Steel/Wood Combined Beam 8

[DYork]

I have to design a beam that must span what a 2-2x12 beam could handle. How do I calculate what a 2-2x6 beam w/ a 1/4" steel plate sandwhiched in the middle can support. Thanks for the help in advance.

Devon

 

[JRGse]

Ratio the flitch plate EI to the beam EI and obtain the new section properties. Reference WOOD: ENGINEERING DESIGN CONCEPTS published by Materiels Education Council for further info.

 

[LPPE]

Simple 10 second answer - See if the plate by itself can handle the load. Then the wood is just along for the ride.
Otherwise, transformed properties.....

 

[SlideRuleEra]

Be sure to check the wood for shear at the supports. While the steel plate will resist most of the moment, it is typically of little value for shear (the steel's contact bearing area, if any, is very small).
See the following thread for a similar discussion: thread507-85244

 

[ChipB]

Check deflection too. That's pretty shallow even for steel.

As long as you have enough bolts over your support to transfer the steel reaction to the wood, you could use the wood bearing area without having to take the full shear of the combined section in the wood.

 

[CSEllc]

I would not recommend using anything less than 2x8 for flitch beams. With 2x6's , realistically, the bolts will be in line. You won't be able to propperly transfer the forces.

If you need to use a 2x6 flitch, then I would not count on the wood and design the plate to resist the full load.

 

[TFL]

i agree if you need a 6" depth the bolting will be difficult to develop bolting. as stated ealier i usually design the steel to take the whole load. the wood is just bracing.

be careful with how you design the bolts. you have use the bolts in the center of the span to transfer the force into the plate from the wood. don't forget the plate is usually not flush top with the bottom of the joist and the steel does not supply adequate bearing area for the joists. you then have to transfer the shear back into the wood at the support. i usually try to transfer the shear back in within d of the support. this allows you to transmit a greater shear into the supprt then the beam's capacity. as a side note i understand the 2002(?) NDS has essentially eliminated the exemption for point loads within d of the support. however it seams to be a logical way to do things that has worked for years

 

[PT999]

Check out "Better Header", a manufacturer of flitch plate headers. Their catalog sizes their products for span and live load

 

[boo1]

Typically the steel is undersided so first instance the load is taken by the timbers, the calculated proportion then being transferred from the timbers to the plate by the bolts over the main span, and then transferred back from the plate to the timbers at the bearings. The two guides below agree with 8" min size.

You MUST ensure that any flitch beam is provided with adequate bearing, fastening methods, and lateral restraint.

See:

http://www.destefanoassociates.com/tech/flitch.pdf

http://www.toolbase.org/docs/MainNav/WoodFrameConstruction/2947_flitchplate.pdf

 

[jike]

Here is another idea...for what it is worth. How about taking a 1 3/4" x 11 7/8" microlam and cut it in half? I think that (2) 1 3/4" x 5 7/8" microlams (with higher strength) might be close in strength to what (2) 2 x 12's gave you. You will have to check the resulting strength and deflections.

 

[DYork]

jike - microlams would not work - by cutting in half you are losing more than half of the strength of the microlam. Thanks

 

[jike]

DYork -
My mistake. You are correct. You would need 4 of the mini-microlams 1 3/4 x 5 7/8 to equate to the (2) 2 x 12's. It still might be an easier option if you can live with the width.

 

[boo1]

Equivalent section charactistics are generally designed for bending stress and then checked for shear and deflection. Cutting does not decrease strength (ignoring size factors) it decreases section modulus.

Using allowable (permissible) stress design, the required elastic section modulus, Z, is found by dividing the maximum bending moment by the adjusted allowable bending stress, Fb. As we all should know Z is a function of height squared and width. You can not use an area fuction, but must use Fb * Z equivalancy.

Live load deflection = 5/384 [wL / (EI)] gives us the functional inverse realtionship E * I

 

[LPPE]

A quick aside - I think Elastic section modulus is typically noted as "S", while Plastic section modulus is noted as "Z".

 

[boo1]

yes I agree
S=section modulus
must have had a senior moment...

 

[RASmithPE]

If width is not a problem, You may want to consider (2) 5" or 6" Channels back to back. The webs are easier to drill and the flanges greatly increase the stiffness compared to the same weight plate.

How about a W6 steel beam with wood nailers and hangers? Leave a 1/4" space above the top flange to prevent a lump in the floor when the joists shrink.

Check the deflection. I recommend span/480

 

[DYork]

Bad choice of words - did not mean the strength charecteristics of the microlam.

 

[boo1]

should have been
Live load deflection = 5/384*[w(L**4) / (EI)]

 

[KarlT]

Another thing you could look at is welding a 3" wide plate along the bottom of the vertical web plate to make a "T" shape. Then the steel member becomes much stiffer and the steel can bear directly on the supports.

 

[ChipB]

Just a thought.....

Can you provide additional support elsewhere? i.e. breaking up the spans of the joists framing into this beam with intermediate supports? Then you may be able to reduce the req'd size.

If there happens to be a wall above the beam, in the past, I've cut into the wall 1/2" depth and used a single microlam with simpson tension straps wrapping the beam below to carry the loads.

Chip