I don't intend to argue these points of glue strength with Ern because
the subject was never about absolute joint strength measurements, to
begin with.  This discussion (which I still very much appreciate being
able to answer, because things are learned best and surest in an
earnest comparison of facts this way) began when he said that hot hide
glue is a thermoplastic, and that it creeps, just like any other glue.
He could not have possibly used hot hide glue on anything that is
gasketed and spring-loaded, like valve blocks that slide apart, and
still make that claim.  All veteran users of hot hide glue (not to
mention chemists) would instantly agree that Ern is wrong about his
basic premise from which everything else he says is taken.

Tests of hot hide glue strength years ago were run very exactly in labs
all over the world and decided to be directly connected to viscosity,
from which comes the "gram strength test."  They too probably realized
a large variance by gluing collectively thousands of boards together,
and would have decided that their variances found had a lot to do with
the board, itself.  You cannot control that factor.

Today, hot hide glue strength is sold on the basis of its strength,
and is the only glue that has ever been done that way, scientifically.
Remember this: When you buy a glue labeled 165g, they are telling you
what the strength of that very batch of glue you bought actually tested
to in the lab.  These factoid label 2000 lb./sq. in. stuff is in your
dreams, and meaningless for the same reasons I have already given.

However I will answer, again, the crux of Ern's last comments, and then
let him finish the discussion on hot hide glue.  I've already made my
point that it isn't strength I am primarily interested in but glue
latitude (friendliness to use), rigidity of the joint, and adhesion,
that matters most.  I have never addressed absolute glue strength,
except relatively speaking.  I have said, and I believe it to be so,
that if your glue is stronger than your wood, why worry about how much
stronger?

> What were the results for the lap-joint tests?  If both sets of
> adhesive were applied and set correctly, and if the moisture contents
> and species glued were the same, I would expect that similar amounts
> of wood failure would have been obtained and that the joint strengths
> would have been similar.  The only variation would have come from the
> wood strength variability, if the tests were accurately controlled.

Glue application, in this instance of lapped joints, was "identical"
application, and for my eyes, only (at the time).  That was, I took
a wide 1" brush, quickly flowed on some glue on one board and pressed
them together at the 1" line, and left them.  The PVA was done the
same way, except flowed from a nozzle and spread with my finger.
(They weren't done with mechanical robots, each accurate to a mm.)

Then I squeezed then together and clamped them.  The wood was just
pine lathe that had been ripped on a saw-- not planed smooth and
surface-sanded.  It was also not selected for density, smoothness,
dampness, grain lines/inch, or anything else.  So here we can say that
Ern finally has the answer to his question why there were differences:
"It was not a legitimate laboratory test."  That's because I don't have
a laboratory or a license to test anything.

Well, to me, knowing what I had done, knowing what I was using, and how
I applied it, I got the answers I was looking for, and as it turned out
later, the answer my test gives is correct, anyway.

Because we are talking a general shop situation, not an industry --
in which 50,000 chair legs of a particular wood, and everything
specified are going together with a certain joint for a certain length
of time, having a specified finish at a certain temperature range, etc.
In a general wood shop,  you must have a glue you can depend on for any
and all kinds of wood which you cannot predict and whatever finish that
come to us in player pianos.  Many is the time I have reglued pneu-
matics of all different finishes, and some were just rip sawn.  So with
only sawn wood, I wanted to know (for my own information) what the
best, most reliable bonds were going to be, period.  Remember, if you
aren't measuring differences, then you aren't measuring correctly.  And
to say that the adhesive should not have sheared is argumentative, in
my mind, since every hot hide glue joint broke its wood at least 50% or
more of every joint, and most 100%, but only two PVA joints performed
that way.  The rest showed less adhesion that the hot hide glue.

That is because of the "lacy dried glue pattern" of PVA glues that
require evaporation and local absorption before setting up.  Which
means, PVA glues must move and seep around by capillary forces as some
parts dry quicker than others within the joint.  It weakens itself this
way, not to mention the fact that it invariably warps and curls thin
flat boards with its own moisture leaching through them that not even
clamping is able to fully prevent.

So let's please keep on the subject; that being, "Is hot hide glue a
thermoplastic, and creeps just like any other thermoplastic glue," as
Ern claims, or not?  I will let others experienced in the use of hot
hide glue finish this moot point if they're interested.  Ern should now
honestly address his claim, as an expert on glues, that hot hide glue
is a creeping, plastic glue, just like the PVA's.

Biodegradability of (fully cured) hot hide glue?  Surely, you're
kidding.  I would suggest that Ern place a phone call to Jacob Utzig,
Technical Director of Milligan and Higgins who make hot hide glue in
this country and tell Jacob that he is worried about the thermoplastic
creeping properties and possible biodegradability of this colloidal
proteinate, and how bugs, mold, and bacteria are likely to devour it,
not to mention wood acidity degrading it.  He is going to learn
something, but obviously not from me.  Hopefully he will report his
findings to everyone here.  If we want actual facts, then it's high
time we hear them about these counter-claims, which I have not made.

Craig Brougher

 [ These definitions given at  http://www.custompak.com/glossary.htm
 [
 [   Thermoplastic:  A glue or resin having the property of softening
 [   or fusing when heated and of hardening again when cooled.
 [
 [   Thermosetting Glues and Resins:  Glues and resins that are cured
 [   by heat, but do not soften when subjected to high temperatures.
 [
 [ Glass is probably considered thermoplastic, and it is sometimes
 [ used as a seal between metal components, but it's not normally
 [ considered an adhesive.  -- Robbie