Overstressed Wood Beam to Composite Wood Beam 1

[AlexREI]

I am working on a simple span wood beam that is overstressed and I'm toying with a few different ideas. One interesting one that I've come up with would be a good solution, but I'm looking for someone who has possibly done something similar to this.
Obviously, the wood beam (7X18 psl) is in place and has been stressed with dead loads and partial live loads. I'm designing as if it is seeing DL + .15LL (ASD). At this point it is stressed 2400 psi out of 2900 psi. The interesting idea that I've designed to fix the bending issue is inserting a 7/8" dia turnbuckle rod 2" under the wood beam and stressing (putting tension in the rod) it up a certain amount. With proper connections at the wood beam ends, the tension added in the turnbuckle would act as "post tensioning" the wood beam and create an upward moment from its eccentricity, which would in turn bring my DL + .15LL stress in the wood beam down to 2250 psi. From this point on (for the rest of the LL), I've assumed composite action with an increased Ix from the transformed section. I can make the bending stress in the wood come down to under 2900 psi, while keeping the tension in the turnbuckle within the allowable amount.
Here is the question: does this solution conform with VQ/IT? If I'm assuming a composite section with a 2" gap between the wood and steel rod and only connections at the wood beam ends, is any horizontal shear stress transferred to the rod, or can I assume the shear stress is taken in full by the wood beam and the LL bending is taken by the composite composite section. This solution works well in theory for bending, but this shear issue is throwing me for a loop. If anybody has done something similar or has comments, they would be greatly appreciated.
Thanks!!!

 

[frv]

Have you considered axial and flexural interaction?

 

[BAretired]

A very old technique is to slope the rods from the ends down to the center or to the third points with a straight rod in the middle third. In this case, the shear is partially carried by the rods.

Another technique is the use of flitch plates where steel plates are anchored to each side of the timber beam.

Another possibility is to add a steel plate on the bottom and fasten to the PSL with glulam rivets.

If you place rods 2" below the member and anchor at each end, you can satisfy the bending moment requirement because, essentially you are adding a negative moment at each end of the beam which relieves the midspan moment.

I believe that you reduce horizontal shear stress in the beam by introducing axial compression in the bottom fibers, moving the neutral axis downward. It's getting late...need to think about it some more.

BA

 

[Lion06]

I know that tensioning rods has been mentioned on this forum before, but I've never looked into how this is done. The idea of using a turnbuckle 2" below the beam gives me a little pause and here is why - In this case, the turnbuckle is not just delivering an axial force to the wood beam, it's also delivering a moment (the pretension force in the rod times the 2" moment arm). I don't know the magnitude of your pretension force, but it seems to me that if the turnbuckle rotates (even just slightly), it will relieve the pretension reducing the effectiveness. This can only get worse over time as the rotation of the turnbuckle creeps.

 

[BAretired]

The beam is 7"x18", so the eccentricity is 2" + 9" = 11".

A = 7*18 = 126in^2
S = 7*18^2/6 = 378in^3

If the rods are prestressed to 1000#, P/A = 1000/126 = 7.9psi and M/S = 1000*11/378 = 29.1psi.

f(top) = 29.1 - 7.9 = 21.2psi (tension )
f(bottom) = -29.1 - 7.9 = 37.0psi (compression)

For each 1000# of prestress, compression in the top fibers will be reduced by 21.2psi while tension in the bottom fibers will be reduced by 37.0psi. The beam will arch upward and the neutral axis of the beam will drop slightly.

Since neither the constant bending moment nor the axial force produces shear, I think the timber section must carry all of the shear, i.e. VQ/I.

BA

 

[dhengr]

Alex:
Your OP seems to imply a full length horiz. round bar from end to end with a turnbuckle in the middle, all 2" below the bottom of the wood beam. This arrangement will help with the bending stresses, by prestressing, but does nothing to change the shear stresses in the wood beam and you must check bm./col. interaction. The end connections are the difficult part of this design, around the existing beam bearing and all.

