NetNews Usenet Archive 1993 #3/NN_1993_3.iso/spool/rec/railroad/10895

Text File | 1993-01-24 | 4.5 KB | 94 lines

Newsgroups: rec.railroad
Path: sparky!uunet!munnari.oz.au!mel.dit.csiro.au!squid.mel.dit.CSIRO.AU!ajw
From: [e-mail] (Andrew Waugh)
Subject: Re: European track
Message-ID: <[e-mail]>
Sender: [e-mail]
Organization: CSIRO, Division of Information Technology, Melbourne
References: <[e-mail]> <[e-mail]>
Date: Sun, 24 Jan 93 23:56:59 GMT
Lines: 82

In article <[e-mail]> [e-mail] (Jobst Brandt) writes:
>Andrew Waugh writes:
>
>> To answer the original question, however, it was definitely U.S.
>> railroad practice (at least until the introduction of CWR) to
>> provide expansion gaps at the end of each rail length and to allow
>> rails to expand and contract as necessary.
>
>The use of clearance between rails resulted from a misunderstanding
>of rail buckling in hot weather.  In fact it occurred on any average
>day as well except that the symptoms pointed toward thermal
>expansion.  You will notice that all new rail laid these days is
>longitudinally anchored at as many as every other crosstie in some
>cases.  The cause of rail buckling in the days after anglebars with
>spike holes ended (long ago) rails moved relatively freely under the
>rolling action of trains similarly to cookie dough under a rolling
>pin if you roll only one way.  Rail piles up at the next switch
>because these cannot move.  Buckling invariably occurs at such a
>rail feature.

I'd agree with the contributory cause of rail creep, but I'd disagree
that the railways had any misunderstanding about its relationship with
rail buckling.

The fact that rail crept was well known by the turn of the century,
although the cause was much argued. I rather like your analogy about
cookie dough; it describes the underlying reason beautifully. On
lines which were used primarily in one direction, the rails crept
in the direction of traffic. On bi-directional track the rails usually
crept downhill, although this was not constant. Some rail features
cannot creep as they are too well fastened down; switches and
diamonds for example. If the rail was creeping towards a switch,
the expansion joints behind the switch would close up. This meant that
the rails closest to the switch had no room to expand and consequently
buckled.

The solution to rail creep -- rail anchors -- was routinely applied
by the middle 1920's to jointed track. Maintenance of way cylopedias
of that era show numerous patented designs, all claiming to be easier
to install and remove than the rest.

The underlying cause of buckling was the lack of lateral stability
in the track. Earth ballasted track was probably the extreme example.
Good practice with earth ballast was to pile the earth _above_ the
ties in the centre of the track. The top of the earth was then graded
downhill towards the ends of the sleepers. The ballast was below
the top of the ties at the rails (to keep the rails clear of the
soil) and was below the _bottom_ of the ties at the tie ends. This
structure allowed water to quickly drain away from the track and
prevented pools of standing water forming. The ends of the ties, being
above ground, reduced the incidence of rot and so extended the life of
the ties. The cost, however, was almost no lateral support for the
rails.

While earth ballast was mostly used on branch lines (this century,
anyway), many 'main' lines had poor quality ballast, small section
ties and light rails -- consequently they had problems with track
buckling.

By the 1920's, railroads were beginning to realise that if good
quality rock ballast was generously applied, particularly with
banks of ballast outside the ties, the lateral stability of the track
was so much improved that buckling became a neglible problem.

This, in turn, allowed expansion gaps to be dispensed with and made
continuously welded rail feasible.

>> According to the PWI book, during a typical yearly routine for a
>> track gang, they would spend about 3 weeks greasing the
>> fishplates.  In addition to greasing the fishplates and bolts, the
>> gang would check the ends of the rails for cracks, burns, etc.
>> They would also adjust the expansion gaps if necessary.
>
>This must be from the days of undocumented foreign labor when such
>useless activities were the norm.

I should have been a little clearer; this was in the U.K. Wages were
lower in the U.K. than in the U.S., but the cost of the track force
was still stupendous. Track circuits, of course, were relatively
uncommon.

andrew waugh