Strengthening an existing glulam beam

[adriank]

I have a 19" glulam beam that will see an increase in demand due to a heavy mechanical equipment. We can deepen it to a 24" beam, but the exterior of the beam is architecturally sensitive. That is, we cannot sister steel or wood to the side for strengthening. I am calculating about 6 kips/ft of shear flow using v=VQ/I. A preliminary design would be 1" lag screws at every 4" which is way excessive.

My question is, are there any products or design guides that can allow me to increase the shear flow capacity at the interface between the 5" add-on member and the existing glulam? Either adhesive, or some kind of toothed plate that has approved capacity? Or even a split ring connector to act as a shear lug? Thanks in advance.

 

[JAE]

Why don't you try looking at a steel rod, pretensioned along the bottom of the GLB. I know that bridge beams are "improved" using this technique quite often.

Bolt brackets to the bottoms of the GLB at each end and run a rod along the bottom - pretension it.

 

[JedClampett]

If you're just worried about the connections, you're on the right path. See:

http://clevelandsteel.thomasnet.com/category/shear-plates-split-rings-spike-grids

There's design guides in there, too. They were written in the year one, but they still work. NDS also has a section on them.
I did a design a couple of years ago and used split rings to improve the connection capacity. I specified them, drew them into details, did everything but hold the contractor's hand. He of course, "forgot" them. Fortunately, the capacity still was enough without them.

 

[msquared48]

So... the rings just split, huh? Did they ever come back?

Seriously though, what about just adding beams to either side of the existing, not adjoining, but a distance away, in order to share the load?



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

 

[csd72]

This is the sort of thing they would do over here:

http://www.diydoctor.org.uk/projects/weakbeam.htm

Only problem with this option is that it will increase the shear flow on the existing glue joints possibly to above their capacity.

 

[LonnieP]

Yes, there is an adheasive solution. I've used Simpson two part epoxy to glue PSL laminations to the bottom of 4x and 6x purlins in warehouses, and an occational residental ridge beam (exposed). The issue is working time once the two part glue is mixed (usually 30 minutes), and of course the mess as it squeazes out of the glue joint. I use Simpson SDS screws (in pre-drilled holes) to hold the lamination in place til the glue cures (24 hours). Shore the whole thing up til the epoxy cures. You'll need to check the horizontal shear flow near the supports if you have to stop the lamination before the bearing.

Simpson has an unpublished report of their SET epoxy being used to anchor an allthread rod into a wood beam. Not really the application you've got but it does indicated the glue will bond to wood. Of course you could use white wood glue (the drying time would be unacceptable) but at least nobody would be bitching about using concrete epoxy on wood.

 

[Brad805]

We did this years ago. We were working with the Wood Science Technology Institute in Oregon (WSTI). We purchased additional laminations from a glulam supplier and had them laminate WSTI carbon fiber strips to the lams. The advantage of the carbon fiber was we could increase the beam strength dramatically (100 - 250%)without adding a great deal of depth. If deflection is part of your problem, then you may need more than one lamination. To adhere the parts to the existing, we first power planed the existing glulams to remove the original finish and then bonded the parts using Gougeon Epoxy. The advantage to this epoxy is that it has a set time around 5hrs, and you can achieve the rated bond strength with very small clamping forces (10psi). We used lag screws to effectively clamp the lams to the existing. The downside of the epoxy, is the difficulty cleaning the adhesive that squeezes out. It is solid and it does not sand easily.

Gougeon has a new product called the West Six10. I have personally used this and it would be my choice if I were doing this again. They developed a mixing tube that mixes the epoxy correctly using a standard caulking gun. Mixing the epoxy was a bit of a nuisance when we purchased this in 5gal pales.

Another adhesive I have done some simple tests with is the PL Premium. It is rated as a structural adhesive, but I was never able to find sufficient technical data. In my little tests I did find it performed as well as the epoxy, and it is far less expensive as well as easier to procure.

Brad

 

[adriank]

Thanks so much for everyone's input. The info on the Simpson adhesive and Gougeon Epoxy is very useful! I'll weigh all these options along with the shear plate connectors. Adhesive would be nice and reliable, but we'll have to factor in finishes, lead times, and labor vs the shear plates. There's a lot of labor either way, so it may be a wash but at least I have some options.

 

[BAretired]

A number of years ago, I strengthened an undersized glulam beam using a steel plate on the bottom. The plate had the same width as the beam. The plate was pre-drilled in the shop to receive glulam rivets. The beam was jacked up to remove deflection and the rivets were driven up into the glulam by hand. It was a lot of pounding, but it worked.

BA

 

[dhengr]

BA:
Timber rivets are an interesting and under utilized means of making connections in some wooden structures. Your’s was a clever use of them. Did you have any trouble rationalizing the concentrated end reactions of the plate tension near the ends of the glue lams where the horiz. shear is also at its max.? You would have pretty high shear transfer through the bottom laminations into the rest of the member depth. Or, did you distribute the rivets along the entire length of the cover plate?

 

[BAretired]

dhengr,

In my case, the beam had adequate strength, both in shear and flexure as designed but its deflection was excessive. The cover plate was primarily intended to increase the moment of inertia. The rivets were installed throughout the length of plate, but the spacing varied according to the expected shear flow. A concentration of rivets was added near each end for good measure.

BA