Hi – I’m a relative newcomer to the hobby, and I just ran into a problem that I’m hoping I can get some feedback on: I just got a Kato SD40 that routinely derails at a place where none of my other locos have ever had a problem.
I have a small N scale layout (an oval with some sidings, staging, etc.). It’s mostly Atlas code 80 track (including where the derailment happens). The curves are 11” radius sectional track, with a 19” radius piece at each end of the curve as an easement. Anyhow, at one of the joints between an 11” and a 19” track section, the Kato SD40 derails pretty much every time (with or without cars coupled, either direction). None of my other locos (Atlas and InterMountain four-axle) have ever had a problem at that point on the track. Oh, yeah – there aren’t any grades where the derailment happens, either.
Obviously there is some problem with the track there, but it’s pretty subtle – things check out with my NMRA gauge, and there isn’t anything visibly different between that joint and other joints that don’t cause any problems. The loco wheels also check out as being in gauge.
I have only one pair of six axle diesels. SD35’s from Atlas in N gauge. I have had no problems with them on straight or curved track or running over switches. My minimum radius is 15". You may have some trouble with them because of the number of axles. I also have three of the same loco’s in HO and I do have some problems with them derailing on track with some minor defects that four axle locos traverse ok. The front axle on each truck will hop over the rail on a low spot and the rear axle will hop over the rail at a high spot. Also check to see if the trucks are free enough to swivel in curves. The difference in curve radius from 11" to 19" could do it too. Maybe use flex track on those curves.
It really sounds like you have a pretty good handle on things and have done everything you should have. I assume it handles other 11" curves fine right? I don’t do Kato locos, so I might not be of much help, but I have a number of N six axle locos from other manufacturers that handle 11" and 9 3/4" radius curves without a hitch.
Only things I might suggest are 1. looking closer at your road bed and making sure there isn’t a slight dip or bump in it and 2. look closer at the loco and see how the arc of the trucks compare to the arc of your 11" curves.
As far as handling the other 11" curves okay, yes, it does. The curve on the opposite end of the oval is exactly the same (within the tolerances of my tracklaying ability) and it hasn’t derailed on that one.
I have a Kato SD40 and it does just fine. However, I can recall my Kato E8 acting the way your SD40 is. Turns out the gauge in the wheels was a little wide, so I pressed the wheels in on the axles just a hair. Never had another problem.
If you don’t already have one, get yourself an N scale NMRA standards gauge. It’s a very, very useful tool. It will tell you if your wheels are in gauge or not. I can almost promise your problem is caused by one or more wheelsets being slightly out of gauge. Because the wheelsets are press-fit on the axles, it’s a VERY easy fix. Just make sure you’re not changing the gauge willy-nilly; use your NMRA standards gauge tool to make sure you’re adjusting it correctly.
The Kato SD40 is an excellent engine. Once you correct the gauge issue you will be very pleased with it.
Hmm – I checked the wheel gauge initially, but I’ll check it again more carefully. One other thing I thought of that I haven’t checked yet is that each axle could be in gauge but they could be out of alignment with each other.
Thanks to everyone who’s offered advice so far – I’ll try to follow up once I’ve tried some of the things people have suggested.
Well, the problem seems to be solved. I replaced the two sections of track comprising the joint where the derailment happened, and that really seems like it has solved the problem. I guess that at least one of the original sections had a defect that was subtle enough to not be obvious when checking with the NMRA gauge, but big enough to cause problems. I haven’t found any problems with the loco itself so far.
I guess the lesson is that six-axle locos are just more sensitive to any track imperfections than four-axle locos, at least on fairly tight curves. Learning all the time! [:)]
You are right in saying that six axle locos (and cars for that matter) will pick track defects much easier than four axle ones.First,they are less tolerant of track pinching in curves as their truck’s wheelbases are slightly longer and since the loco’s weight is distributed through more wheels,each wheel rides slightly lighter on the track,making it more susceptible to jump off the rail.
