Maximum permissible wheelbase for two axles

Have just been watching a a film clip showing the trials of the Properllertriebwagen Kruckenberg Schienenzeppelin (the Railzeppelin) in 1930; the aero industry engineering is evident in the early, unpanelled views (and is remarkably beautiful - 01:00-01:20 - could live with that as interior design).

(Link: )

What I hadn't realized before is that, as originally built, the RailZeppelin was a straightforward, two-axle four-wheeler with a fixed wheelbase of some 19.6 metres (64' 5 3/8") - or more than double that of a Class 141 Pacer.

This has set me wondering - what is the maximum theoretical wheelbase for a diesel / electric locomotive or self-propelled vehicle running on standard-gauge mainline, before it falls foul of the track geometry and ends up on the floor?

Am assuming that the maximum permitted length would be a safe margin under that figure - but what else has pushed the boundaries and played havoc with the p/w?

Apologies in advance if this is a well-trodden path; any recommendations for reading gratefully received.

 

That depends on your track geometry! But a four-wheel vehicle will have a greater permissible wheelbase than a six-wheel vehicle.

In the UK - particularly in the past - signalling was probably a bigger restriction. When train detection at FPLs was provided by lock bars (well, it still is on at least some heritage railways of course), the maximum distance between two axles on a vehicle had to be less than the length of a lock bar, to ensure that there was no risk of the FPL being unlocked with a train stood on it.

 

Current equivalent is the specified minimum length of track circuit, because you don't want a track showing 'clear' in the middle of a train or you get the appearance of a train splitting. On Network Rail I understand this figure to be 18.3m.

There was some debate about this at work recently; I speculated that the figure was simply a conversion from a standard 60' rail length but a colleague came up with some sort of car transporter wagon from a few years ago which effectively set this limit. I wonder if anyone can confirm this?

 

The distance between inner axles of adjacent bogies - in other words the distance between bogie centres minus the mean wheelbase of the two bogies - must also be less than the locking bar length for the same reason. This will generally be greater than the distance between the outermost axles of adjacent vehicles in a train.

The IRSE Minor Railways Section's "Guideline On Mechanically Operated Points" gives a length of "45 to 50 feet" as a typical locking bar length, which would preclude a Schienenzeppelin given the dimensions above.

 

Firstly, that's a fascinating bit of video, thanks for posting. I've only ever seen photos before.

For your actual question, as to when a vehicle of given wheelbase would actually fall off the track, the answer is "It depends." With a sufficiently large curve radius, you could get away with anything, but that wouldn't make for a very practical railway. I did a bit of messing around with a miniature tender wheelset with badly worn flanges, and was slightly surprised to find that you can't just rotate it right off the rails without the flanges fouling anything (doesn't mean it won't derail, of course, and with a sufficiently small wheel diameter I reckon you probably still could just pivot off without fouling the flanges).

This suggests two potential limit states for feasible wheelbase, one where the flanges are binding on both rails at once (or worse, riding up them) and one where the flanges bind up in check rail grooves.

For a simple four wheel unit on a constant radius curve, each wheelset is at the same (opposed) angle to the centreline of the track. Calling wheelbase W, and curve radius R, this angle is (assuming consistent units):

a = arcsin(W/2/R)

Where things start to bind up depends very much on the wheel diameter and flange profile. The taper on the rail side and radius on the tip of the flange make it quite awkward, the 2D profile at rail height is a mix of straight lines, ellipses and a hyperbola or two, assuming a non-complex profile, so I'm not in a hurry to start calculating the angle at which it might bind for any particular case....

 

My thanks to all for their responses - they've all been informative and well-considered, and there's some elegant thinking in there.

The explanation of the geometrical issue (#8126) resolves my initial query - i suspected there would be a 'pinch point', either figuratively or literally, but couldn't put it down on paper to comprehend it - the problems and limitations around locking bars, etc, had, I freely admit, eluded me altogether.

This has been very enlightening; once again, thank-you, everyone.

 

And a curve with a bit of a twist fault in it will throw it off as well.

 

No mention of gauge widening, which can allow a longer wheelbase around a curve. Gauge widening used to be common practice in yards with sharp curves but these are now largely confined to history. Is gauge widening practised on today's big railway?

 

It has been known to be used on certain tight curves.
Crimple near Harrogate was relaid approx 1/4" wide to gauge in the 90s