I have always heard that the gauge of mainline tracks is widened on curves in order to reduce rail/wheel wear. How is it determined what degree of widening is used, and over what distance the transition takes place? From a practical standpoint, how is the gauge measurement accomplished during the transitional range? How can gauge widening be possible with the use of concrete ties which feature pre-cast gauge shoulders spaced for straight track gauge?


KEAGLEDESIGN@VISI.COM

> I have always heard that the gauge of
> mainline tracks is widened on curves in
> order to reduce rail/wheel wear. How is it
> determined what degree of widening is used,
> and over what distance the transition takes
> place? From a practical standpoint, how is
> the gauge measurement accomplished during
> the transitional range? How can gauge
> widening be possible with the use of
> concrete ties which feature pre-cast gauge
> shoulders spaced for straight track gauge?
Gage widening on curves was used to accommodate steam locomotives with long rigid wheel bases. The amount of widening was determined by the degree of curvature and the lenght of the longest rigid wheel base. The transition was also variable based on the speed limit and the degree of curvature. To the best of my knowledge the practice is no longer used on mainlines because diesel locomotives do not have nearly as long of wheel bases as large steam locomotives. If the gage were to be opened on concrete or steel ties they would have to be manufactured accordingly.
The practice of opening the gage on curves did last well into the diesel era but was mostly unnessary. On the Aberdeen And Rockfish we have 24 degree curves on both legs of the wye in Fayetteville, NC and regularly take 60 foot boxcars of paper around them. We also have 18 degree curves in the mainline and operate 90 foot flatcars loaded with steel around them with no open gage. Obviously we are not going fast but you get the idea. Mainline curves of 12 degrees are considered very sharp. Wheel and rail wear is now controled mostly with rail or flange lubrication on both mainline and branches. Also most of the new 6 axle locomotives built in recent years have self steering trucks that articulate when going around curves greatly reducing both wheel and rail wear.

jhbohon@yahoo.com

You mention x degree curves in your post. How are the degrees of curvature calculated?

Thanks in advance for your response

wyld@sbcglobal.net

> You mention x degree curves in your post.
> How are the degrees of curvature calculated?

> Thanks in advance for your response

The degree of curvature is the size of central angle of a 100' long cord. This also equals the angle of deflection from the bearing of one 100' long cord to the next.

This goes back to surveying the railroad alignment. And back to the era when surveyors used real 100' long chains. This is the reason that engineer's stationing along a railroad alignment are at 100' intervals.

Brian Norden



bnorden49@earthlink.net

Go to http://www.steamlocomotive.com/model/curve.html for details.


hankmorris@earthlink.net

To finish off Brian's comment, it turns out that the radius of a 1 degree curve is 5730 feet. To get the radius of a curve expressed in degrees, divide 5730 by the degree of the curve. So for example the 24-degree curve mentioned earlier has a radius of 239 feet, very sharp for standard railroad practice, but not for transit.

Modern rail transit design usually calls for gauge widening of up to 1/2" to 3/4" on sharp curves to prevent truck binding and reduce rail wear. Gauge widening on the curved side near the points of AREA-style turnouts is also typical for many transit properites.

The old streetcar properties, with curve radii down to 40 feet or less certainly widened gauge on curves. Having laid out the loop tracks at the Baltimore Streetcar Museum years ago, I am convinced that the "typical" three-point track gauges inheritied from Baltimore Transit were designed specifically to automatically widen gauge on curves the appropriate amount when the two-point end was used on the outside rail.

pjslks@ix.netcom.com

Degree of curvature is the measure of deflection from a tangent measured at stations 100 feet apart. One degree of curvature is equal to a radius of 5730 feet.

jhbohon@yahoo.com

> The old streetcar properties, with curve
> radii down to 40 feet or less certainly
> widened gauge on curves. Having laid out the
> loop tracks at the Baltimore Streetcar
> Museum years ago, I am convinced that the
> "typical" three-point track gauges
> inheritied from Baltimore Transit were
> designed specifically to automatically widen
> gauge on curves the appropriate amount when
> the two-point end was used on the outside
> rail.

For streetcars, the wheelbase of the truck determined the minimum radius of the track as opposed to the truckbase on the body. A rough estimate was eight time the longest truck wheelbase for the minimum. Of course the interurbans that used city streets had to conform. Hence you saw cars like LSE167 62ft long with seven ft wheelbase trucks grinding around the 40ft radius curves on Clevelands public square. The only reason it worked was the slow speed allowing the flange to drop back in the slot rather than climb out. Depending on the availability of property easements Ohio interurbans did have at-speed curves if they could afford it and I can show you at least one road today that was built to follow an interurbans curve in what was then countryside.


lamontdc@adelphia.net

I think you'll find that the true surveyor's chain (also known as a Gunter's chain) commonly used in the nineteenth century was 66 feet long. This was equivalent to four rods and was the reason so many single track rights of way were 66 feet wide. The surveyor's chain also conveniently related to the mile, there being 80 chains to the mile.
The engineer's chain was 100 feet long, with each foot divided into tenths, but this was not usually used in land measure.

> The degree of curvature is the size of
> central angle of a 100' long cord. This also
> equals the angle of deflection from the
> bearing of one 100' long cord to the next.

> This goes back to surveying the railroad
> alignment. And back to the era when
> surveyors used real 100' long chains. This
> is the reason that engineer's stationing
> along a railroad alignment are at 100'
> intervals.

> Brian Norden


dnoyes1@comcast.net

> I think you'll find that the true surveyor's
> chain (also known as a Gunter's chain)
> commonly used in the nineteenth century was
> 66 feet long.
> The engineer's chain was 100 feet long, with
> each foot divided into tenths, but this was
> not usually used in land measure.

When I worked in civil engineering and surveying, all we used was a 100ft chain and our 100ft tape was marked off in feet and tenths of a foot. When I worked in a steel plant the vendor accidently sent in some engineering tapes and it drove the inspectors nuts because all of the tube rounds suddenly were the wrong length ...then somebody noticed the 10 divisions per foot instead of twelve.


lamontdc@adelphia.net

> I think you'll find that the true surveyor's
> chain (also known as a Gunter's chain)
> commonly used in the nineteenth century was
> 66 feet long. This was equivalent to four
> rods and was the reason so many single track
> rights of way were 66 feet wide. The
> surveyor's chain also conveniently related
> to the mile, there being 80 chains to the
> mile.
> The engineer's chain was 100 feet long, with
> each foot divided into tenths, but this was
> not usually used in land measure.

Yah, you're right. We have a deed at work (I'm the keeper of such stuff at a private domestic water purveyor) that was written in 1906 using 66 foot long chains. With a rod being 16.5 feet that is a reason you may see an old railroad (or even public roadway) right-of-way being 33 foot wide (like just down the road in Covina, California).

Nowdays we see deeds and subdivision maps done in feet and hundredths of feet.

The earlier post about dividing a fixed figure by the degree of curvature will give an APPROXIMATE answer. To get the correct figure you need to use trig, especially as the degree of curvature gets higher (tighter curve).

Brian Norden

bnorden49@earthlink.net