Need a book? Engineering books recommendations...

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

RE: Unbraced length question

[Subject Prev][Subject Next][Thread Prev][Thread Next]
Hi Mike,
The radius GLB is in-plane to the lateral load. The diaphragm is
continuously connected to the GLB and the transfer to the columns in not
eccentric. I can not use a straight drag strut since the projection of the
roof where the shear is transferred relies upon the radius inside and out
for architectural appearance.
I think that your suggestion to ignore the 1/2" displacement is best. At
worst, I can increase the column thickness to increase column stiffness and
create a greater margin for deflection and column failure potential.
The GLB is, unfortunately, not oversized for appearance. It is only 13" deep
(2 spans over 64 feet total)and intended to be hidden in the roof framing to
create a curved, thin, line from the outside view and a smooth plaster
ceiling from within. Timberweld in Montana has designed the GLBs to my
To make matters worse - from the exterior of the building the roof assumes
the appearance of an "S" on it's side. The dropped radius is hidden behind a
parapet for all but the portion extending from the face of the structure and
resting on the columns that diaphragm shear is transferred to. The drainage
is handled behind parapet walls with expanded polyurethane foam built up to
handle proper drainage.

Mike, I thank you for all your help. Your advise has helped me to make a few
changes and has justified the much of what I have done. I am confident that
the structure will perform well despite the unusual design. If you are
interested, I can send you a roof framing plan to see what I had to contend
with and an architectural elevation to see what the architect wanted to
The project is under construction in a development called Mirada - an area
developed with the Ritz Carlton. It's one of our more creative pieces and I
am constantly questioning my design as I watch it being erected. This is
also a project that I used the Hardy Frames on. Fortunately the owner loves
the technology and we are watching this one go up very closely.

Thanks again for your support and help - as usual I really appreciate it.


-----Original Message-----
From: Mlcse(--nospam--at) [mailto:Mlcse(--nospam--at)]
Sent: Friday, November 13, 1998 7:06 PM
To: seaint(--nospam--at)
Subject: Re: Unbraced length question

In a message dated 11/12/98 11:21:50 PM EST, wish(--nospam--at) writes:

<< Mike,
 The columns are connected to beams radiating out of the building and
 on the columns. The beams are only tied by a radiused GLB which bear above.
 The links are mechanical - ie, welded column caps that bolt to the beam,
 radius glb and weld to the columns.

 Considering the radius in the GLB, there is not direct tie and I would
 expect deflection to occur within the GLB.
 The only way to adjust stiffness to insure different story drift would be
 change column thicknesses and force each to take the same load. Therefore,
 the longest column will control. The difference in deflection using the
 members is expected to be about 1/2" although some of it will be absorbed
 the GLB.
 Do you think this is a problem worth worrying about? or should I just
 for the taller unspupported member and keep it as stiff as possible (under
 code drift criteria of 0.005h)?

 Thanks for the response, Mike

I would probably use the same size for both columns and not worry about the
1/2" deflection (probably less expensive to buy more of the same size
and less expensive for fabrication if the connection dimensions are all the
same) assuming the glu-lam beam can handle the out-of-plane bending.  If we
get an earthquake large enough to cause displacements greater than 0.005H,
there will probably be just as much nonstructural damage if you have used
different sized columns.

How much of a bending moment is induced in the glu-lam if you get 1/2"
deflection in the out-of-plane direction.  I assume the glu-lam is probably
lightly loaded for gravity load, and is sized more for architectural
appearance ( I picture the radius glu-lam beams extending out beyond the
of the building or to the exterior wall of the building).  If the glu-lam is
oversized for gravity, I imagine it can handle the out-of-plane bending.

Does the diaphragm extend over the entire length of the glu-lam beam or only
portion of the glu-lam (portion of glu-lam is a flying beam to the flagpole
column).  If a portion is a flying beam,  try and provide at the edge of
sheathing over the glu-lam a drag struct perpendicular to the glu-lam to
distribute the cantilever reaction back into the diaphragm.