How fast can PVA glue creep?

[David Kuzdrall]

So glued up a few test laminations in an attempt to calculate exactly how much spring back I am getting when bending and glueing 3 layers of 1/4" mdf so that I can redo my bending forms and result in the correct radius on the final part.

In the interest of cost and time I made my test panels with titebond iii and measured spring back around 12-hours after the parts came out of the vacuum bag. Now, 4-days later they have relaxed a bit more and I wonder if this is from glue creep or the material itself that need to account for? My final parts will be bonded with West Systems Epoxy, so the Pva is the only variable right now.

thx

[Joe Angrisani]

.....In the Internet of cost and time.....

Is this anything like the Internet of Al Gore?

[David Kuzdrall]

Is this anything like the Internet of Al Gore?

Ha, not unless he invented "auto fill" as well.

[Bob Vavricka]

It is likely the glue you used allowing the creep. I use pre-catalyzed powered glue from joewoodworker for the laminations I do. I have a cutoff from one I did over a month ago and it still matches the mold with very little springback. I left it in the vacuum bag close to 24 hours with a heating blanket over it to speed the cure. I don't think you will get much creep with the epoxy glue if it is well cured before you take it out of the mold--it dries much harder than the Titebond you used for your trial run. You can see the 1" cap piece I formed for a crib using 1/8" lamination in this post http://www.sawmillcreek.org/showthread.php?195434-Walnut-crib&p=2017230#post2017230

[george wilson]

1/4" MDF? I'd think that stuff could stretch a little from its own structure(does it have structure?). So,it may not be the glue at all.

[Mel Fulks]

David,regarding the info I sent you,we were usually making trim that would be installed nailed to something,so a little error did not matter. For anything freestanding ,sometimes segmented (with or without veneer )works better than lams.In most cases we used yellow glue thinned with 5 to 10 percent water and brushed on "both sides".When the pieces were removed from form they would first spring back ,then over a few days in a warm dry room ,diameter would DECREASE with the loss of the water we added. My tweaked figures were based on that experience .Yellow glue without added water acts differently. Glue creep refers to an actual shrinking as glue cures ,often causing an actual shift in surfaces.White glue creeps. I once glued 2 pieces of 1/8 inch aluminum together with white glue as a test. When dry,I sanded the edges flush ,aday or two later, they were misaligned. Most modern glues have no creep.When you need more accuracy with lams,use more of them....One more thing , that formula depends on a material being springy,so any real bending would matter.

[Mel Fulks]

Bob, sounds like you are using plenty of lams.That glue you use is good ,but the water in it definately can cause some over bending as the moisture leaves it. That's why many are using the 2 part liquid version of it

[Mike Henderson]

I've done a fair amount of laminate bending and have not seen creep with Titebond glue. You do get some springback, but that's not due to the glue. I've done laminate bending with cherry, walnut, oak and maybe a few other woods.

Sam Maloof did laminate bending for the rockers on his chairs and he used Titebond. He used Titebond I on light colored woods and Titebond III on dark colored woods. I assume that before Titebond III he used regular Titebond on everything.

Mike