My experience with these bridges is similar to what you describe. Also I
have found that the wind loading governs over seismic (even here in
California) due to the lightness of these old trusses and the application of
wind loads specified by the code.
The AASHTO LRFD manual (Section 6.7.6) has an interaction equation for pins
which may be what you are looking for. The commentary references this:
Kulicki, J.M. "Load Factor Design of Truss Bridges with Applications to the
Greater New Orleans Bridge No. 2" National Transportation Research Board,
TRB 903, 1983. I suspect the Canadian Code (which has lead the way
re:bridges and LRFD) has the same equation in it. Perhaps this reference
has some useful info.
There has been alot of research on NDT of pins, due to the spectacular
failure of some girder bridges (in the 80's). I am surprised pin failures
on old truss bridges are not more common, due to the findings on load raters
that the pins are usually overstressed under live load.
My humble suggestion: Analyze it the good old-fashioned way you have and if
it doesn't work the owner has to bite the bullet and replace with, say 100
ksi pins, bypass the pins with yokes, or change the bridge posting. Failure
of just one pin brings the whole thing down, so I wouldn't rely on anybody's
From: Steven Mallett [mailto:smallett(--nospam--at)dillon.ca]
Sent: February 07, 2001 7:31 AM
Subject: Connection Pins in Old Steel Truss Bridges
I recently worked on a detailed structural evaluation of an existing steel
through truss bridge built in 1904 with a span of 160 feet. While all of the
members were determined to be sufficient to carry the specified design
truck load, the connection pins were found to be over stressed in
bending. These pins connect the bottom chord, vertical, and diagonal
members. A U-bolt hanger supports the deck floor beam from the center
of the pin. The steel pins are 3 inch diameter and approximately 14
inches long with yield strength of 30 ksi.
In the bending analysis, I assumed applied point loads by the members
(maximum member thickness is 1 inch). Using distributed loads did little to
reduce the bending moment. The pin bending capacity is as low as 40%
of the applied bending moment (i.e. Live Load Rating Factor of 0.4, as per
Canadian CSA S6 bridge code, Clause 12 evaluation procedures). Shear
capacity is no problem.
Now that the lengthy intro is complete...to the meat of the message.
Obviously, these pins have worked well for almost 100 years under the
design loads. This bridge does not have posted loads and is exposed to
normal truck traffic. Can someone suggest other analysis techniques or
past research on the subject of bridge pins in bending that would explain
this phenomena? i.e. acts like a short deep beam, etc.
Steve Mallett, P.Eng.
Dillon Consulting Ltd.
Halifax, Nova Scotia