Glue Creep

A long discussion of wood movement, adhesive rigidity, and finishing and performance issues. June 29, 2008

I have seen how a once smooth PVA glueline joint can change with the seasons to really show itself right through finish. One solution is to use a plastic resin glue, which seems to "freeze" the two edges tight. Do any of you bondsmen know what other options are available? I am hoping to learn of some that are easier to use.

Forum Responses
(Adhesive Forum)
From contributor T:
Try polyurethane mastic, such as PL Premium construction adhesive. I don't mean polyurethane glue such as Gorilla Glue. There is a difference. PL Premium expands slightly which fills the joint, Gorilla Glue foams like crazy.

This glue comes in a caulking cartridge so just squeeze out a bead and clamp. It does need 3-4 hours to cure before stressing. I won't use anything else for tabletops where glue lines need to stay flat. Sometimes end grain orientation, quarter sawn next to flat sawn can cause surface levels to change.

From contributor V:
I started using plastic resin for the same reason. I use to use Titebond II but had a lot of glue creep. As an example I have a crotch walnut end table I built for myself years ago when I first opened my shop. It has been sanded down and refinished two times and it still gets glue creep. I recently went back to Titebond I because they said it does not have creep like II and gave a lengthy explanation. I will soon find out if they are right.

From contributor R:
To contributor T: Is PL Premium compatible with waterborne finishes such as FUHR?

From Gene Wengert, forum technical advisor:
Rather than glue creep, it is the wood that is moving, but the wood right next to the joint is not as free to move. Do you read FDM magazine? There was a discussion of this about a year ago, if you want technical info.

From contributor T:
To contributor R: Absolutely no problems with any finish. Fuhr finishes are what we use too. To keep the glue from sticking to bench tops or clamps use pieces of plastic drop sheets, vapor barrier or garbage bags. It won't stick to that type of plastic. I use this glue for veneered panels in a vacuum press with great results. An old fine tooth saw blade spreads it out on the substrate, much like troweling thin set for tiles. Its totally waterproof so water-base finishes won't affect it.

I think water-based finishes are going to change the way we do woodwork. I am very glad I switched from solvent based finishing. Yellow and white glues can become re-dissolved by water-base and they aren't rigid, and can creep. Not good enough for finer work, but great for general woodwork and carpentry.

All adhesives have a place in a wood shop. Have you tried the new PowerGrab mastic for instant grab? It is different again and only good for stuff like glue blocks, small trim, parts where our grandfathers used hide glue. It's designed for sticking up crown moulding without nails and it works in my house! Its a fantastic time saver.

There are lots of gimmicky products out there but these two adhesives are good ones. I'm not sure, but I think the PUR hot melt system is the same as PL Premium, just faster. I buy the large cartridges to save money. For a large shop I'd look into boxes of the large cartridges.

From Jeff Pitcher, forum technical advisor:
There seems to be a lot of mis-information regarding "creep". In general, thermo-plastic adhesives such as PVA (white and yellow glue) can allow "creep" in a glue joint that is under stress. Often shops that do curved laminations will experience this. If the glue line is not under any particular stress (i.e. an edge glued panel) then there will not be any "creep". In order to avoid this in stressed glue lines shops should look at using a thermo-setting adhesive such as urea resin.

From contributor C:
I'm a bit confused as to what we're talking about. Edge glued panels such as a solid end table top or a planked dining table top gets a definite line over the glue joints during some seasons of a given year. Microscopically it seems that the glue line is rounded and ever so slightly bulging up for a few weeks and then all is flat again. Is this what were referring to as glue creep?

From the original questioner:
Yes Contirbutor C. I have however seen it grow quite obvious and never change back. Use of plastic resin seems to stop this by "freezing" the two pieces together with a grip that won't let it happen.

From Jeff Pitcher, forum technical advisor:
To contributor C: What you're seeing is moisture movement in the wood. It has nothing to do with the glue. Except that the glue isn't gaining or losing moisture and therefore remains the same.

