I can't let such a tidal wave of a discussion go by without grabbing my surfboard and trying to touch some of the high points. Please forgive and bear with me as I do this in one post rather than a series of replies (whose "re:Re:re:Re"'s quickly lose the ability to specify their subject). Dennis Johnson rote (Re: Bechstein Action_Hamm Weight 1/4/95): <<However, wouldn't you agree that in the end a scale cannot tell you exactly how to solve an actions weight problems. This comes from a throughly intuitive understanding of the entire action, its mechanics and geometry..... but I do insist that my end result would be same- with or without a scale.>> I agree, if all that you're going to do with the weights is to look at their numbers. Simply weighing and looking isn't going to change the end result. However once you know the hammer weights, you're one step away from getting the hammer weight where you want it. Where should that be, you ask? This, in David's process, is a function of the action's overall ratio, and whether the action will be spring or lead balanced. Ken Sloane rote (Hammer Weight. etc. Date: 95-01-05): <<There is a definite correlation between deviations from the "ideal" or theoretical 2 to 1 ratio designed into most (all??) contemporary pianos and the touch resistance conditions those deviations produce. I have some interesting conclusions, along with some interesting conclusions about hammer weight and knuckle position. >> Right on the mark, Ken., with this discussion of key and shank leverage. Action ratio is half the battle (hammer weight being the other half). But if we lose this half of the battle, then the whole banana goes down the tubes. As you've documented, Steinway has frequently lost this half of the battle and, in the case of teflon and pre-'84 parts, did so categorically. <<this "black art" of regulating touch resistance.>> Consider this "tuning up mass and leverage". a necessary aspect of action assembly. Michael Wathen rote (Angular Acceleration in a Piano Key Date: 95-01-05): <<you have to consider a variable kinetic friction force and a changing weight component as the hammer-shank assembly moves through it stroke>> I'm interested as to why the mass of parts changes as the assembly moves through its stroke.With this acceleration are we approaching anything near the speed of light? ;-) But I do aggree that the friction will vary through the stroke. Mainly at the outset and only at the knuckle, when the it is at its lowest inclination with respect to the hammer center. As the knuckle's angular motion proceeds, the distribution of vector motion changes from the horizontal (sliding) to the vertical (lifting). It's this phenomenon which I believe is behind the friction gradient. You've seen that the hammer shank's rise with minimal dead weight at the key will require a couple of bumps on the bench. The same happens when running an up weight: the friction gradient will cause the hammer to lodge 1/4" above rest. The NH Chapter has started into an ongoing project, called "Junior Science Project", with a stroboscopic videography exploration of fast/deep repetition as its first subject. I also intend we should find the contours of this friction gradient. <<I typically find that the speed of the shank is the fastest at about half way through its travel. I originally had thought that the acceleration would be constant.>> Maybe this is Bill Spurlock's "flexing of levers". Ken Sloane rote (Re: Angular Acceleration in a Piano Key Date: 95-01-06 02): <<The ability to accelerate a key rapidly is important to a pianist, and what hinders that ability is the presence of too much friction and/or inertia in an action.>> Agreed. In fact this is the crucial transaction between pianist and action. (No, I'm not overlooking another one between hammer texture/soundboard and pianist's ear.) But please let's, look at mass and friction as separate matters. They affect the feel differently and they're dealt with separately. The pianist pours their energy into the action, and this is absorbed by each of these two. Friction is like an operating expense: you're working against constantly, coming or going. Mass is like start-up capital. After you've invested the energy do get mass rolling, the subsequent momentum of the parts is a form of equity: it'll do some of your work for you. The disadvantage of an action with such a high "up-front capital investment", is that it complicates any change of finger pressure/key velocity the pianist may want to make partway through the stroke. A change to the increase will require yet another investment. A change to the negative will be unheeded by the momentum of parts.The only thing help in this case is the constant brake of friction. <<but inertia. No one has come up with a good way of measuring it or even begun to figure out how much is appropriate.