Soundboard stiffening

[Ron Nossaman]

---------------------- multipart/mixed attachment

> >So how much of that is due to restriction from the short back scale
> >lengths of a tuned duplex, and how much is the soundboard?
>
>I don't know.  I suppose one possible way to find out would be to string a
>piano with no downbearing and then apply increasing amounts of downward 
>load on the
>bridge with with some sort of loading devices, which would essentially 
>take the back
>scale angle out of the picture.

Or take the aliquots out and tune it back up to pitch with the resulting 
increased bearing and longer back scale. My point is that the bearing angle 
and soundboard is typically blamed and the short back scale duplexes aren't 
usually considered as possible culprits. The shorter the back scale, the 
more critical and shallow the bearing angle.


>Interesting.  This is the opposite result obtained by Ron Overs, if I
>understand correctly.

So it would appear, though I don't know why.


>I also don't understand what you mean by panel
>compression.  Are we talking RC or CC here, or do you make a distinction in
>this case?

Both, though more so in the CC.


>In a rib crowned and supported board I thought that all the
>load was taken by rib bending.  I thought that the panel didn't enter into
>it.  What would be resisting or supporting this panel compression?  I
>believe you would say it's not the rim.

Take a meter long strip of wood about 25mm square - because the effect is 
easier to see in a big model. Get an accurate measurement of the length 
when it's straight, then clamp it into a bend so the "crown" is about 7mm. 
This corresponds to about an 18M, or 60' radius crown. Now measure across 
the top of the crown and you'll find that it's a millimeter or so longer 
than the straight stick was. If you had a good way to measure the 
underside, you'd find it was a millimeter or so shorter than the straight 
stick. Anything glued to the top surface would be compressed as the curve 
was forced straight. Expansion of a dried panel stretching the top surface 
of the rib is what forms crown in a CC board in the first place. In fact, 
if you glue a dried panel to a flat rib and block the ends so the panel 
can't expand, the rib won't crown at all. This is the other half of the 
experiment I outlined to Thump demonstrating that soundboard crown isn't an 
arch. In a RC board, panel compression could still be supporting crown and 
bearing load to some degree - shared with the ribs. In a RC&S board, there 
will still be some panel compression, but it is not a significant 
contributor to crown and bearing load support. A machine crowned rib with 
no panel will get shorter across the top curve as it is depressed to flat, 
and will add some compression to the panel glued to it as a result.

I expect the progressive spring rate to be least in the RC&S board, and may 
not be evident enough to be of any particular importance. In my boards, I 
use a high crown and a lot of deflection to provide the necessary 
counter-spring for the string bearing, rather than a steeply progressive 
spring rate we tend to see in existing soundboards, or new boards built 
with high panel compression levels.


>I think I'll wait until after measuring some boards in pianos before I decide.
>
>Phil Ford

Always the best way. If it's an original board, take note of what the 
bearing and crown was like before teardown to see what you're starting with.

Ron N

---------------------- multipart/mixed attachment--