[CAUT] Sperrhake Harpsichord wire

[Fred Sturm]

On Dec 6, 2009, at 2:53 PM, Ron Nossaman wrote:

> Now, I'd like to state once more that none of the formulas anyone  
> has are absolute gospel. None of them. The closest thing I have to  
> truth here is from reports I've read through the years from  
> facilities having done empirical testing on wire stating that break%  
> doesn't change with wire size, if you average in enough batch  
> testing. Zero change falls in the middle of the bell curve. Once  
> again, the formulas that we use to calculate tension, inharmonicity,  
> impedance, and break% aren't, and can't be, absolutely accurate.  
> This doesn't render them less useful for their purpose, but picking  
> two at random (which themselves don't nearly agree) isn't a useful  
> indicator of an absolute premise.
> Ron N


Thanks. Obviously I have been relying on false data to come to my  
conclusions.
	Just for the sake of completion, and maybe to make it clear to some  
others, I'll recap a couple things. We will all agree that on any  
working real scale, which means that length increases at less than two  
times per octave, if you string with one gauge, the tension goes down  
as you move down the scale, and up as you move up the scale. Since the  
breaking point stays constant (since you are using the same gauge),  
that means break% is higher at the top and becomes lower as you move  
down.
	More to the point, though, given a single note, meaning a speaking  
length and a target pitch, if you increase the diameter of wire, the  
tension increases (and, obviously, the inverse: decrease diameter,  
tension decreases). So the question is whether the strength of the  
material increases at the same rate, to keep up with the increase of  
tension. I had thought, based on the information I had, that it did  
not keep up. Apparently I was wrong. Of course, the measurement of  
breaking point tension is not an absolutely precise science, and so it  
is likely that my information was based on sampling that happened to  
slant in a particular direction (that is, the breaking point figures  
selected by the spreadsheets I used tended to show more relative  
strength in narrower gauges).
	So, bottom line, I am convinced. Again, thank you for going to the  
trouble.
Regards,
Fred Sturm
University of New Mexico
fssturm at unm.edu