However,from my experience,this is not a Kato (or Atlas’s BTW) loco issue.I’ve had very little problems with derailments with these jewels and it has always been a track problem.I own AC4400’s and SD90’s and they can handle curves as tight as 9 3/4 in. at moderate speeds though.Just wish my steamers would be that good…
It sounds like you’ve solved the problem, but there are still some things to think about.
First, it sounds like you’re using Atlas snap track. This is good basic track, but it’s based on a 45 year old design. The gauge is typically a little sloppy, the rails are tall, and the geometry can be a little imprecise. Also, even with your easements, the 11" curve is putting forces on your big 6 axle, and more importantly, long wheelbase engine that it simply doesn’t like.
Kato SD-40’s are very fussy on questionable track.
I find my (N scale) Atlas/Kato locos are susceptible to track flaws moreso, but due to their low wheel profile! I had to switch out a section of track last night,. a butted soldered meeting of track on a curve (flextrack sections). It looked fine from overhead but after I cut that section out and straightened it away from the roadbed, I could see how jackassedly crooked it was!! WOW what a flaw (both in the track and the soldering haha). After the swap-in of a new curve of flex, uninterrupted by any joints or butts, my track seems to be perfect! (well for me, so far- burnishing comes later). Now I can run my GP7 (Atlas/Kato 1st release) around at full-trottle and not hop a single rail! Incidentally, my Bachmann Spectrum 8.40.C runs the damn 8" radius curves pretty well too. Must be those big goofy wheel flanges keeping everything tucked in between the rails My spectrun F7 did choke up there though, their wheels are a hair wider than NMRA and I know it but am too lazy to fix it! Blah, I’d rather plaster a mountain.
In response to the comments about the Atlas track – yes, right now I’m mostly using Atlas sectional track. But I’m treating my current layout as kind of a test bed – I’m trying out some different operational ideas to see what’s most interesting, trying different track arrangements to see what standards I need, etc. I wanted to get going with something rather than trying to get it all right the first time… My current plan is to start pretty soon on construction of a “phase 2” layout that will basically start from scratch based on what I’ve learned with “phase 1”. At this point I’m thinking of going with Peco code 55 track for that layout, wider curves (12" or 13" radius), DCC from the start, etc. (The current layout is just DC.) I’m certainly open to feedback on any of those ideas – I’d like to take advantage of the knowledge of others as much as possible!
You should be safe with the SD-40’s and six axl’s, however if you might want to reconsider with anything longer and more articulated like MAC’s or large steamers. Anyhow, your problem sounds like a small kink in the rail where Atlas or you might be a little on the radii. The only way to be for sure is if you have any long 5" passenger cars to run them carefully along that curve with on finger on the roof at eye level and feel for any snags or any culprit that might make the engine jump rail on the bend. Regular 3" boxcars are too short to detect and the wheels on your loco do not spin without current. If you find the peice that is giving you trouble after a careful inspection then instead of replacing try to carefully bend out that piece with light pressure at each end, replace, and test. Also, you need to remember that nickel will flex and shift just like steel when heat is applied over time, so when you are running your loco on that railline heat accumulates from resistance and when power is off the track cools and may not go back to it original shape can and this can cause a minor flex in the rail ultimately effecting the curve radius tolerance by a small amount, but enough to cause a derailment on tight bends and especially along joiners, so I suggest soldering every bend and 3rd straight when you become 100% satisfied, plus it will reduce voltage loss if you ever decide to expand. One last note: make sure you have track easments before each bend because any long loco coming into a tight section is liable to derail at high speeds. The easment does not need to be much, but enough to release any excessive inertia before the bend, and you might want to consider using flex-track, code 80 of course if you become really aggrevated with derailments and track piece replacements. Good luck if you haven’t figured it out already.
My spectrun F7 did choke up there though, their wheels are a hair wider than NMRA and I know it but am too lazy to fix it! Blah, I’d rather plaster a mountain.