Spreadsheet info / Jason Kanter

[Frank Emerson]

I have used Excel for many years to do much of my scale calculation;
however, I alternate between my CAD software (SolidWorks) and Excel in
working out the final details of the scale design.  Some of the work in
Excel becomes a "design table" which interactively defines dimensions in
SolidWorks, and some of the work in SolidWorks determines some basic
parameters for the Excel spreadsheet.  If you are serious about scale
design, it is my opinion that you must get into CAD software as well as
spreadsheet software, and SolidWorks is the best IMO.  My spreadsheet is an
ever-evolving thing.  Each time I begin a new project, I tweak the
spreadsheet to factor in things that I have learned from the previous
project.

One thing I failed to mention in my earlier post is that I avoid allowing a
section break to coincide with a wire gage change.  This only exacerbates
the bridge dog-leg problem.  I would move the section break, before I would
allow the wire gage change to occur at the break.  In a
rebuilding-rescaling situation, one cannot change the section break, but
you can avoid wire gage changes at the break, even if it compromises the
tension smoothness curve. 

 You are correct, I do calculate the lengths first, then the diameters, to
satisfy "target tension" parameters.  At this level, I calculate these
values for the unison.  I would prefer this to be the same for all strings
within the unison, but the requirements of the manufacturer may limit my
"preferences."  

Someone posted a comment about equal-tension scale design.  I believe that
I was the first to design such a scale.  The Baldwin 248 upright is an
equal-tension scale, as well as the Wurlitzer grand, G550.  You can look
down the treble bridge and see the wire gage changes, which coincide with
irregularities in the string length increments.  I am still fond of this
design concept, although my current projects do not include this
recalculation of string lengths.

Regarding inharmonicity, my spreadsheet calculates the inharmonicity of the
first six partials, and the longitudinal mode partial.  I used these
calculations to mathematically "tune" the piano before it is built, the
same as I would have tuned it after building it.   My intent was to save
the "stretch factor" in a computer chip installed in the piano, which could
be accessed by any ETD.  The manufacturers I have worked for have not been
interested in pursuing this, so it has fallen by the wayside.

Regarding the Z factor, my spreadsheet calculates this, and it is plotted
in a graph.  It is taken into consideration only to the extent that
irregularities are evidenced in the graph.

Frank Emerson
pianoguru at earthlink.net


> [Original Message]
> From: RicB <ricb at pianostemmer.no>
> To: <pianotech at ptg.org>
> Date: 12/17/2006 10:06:35 AM
> Subject: Spreadsheet info / Jason Kanter
>
> Hi Frank.
>
> I understand this to mean you figure your string lengths first, then 
> figure string diameters to result in as even a tension (unisons or 
> single string?) as you can reasonably get.  How much do you look at Z 
> and inharmonicity... and basically how do you go about this.  Keep in 
> mind we are looking for ways of providing some kind of automation for 
> things usually done manually. A comment or two on how much leeway you 
> give for each of these parameters would also be very helpfull. 
>
> I get the feeling that excell may be too limited a tool for some of the 
> automation we've been tossing around.  But thats what we are going to 
> use so whatever is practical / doable will be attempted.
>
> Cheers
> RicB
>
>     .........
>     Starting with a blank piece of paper, designing a new piano, I prefer
to
>     calculate the string lengths with a pure logarithmic progression,
>     and deal
>     with the dogleg of the bridge in other ways.  By making the distance
>     from
>     the front bridge pins to the back bridge pins longer on one side of
the
>     break, and shorter on the other, the dogleg in the bridge can be
>     reduced.
>     The side bearing angle can be maintained, even thought the distance,
>     front
>     to back, is varied.  The bridge can be undercut, to further reduce the
>     dogleg of the bridge, at least with respect to the gluing surface
>     between
>     the bridge and the soundboard.
>
>     When it comes to restringing a piano, when the treble bridge is not
>     being
>     replaced, recapped, or repinned, I would not hesitate to reverse the
>     wire
>     gage progression across the break, when that would serve to smooth the
>     tension across the scale, in some cases, more than a half-size.
>
>     Frank Emerson
>
>