Pratt vs. Howe boxcar trusses


Denny Anspach <danspach@...>
 

Although I seem to know that in bridge engineering a Pratt truss is inherently lighter than a Howe, and thus in practice, the Howe is relatively rare, how does this play out in boxcar design? Does the engineering of the door openings, or the weight and strength of the underframe have something to do with it?

Denny
--
Denny S. Anspach, MD
Okoboji, Iowa


Andy Carlson
 

I think that the Howe truss designed car was familiar with the RR's back when cars were mostly of all wood construction. The engineering of wood trusses takes advantage of the greater strength of compression for wood- hence the Howe truss. Later, when steel underframes and SS cars became popular, basic familiarity with Howe trusses caused a lot of steel framed boxcars to be built with the Howe truss with it's structural members in compression. As early engineering students quickly learn, steel is much stronger in tension, and bridge trusses take advantage of this property to keep the amount of steel used to the minimum. The enlightened RR engineering departments recognized this when the Pratt truss was selected with it's structural members taking advantage of steel's tensile strength. Not all RR engineering departments were so enlightened- hence the large numbers of Howe steel trussed cars being made.
-Andy Carlson
Ojai CA

Denny Anspach <danspach@...> wrote: Although I seem to know that in bridge engineering a Pratt truss is
inherently lighter than a Howe, and thus in practice, the Howe is
relatively rare, how does this play out in boxcar design? Does the
engineering of the door openings, or the weight and strength of the
underframe have something to do with it?

Denny
--
Denny S. Anspach, MD
Okoboji, Iowa


Anthony Thompson <thompson@...>
 

Andy Carlson wrote:
I think that the Howe truss designed car was familiar with the RR's back when cars were mostly of all wood construction. The engineering of wood trusses takes advantage of the greater strength of compression for wood- hence the Howe truss. Later, when steel underframes and SS cars became popular, basic familiarity with Howe trusses caused a lot of steel framed boxcars to be built with the Howe truss with it's structural members in compression. As early engineering students quickly learn, steel is much stronger in tension, and bridge trusses take advantage of this property to keep the amount of steel used to the minimum. The enlightened RR engineering departments recognized this when the Pratt truss was selected with it's structural members taking advantage of steel's tensile strength. Not all RR engineering departments were so enlightened- hence the large numbers of Howe steel trussed cars being made.
This is a good summary of the bridge perspective, Andy, but it misses two points. First, steel trusses were essentially universal in bridge building after about 1880, and all their superior features were well recognized, even by car designers. To see this, read the MCB proceedings in the first two decades of the 20th century. Yet Howe truss car framing was still being built. Your idea that somehow the bridge guys were "enlightened" and the poor mechanical engineers designing cars were decades behind, just isn't true.
Second, the car side truss has to do a job not required of the bridge truss (or at least not in the same way): it has to resist sideways and twisting forces. The car-design advocates of the Howe for car sides believed it was superior in that aspect--though their colleagues of the opposite persuasion argued vehemently against them. You can see this in the ARA Mechanical Division proceedings all through the 1920s.
Denny's belief that the Howe is lighter is wrong: in either steel or wood it is heavier, though of course, as Andy observed, it has to be that way in wood to provide the compression strength in the long members. But for bridges, the modest difference in weight is of no real consequence.

Anthony Thompson
Dept. of Materials Science & Engineering
University of California, Berkeley
thompsonmarytony@...


Anthony Thompson <thompson@...>
 

I wrote:
Denny's belief that the Howe is lighter is wrong . . .
My apologies for having misread Denny's post. He said the Pratt is lighter, which is correct.

Tony Thompson Editor, Signature Press, Berkeley, CA
2906 Forest Ave., Berkeley, CA 94705 www.signaturepress.com
(510) 540-6538; fax, (510) 540-1937; e-mail, thompson@...
Publishers of books on railroad history


Miller, Andrew S. <asmiller@...>
 

The issue, as already noted, was whether the longest members, the
diagonals, would be in compression or tension. In wood construction,
joints in tension are inherently weak, so putting the diagonals in
compression was preferred. Furthermore, wood required a much larger
cross section for any given design load, so a wood member would be less
prone to buckling than a steel or iron member designed for the same
load. That's why in bridge construction, steel and iron bridges were
Pratt trusses, and wooden bridges were Howe. However, if the primary
concern with a Howe truss diagonal is buckling (in compression,
buckling occurs far before outright material failure), then it should
be much less of a problem in a car side than in a bridge, because the
car side has all the sheathing attached to the ribs, including the
diagonals, and this is a substantial deterrent to buckling, at least in
the plane of the car side.


