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

RE: Wall thickness of HSS sections

[Subject Prev][Subject Next][Thread Prev][Thread Next]
Code answer:
The width-to-thickness ratio for HSS should be taken as flat width to
wall thickness (AISC Seismic, Section 13.2).  Therefore, even if
0.93xthickness is used, HSS10x10x5/8 still satisfies this criterion
(for Fy=46 ksi).

Rational answer:
Dependence on sqrt(Fy) is based on mathematical formulation and,
perhaps, general trends (not an extensive series of tests for the same
section with different actual yield strengths).  The limiting
width-to-thickness ratio is based on pretty sparse data [one of the
test programs reports b/t = 14, 25, and 30 which for the material
tested were about 120, 197, and 228/sqrt(Fy)] and there is not a sharp
transition in the observed behavior (assuming you could see that
transition with three data points).  The section b/t reported in the
research is often based on the nominal thickness (not actual, measured

Based on the limitations of the research behind the prescriptive code
limits, I feel comfortable basing the checks on the nominal properties
and dimensions.  In other words, for the purposes of prescriptive
section qualification, I don't get excited if the actual Fy is a
little high (which, by calculation, would imply a lower b/t limit) or
if the actual thickness is a little low (therefore, by calculation,
b/t is a little high).  Instead, I am more irritated by the reduction
of overall section properties forced on us by unethical,
unprofessional, legalistic, cheap, low-life steel producers; I am
personally and professionally disgusted by the level of chiseling
encountered in our system.  I believe that the correct solution is to
fix the material specs so that we get what we pay for, instead of
moving to a system of "nominal" dimensions.  Of course, what is right
doesn't usually happen.

-Mike Valley

-----Original Message-----
From: Alexander Sasha Itsekson [mailto:sitsekson(--nospam--at)]
Sent: Friday, June 23, 2000 11:01 AM
To: List server
Subject: Wall thickness of HSS sections


For some time now we were aware of the fact that the mills were
rolling out
tube sections on the lower end of the wall thickness tolerances.  The
black book (hollow structural steel sections) indicates the actual
nesses for the TS sections.  It appears that the actual thickness is
93% of
the nominal thickness for tubes.

No, here's my question: how many engineers revised their braced frame
aids to account for the local buckling requirements that limits the
thickness to width*sqrt(Fy)/110. Are you aware that we can't use the
good HSS10x10 in the braced frames at all (it requires the minimum
thickness of .66")?

Any comments?


Alexander Sasha Itsekson
(510) 527-7223 ext. 209