Thanks for that link. George is very knowledgeable and generous with it, and I would give credence to his explanation over most others (including some of my own).
- Paul North.
Keep them 2-3 cribs apart.
End batter, joint batter, joint ties surface mismatch - expensive to fix and joint memory does NOT go away. It’s diminishing returns time once you get there. Secondhand rail gets cropped for a reason.
Been there, done that. (and have seen some pretty weird things happen with areas that temporarily had panel party square joints - it terrifies bridge people)
From the current (March 2011) Norfolk Southern “Guidelines for Design and Construction of Privately Owned Industry Tracks”, Chapter VII. MATERIALS, Section 7.03 RAIL, middle of page 23 (page 2 of 6 of this “PDF” format excerpt), middle of the 4th paragraph: http://www.nscorp.com/nscorphtml/pdf/Customers/ID/2011/07_Materials.pdf
“Joints in opposite rails shall be staggered not less than 8 feet and not more than 14 feet apart, except as close joints may be required at insulated joints or turnouts. To minimize the cutting of full length rails, short rails of not less than 15 feet may be used in adjusting for proper spacing of joints.”
- Paul North.
Haven’t I read somewhere that squared joints have always been standard practice in Russia?
Of course, the whole question is becoming of academic, historical interest only. One has to look pretty hard to find any joints in main track these days.
I have traveled Slovenian Railways line between Nova Gorizia and Bled many times. Originally, it was the main line of the Transalpina Railway built between Vienna and Trieste, Italy, in 1906, but two world wars severed the line at the borders and it fell into disuse. Today it is basically a branch line with about a 50 MPH top speed, small 2 car DMU’s, and a few very light freights. The track is well maintained with rock ballast, wood ties, and good rail, but the joints are all square, rather than saggered per US practice. The ride is relatively smooth with no low joints or kinked curves noted.
Another part of the Transalpina runs from Campo Marzio Station in Trieste, up to Opicina, then across the border to Sezana, Slovenia to eventually connect to the above noted segment just south of Nova Gorizia. I’ve ridden the section between Trieste and Opicina and found it to be in the same condition and make-up as the above mentioned line.
It should be noted that the axle loadings are low with speeds are low to moderate, but the track is smooth without low joints or kinks.
What I find interesting is that even after all the historical development, there seems to be no clear consensus about which rail joint pattern is better. Instead, it appears that there are many pros and cons to each pattern and one or another emerges as standard in each country or individual system within each country. The choice does not seem to have anything to do with gage, axle loading, speed, or whether the cars are four-wheel or eight wheel.
In a way, the joint pattern is like choosing gage in that it becomes a standard. However, unlike gage, the joint pattern is easily changeable if one wishes to explore an alternate pattern.
At the mine I used to work at, we used LOTS of panel track in the pit. Those joints are, of course, square at 39 feet. Most of the track was too rough and slow to tell the difference, but where things smoothed out a little it rode horribly with constant pounding. Much worse than on permanent track where the joints were staggered.
So, the wheels hitting two joints together increased the pounding? That does seem to indicate the basic advantage of the stagger pattern.
The basic advantage of the square joints is the prevention of rocking (and also for convenience of panel track). But the combined bounce effect at each pair of square joints also a disruptive force, so it is hard to decide the advantage over staggered. I conclude that historically, the two patterns competed with each other to the extent that tracks were out of alignment. With good track, I suspect that the staggered versus square question becomes moot.
It’s called impact loading and it’s much more a threat. Does a heck of a number on timber pile caps (splinters), railcar ride and freight car lading.
Wonder what some of that foreign practice requires in terms of joint support and fastening.
Bucyrus:
I don’t have any scientific or engineering reason, merely an intuitive one. This comes from an old William Henry Jackson photo of the newly-laid Colorado Midland tracks near Buena Vista, Colorado. The rails joints are clearly square on a tangent section of track. The reason I can come up with is that the track was laid by hand by crews of men puling/sliding rail sections off a flat car. The flat car was pushed along from behind as the track advanced. As the rails were pulled off the flat car, with square joints, the crews could work both rails at the same time with minimal effort since the flat car would be at the end of the two previously laid rails which had evenly spaced ends. As the new rails were laid down and spiked the material car could move forward 39-feet to allow the next set of rails to be laid, etc.
How they eventually went back and staggered the ends is something I have not figured out yet. I agree with your assessment of the simultaneous pounding of the wheels hitting the two matched joints as being detrimental to the ride and rail life.
One observation. On the primary route used by train arriving and departing at the west end of CP’s (former) hump yard at Alyth, two facing point turnouts mean the joints are very nearly square. There is usually a very pronounced dip in the surface right at that location, and we can watch each truck tilt up and down as it crosses. That develops where speeds are very slow; I’d hate to think what would happen with the dynamic loading of more normal track speeds. Square joints are most viable where loads are light and/or meticulous preventative maintenance is the rule. The latter requires plentiful manpower, not usually available in North America.
John
Methinks that the loading at higher speeds would actually be less of a problem - at least beyond a certain point.
The factor being the time a given weight spends on a given point.
I would suspect that if the approach to a given spot is smooth, and that “smoothness” continues through that spot, then a train at “speed” will have a limited effect.
The best examples I can think of are first, a car being driven on a beach, wherein a slow moving vehicle leaves a rather defined tire track, while a vehicle moving at speed will reach a point where it leaves only a minimal track; and second, a washboard on a gravel road. It starts with some irregularity in the road, and grows as more and more vehicles pass over it, as more and more tires bounce and “stress” the road surface.
I think one good example of square joints is a 90 degree diamond. The one in Durand, MI was getting worn, and I’ve seen wheels an inch or so above the rail on a train crossing at track speed.
Could also be done with staggered joints and the same minimization of effort by just pushing the flatcar ahead 1/2 rail-length at a time (which was likely 30 ft. or so, not 39 ft., but that’s nit-picking). Actually, that might work better, because a spiking crew could alternate from one side to the other. With spiking down 2 rails simultaneously, the guys back-to-back in the middle might be interfering with each other - there’s not that much room.
Anyway, find and read the detailed step-by-step description of the 10+ miles of track was laid in one day on the first transcontinental railroad. See, for example:
http://cprr.org/Museum/Southern_Pacific_Bulletin/Ten_Mile_Day.html
I don’t recall that any mention was made of whether the joints were staggered or not - but those guys clearly had the logistics and “time-and-motion” analysis of the process to build that track with minimum yet coordinated effort down to a science. If the joint pattern made a difference, I think they would have recognized and exploited that advantage. Nevertheless, they seemed to accept the cost and burden of handling the rails a couple of times during the track-building process.
- Paul Nor
Paul,
Railroad history is about many things. This article is one of them and I think you for it. The writes shows the sheer human effort involved in laying track for the Union Pacific. It sounds to me like the joints tended to be parallel but to the extent the rails curves there may have been haphazard staggering of them. Here is what it says:
“A single horse pulled the car up to rail-head, where it was blocked by a wooden-framed iron track gauge. Four men worked on each side of the track. Two men seized the forward end of the rail with their tongs while the two rear men slipped the rail to the side of the car so it rested on iron rollers. The two forward men trotted ahead the length of the rail, thirty feet, the rear men dropping the rail in place, where it was bolted and spiked by the track gang. The car was then pulled forward to the next track gauge and the procedure repeated.”
John