Model Railway Forum banner

Locos derailing

11373 Views 53 Replies 16 Participants Last post by  Ozzie21
QUOTE The diesels and DMUs are ok but some Hornby steam derail quite regularly. None of my German locos derail at all!

How common is this. The only issue I ever had was with a Bachmann Class 4MT when the front bogie kept coming off on a certain point. The answer was to not run the loco across that part of the layout.

The thing is Hornby and Bachmann don't deliberately design trains to derail, and surely they carry out extensive testing prior to putting a model into production.

And Hornby have their roadshow and their trains run for many hours each day and do not derail.

So lets have a definitive discussion on this subject.

Why do trains derail and how can you prevent this from happening?

Convince MRF members that it is the loco that is at fault and not some other aspect.

And what makes German steam locomotives stay on the track better than British locomotives?

German steam locomotives have smaller wheels and run more slowly as HO scale speeds are slower than OO so is this a factor?

Do German locomotives have bigger flanges than their British counterparts?

Happy modelling
1 - 6 of 54 Posts
Unfortunately, no discussion on these lines (!) can ever be 'definitive', because so much of what is said is based on age-old entrenched beliefs and personal opinions rather than clear, measurable facts.

A good example of this is the constant jibing about German/British quality, which I do not want to progress any further. I'll just say that I have a lot of experience with both (and many others besides), and until quite recently, almost anything not of German or Japanese origin was unmistakably inferior in almost every respect. I would add that I HAVE had the odd duff German item too, but those have been very few by comparison. Hopefully, quality improvements will continue, though it needs to be said that that is not much help to people who have already made their investments! Enough of that.

QUOTE . . . The answer was to not run the loco across that part of the layout.
It might have been AN answer but not THE answer. It's clear that there was a problem with that part of the layout, equally obviously track related, and THE proper answer would have been to investigate and fix the problem at source, rather than running away from it. How many bits of track does one designate a No-Go area, before deciding to identify and fix the REAL problem, the root cause?

But that anecdote is a useful pointer towards the most usual reason for derailments being track that is faulty in one or more of several ways. It could be poorly designed in the first place, badly manufactured in the second, damaged at some point in time thereafter, incompetently laid or, more than likely, some nasty and hard to identify combination of SEVERAL of those factors and perhaps ALL of them! While one factor alone MIGHT result in a derailment, it is more likely to be an accumulation of several small faults which, combined together, eventually produce a crash.

While track problems are the most probable causes for derailments, all of those track factors are very often also combined with a lot of additional factors in the vehicles running upon it. The same factors noted against the track can also be pointed at the vehicle running upon the track plus some extra ones too. Unfortunately, at this point, there are so many accumulated variables involved in the combination of these that it can be extremely difficult, close to impossible in some cases, to pin down the exact causeS for the derailment - another good reason why a discussion just cannot be 'definitive'. I haven't mentioned idiots driving their trains like maniac racing car drivers, but that is a common factor too!

I laughed out loud at the suggestion that any difference in 'scale speed' between OO and HO had anything useful to contribute! As if the average home train player, whom Gary seems to feel he represents, would have the slightest idea as to what scale speeds either actually are or what they should be, in either scale, let alone possess either the ability or the inclination to drive carefully in accordance with them! Oh, I needed a good laugh today!

It is my personal opinion that most (but NOT all) derailments are primarily track related, but these are exacerbated by locomotives and other rolling stock, each of which can have a myriad additional tiny faults, that contribute to one particular vehicle derailing more than others on the same track.
See less See more
Excellent post by Ozzie21!
QUOTE Model locos can derail when there is an obvious fault like an oversprung pony truck or wheels out of guage. But in the end it all comes down to trackwork and benchwork.

You have to start with the basic foundation and make it good. Once that is established and eliminated as a cause, then trouble shooting the rest becomes much easier and more effective. Be methodical.
QUOTE I took time to lay my track properly
Obviously another 'exhibitionist'!
See less See more
QUOTE but then you may as well go the whole way and just use flexi.
I do agree!

Simply getting rid of track joints is a great advantage, both electrically and mechanically, apart from which, sudden kinks (between joints) also disappear from the visual annoyance list. For unavoidably tight curves, I would sometimes use a few set-track curves first, but only as a guide and possibly running trial towards developing the best alignment of flexi-track for the final production. When using flexi-track, I would also be likely to use tracksetta plates in the main body of the curve, certainly on the tightest part of it, while leaving the ends free for hand adjustable easements. This rigid centre/flexi ends combination gives considerable latitude for experimentation and I find it works very well for me.
See less See more
Perfectly simple.
  1. Chock up outside rail until all trains can traverse at full speed without flying off the outside.
    Stop train on super elevation.
    If it doesn't fall off the inside, you're done.
  2. If it does fall off the inside, gradually lower outer rail until it doesn't.
    You're done, but drive trains a bit slower round this this curve.
  3. When laughter (or rage) has subsided, use Google to search the Internet!
See less See more
For a slightly more technical response than my previous post, readers might like to take a peek at

Central Pacific Rail Road Photographic History Musem

There, you can see some rather nice photographs graphically demonstrating old, simple tools used to check and adjust superelevation. The Americans really do excel with this sort of site and I feel safe in suggesting that the calculations are close to universal, regardless of country of origin - maths and geometry don't change much, although specific engineering techniques can vary quite a bit!
QUOTE Superelevation is derived by the equation E = CDV^2 where

E = superelevation of outer rail in inches
C = 0.0005 for curves less than 3 degrees OR
C = 0.0004 for curves 3 degrees or more
D = the actual curvature of the track in degrees
V = maximum allowable speed in miles per hour

A more austere presentation of simple track geometry is available from
"Railroad Track Standards": US Army Technical Manual # 5-628, dated 1991

If you want transitions (getting a bit complicated now), try
Trailing Edge Technology's Superelevation and Transition Spiral Web Pages

Trailing Edge provide a very handy Superelevation and Transition Spiral CALCULATOR
Unfortunately, the calculator was exhibiting a "Server Runtime Error" at the time I posted this, but might be recovered by the time anyone else tries it. Maybe after the weekend.

However, their Transition Spiral PRIMER was working just fine and is truly fascinating for the technically minded.

The whole business becomes much more complex when forced to prooduce some form of 'optimum' banking for trains travelling at quite different speeds. It's considerably simpler if the track is dedicated to one, standard type of traffic.
See less See more
1 - 6 of 54 Posts
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.