Glue line creep -- yellow glue

Same reason that rocks do...they are flexible and time. In the case of rock strata, LOTS of time. In the case of aliphatic glue, I have never had a problem.

I have no scientific evidence, but a sneaking suspicion that it is a combination of many factors, among others: type of wood and its dimensional instability due to relative humidity/temperature; cut of wood and grain (flatsawn moves in width, quartersawn in thickness); type of glue and amount used; clamping pressure, et al.

IOW just about anything that causes dimensional instability in a glued up joint, including finish.

I took a look at several websites describing it and they can't even agree on what it is, let alone what causes it. Some say glue "bumping" out of the glue line over time is creep. Others talk about two pieces of wood glued together changing position relative to each other. I didn't find anyone mentioning both, which is really the case.

But as far as I know, it's caused by the fact that the yellow glue (and white as well) is "plastic" i.e. a petroleum based polymer. It never gets really hard so it can deform or squirm under pressure over time.

If you suspect creep could be possibility in a project use hot hide glue or resourcinol or epoxy. I'm sure there are others but those will do.

Some say that liquid hide glue will creep, others say no - no definitive answer I could find but I suspect it at least creeps less than PVA glues. Someone else may have more info on that.

I'll throw in my 2 cents, this type glue remains flexible, wood expands and contracts, no two pieces that form a joint expand or contract exactly the same.

The definition of "creep" used in engineering & materials pretty much expalins it:

"In materials science, creep (sometimes called cold flow) is the tendency of a solid material to move slowly or deform permanently under the influence of mechanical stresses. It can occur as a result of long-term exposure to high levels of stress that are still below the yield strength of the material."

(wikipedia)

At high enough temperatures, or given enough time, even steel structures will "creep"

The reason the 'bump' is called creep is that it is really extrusion under pressure -- plastic deformation. Whether the two jointed pieces also move relative to each other in net displacement depends on whether and how well they're constrained and in which direction there are sufficient forces.

Thinner gluelines are better for minimizing the effect because there's less bulk material there. Which is why higher clamping pressure is _a_good_thing_ (tm) besides the empirical observation from testing that it improves ultimate joint bond strength.

"Fine Woodworking" magazine did a clamping pressure vs. strength test some years ago and found that in almost all case, more pressure was better. Reading that article changed my clamping life and my joints are much better than they used to be.

...

I recall it...and it only confirmed what previous work at US Forest Products Lab and elsewhere has shown...

On 10/27/2013 9:02 AM, Larry W wrote: ...

To be observable, it'll still need the loading until temperatures are quite high... :)

Ok, I understand that.

So how is yellow glue capable of high stresses structurally but not in bent laminations? I would think the stresses structurally would be subject to creep to and therefore lower the structural value.

But only bent laminations are devalued.. I understand that bent laminations are under constant stress, but we don't know what or how the glue is used structurally. Imagine a bench that someone sits in 16 hours a day... two shifts... is'nt that under a similar stress?

Agreed. I've always wondered about selling dark vs light glues depending on the color of the wood, because I've never been able to see a glue line after clamping.

I did once have a few glue bumps on a project - still don't know why and I've never had it happen again.

Not really, the boards of the top are not stressed in reference to each other.

Now if the legs were bolted to the floor and braced and the top glued to those legs, you'd see sag over time. How much would be creep of the top vs the legs and how much wood fibers elongating is beyond me.

On 10/27/2013 12:30 PM, woodchucker wrote: ...

The difference is that in the bent lamination example there's residual stress (and quite a lot) of the bent material trying to return to it's original shape. This is constant and unrelenting; hence over time the plasticity begins to show.

In the bench there's some intermittent loading but there really isn't a comparable load between the glue joints that is continuous nor in the tangential direction that is there as the difference in lengths in the curved lamination.

On 10/27/2013 2:26 PM, Mike Marlow wrote: ...

Basically, research has shown that there really is no such thing as being able to "starve" a joint by excess pressure with anything approaching a practical application of clamps and cauls that one would use for normal glue-ups. Hmmm....I can't seem to find the link just offhand and don't want to take the time to really look now, but the kinds of clamping pressures that were used are in the _way_on_up_there_ (tm) ranges--actually in the incredible kinds of numbers compared to what we normally think of as clamping pressures.

I'll throw in my 2 cents, this type glue remains flexible, wood expands and contracts, no two pieces that form a joint expand or contract exactly the same. ============================================================================================ Not saying that glue creep doesn?t happen but I've never seen it. I have a few pieces that my father built in Japan in the mid 70s. They went from there (very humid) to the desert (bone dry) then to Seattle (humid) and have seen no creep at all.

I have no idea what kind of glue your father used. It doesn't make any difference unless the conditions we've been talking about here exist. That's a situation where the pieces of wood are under strain in reference to each other and the glue is the only restraint.

Any of that furniture have bent laminations in it?

--------------------------------------------------------------- "Sw> I have no scientific evidence, but a sneaking suspicion that it is a

------------------------------------------------------------ You planning on running for public office any time soon?

Understand the Texas governor's office will be available next election.

Lew

begin 666 Calculator.lnk M3 ````$4`@``````P ```````$:?`@``( ```):A_%?EX&4` M`"@``P`$`.^^Z3@:B"HY1Z@4````8P!A`&P`8P`N`&4`> !E````& ```$L` M```?*UU