regards,

Andy Miller

-----Original Message-----
From: STMFC@... [mailto:STMFC@...] On Behalf Of
Anthony Thompson
Sent: Monday, August 20, 2007 8:06 PM
To: STMFC@...
Subject: Re: [STMFC] Pratt vs. Howe boxcar trusses

Andy Carlson wrote:
I think that the Howe truss designed car was familiar with the RR's
back when cars were mostly of all wood construction. The engineering
of wood trusses takes advantage of the greater strength of
compression
for wood- hence the Howe truss. Later, when steel underframes and SS
cars became popular, basic familiarity with Howe trusses caused a lot
of steel framed boxcars to be built with the Howe truss with it's
structural members in compression. As early engineering students
quickly learn, steel is much stronger in tension, and bridge trusses
take advantage of this property to keep the amount of steel used to
the minimum. The enlightened RR engineering departments recognized
this when the Pratt truss was selected with it's structural members
taking advantage of steel's tensile strength. Not all RR engineering
departments were so enlightened- hence the large numbers of Howe
steel
trussed cars being made.
This is a good summary of the bridge perspective, Andy, but it

misses two points. First, steel trusses were essentially universal in
bridge building after about 1880, and all their superior features were
well recognized, even by car designers. To see this, read the MCB
proceedings in the first two decades of the 20th century. Yet Howe
truss car framing was still being built. Your idea that somehow the
bridge guys were "enlightened" and the poor mechanical engineers
designing cars were decades behind, just isn't true.
Second, the car side truss has to do a job not required of
the
bridge truss (or at least not in the same way): it has to resist
sideways and twisting forces. The car-design advocates of the Howe for
car sides believed it was superior in that aspect--though their
colleagues of the opposite persuasion argued vehemently against them.
You can see this in the ARA Mechanical Division proceedings all through

the 1920s.
Denny's belief that the Howe is lighter is wrong: in either
steel or wood it is heavier, though of course, as Andy observed, it has

to be that way in wood to provide the compression strength in the long
members. But for bridges, the modest difference in weight is of no real

consequence.

Anthony Thompson
Dept. of Materials Science & Engineering
University of California, Berkeley
thompsonmarytony@...




Yahoo! Groups Links


Anthony Thompson <thompson@...>
 

Andy Miller wrote:
. . . if the primary concern with a Howe truss diagonal is buckling (in compression, buckling occurs far before outright material failure), then it should be much less of a problem in a car side than in a bridge, because the car side has all the sheathing attached to the ribs, including the diagonals, and this is a substantial deterrent to buckling, at least in the plane of the car side.
This is exactly right. But the early truss-frame, single-sheath cars were explicitly designed WITHOUT letting the truss sides do any work, and the underframe was beefy enough to carry all the load. This is partly true even as late as the USRA single-sheath cars, whose underframe is substantially overdesigned if the car side trusses were credited as carrying much load.
This VERY conservative approach was decried by those who wished to greatly reduce car weight, in part by reducing underframe sections, and by the first ARA underframes of the early 1920s, that's exactly what happened. Obviously any later car DOES credit the side truss with carrying capacity.
The stouter posts in a Pratt truss help resist outward bulging, and accordingly were advocated as superior to Howe trusses for car sides. But the converse is that Pratt braces (diagonals) are relatively lighter, and those play some role in resisting torsion of the car body. I think if you read the car design literature (which was very active in the 1920s), you will see not only that car designers understood bridges perfectly well, but also that they had concerns about car frame performance which goes beyond bridge issues.
This entire matter is NOT simply understood in terms of bridges, nor of "wood vs. steel" ideas, nor of "conservatism" of railroad car designers (apparently relative to bridge designers, in some accounts).

Tony Thompson Editor, Signature Press, Berkeley, CA
2906 Forest Ave., Berkeley, CA 94705 www.signaturepress.com
(510) 540-6538; fax, (510) 540-1937; e-mail, thompson@...
Publishers of books on railroad history