Adhesive for Exterior Doors

Dark-stained wood exterior doors can get very hot in service. Here, woodworkers discuss choices for glues that will perform under those tough conditions. September 6, 2007

We have recently had some glue line failures on raised panels in our exterior doors. We use double floating panels and the failures were only on the exterior face of the exterior panels. We use TB3, which I've recently learned from their tech department can start to fail at about 120 F. The doors are stained a darker color, which is not helping the situation. They all did get some sun, and one did have a storm door. The panels were finished on all six sides before being installed. The material is Sapele and Spanish cedar, both with a MC of about 7 1/2 to 8%. All joints were properly machined with sharp insert tooling, square, and tight. Should we be looking at other glues? Do we need to get our product drier before finishing? We have about 150 doors in service with no failures till now. Is a wood door that gets some sun simply doomed? We care very much about the quality of or product and would greatly appreciate any help with this situation.

Forum Responses
(Architectural Woodworking Forum)
From contributor T:
That's a tough situation. Dark exterior doors are never a good idea, and the glass storm door allows a tremendous amount of heat to be trapped in a small area between the two doors.

If your panels are splitting at the glued seams, and TB III fails around 120 degrees, you might consider the old Weldwood Plastic Resin Glue. It's approved for use in fire-rated doors that are tested at a 250 degree temperature rise. Those doors have to burn for as long as 90 minutes to pass the test.

A reminder about the glue - it has about a one year shelf life, so make sure your supplier can offer you a fresh batch.

From contributor B:
I recently spoke with Franklin techs about some 2+ year old TB3. They told me that as long as the glue wasn't gummy coming out of the jug, and it dried as you would expect, it was good.

I would look at adhesives like West System epoxy if the TB3 is failing. TB3 acts more like TB1 in terms of radio frequency drying. That is, you have to let the joint cool before it can be moved. Heat it again and it can be separated (theoretically anyway). TB2, on the other hand, crystallizes hard immediately upon the radio frequency drying. In hot house situations like the overheated door, TB2 might be the better choice. Or epoxy or other catalyst adhesives.

From contributor J:
I agree with contributor B. West System epoxy is the way to go. I worked in the Caribbean for a while making exterior doors out of Spanish cedar and mahogany. We used West System epoxy down there, mainly due to climate, etc. It worked great.

From contributor K:
Be aware that epoxy, too, weakens with heat, though up around 180 F. I would recommend checking with the formulator for specifics.

From Professor Gene Wengert, forum technical advisor:
I agree with contributor B too. Epoxy deteriorates with UV light, so make sure that you use it only on doors that will be painted.

From contributor I:
I like West Epoxy for my exterior glue joints. Make sure your topcoat provides UV protection. Other reasons for panel failure could be too tight a panel fit in the groove, or poor gluing technique. Could glue be creeping out to the panel and making it stick during the gluing process? This would prevent the panel from floating properly and be a definite candidate for failure.

From contributor O:
I'm curious why no one has any further interest in using plastic resin glue. It seems to be unaffected by heat or cold when set up. It does require special care when doing the glue-up, but don't they all?

From contributor K:
I would like to point out as well that epoxy does not adhere well to smooth surfaces. The glue lines should be sawn or sanded with 80 grit, especially with dense hardwoods. We use a lot of epoxy, but plastic resin is a good adhesive as well. The epoxy has the longest open time of any glue we use, and the best shelf life as well as being compatible with different fillers for varying purposes such as thickening, fairing, coloring, etc. It is also more tolerant of varying glue line thickness. The heat factor is not significant in most situations, but I mentioned it because of the storm door condition the poster alluded to. I doubt that the heat buildup there would approach the softening point of epoxy, but it's something to think about.