Laminating a Curved Exterior Beam
Here's a detailed discussion of adhesive selection and clamping methods for a large, curved, laminated beam for an outdoor structure. March 12, 2014
I'm working on a project that calls for 2 radiused western red cedar beams for a pergola. They require a 10' radius in plan view and are approximately 20' long. The beams need to be 6x12. My plan was to laminate (12) 1/2" x 12" boards using a vacuum bag that opens on the long edge and clamp to an exterior form.
I'm still debating what type of glue to use. I typically use polyurethane on cedar laminations but am contemplating resorcinol for the longer open time. Any suggestions? This is the largest radius lamination I've done. I assume the vacuum will be able to apply enough force for (12) 1/2" layers?
(Architectural Woodworking Forum)
From contributor K:
You ought to read the specs on suggested PSI for resorcinol. I think it is up around 175 PSI. I normally pull 28" of Hg on my system, which is only about 14 PSI. For that reason, I would use epoxy, which also gives you plenty of open time.
On projects like this, I have one person just mixing, and I just pour it onto the board ahead of the paint roller, rather than picking it up on the roller from a tray.
As you suggest, I have done long projects like this where you could never stuff a slippery stack of parts into a long narrow bag. I use the cloth core carpet tape to seal the bag. Also, it is a good idea to have some rope caulk, to seal where the hose comes into the bag, and for any place where there may be a wrinkle in the bag where it crosses the tape. Just stuff a wad into the wrinkle and press it down.
I use the 1 : 1 ratio mix from Fiberglass Coatings Inc. It is a thicker viscosity than most, and about half the price of West, or System 3, and much more forgiving than all others I've used. Also, it is a good idea to mix in some colloidal silica to thicken it to about latex paint, or catsup.
From the original questioner
Thanks. The only info I could find about polyurethane was from the Franklin site which was similar, 30-80 psi for HPL, 100-150 psi for softwoods, 125-175 psi for medium woods and 175-250 psi for hardwoods. That pretty much leaves epoxy unless someone has another suggestion.
Any ideas on places to get long bag material cheap? I see McMaster-Carr has 1/32" vinyl 36" wide roll for around $14/lf?
From contributor M:
Have you done a dry bend with some 1/2" cedar? My guess is this will be difficult, and you should shoot for 1/4" laminates. As far as clamping, I'd add several extra laminates (2" worth) and some 12" cauls (with pipe clamps on either end) and clamp directly to the form, working in small increments from the center out. Suck it down after you've got it all clamped to the form if you want. It can't hurt, and it might help with the mess, but it's silly to think that the bag is creating the bonding force here. The spring force of all of those laminates is going to give you a workout. Resorcinol is the glue that I'd use. Epoxy would be fine if you could limit the clamping pressure, but that would either mean thinner laminates or bagging on one laminate at a time for the glue up.
From the original questioner
We tried bending a 10' radius on a 1/2" board and felt comfortable. That might change once we get 12 together but we'll have to see. So you would skip the bag? We thought about this but were worried about getting even glue lines without voids using just clamps since it will all be visible once installed. How close together do you think the clamps would need to be to get a consistent glue line? I would think we would need every 6-8".
From contributor A:
I would use a slow cure epoxy. Just clamps, the bag is a waste of time. You are going to have to clamp it anyway until the vac sucks it down. I would think pairs of clamps every 12" would be adequate. The bending action places a fair amount of pressure. You are going to need a protective outer layer like a piece of steel to protect the wood from getting crushed. Keep in mind epoxy likes thick glue lines. Also you need to run the wood through a sander at 60-80 grit when using epoxy. It doesn't matter which species.
From contributor B:
I would not do a 12" wide lamination without cross sticks every 8" to 12", and a convex clamp block would be best.
A vac bag lamination is going to do a fair job on pressing the edges but you are likely to end up with internal gapping in the center of the 12" width. I agree with the others that it is going to lead to inadequate pressure.
We did a 2" thick mahogany lamination at 8" wide and about 12' long a few years back. Even with the above clamping scheme we had some gap issues. And that was with 1/4" strips. Our clamps just were not large and strong enough to do the job.
If you are committed to doing this with the 1/2" strips based upon your earlier test, then do 2 strips at a time. Glue up the first 2 then add 2 more. I think this will give you a better chance of success.
As to which glue I think the gap filling properties of epoxy outweigh the ease of use of the resorcinol.
From the original questioner
Thanks. So assuming I forget the vac bag and go for 1/4" strips with clamping blocks and cauls every 8" using epoxy, would you recommend trying to do all 24 layers at one time? Does anyone have a maximum number of layers they are willing to do at one time? We will probably have 4 people available to glue, place and clamp but could get more if needed.
From contributor M:
Yeah, I'd probably skip the bag. The only benefit of the bag that I can see is the force that it applies to the outermost layer. This is a difficult layer to get even pressure on, but the bag seems like a lot of hassle just for this. Think about gluing it all up straight first (without the radius); you'd need a certain amount of depth to the clamping cauls to have even clamping pressure on the outside layers (using the 45 degree rule). Balance the depth of the cauls with the clamp spacing to give even pressure along the length of the beam. To give even pressure across the width of the beam, I'd use a pipe clamp on the top and bottom with two additional cauls bridging the clamps on either side of the beam. These cross cauls would have holes for the pipes to pass through and be curved a bit to add pressure first to the middle of the cross section. Use these same principles in gluing it up on the radius form. The inner longitudinal caul is created by the form. This form has to be solid, or else it won't be able to transfer an even pressure along the beam's cross section. The outer longitudinal caul is created by a bunch of extra laminate layers that don't get glued to the beam; above I suggested 2" worth, but really this overall thickness has to be balanced with your clamp spacing (again, using the 45deg. rule). The cross cauls are the same as for a straight glue up.
I would do the glue up all at once, but I'd do a dry run first, and might even leave it in the clamps overnight before the real deal. And give yourself some way of checking (glue, clamps, and all) that all of that spring pressure doesn't deform the shape. Good luck; I see a really fun day in your future.
From Professor Gene Wengert, forum technical advisor:
Are these structural beams? If so, your options for adhesives are limited. Although epoxy is a good adhesive, you need to ask the manufacturer if their adhesive is approved for structural use. In any case, if you use epoxy, make sure you learn about its use, as too much pressure will weaken the joint - a strong epoxy joint is a thick joint.
Epoxy adhesive is easily deteriorated by UV light, so it is very important that epoxy used outdoors be UV stabilized. Such formulations are available for a slight increase in cost.