>quoted some of the results earlier. However the analytical and >computational techniques needed to study this seemingly simple >problem with any degree of realism have only really just become >available within the last few years. I'm hoping that some new >computational techniques for analyzing multi-body mechanical systems >will be applicable to this problem (that's the main approach in the >computer model I'll be developing). To achieve anything resembling >realism has always been virtually impossible because of the >computational difficulties arising from the nonlinearity...but >these new methods were devised precisely to make this kind of >work tractable. >Stephen Birkett (Fortepianos) I have a transducer that collects the signal of the motion of the string excited by a grand hammer of an action model. My device consists of a monochord with piano like terminations. The string passes by two copper plates each about three inches long. The string is about 23 inches long. The two plates are parallel to the string and at right angles to each other. A power supply sets up a charge on the plates which act as two sets of capacitors ; one plate and the string, the other plate and the string. Since the plates are perpendicular to each other they have the ability "see" the string motion in two independent planes. The signal is sent to an Analog to Digital Converter Board in my computer. The Software package is able to process the signals to help extract meaningful information. The hammer strikes the string and cause it to initially displace in a vertical direction only. After a few milliseconds the motion begins in the horizontal plane. This horizontal movement delay is when the string moves after it has been reflected from the nearest termination point. I can see that time delay vary as I change the strike point of the hammer in relation to the string. I can also detect differing decay rates between the two planes of motion. This verifies some earlier work of Weinrich (1977 JASA, I believe). I do plan to publish this when I polish up the experiment. I need to eliminate some confounding errors, organize my data, and prepare a presentation. I do not plan to have it published in a scientific journal so it will probably end up in our "Journal". Michael Wathen
This PTG archive page provided courtesy of Moy Piano Service, LLC