Klarinet Archive - Posting 000126.txt from 1997/04

From: Mark Charette <charette@-----.COM>
Subj: Re: Mouthpiece Facings
Date: Fri, 4 Apr 1997 12:22:02 -0500

Jerry Korten wrote:
> For instance I think it would be important to do a curve fit to the facing
> curve (say a 4th order polynomial or some other model) and describe the
> curves in this way so that we could study the rate of change of curvature.
> This would require measuring the curve at a fairly high frequency such as
> every 1 mm or so.

I think a NURB would be the right choice for your curve here,
not a regular polynomial. Your frequency of measurement would be
much lower. Also, by measuring a bit more, you could create a
NURB surface (or set of surfaces) and really analyze the properties.
Curiousity, however - I wonder if the mouthpiece designes are
patented? If not, then you're really going to be doing the competing
mouthpiece manufacturers a great service. Maybe you can get
funding ...

There are a large number of CMM machines out there that could take
the measurments relatively fast. Again, interesting someone at
the U of Michigan Engineering school or Lawrence Tech or Wayne
State (all in Michigan, USA, and near the Big 3 US car manufacturers)
would be the way to go - they all have this type of machinery laying
about.
--
Mark Charette Webmaster, http://sneezy.mika.com/clarinet/ - the
MIKA Systems, Inc Clarinet Info Pages . For Klarinet help, send
charette@-----.com