Need a book? Engineering books recommendations...

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

Re: Compression flange bracing

[Subject Prev][Subject Next][Thread Prev][Thread Next]

The concept of bracing the compression flange of a flexural member is intended
to provide external boundary conditions for the lateral torsional buckling limit
state (failure mode for ASD types).  The brace must be strong enough and stiff
enough prevent the entire beam cross-section from either twisting or laterally
displacing.  The question you need to ask is whether or not the roof plates are
heavy enough and stiff enough with enough normal force to prevent the
compression flange from twisting or laterally displacing.

I don't think that friction would be your first choice as a load path for the
required bracing stiffness and strength.  However, we can't ignore the economics
of the situation also.  If a tank roof rafter actually overcame friction and
laterally torsionally buckled, its flexural capacity would be suddenly reduced
and the roof system would sag.  However, if the rafter end connections are
strong enough, the rafters would now start to act like cable structure tension
members.  The situation would be stable, although with greater vertical
displacements.  Is this acceptable to your clients?

Another way to look at it is the famous list server line "Where are the bodies?"
Have any AWWA designed roof structures "failed" because of the current design
procedure?  If not, the curent design procedure is fit for purpose.

Rick Drake, SE
Fluor Daniel, Aliso Viejo


"Lutz,James" <JLUTZ(--nospam--at)> on 10/29/99 08:33:16 AM

Please respond to seaint(--nospam--at)

To:   seaint(--nospam--at)
cc:    (bcc: Rick Drake/AV/FD/FluorCorp)

Subject:  Compression flange bracing

The issue of compression flange bracing has come up in my practice on
numerous occasions with respect to rafters supporting the roofs of steel
water storage tanks. Typical practice is to not weld the rafters to the roof
plates, so compression flange bracing is only provided by friction of the
roof diaphragm. This is specifically permitted by AWWA D100, which allows
rafters to be designed as if continuously braced.

The argument I get into when reviewing the work of other designers is that I
think UBC (AISC Specifications) should overrule the AWWA procedure. The UBC
permits the use of "approved national standards" (e.g. AWWA) for the design
of ground storage tanks (UBC 1634.4), but I read this as applicable to
seismic design only. In all other aspects, the UBC should govern as is.

Anybody have any thoughts on the validity of counting friction for lateral
support? I can't imagine AWWA recommending this design approach without some
sort of basis.