Shear in Flitch Plate Beams

[greenfinger]

I have designed a flitch plate beam to be used in a wood framed house, but am baffled by a phrase in an article, which I found trough this forum.

THe clear span is 13'1", loaded at 1470#/lf for a total of 19,228#. THe beam is (3)2x12 with (2)1/2" x 11" steel plates, held together by (2) 5/8" bolts 16" o.c. plus 4 bolts at each end. The ends will be supported by 4" of wood posts the full width of the beam.

THe article "Flitch Plate Beams Design Guide" from STRUCTURE, includes a similar beam in a table of beam section properties. The phrase which worries me is "The shear capacity indicated in table 1 is the shear strength Of the wood alone since the full beam reaction must be transferred through the wood side pieces at end bearing supports".

This qualification would make my beam fail. As I understand it, shear in the wood will be shared with the steel through the bolts, and so not be a problem. The bit of wood which extends below the depth of the steel plates would be subject to compression perpendicular to the grain, which it passes, so this beam is OK for the application.

Am I missing something?

 

[msquared48]

Remember that if the beam has no checks, splits or wanes you can double the allowable shear.

I would also consider the d/2 allowance for shear reduction, as with concrete.

You can also install a steel bearing plate between the column and beam of sufficient length to negate the shear problem.

Mike McCann
MMC Engineering
Motto: KISS
Motivation: Don't ask

 

[Lion06]

The way I see it if you are dumping all the shear from the steel bak into the wood NOT directly over the support then you have to look at all of the shear being taken in the wood. I don't think you can use the shear at a distance d from support if you have a concentrated load very close to the end.

Mike, how do you get to double the shear capacity if there are no checks, splits, or wanes? Isn't that accounted for in the grading?

 

[msquared48]

Yes. In the IBC, there is a note in the table for stresses in wood beamns that allows for doubling the allowable shear stress for better materials of the same grade. I will look it up for you...


Mike McCann
MMC Engineering
Motto: KISS
Motivation: Don't ask

 

[msquared48]

The IBC defers to the NDS so you need to check there. I do not have a current copy. Check there.

However, in the old 1994 UBC, Note #5 to Table 23-I-A-4, there is a factor listed for "No Splits" and "No Shakes" for the shear stress factor Ch. This is what I am talking about.

I have specified "no splits or shakes" on the drawings for years for beams in a shear-critical condition to avoid having to oversize the beam in bending.

Mike McCann
MMC Engineering
Motto: KISS
Motivation: Don't ask

 

[BAretired]

I don't understand the phrase "The shear capacity indicated in table 1 is the shear strength Of the wood alone since the full beam reaction must be transferred through the wood side pieces at end bearing supports".

The flitch plates share the load with the wood members in accordance with their stiffness. If the flitch plates resist moment, they must also carry shear. The end reactions of the flitch plates must be transferred to the wood members by fasteners and the wood members must deliver the reaction to the support. This does not stress the wood members in shear if the end fasteners occur over the support. It stresses them in compression perpendicular to the grain.

I believe the quoted article is wrong and I do not think you are missing anything. Just make sure that you account for how the end reactions of the steel plates get into the support.

BA

 

[hokie66]

It would be preferable, and I think simpler, to bear the steel plates directly on the end supports. You could just notch the 2x12's and use a steel bearing plate.

 

[msquared48]

And why does it speak of "wood side pieces"?

I thought these were steel flitch plates...

Oh, and from your calcs, if the Fv on this is 95 psi or greater, and if the joists have no splits or shakes, you are on the line of not needing any flitch plates.



Mike McCann
MMC Engineering
Motto: KISS
Motivation: Don't ask

 

[msquared48]

... at least considering shear...

Mike McCann
MMC Engineering
Motto: KISS
Motivation: Don't ask

 

[BAretired]

hokie is right. If steel bearing plates are used, flitch plate reactions can be transferred directly to the support. Then, there is no need to transfer the load to the wood side members.

BA

 

[msquared48]

I must be missing something here.

For wood side plates then it seems the configuration is wswsw and not s

http://wwws.

Correct?

If it is wswsw, then this must be a new beam and not a retrofit beam.

Mike McCann
MMC Engineering
Motto: KISS
Motivation: Don't ask

 

[hokie66]

I didn't read anything suggesting it was a retrofit. As far as I know, flitch plate beams always have the wood on the outside.

 

[Lion06]

Hokie is right, if you make bottom of plate equal bottom of wood and use a steel bearing plate then the whole issue goes away. If you do that you don't even need the (4) bolts concentrated at the end. The whole point of the extra bolts at the ends is to take all the load that was tranferred into the steel over the length and dump it back into the wood. If you can bear the steel directly on the support then there is no need to transfer that load back into the wood.

 

[woodman88]

Doubling the the shear parallel to grain is no longer allowed. The Ch factor has been removed as of the NDS 2001 supplement due to the almost doubling of the shear values. Take Douglas fir-Larch the Fv pre 2001 was 95psi after 2001 it is 180psi.
Also, I am concern about your required bearing. reaction/((compression perpendicular to grain)x(beam width))= bearing length gives for DF-L 19228/(625x4.5)= 6.84" required bearing length for the beam. What species of 2x are you using?

Garth Dreger PE
AZ Phoenix area

 

[greenfinger]

Wow, you people are fast! Thanks so much for the feedback.

The construction is new. The beam design is WSWSW. THe species under consideration is hemlock #2. If the shear is carried by the wood only, as the article indicated, it ends up being 285 PSI, so even doubling the 85 PSI I was using won't save me.

I was thinking along the lines of what BA retired wrote. It seems like my options are to increase the bearing at the ends so I can put enough bolts there to carry the load from steel to wood - which will also eat away at the amount of wood and so weaken it - or to put in a steel bearing plate in direct contact with the flitch plates, so the wood only has to carry it's own proportion of shear. I had speculated about that before, and seems the easiest thing to do.

THanks again!

 

[msquared48]

Thanks too here. I stand corrected.

Mike McCann
MMC Engineering
Motto: KISS
Motivation: Don't ask