Case Closed? RE: Origins of US Standard railroad gauge

[James Love <love@cptech.org>]

Below is Richard Solomon's response to comments about the origins of US
Standard railroad gauge.
jamie

--------------
Subject: Case Closed? RE: Origins of US Standard railroad gauge
   Date: Mon, 13 Apr 98 08:57:00 EDT
   From: "Lee W. McKnight" <mcknight@rpcp.mit.edu>
     To: love@cptech.org
     CC: Pearah@rpcp.mit.edu

Hi Jamie,

Richard Solomon's history is below, including what the Encyclopeda
Britannica of the early 1800's reported on this issue, and the rut
widths in Corinth - turns out if you dig deep enough it's a Greek spec,
made a Roman standard. For more arcana and insights into railroads,
standards,mythlogy, and the Internet, I again refer you to The Gordian
Knot.

Feel free to pass this along to your other correspondents.

Lee McKnight

>Mime-Version: 1.0
>Date: Sun, 12 Apr 1998 21:44:18 -0400
>To: "Lee W. McKnight" <mcknight@rpcp.mit.edu>
>From: "Richard J. Solomon" <rsolomon@dsl.cis.upenn.edu>
>Subject: RE: Origins of US Standard railroad gauge (fwd)
>Cc: David Pearah <pearah@MIT.EDU>,
>        rneuman@pobox.asc.upenn.edu (W. Russell Neuman)
>
>At 12:17 PM -0400 4/7/98, Lee W. McKnight wrote:
>>Richard,
>>
>>So who is right, if anyone?
>
>There is a short and a long answer about the Roman origin of 4'8.5".
>
>The short answer: it is quite easy to trace the standard gauge from Ancient
>Rome (& especially Ancient Greece) to today's railroads; 4'8.5" is not a
>bad choice; and there is no such thing as an "unusual" number (but that is
>another joke for another time). Numbers are odd, even, integral, prime,
>etc., but not unusual.
>
>That said, I'll give you my longer, and probably just as winded, but more
>easily verifiable story about the gauges:
>
>Somehow your correspondents confused the "tens of thousands" of guided
>pathways by the 1880s (I have no idea where they get that number),
>streetcar lines, quarry roads, and weirdo things that didn't connect, at
>least for through freight traffic with conventional steam railroads.  There
>were only a few hundred steam RR companies; and even if you include the
>fact that by the 1880s consolidation had begun from much shorter roads, you
>only get a thousand or so RRs. By the 1880s, the dominant (by mileage)
>common carrier gauge in North America & Europe was 4 foot 8-1/2 inches.
>There indeed were other gauges for steam common carriers -- wider and
>narrower -- but all of them came AFTER 4'- 8.5" had been adopted by the
>first steam roads in the U.S. and England. (There is an excellent book by
>Taylor & Neu on the subject of the 19th Century American gauge wars, which
>we reference in Gordian Knot.)
>
>How did this not unusual measure of 4'8.5" come about? The Roman cart
>gooves (and they were indeed Roman -- I'll get to that later) in stone
>village pathways all over Europe were roughly 5 feet apart measured from
>the OUTSIDE edges of the grooves or ruts. The ruts themselves (some of
>which still exist to be examined) were each about 4" to 6" in width. (Note:
>being grooves in stone to carry off mud, excrement, & etc. they weren't
>exactly calibrated in micrometers.) Rough widths were good enough for
>cartwheels to transverse the towns without breaking their axles on the
>"crosswalks." Indeed, a wide width for the grooves was necessary since the
>wheels were NOT guided with flanges -- the next key engineering point.
>
>Hence, comes the second, and most interesting part of re-engineering Julius
>Caesar's mandated gauge for carts (Yes, he did mandate this -- I'm getting
>to that). Lo & behold, some genius in the  16th Century noticed that if a
>flange was applied to a guideway mechanism, an animal-pulled cart could be
>run on a wooden rail higher than the ground, instead of being guided by the
>rutted and uneven grooves in the dirt or wooden plank road. (Guidance plus
>a smoother surface lowered rolling friction so the animal would eat less
>and pull more.)
>
>Now, flanges can be inside or outside of a wheel, but the flanges also
>could be part of the RAIL. Without access to a major government R&D program
>to judge the most effective architecture for infrastructure-reducing
>friction, the first-generation railroads ("tramroads") picked the flanged
>or "edge" RAIL for their first systems, instead of today's common flanged
>WHEEL. Like I always say, little things can make a big difference. But
>first the gauge width:
>
>The conventional carts (trams) of the day had their wheels about 5 feet
>apart -- steming from the Roman cart gauge (don't worry, I'm getting to the
>Romans), and measured from their OUTSIDE wheel edges, with tolerances noted
>above for rut widths. In 1789, one William Jessop of the Butterley Iron
>Works in Derby, England laid a new colliery tramroad on a heavily used
>conventional roadway between Loughborough and Nanpantan with a novel,
>L-shaped iron "edge" rail. The rail had its riser on the inside of the
>track, but the "track" had to be depressed so the top of the riser was
>flush with the surface of a conventional roadway permitting other vehicles
>to continue to use the right-of-way. Since the guidance part of the L was
>below the roadway surface, two sets of wheels on each axle were used: one
>set riding on the top of the iron L-beam bearing most of the weight of the
>cart, and another set riding on the bottom of the L to guide the cart and
