Theoretical Question Related to Flitch Plates

[Archie264]

I understand that Flitch plate beams are usually constructed with a double row of bolts but what if they weren't? What if it were only a single row in the center? Further, what if the steel portion were only in, say, the middle third of the beam? Or equivalently, I suppose, a steel channel reinforcement on the middle third of the beam, again, bolted with only one row of bolts?

It seems to me that compatible deflections and compatible strains would require that the steel be engaged in load sharing and would therefore have value but am I missing something here? Or is the double row primarily for stability, i.e. to prevent buckling?

Thanks.

 

[msquared48]

With a single row of bolts, any moment needed would have to be developed by vertical bearing of the bolts rather than both vertical and horizontal bearing, and with a more concentrated horizontal splitting force generated if they were not staggered vertically.

Mike McCann
MMC Engineering

 

[Archie264]

That makes sense. Thanks.

 

[focuseng]

I am assuming this is an academic exercise. I can't see this being a very efficient section configuration to deliberately design. If the steel were only in the middle 1/3 then you would have to analyze it three ways me thinks.

1) as a composite section with calculated "I" and "S" where the steel is almost ineffective for the bending strength due to the neutral axis. Basically the bolting and steel are used for shear strength only.
2) as composite section similar to concrete where the steel is now at a set depth from the compression zone and acts to engage tension forces only. Bolts now engaged horizontally
3) Steel is acting alone for bending strength because by the time it engages, the wood is beyond its strength capacity. Bolts are used to transmit loads only into steel.

______________
MAP

 

[johnbridge231]

Flitch plate beams with a single bolt row must be avoided, as these consist a kinematic mechanism, due to the fact that more than 2 hinges are collinear within the beam length. The beam will need to develop considerable defelction so that these connections will work. Then the serviceability limit checks would not satisfy.

Analysis and Design of arbitrary cross sections
Reinforcement design to all major codes
Moment Curvature analysis

http://www.engissol.com/cross-section-analysis-design.html

 

[BAretired]

johnbridge,

A single line of bolts at the neutral axis does not constitute a mechanism, kinematic or otherwise. If the Flitch plates run full length, the wood portions and the steel portions would carry load in accordance with their stiffness, provided all parts can do so without buckling.

Archie,

Using Flitch plates over the central third of a beam would not benefit beam strength very much as the moment at each end of the plates is nearly equal to the midspan moment. However, it would reduce deflection.

BA

 

[gobsmacked]

Please may I ask: If I had two rows of bolts and wanted to know the forces on the bolts I suppose I would use an adaptation of the shear stress formula f=QAy/Ib
If I have a single row of bolts at mid-height, for the whole length of the beam, how do I calculate the force on each bolt then?

 

[hokie66]

Flitch beams are not designed as composite beams, but rather as wood/steel beams which share the load in proportion to their stiffness. The gravity load is normally delivered first to the wood, then through the bolts to the steel. Therefore, the force on each bolt is the proportion of the load taken by the steel, at that location.

 

[Archie264]

Thanks. Yes, from my end the question is theoretical...but only to a point. I was trying to get a grasp on the mechanics involved to see if was applicable to a situation at hand. There are some timber knee braces that might need to be reinforced. I was considering bolting channels to the sides of the beams, columns, and braces, and connecting the channels together with steel connection plates bolted together. I was trying to figure out how far to extend the steel channels along the beams and columns.

That is, I was trying to figure out if I need to essentially rebuild the entire frame with steel, effectively ignoring the timber? Or can I cut stop the steel a few feet beyond each connection, with periodic thru-bolts along the way?

I think with the later the moment the steel imposes onto the timber by way of the bolts might prove to be a bit much to bear, literally.

It will take further investigation but in the mean time, any thoughts or suggestions? And sorry, no, I don't have a sketch, unfortunately.

 

[XR250]

Not that this is usually checked, but the wood prevents the steel plate from buckling. With the bolts in the middle, it is less able to brace the "flanges" of the plate. When I have a flitch that is highly loaded, I add a steel bearing plate at each end. Keeps the steel from knifing into the top plate if the bolts aren't up to the task of transferring the load back into the wood at the ends.

 

[Lion06]

I would steer clear of using a plate or channel that isn't full length (or just short of full length). Unless you want to consider a different moment diagram for the steel and wood then you would have to have moment transfer at the end of the steel plate or channel. At 1/3 of the span the moment is reasonably high and getting the bolts at the ends of the plate to transfer that moment back into the wood will be difficult to get to work out.

My understanding of the primary reasoning for staggering the bolts is to allow the wood to brace the plate.

There is no shear flow (VQ/It), as hokie points out. This is all stiffness based and the load is transferred from the wood to the steel (typically, it can be the other way around) by bearing through the bolts.

 

[RFreund]

Lion makes a good point and summary although since this was a 'theoretical' discussion I will add (and I may be wrong here) -

You don't necessarily have to develop the moment at the end of the steel portion. You could just dump the reaction at the end of the steel back on to the wood. So you would then have a point load on your wood beam at the end of the steel portion but you would still have a reduced load where you have steel (because the steel supporting some of the weight). Or do I have this wrong?

EIT

http://www.HowToEngineer.com

 

[hokie66]

You are correct, but getting the load from the steel back into the wood is often cumbersome...too many bolts, or the need to resort to shear plates or split rings. Shear on the wood may be a problem as well.

 

[Archie264]

Thanks, I feel I have a much better grasp on it now.

As for the project that caused me to start thinking about this, it sounds like full reinforcement might be the order of the day.