Need a book? Engineering books recommendations...

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

Fwd: Re: Pushing Steel Design with Mill-Certs

[Subject Prev][Subject Next][Thread Prev][Thread Next]
Recently Harry wrote:

>Does anyone have any specific information (or opinions) regarding the
>design of steel moment frames by using actual fy yield values based on mill
>certs rather than using code minimum values?

Yes, I have an opinion: It's wrong, wrong, wrong!

The Steel Design Codes say we should design to the yield stress of the
material specification, not to the tested yield strenth of a sample. You're
reducing the intended safety factor if you design right to the limit.

I believe the reason mill test certificates ("mill-certs") are not normally
used for setting allowable stresses is that they are only SAMPLE tests. How
do you know that the most highly stressed part of your bending member
happens to be  as strong as the tested part of the batch?

After a well known structural collapse of a roof structure here in
Vancouver, the investigator learned that a reviewing engineer had been
talked into accepting a beam which, in his initial review he wanted
reinforced. The argument for acceptance was that the mil-certs showed a
higher yield stress than the nominal yield stress in the material spec. The
beam would have been overstressed 20% based on the normal limit of 44 kips
per square inch, but the mil-cert said the yield had reached 55 ksi iin the
test. The engineer involved rationalized that the beam was strong enough,
and told the client the "good news" - the beams didn't need to be reinforced
after all! 

After the collapse, samples were taken fron the failed beam and tested. Sure
enough, the sample from the web of the wide flanged beam tested out at 55
ksi. That was where the sample in the original mill test had come from. The
sample from the flanges, however, proved to have a yield stress of only 47
ksi. Apparently steel in webs is likely to be be strain- hardened by the
rolling process, more than the flanges, so the web will often proove to have
a higher yield than the steel in the flanges.

If you're designing frames for a metal building manufacturer, be careful.
That seems to be a highly competitive market where saving a few pounds here
and there can win a job... but how much can you shave before there's no
reserve? Maybe you haven't checked every condition, and maybe the snow
builds up differently from what you assumed.

There have been reports of many failures of pre-engineered metal buildings
when the big snowstorm hit Victoria, British Columbia on December 30th. The
snow load did exceed the code value, but other building types seemed to be
more tolerant. And the effect of the collapse on the businesses which
occupied the buildings was, of course, catastrophic.

Who are you really helping by talking yourself into more optimistic
assumptions?
Maybe not the building owner, and certainly not the occupants. A safety
factor that lets you sleep at night probably costs under 1 percent more, for
the structure, than if you used the bare safety factor that - you hope -
just meets minimum requirements. It may be good for our ego to brag that we
are smart enough to design a frame that weighs less than some other
engineer's design. But what if some of our assumptions were wrong, and it
fails?

                                                          Jim Warne,
Vancouver




--- Internet Message Header Follows ---
Received: from server1.seaoc.org (bqe.com [204.140.166.34])
	by host1.texramp.net (8.8.5/8.8.5) with SMTP id XAA29078
	for <rlewis(--nospam--at)techteam.org>; Sat, 1 Mar 1997 23:53:31 -0600 (CST)
Received: from aphex.direct.ca by server1.seaoc.org (NTList 3.02.10) id
la005133; Sat, 1 Mar 1997 21:52:01 -0800
Received: from van-52-1119.direct.ca (van-52-1119.direct.ca [204.174.250.19])
by aphex.direct.ca (8.8.3/8.8.0) with SMTP id VAA22320 for <seaoc(--nospam--at)seaoc.org>;
Sat, 1 Mar 1997 21:49:24 -0800 (PST)
Date: Sat, 1 Mar 1997 21:49:24 -0800 (PST)
Message-Id: <199703020549.VAA22320(--nospam--at)aphex.direct.ca>
X-Sender: seaoc-ad(--nospam--at)seaoc.org
X-Mailer: Windows Eudora Light Version 1.5.2
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
To: seaoc(--nospam--at)seaoc.org
From: Jim Warne <jwarne(--nospam--at)direct.ca>
Subject: Re: Pushing Steel Design with Mill-Certs
Reply-To: seaoc(--nospam--at)seaoc.org
Error-To: seaoc-ad(--nospam--at)seaoc.org
X-Loop: seaoc(--nospam--at)seaoc.org
X-Info: [SEAOC]
Owner: seaoc-ad(--nospam--at)seaoc.org
X-POP3-Rcpt: seaoc-ad(--nospam--at)seaoc.org
Precedence: list
X-ListMember: rlewis(--nospam--at)techteam.org [seaoc(--nospam--at)seaoc.org]


__________________________________________________

Richard Lewis, P.E.
Missionary TECH Team
rlewis(--nospam--at)techteam.org

The service mission like-minded Christian organizations
may turn to for technical assistance and know-how.