Need a book? Engineering books recommendations...

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

Re: WOOD - Engineered Lumber vs. Sawn Lumber

[Subject Prev][Subject Next][Thread Prev][Thread Next]
I guess a lot of folks think that dead load deflections are unacceptable
for GLBs while they are acceptable for sawn sections.

Compared to sawn sections (which are allowed to deflect under dead load),
wouldn't complying with UBC Table 16-D (maximum allowable deflection for
structural members) suffice if one took into account this 0.5 *DL
Deflection for creep of the member?

Keep in mind that, until this creep occurs, the floor will have a "bow" in
it. If there is a framed wall above this member, it is very difficult to
frame this wall. I know. I've gotten the telephone calls. So, for the first
ten years, the deflection is "negative". Is that better than being zero at
day one and negative in ten years? It's your choice.

I have designed beams based on a stiffness requirement of UBC Table 16-D so
that the moment of inertia is sufficient to make the following true:

(LL+K*DL)Deflections<L/240 where K=0.5 for seasoned lumber such as GLBs and
=1.0 for sawn sections.

I haven't gotten a call from sagging floors yet even though I've been doing
this >10 years. However, I have gotten calls from framers who have trouble
framing on a beam with a crown in it.

With regards to roof members, I am compelled to comply with the Code
requiring the 1.5*DL Defln even though I would rather slope the roof 3/8"
per foot and specify "standard" cambers. Standard cambers are equal to a
radius of 2000 ft. Most GLB suppliers will provide a chart of camber vs.
span based on standard camber. The relevance of specifying a standard
camber is lead time.

Most GLB sizes specified with a standard camber can be delivered to the job
site within two days. Cambers in excess of standard require a lead time of
4 to 8 weeks. This issue is the same when specifying 24F-V8 grades as well.
If the cantilevered stresses are within the allowable tensile stresses in
the compression zone (usually about 1/2 the allowable tension stress in the
tension zone), 24F-V4 could be specified thereby saving the lead time.

Bill Allen

> From: Dennis S. Wish PE <wish(--nospam--at)>
> To: 'seaoc(--nospam--at)'
> Subject: RE: WOOD - Engineered Lumber vs. Sawn Lumber
> Date: Thursday, September 11, 1997 1:03 PM
> On Wednesday, September 10, 1997 7:25 PM, Bill Allen, S.E. @ 
> ALLEN DESIGNS [SMTP:BAllenSE(--nospam--at)] wrote:
> | I'm not sure how to explain this. If you camber the beam for
> | 1.0*DL
> | deflection then, theoretically, when the dead loads are 
> applied
> | the beam
> | will be level. If you camber the beam for 1.5*DL deflection,
> | after the dead
> | loads are applied the beam will still have a negative
> | deflection (camber)
> | or "crown" equal to 0.5*DL deflection. Right?
> |
> | Regards,
> | Bill Allen
> |
> | ----------
> | > From: Stan Johnson <hawneng(--nospam--at)>
> | > To: seaoc(--nospam--at)
> | > Subject: Re: WOOD - Engineered Lumber vs. Sawn Lumber
> | > Date: Wednesday, September 10, 1997 5:50 PM
> | >
> | > Bill Allen, S.E. wrote:
> | >
> | > > I believe one should be careful about specifying
> | > > a camber=1.5*DL
> | > > deflection. While this is appropriate (and
> | > > required by Code) for roof
> | > > members, it will leave a crown (theoretically
> | > > equal to 0.5 * DL deflection)
> |
> |
> | > > on floor members.
> | >
> | > I don't understand why this would be the case?
> | >
> | >
> | > Stan Johnson
> | > BS, EIT, Are we talking gold crown or porcelain
> Bill,
> According to the AITC 3rd Edition, Table 5.9, the recommended 
> Minimum Camber for GLB's is 1.5 times Dead Load Deflection. 
> There are two notes attached:
> "a) Roof Beams, the minimum camber of 1.5 times dead-load 
> deflection will produce a nearly level member under dead load 
> alone after plastic deformation has occurred. Additional camber 
> is usually provided to improve appearance and/or provide 
> necessary roof drainage. Roof beams should have a positive slope 
> or camber equivalent to 1/4" per foot of horizontal distance 
> between the level of the drain and the high point of the roof, 
> in addition to the minimum camber, to avoid the ponding of 
> water. In addition, on long spans, level roof beams may not be 
> desirable because of the optical illusion that the ceiling sags. 
> This condition may also apply to floor beams in multi-story 
> buildings.
> b) Floor beams - the minimum camber of 1.5 times dead-load 
> deflection will produce a nearly level member under dead load 
> alone after plastic deformation has occurred. For warehouse or 
> similar floors where live load may remain for long periods, 
> additional camber should be provided to give a level floor under 
> the permanently applied load."
> I believe the 0.5 time Dead Load that you mentioned has to do 
> with creep. Also according to the AITC page 4-90:
> "b. The inelastic deflection, creep or permanent set of glued 
> laminated timber averages approximately one-half of the 
> calculated dead-load deflection. This deflection can be offset 
> by camber, as previously stated. The permanent set of 
> un-seasoned sawn lumber often used for secondary framing is 
> approximately equal to the calculated dead-load deflection. 
> Therefore, the total long-time deflection is approximately 2 
> times the calculated initial elastic deformation. The additional 
> ponding load resulting from the permanent set of all members 
> within a particular framing system should be considered in the 
> design analysis."
> Hope this clears it up a bit.
> Dennis Wish PE