On Aug 28, 2007, at 8:29 PM, Jim Busby wrote: > We did devise a thumper, but I’m wondering is a Disklavier would be > the way to go. Consistency was definitely an issue with our meager > spectrum studies. I’m really hoping that these “science professors” > will take the ball and run with it. That is what they do. Hi Jim, Let me give just a few quick thoughts about "thumpers" and other devises for activating a key, to give your physicists an idea of some of the problems involved. The tuning test thumper drops a weight from above the key. There is a free fall of 6 inches or so before the weight hits the keytop. It has now achieved a certain velocity, and is still accelerating with gravity (a simple calculation can give the specifics). So far so good. Now comes the moment of impact. The key is in a state of inertia, as is the whole key/action assembly. During impact, there is a period where the colliding bodies "bounce against each other." The weight slows suddenly, or possibly stops falling altogether or bounces upward (we'd need some high speed videography to know for sure what happens). The keytop absorbs this enormous force, which has impacted it suddenly. This scenario is far different in many ways from the way a pianist plays a key. If he/she plays the key with a great deal of force, slamming down from above the key (more or less like the falling weight), the difference is that the mass of the body behind the finger will push all the way through the keystroke with only a minimal "rebound" effect. And, as we have seen in various high speed films, the front of the key will hit bottom before the hammer begins to move (flex and compression allows this). But this is an unusual piano technique. Far more common is simple pressing of the key, starting with the finger contacting the keytop, and accelerating the key. This technique can be used to create a full range of the available volume of the piano (not counting percussive sounds from the key crashing into the keyframe/keybed, in that slamming action described above). And I think that any study involving tonal output of a piano ought to be based as nearly as possible on normal key activation. The problem with using a real live pianist is that you can't be sure of exact repeatability, and being exact is tremendously important in this area. Joggle a mike, play a little louder or softer, and the spectrum and even measured pitch will change. Askenfeld (did I get that right? the guy in Sweden who put together the 5 Lectures in PIano Acoustics) said in his lecture that he used a pendulum to activate a key. This makes some sense, as its velocity at impact can be controlled by how far it is allowed to swing (position before letting it go), and bounce is minimized because of the oblique angle with which it strikes the key. But it is still a rather artificial setup: a body in motion suddenly impacts the key, and the arc of movement means that the downward movement of the pendulum weight is probably decelerating (a pendulum weight moves downward most when it is at the same level as the pivot point, and it moves downward less and less as it approaches the bottom. Even though the weight is accelerating to that point, the geometry is such that its downward motion is becoming less and less). So I'm not convinced a pendulum is a good substitute for a finger. I guess the disklavier/pianodisc/pianomation systems with their controlled solenoids do a reasonable job. I haven't really fooled with them enough to have an opinion. It's going to be a sudden electromagnetic charge of a controlled amplitude creating a an electromagnetic attraction of a corresponding force. Does this replicate what a finger does? I think a finger has more possibilities, in terms of how much acceleration it imparts to the key in any given portion of the key travel. (I have puzzled a good deal about how it is possible to make one finger's note stand out when playing a chord. Somehow that hammer had to be given more velocity. But it all happens so fast, and is so hidden in psychology, that it is hard to get a handle on it). Anyway, I will again say that I think a mechanical devise, with a system of accelerating levers activated by a weight, and with the mass and placement (leverage advantage) of the weight being variable, would be a great contribution to study of any number of things to do with pianos. Regards, Fred Sturm University of New Mexico fssturm at unm.edu -------------- next part -------------- An HTML attachment was scrubbed... URL: https://www.moypiano.com/ptg/caut.php/attachments/20070904/03607cb7/attachment.html
This PTG archive page provided courtesy of Moy Piano Service, LLC