Epoxy Man Strikes Again

[Delwin D Fandrich]

----- Original Message -----
From: "Isaac OLEG" <oleg-i@wanadoo.fr>
To: "Pianotech" <pianotech@ptg.org>
Sent: March 08, 2003 1:00 AM
Subject: RE: Epoxy Man Strikes Again


> Hello,
>
> I can't see why the glue joint or glue zone between bridge and
> soundboard can't act as a sound modifier, or filter. A too thick  glue
> joint certainly should.

I suppose if one were to really use a pva adhesive (commonly referred to as
"white glue") and if one were to make the glue-line thick enough (a mm or
so) it might be possible to introduce enough compliance to affect the
mechanical response of the soundboard. There would also be some structural
problems since there are no adhesives except epoxy that have any mechanical
strength at these thicknesses.

Typical glue lines are in the neighborhood of 0.075 to 0.150 mm in
thickness. Much less and there is the risk of starved glue joints, much
thicker and the structural strength of most woodworking adhesives begins to
break down. Modern adhesives manufacturers have improved the gap-filling
qualities of their formulations with various additives so that adhesives
such as Titebond and Titebond II work reasonable well even when the
wood-to-wood joint fitting is not perfect.

While technically thermoplastic, both of these adhesives dry hard. As Ron O
has pointed out, they chip when the squeeze-out is cleaned away.

The idea in fixing the bridge to the soundboard is to get a reasonably good
wood-to-wood fit and a reasonably good glue bond so that the motion of the
bridge is closely coupled to the soundboard. This is more easily
accomplished that commonly thought. Remember that the bridge base has a
relatively large surface area. The motion of the bridge is not isolated to a
single point. It is, after all, a bridge. When a specific string sets the
bridge in motion, that motion is felt, mechanically, over a broad expanse of
the bridge. It would take an unbelievably viscous adhesive to audibly or
measurable affect the energy transfer from the bridge to the soundboard.


>
> And the choice of the glue may change the resonnance frequencies of
> the assembly too.

Keeping in mind that we're really trying to keep the soundboard/rib/bridge
system relatively free of specific resonances. Many things will affect the
resonances found in a soundboard. Among them are changes in humidity and
temperature. Even tuning the piano will cause changes in the resonances of a
soundboard assembly.


>
> Beside I understand that the bridge and the board  have to be warm
> (and dry) to be glued. Do you use the hot box before gluing ?

This is only true if hot animal hide glue is being used to bond the ribs
and/or bridges to the soundboard. Historically, soundboards and ribs were
heated to temperatures up to 120º F (approx. 50º C) to keep the hot glue
from gelling prematurely.


>
> Epoxyes seems to be cook sometime, to attain their final hardness
> temperatures as high as boiling water . I suspect we don't want to
> cook our  soundboards as that !

So, it's OK to heat the soundboard to 120º F (approx. 50º C) when using
animal hide glue but it's a problem to raise the temperature of the wood by
some fraction of a degree (by either scale) when using epoxy? Let's think
this through.

Epoxy adhesives are thermosetting. They cure by chemical reaction and they
do generate heat as they cure. This becomes most obvious when you mix up too
much and it begins to kick over (chemically cure) in the pot or can in which
it was mixed before you can use all of it. If you have really mixed up too
much it can begin to bubble and boil, turning itself into a real witch's
brew. The heat generated by its own chemical reaction feeds on itself,
accelerating the chemical reaction and the mix goes into thermal runaway.
The more epoxy mass you have the faster this process takes place. Once it
starts, it's too late to do anything about it. Set the container on
something nonflammable, wait for it to settle down, throw the batch away and
start over.

This is one of the most obvious differences between slow-setting and
fast-setting epoxy. Mix up equal batches of each and the fast-setting batch
will start bubbling much faster than the slow-setting batch. And it's one
reason why, in our shop, we use slow-setting hardeners almost exclusively.

One way to control this--if you really must mix up a large batch--is to mix
the resin and hardener in a wide, flat pan. This spreads the mix out, giving
it more surface area to dissipate the generated heat and slow the process of
thermal runaway. It also illustrates why it is not a problem in real life.
The epoxy goes on so thin that, as heat is generated by the curing process,
it is immediately dissipated into the wood (or whatever). Yes, the curing
process will raise the temperature of the wood. But, the mass of the wood is
so large compared to the mass of the epoxy that the actual increase in
temperature is nominal. In other words, the wood will get warmer but the
temperature increase will be so small it would be difficult to measure just
how much warmer.

Confused yet? I can elaborate....

Regards,

Del