more is required, but not enough to deprive the 3rd set of traction.

Thanks everyone who is a puzzled as I am.

I will, over the next few days, remove the tender and trailing truck and experiment.

I will keep you all posted.

Cheers, Bob.

[victor miranda]

photos can have distortions so I am not stating this is "TRUTH"
just me looking things over.

I made a track gauge for the photo.
the rails on the left are wider than the rails on the right.
they get closer together starting between 1 and 2 and stay about that same gauge
for the rest of the way.

I am trying to figure out how the flanges get forced up.
if the gauge narrows there, it will force the wheel into the outside of the curve rail
a little after the narrowing...

victor

[bobthebear]

I will measure it tomorrow Victor.

[victor miranda]

Hi bobthebear,

you have said you have checked it,  what I see does not look out of tolerances...
just maybe one side of the spec and then the other side of the spec.

perhaps where it changes causes the problem.
in any case, something is allowing the flange
to get enough traction to go up the side of the rail.

victor

[peteski]

Bob, did you ever get this problem resolved?  Is this the same layout where you are having problems with the BLI locos?

[bobthebear]

Hi Peteski.
Yes, same layout, but surely unrelated problems.
I have replaced the front and rear truck wheels with nearer to NMRA wheels than the overfine Kato ones, and forward travel is fine now. Still problems reversing through curved turnouts, so I just don't do it!
My clubmate "chinapig" had similar problems (ran fine as an 0-8-0 for a while!!!) and has got a replacement rear truck, which has solved his trouble. We both ordered from the same dealer, so maybe there was a batch with a problem. He has got me a replacement rear truck too, but it's not fitted yet.

[Doug G.]

I think it's an artifact of the Kato design. All that vertcal play allows the driver(s) to climb the rail due to friction between the flange and rail rather than just getting knocked into the correct path.

I have always thought too much pseudo-equalization in N scale is not a good idea because there isn't enough weight involved to counteract the friction. It's better to just depend on the normal bushing/bearing tolerances to adapt to irregularities in track.

Doug

[victor miranda]

Hi Doug G.

I read your post as advocacy for allowing equalization AND
limiting  the amount one wheel can move to less than the height of one flange....

that may work.  difficult to get right... buuuuut it could work.

victor

[Doug G.]

Yes, or if you are going to have considerable vertical movement, use enough spring force to keep the drivers from rising up due to friction. Of course, then you are pretty much negating any advantage equalized drivers may have, whatever that may be.

The end drivers, exacerbated in long wheel base steamers, hit the inside of the outer rails in curves, of course, and something's gotta give. Either the rail will knock the driver into the corect path or, if the driver is allowed to rise, the driver will go up over the rail. I believe that's what's happening to Bob's FEF.

[peteski]

The end drivers, exacerbated in long wheel base steamers, hit the inside of the outer rails in curves, of course, and something's gotta give. Either the rail will knock the driver into the corect path or, if the driver is allowed to rise, the driver will go up over the rail. I believe that's what's happening to Bob's FEF.

Um, Bob uses standard Peco turnouts, supposedly laid perfectly flat.  It is not some custom-made turnout. If what you say was the case, wouldn't all the people who have the FEF-3 and use Peco turnouts have the same problem?  Wouldn't we see a large outcry about this on the model RR forums?

I often frequent a layout with only Peco turnouts on it.  But it is DCC. Once I install a decoder in my FEF-3 I'll take it for a test run there.

[Doug G.]

So the corollary is if Bob is using standard Peco switches like many others and it's all in spec, gauge-wise, why is he having a problem with that one loco derailing? How many FEFs are there out there being run on Peco code 80?

Doug

[peteski]

So the corollary is if Bob is using standard Peco switches like many others and it's all in spec, gauge-wise, why is he having a problem with that one loco derailing? How many FEFs are there out there being run on Peco code 80?

Doug

Well Doug - at least one. 
Tonight I visited my friend. His layout uses exclusively Peco track (mostly C80, with some c55).  If you remember from the A-board, I posted some photos and video of the layout.  To see all the mainline switches from engineer's perspective, check out http://forum.atlasrr.com/forum/topic.asp?TOPIC_ID=62981 .

Anyways, I kept wanting to run my FEF-3 on his layout since this thread started, but the layout is DCC and I still don't have a (sound) decoder installed.  But tonight I decided to take a simple DZ126 decoder and rig it up in the locomotive, so I could take it for a test run.  In order to do that I took the loco's shell and and temporarily wire the decoder in.  I taped the decoder to the chassis and ran the loco without the shell.

I ran it, pulling just one auxiliary water tender, few laps the layout in both directions.  From very slow to fast (but not at maximum throttle  that is way too fast). The loco went through facing and trailing routes on curved, large and medium radius switches. Mostly electrofrogs.  Few turnouts had the guard rails shimmed, most are unmodified.

No derailments.  But since I had the shell off it was easy to observe the wheels. The narrow-thread low-flange wheels were really falling deep into the frog.  It looked really frightening!  I'm actually surprised that the loco wasn't derailing.  The leading and trailing trucks were the worst, but #1, 2 and 4 drivers (the ones that are equalized or sprung) were also diving into the frog.  The wheels made rather loud clicking noises while goring through the frogs.  Those noises also sounded frightening.  But the loco stayed on track.

SO now, from a first-hand experience, I can state that my FEF-3 has no problems going through Peco switches (although it is not happy going over them).