dissent on compression

[Delwin D Fandrich]

Stephen Birkett wrote:

> Del:
>
> > Perhaps you can understand why I'm not willing to let your theory stand
> > in place of the actual soundboard damage that I have seen and measured and
> > documented.
> >
> Please don't get me wrong here. I have no doubt in the seriousness of the
> damage you have seen in the pianos you have to repair. That is without
> question. I have no theory about this. In fact I said in the last letter
> that I don't believe theoretical reasoning can determine which soundboard
> will develop problems and which will not. The only thing that I said in
> dissent about this was that the inevitable failure of every
> compression-crowned board in a modern piano is unlikely...it is more
> complex than that, and an overall design issue.  That doesn't imply it will
> never occur, nor that it won't occur frequently. And certainly your
> evidence is not in doubt insofar as those pianos are concerned.
>
> This is not an area I have studied, so my thoughts are somewhat ad hoc.
> Here's how I see the mechanics.
>
> Soundboard panel flat, ribs flat, differential drying produces that
> curved surface...the surface of minimal potential energy. Because the
> panel cannot shift in relation to the cross-constraint any difference in
> size must be taken up by cellular, or intra-cellular forces at the glue
> surface. The panel begins flat, but finishes curved, therfore surface area
> has increased, both the top and bottom surfaces, the top somewhat more
> so because of the non-zero thickness of the panel.
>
> The surface area of the board has increased as a consequence of
> differential drying, namely the drying of the board to an EMC much lower
> than will be experienced in ambient conditions....  the natural surface
> area of the board will increase as the board regains its moisture content.
> The final surface area will depend only on the relation between dry board
> EMC and ambient EMC.  The cellular and intra-cellular forces are
> compression forces, as the moisture content between cell walls increases.
> The cells will collapse if these forces exceed a critical value. The
> extent of the compression will depend critically on the degree of drying
> the board undergoes before gluing on the ribs. Compression is somewhat
> lower on the upper surface than the lower.
>
> Finally there is the balance of forces achieved from the distributed
> down-bearing components once the board is installed. These act to
> decrease the curvature of the board, therefore reducing its surface area,
> and thereby increasing the compression further, from purely mechanical
> reasons.
>
> To achieve a balance that keeps compression below critical levels will
> require consideration of quite a few design parameters. From what I've
> seen, the dry-board EMC is rather critical and also rather difficult to
> control in typical manufacturing conditions.  In addition, excessive
> downbearing forces may increase compression beyond material capacity. My
> "theory" (if it is to be called that) is that careful consideration of
> these two factors can produce a purely compression crowned board with flat
> ribs on a modern piano that maintains compression within the critical
> material parameter for failure. On the other hand, I would imagine that
> either, or both of: failure to maintain dry board EMC within specified
> manufacturing parameters, or poor design with respect to downbearing,
> might explain many observed soundboard failures. Those Grand Canyon ridges that have been mentioned from time to time on
> the list.
>
> Beginning with curved ribs will give the board a head start so to speak,
> since its "natural" condition after glue up involves a somewhat increased
> surface area, without taking into account the moisture-related surface
> increase.

Now you're getting the idea! But a board designed to function this way will no longer be a pure compression-crowned design.
(It will be a compromise design of some sort. I've not yet thought of a catchy name for it.) Statements I have made in
reference to the stresses found in pure compression-crowned designs will not apply to this design. The load of supporting the
crown will now be shared by the panel's internal compression and by the rib set. And the stresses built up within the panel
can be considerably less because it is no longer necessary to dry the board to such extreme low moisture content levels. The
idea is to reduce the levels of internal stress within the panel so that they are always below the levels that cause
permanent damage to the wood fiber.


> Therefore some expansion can be taken up without producing any
> compression at all, simply cancelling the tension from glue-up on curved
> ribs. I can't really see any advantage to not using curved ribs...these
> were almost universally used on historical pianos with less down-bearing
> forces to boot.

There is no acoustical reason for not using pre-crowned ribs. I don't know about historical instruments, but Wolfenden was
talking about them as early as 1916 in pianos of contemporary design. I strongly suspect that some of the survivors that you
insist on using as evidence that compression-crowned boards can last indefinitely, have, in fact, compromise designs of this
type.


> I suspect the reasoning is that a board already under
> extensive compression, which will be further increased by downbearing
> forces, will take less time to "develop". Meaning the time period for
> settling, as internal forces reach equilibrium, will be reduced. Faulty
> reasoning though if the board goes poof as a consequence. However for
> short term rental instruments it may be an expedient to avoid that
> waiting period for tone to develop. Trouble comes if they sit around too
> long in the warehouse.
>
> Stephen
>
> Stephen Birkett Fortepianos
> Authentic Reproductions of 18th and 19th Century Pianos
> 464 Winchester Drive
> Waterloo, Ontario
> Canada N2T 1K5
> tel: 519-885-2228
> email: birketts@wright.aps.uoguelph.ca



William Braid White seems to have been under the impression that the panel had to be under "tension" to promote the "power of
resonance" in a board. (This may have been just an archaic use of the concept of compression, however. See Part II of my
series on soundboards in the Journal.) (Ugh, sorry, that's not till next month.)

With a well designed soundboard system, it shouldn't take much time at all for a soundboard to reach equilibrium in terms of
tone. We expect the tone to have settled down and stabilized pretty much by the time it leaves our plant. Usually, this is
about a month or so after it has been bellied. It will continue to go through some changes for a few months after that, but
they will be fairly minor and subtle compared to what it goes through in the first few weeks..

Del