BA’s 2 rods, one on each side, horiz. and below the beam by 2" or 3" over the center half or third, and then sloping up to the beam ends, is likely a better solution. You can manipulate the elev. of the end axial reaction point, BA said 9" + 2", and you can adjust the distance below the beam and the 1/3 or 1/4 rod bend points to your benefit. This arrangement imparts two upward vert. loads at the rod bend points and downward concentrated loads, through the beam to its bearing, These cause a reduction of shear in the wooden beam, with the stl. rod carrying that portion of the shear. Draw the shear and moment diagrams for the two conditions, a simple beam with gravity loads, and a simple beam with two upward loads at the rod bend points. You still have to check bm./col. interaction, but end up with lower bending and shear stress in the wooden beam. The hardware might be something like: 2x2x11" stl. tubes with simple saddles at the rod bends, screwed to the bot. of the beam; 4x8x11" stl. angles at the ends with 4" horiz. leg over the top of the beam, and beveled washers or seats for nuts and washers; once the slack is taken out of the rods, knowing the pitch of the threads, turn of the turnbuckle sets your calc’d. rod tension; jack the beam up a bit if you can to help get the slack out of the rod and hardware. I think BA’s calcs. will apply to this situation at the beam center. I don’t understand SEIT’s concern about creep in the turnbuckle. Put a bolt through the two turnbuckles so they can’t be changed without some forethought.

 

[dhengr]

Alex:
Can I please steal your thread for a post or three, while you’re not looking?

BA & frv: RE: Basic ....Help, and lost posts.
I gotta think that when you’re as far off-base as this OP’er seems to be; and so dumb about the problem at hand or the facts of engineering life that you don’t realize the beating you’re taking; the powers-that-be step in and stop the fight (a TKO ?), rather than watching any further carnage. I think they just wanted this thread to go away with a ‘thanks for all your help guys,’ from the OP’er, and your post just got in the way of that, and got zapped. As Paddington put it, (two or three posts before your’s of 13NOV, his had already been obliterated when you two posted), ‘this guy can’t even figure out how to start the car, he sure isn’t qualified to be driving this car or this fast.’

BA, your post of 12NOV10, 22:43, RE: bickering and helping young engineers, was one of the best posts on the whole thread, and I pretty much agree with your feelings about helping willing young engineers, and letting them all participate. But what bugs the hell out of me , and I think this was JAE’s point too, we shouldn’t be teaching elementary physics, statics or strength of materials here. They should go to school for that and have a few text books of their own. When some guy pretends to be an engineer and then asks a dumb question which pretty much shows his experience level, and then he wants to argue with those who are trying to help him; I do get pissed off at that, why waste my time, all of our time. I will admit, I am not quite as diplomatic and patient as you are, but if that same person came to me, and said he was a draftsman or a tech., and wanted some help understanding how you would tackle such and such a problem, because he/she wanted to start learning why and how the engineers do some things, I’d try to work with him/her. If they are honest with me, at least I know where to start working with them; if they are B.S’ing. me, then I think they are wasting my time. In fact, over the years I’ve had a couple draftsmen who, with a little nurturing turned out to be better engineers, than some of the graduate engineers working around us. Once they got a little direction, I trusted their work more than that of some engineers, and I knew they would come to me with questions when in doubt, not going off half cocked like this OP’er is, and getting us in trouble. I’ve had good mentoring experiences in both directions, a few bummers too, but these forums are really an inefficient means of doing that. I just can’t type as fast as I can talk and sketch when we are face to face and looking at the same plans and specs., and that’s why it’s so important that they get a local mentor who they can work with and trust, and develop a meaningful working relationship with.

 

[Lion06]

How would the turnbuckle be attached othe wood beam at the ends?

 

[BAretired]

dhengr,

Thanks for your comments about the other thread. I see your point and JAE's too. 'Nuff said.

BA

 

[Cadair]

Post-tensioning of timber is done in Canada for bridges quite a bit.

http://pubsindex.trb.org/view.aspx?id=80068

One caution though. Since wood is an elasto-plastic, your post-tensioning is going to slowly reduce over time. In other words, someone is going to have to go twist that turnbuckle every now and again or it won't work.

 

[hokie66]

Cadair,
That is post-tensioning for a different purpose than discussed above, i.e. transverse post-tensioning as opposed to longitudinal.

 

[Ron]

Post-tensioning of a wood beam can obviously be done. BA's suggestion would alleviate some of my concerns with localized crushing at the anchor points. It would still be much simpler and easier to flitch the beam or add a plate to the bottom.

 

[rscassar]

The concern I have wherever rods are used to strengthen existing structures is the loss of tension due to thermal expansion.