Need a book? Engineering books recommendations...

Return to index: [Subject] [Thread] [Date] [Author]

Reinforcing existing steel members

[Subject Prev][Subject Next][Thread Prev][Thread Next]
Hey, I love a good argument.

Actually, we probably don't have much to argue about. 
Residual stresses don't invalidate elastic
superposition, they just render it imprecise.  You can
observe the same phenomena in residual-laden members
that you can theorize about in non-residual-laden
members;  you just get a lot of scatter in your data,
and rounded curves rather than cusps in your graphs of
said data.  For example, if you were able to test
steel columns with no residual stresses, I suspect
that the concave-down portion of the column curve
between squash yield load (flat portion) and Euler
buckling (concave up) would be a lot shorter and
sharper.  Similarly, coupon tension tests on annealed
samples have sharper transitions between the elastic
portion and the yield plateau than those of
un-annealed samples.

And, I don't think weld shrinkage will ever unload a
core column, at least significantly, especially with
skip welds.  In the usual case where the reinforcement
area is less than the core column area, the heat input
would expand the reinforcement more than the core,
resulting in (very small) net tension on the
reinforcement and compression on the core.

Neither of these points matters much.  What matters
most is your second argument--OK, discussion
point--which points out that Brown's theory has no
test data in the range that we care about.  You're
right.  We're just whistling in the wind.  Maybe
somebody wants to make a name as the next Reidar
Bjorhovde and actually do some tests for a Ph.D.

Mike Hemstad, P.E.
Saint Paul, Minnesota

Charlie Carter wrote:

Not meaning to be argumentative, of course ... the
Brown paper is a fine
paper with an excellent way of looking at the problem
if one cannot ignore
initial loading. But two things:

1. Brown assumes residual stresses can be ignored in
establishing his model.
There are always residual stresses and in reinforced
members there are two
kinds: those in the shapes and plates due to the
cooling during production
of the elements of the cross secton and those induced
by weld shrinkage
after the reinforcement is welded. The residual
stresses due to shrinkage of
the welds attaching the reinforcement may even be
enough to unload the
initial member. Rick Drake already explained very well
how residual stresses
affect column behavior. I believe any elastic
superposition of calculated
stresses due to loading would also have to account for
the effects of the
residual stresses to be correct. But then you are left
with two problems:
you do not really know what the magnitude of the
residual stresses are
(hence the reason Brown did not consider them) and
elastic superposition is
not useful once the inelasticity occurs that residual
stresses are likely to

Second, the only data point for comparison of the
Brown analysis to actual
test results is one that falls well below the values
of L/r for which the
reduction in strength he postulated applies. In the
author's words, his
analysis is not substantiated by testing.

So, I guess I was argumentative. (-: Sorry.


Do you Yahoo!?
Yahoo! Mail Plus - Powerful. Affordable. Sign up now.

******* ****** ******* ******** ******* ******* ******* ***
*   Read list FAQ at:
*   This email was sent to you via Structural Engineers 
*   Association of Southern California (SEAOSC) server. To 
*   subscribe (no fee) or UnSubscribe, please go to:
*   Questions to seaint-ad(--nospam--at) Remember, any email you 
*   send to the list is public domain and may be re-posted 
*   without your permission. Make sure you visit our web 
*   site at: 
******* ****** ****** ****** ******* ****** ****** ********