rescaling tension

[Newton Hunt]

Hi Del,

Nice to have seen you at Arlington.

Some good ideas here, and you are right, 3.3% is a bit
much.  I have GOT to get the cobwebs out of my thinking cap
and from between my fingers.

Talk to you later.

		Newton

Delwin D Fandrich wrote:
> 
> ----- Original Message -----
> From: Newton Hunt <nhunt@jagat.com>
> To: <pianotech@ptg.org>
> Sent: June 29, 2000 7:06 AM
> Subject: Re: rescaling tension
> 
> > Tension is but one of the important factors in trying to get
> > a nice sounding scale.  You can play with the tension a lot
> > by reducing the wrap size and changing the unwrapped ends
> > around.  Balance is the key factor.  1500 pounds is only
> > 0.033% of 45,000 pounds so I think you are well in the safe
> > range but think about reducing it by 500 pounds or so to be
> > in the save range.
> >
> > Newton
> -------------------------------------------------------
> 
> Did that apple fall a bit hard on your head this morning, Newton?  It will
> work better after a couple of cups of coffee...
> 
> That is 3.3%, I think.  Even so, it is true that this is not a significant
> increase in overall tension.  However, it really depends on where the
> increase occurs.  During the rescaling process it is not likely that 3.3% is
> added to each individual string of the scale.  It is more likely that
> somewhat more than this was added to the treble section.  Some would be
> taken off of the mid- to upper tenor.  Some would be added to the lower
> tenor and what goes on in the bass is anybody's guess.
> 
> I pay much more attention to what is added to the low tenor and upper bass
> section than to the overall scale tension.  Usually the #1 strut and the
> struts between the tenor/treble, the treble/treble and the top treble are
> fairly lightly loaded relative to their size and shape.  But, as I said
> earlier, the X between the bass and tenor sections is (structurally) the
> weakest part of any plate design.  I get real cautious here.
> 
> I have encountered more than a few pianos in which the 'mathematically
> ideal' scale -- i.e., the one that has the so-called best fit of all those
> nice curves and things -- has actually reduced the overall string tension
> load while increasing the localized load considerably across the break.  One
> of the worst I can recall would have added about 30% to 35% to the stress at
> the X while keeping the overall tension load approximately the same.  (The
> bi-chords in the bass and the low tenor would have increased quite a bit
> while the low bass and the mid- to upper tenor would have decreased a lot.)
> That is a significant increase of stress on the X.  Easily enough to break
> some of them.
> 
> With my scaling spreadsheet (a simple little thing running in Excel) I have
> it set up so that I can easily sum the tensions through the upper half of
> the bass section and the lower 10 unisons of the tenor section so that I can
> compare the new scale with the original at this point.  I then compare any
> increase in scale tensions through this region with my evaluation of the
> plate.  If there is any question at all in my mind about the strength of the
> X I will either back off on the tensions or add a stress coupler between the
> plate and the belly rail.  Assuming the piano is worth the effort.
> 
> Regards,
> 
> Del