more heresy

[Delwin D Fandrich]

Robert W. Hohf wrote:
> 
> As long as we are debunking dogma, how about this:  loading a soundboard
> does not simply put soundboard into compression, as has been implied in
> several recent posts.  I am no mechanical engineer, but I have questioned
> several over the years on this topic in order to understand some observations I
> have made on pianos.
> 
> In order to put an arched system into compression, the load must be
> _uniformly distributed_ across the arch.

Not at all. It can be rippled like a potato chip and still be under
compression. Remember it started out under compression. It was
compression that formed the arch in the first place. That compression
doesn’t go away just because the panel warps in funny ways.


> This is not the case in pianos. If we
> consider a cross-section including rib, board, and bridge, the load applied to
> the bridge is a _point source_.  This type of loading does not produce
> simple compression.  It causes the bridge to sink into a valley with bulges on
> both sides.  The greater the loading, the deeper the valley and the more
> pronounced the bulges.

The compression itself is fairly simple. The results—the effect on the
soundboard assembly—that develop from that compression are not.


> Of course, real soundboards are not cross-sections, but have an extremely
> complex shape in 3-dimensions.  Real soundboards can have bulges on either
> side or both, and the side of the bulging can change in different areas of the
> same piano.  This sounds a lot like "bridge roll", doesn't it?  Do we need
> another explanation of bridge roll?
> 
> This valley/bulge creates a "wave"  shape in the board.  The resulting pattern of
> tension and compression is complex.  Sometimes it can cause a characteristic
> pattern of  compression failure and cracks.  It can cause paradoxical cracks on
> the bottom of a soundboard.  It can also make measuring the crown on the
> bottom of a board inaccurate.
> 
> Bob Hohf
> Wisconsin

Yes, the movement of the piano soundboard is a highly complex subject.
Much to complex to go into here. In case you, or anyone else on the list
is interested, most of what I have to say on this subject appears in two
separate articles that were published in the February, 1995 and the
February, 1996 issues of the PTJ.