>keep the other wheel on the top of the L-beam.
>
>The track had its iron L-beams depressed in the roadway like this:  _|
>|_
>
>Hence the flanged wheel was born -- but the guiding mechanism (which became
>a flange when the wheel was cast as one piece of iron) was on the OUTSIDE
>of the supporting wheel, not the inside as on today's integral flanged
>railroad wheel.
>
>Jessop found that the outside, double-wheel, flanged system was
>unsatisfactory (the reasons being complicated enough to require an entire
>book to explain the math behind the guidance, and it wasn't written until a
>Century later, so I'll skip this part of the history). Therefore, Jessop
>rearranged the double, now integral iron wheels so their flanges were on
>the INSIDE, but initially kept the L-shaped rail track architecture as
>above (the flanges running in the mud -- I didn't say the roadway was
>paved). It was reasoned that with this arrangement the L-shaped "edge"
>rails would better keep the wheels in position on the AXLES and solve other
>problems which I will not get into here. This twist was the HTML of
>railroading -- little details can make a BIG difference.
>
>Now let's do the numbers:
>
>Since the L-beams were about 1.5" wide in total (the L-beam was formed from
>a 3/4" plate bent 90 degrees), plus a 1/2" clearance for mud & dirt, the
>so-called standard gauge of the world became:
>
>60" - (2 x 1.5") - 0.5" = 56.5" = 4' 8.5".
>
>Not an unusual number at all. This took place in 1800.
>
>My source is an old account in the Encyclopędia Brittanica written more or
>less when all this happened, not knowing that 170 years later there would
>be controversy on something called the Internet. I have read the same
>accounts in several contemporary books published as early as 1810, but I
>don't have them ahand right now. I do own an old Brittanica.
>
>The first steam roads in England, the U.S., Germany, France and Belgium all
>followed Jessop's 4'-8.5" template. Since the first roads also got their
>steam engines from England (or tried to), and since they all used
>conventional coal carts and stagecoaches for their first rail vehicles,
>following the 5' (minus rutted width) was not really a major paradigm
>shift. Shortly after the first steam roads of the 1825-40 period, the
>semi-scientific debate over gauges began, and a large number of different
>track and loading gauges were tested and built in different places. But
>4'-8.5" dominated for various reasons, some of which we discuss in Gordian.
>(See Taylor & Neu for the U.S. history which was indeed very messy.) It is
>not a particularly bad choice for even modern, super highspeed lines as it
>turns out, lucky as the choice was without research. What has been a
>problem is the wide variety of LOADING gauges, and the interconnection in
>some places to narrow or wider gauges built for other reasons, not to
>mention eclectic signalling systems.
>
>And there is no mystery of how 5' was derived from the Roman rutted path
>width.
>
>Back to Julius Caesar. Shortly before he was stabbed by Brutus et others in
>44 B.C., the General visited his new possessions in Greece, in particular
>the military base at the Isthmus of Corinth. He observed the stone rutted
>cartway in operation there to transfer specially-built, mostly military,
>wled boats from the  Gulf of Corinth to the Saronikos Kolpos (gulf in
>Greek), avoiding some 400 miles of sailing in dangerous waters roundabout
>the Peloponnisos. The cartway had been built with carved stones roughly 5
>feet apart, with each stone having a Greek letter carved on it for some
>mysterious reason. This was a fairly meticulous-built guideway, carts and
>all, though very little is known of who designed it & when, though it is
>pretty clear why. Jules, being pretty swift about engineering if not
>assassins and astrology, saw that the key was that all the cart/boats which
>used it were STANDARDIZED by the Hellenic military for wheel and axle
>dimensions. He applied this version of HTML to a growing problem in the
>Empire -- chariots that had to slow in towns because the ruts were all over
>the place. His edict for standardized rut widths lasted a long time, and it
>is hard to argue with the dimensions.  (BTW, measure the distance between
>your auto's tires; you pick the points.)
>
>I was in Corinth 2 years ago and of course I visited the excavated
>guideway. It apparently was in use into the early Middle Ages and then was
>forgotten and got filled with debris and dirt. When the famous Corinthian
>canal was built in the 19th Century the guideway was re-discovered (it was
>thought to be a myth) and a few yards not demolished by the canal have been
>restored, mysterious Greek letters and all. I measured the ruts (actually
>they are quite neatly laid and cut stones). 4'-8.5". You betcha.
>
>Richard
>
>PS. I own a scarce, 1st Edition of Oliver Jensen's lovely American Heritage
>book on railroads cited as the source by your correspondent. The original
>edition is full of beautifully printed, two-tone photographs, and also full
>of errors and factoids -- things which sound like facts but aren't. Jensen
>is hardly a reference, but he sure knows how to choose pictures out of the
>Library of Congress' marvelous collections. The later, more popular
>editions muddied the wonderful photos and didn't change the factoids.
>