From contributor C:
Thanks, Im with you now. When I am concerned about this in a critical situation such as strip laminations, bent laminated pieces I always try to use the most rigid glue. I use a specific Jowat white glue, powdered resorcinol glue and gorilla glue, save the epoxies and Tightbond where static tension is nil like tabletops, raised panels, and wainscot panels.

From the original questioner:
I have been in millwork 25 years, and have seen many strange glue related situations. It seems that softwoods are much more prone to reveal a glue joint by changing with the seasons differently on each side of the line.

Over time, I have used many different glues, so the results I speak of are from experience, not theory. Plain carpenters glue of any brand will retain a tiny degree of flexibility, though I cannot say for how long after the initial glue up. I know it is at least a year.

This is not only evident in butt joints, but by the greater degree of spring-back one sees with bent laminations. Plastic resin cures rock hard, so spring-back after clamp release is very slight in comparison.

When plastic resin is used for butted edge panel joints, it really does limit the two sides in their tendency to swell at different rates, as they seem to lock together very tight.

From contributor A:
Cold creep, or simply creep is a physics dynamic and complicated process that may or may not be the problems that some of you are having. As both a wood worker on Megayachts and a dabbler in polymer chemistry, I have seen the difference between being a theorist and a problem solver. My hope is to find solutions through science to my problems, instead of guessing.

Water based adhesives work excellent in solid wood and veneer conditions when the glue is allowed to organize uniformly on a molecular level. Pressure control allows the polymerization to be complete, instead of having isolated pockets of bond and crystallization in areas where the polymer hasn't received its negative ionic charge from the introduction of the catalyst. Adding heat can not only speed up clamp time, but also assist in uniformly accelerating the atomic particles in the polymer to give a more complete glue process, reducing what is known as creep.

Water based polymers work, plastic resins work when the right conditions occurred at glue up, try a vacuum press and heat blanket, and test the results for yourself. The nice thing about plastic resin is that it's stiff and impervious to moisture transfer. That also is its downfall. Its more brittle than water based polymers. It all starts with the right products - try West Systems, or Unibond 800 for starters and the Unibond Blocker for filling spread joints.

From Gene Wengert, forum technical advisor:
Note that creep of an adhesive or any adhesive joint requires stress. For this reason, we do not use "soft" adhesives for a laminated beam that is under continuous stress and strain. But we do not hesitate to use PVA and other similar adhesives for chair legs, tables, and so on, even though these items may have some occasional stress.

Note also that the original question involved the change in a joint with seasons. The adhesives are not hygroscopic, compared to wood and it is well know (not a guess) that the RH changes with the seasons, causing moisture changes in wood which causes size changes.

The role of pressurization to polymerize an adhesive is interesting. I have never seen this comment before. Certainly, pressure is needed to put the two pieces of wood close to each other and to squeeze out excessive adhesive and to push the adhesive into the very small nooks and crannies. I look forward to hearing how pressure causes polymerization of PVA and similar adhesives.

Perhaps the three items mentioned above mentioned in practical terms are actually what you are referring to when you talk about negative ionic reception. In any case, note that these adhesives are not catalyzed, so I am not sure what you mean about "introduction of the catalyst."

I am also surprised that you imply plastic resins are better than PVAs when both make a joint stronger than the wood itself and both have no significant creep in furniture, cabinet and similar items. Further, PVAs are plastic resins. Plastic resin adhesives are not, as a group, impervious to water. Nor are they more brittle.

Incidentally, Unibond 800 is a two part urea resin adhesive. The catalyst is formaldehyde, which is often avoided in furniture and cabinet production. The glue is a gap filling (implying that high, uniform pressure is not needed) and is rigid when dry.

West Systems is know for their two part epoxy adhesives, which are usually too expensive for wood and this higher expense must be offset by high benefits, which again for furniture and cabinets does not happen.