>> In fact David Stanwood did in early '92. The process has already been granted its initial patent. Bill Spurlock rote (Re: Angular Acceleration in a Piano KeyDate: 95-01-06 01) <<Consider two keys on the same action: A13 with a 9 gm hammer and 4 key leads, and A73 with a 4gm hammer and no leads. The difference in inertia between these two keys is tremendous, much greater than the change in inertia we might make when customizing an action, yet both keys are capable of nearly equal repetition speed. >> I don't believe that repetition is where inertia will make itself felt, though anything involving reversal of motion would be an obvious place to look. Keep in mind the speed with which the hammer comes flying off the string. In the closed loop cycle that is fast repetition, this rebound should certainly override any inertial "tax" incurred in such a reversal.. Inertial resistance is certainly felt with increasingly harder attacks by the pianist's finger. The finger hits the keys (and action) at a dead rest, if not heading in the opposite direction. When the key's ability to accelerates fails to respond to and match the finger's impact velocity, the finger may seem to be pushing on a dead weight. It's this temporary mismatch in velocities, as the key and action pick up speed, which can be so tiring to a pianist. (Like dancing with a partner on thorazine.) <<So it is action mass, geometry, voicing, power of the piano structure, and of course regulation that combine to give an action a certain feel. I doubt a single formula for ideal hammer mass would apply to all actions given all the other variables. In general, I feel the biggest and easiest gains in piano performance come not from hours of work on special modifications, but rather on getting the basics right: choose appropriate hammers, shape them properly, attend to all pinning and alignments, bed the key frame and everything else as solidly as possible, and regulate well. Then, nit-pick the regulation and voicing some more. Not that there aren't pianos that came out of the factory needing modification, but very often it's the dull stuff that really pays off, and often gets overlooked in the rush for the magic formula.>> The point is well-taken, and is fine advice for someone who might dream that some miracle "GL-70" additive can obviate all the basics, "the dull stuff". However here, you're among people for whom the latter is a premise. Ken Sloane wouldn't be complaining to us about "some pretty unwieldy pianos", if he knew the answer he was missing lay somewhere in the basics. That goes for a whole bunch of us who are fastidious on the basics. We may charge good money for it or lose the same on these "basics", but we do them. With all due respect, Bill, what we're talking about does lie beyond this tedious realm. Michael Wathen rote (Re: Angular Acceleration in a Piano KeyDate: 95-01-06 01): <<After the gram weights begin to move the key how long does it take to reach the bottom of its travel? Answering this question is not really the same as finding the acceleration but it is rather finding the average speed. So how do we find the acceleration?>> Is not acceleration simply a matter of leverage ratio (taking into account the distortions induced by Bill Spurlock's flexing of levers and softness of bearings)? I wouldn't try to determine acceleration from observations of shanks speed as driven by a dead weight at the key front. To do so adds friction to the already complicated product of how the mass of the parts throughout the lever train is translated by overall action ratio into the balance weight at the key front which opposes the driving dead weight. <<So where do we start?>> You start with a simple requirement that as many of these force measurements will be taken at a standard point in the leverage. Why not the front end of the key where we now do our down and up weights? << In other words, if the strike point along the shank is too close to the center you would expect that the Weight of the action would decrease and if the Hammer were placed on the shank too far away the Weight would feel two heavy. So where is or what is the theoretically correct way to locate?>> To paraphrase Beavis and Whats's'name, Sound Rules! If you check it out, you'll find that the shank length isn't nearly as sensitive to 1/8" as some other junctures in the lever train which Ken Sloan has already mentioned. I don't see this bearing much on leverage. <<One more point, how do you think reproducing players work? They must at least measure the angular speed of the shank as it moves through a fixed point.>> The digital ones surely must. The original pneumatic renditions of shank acceleration must have chopped the curves into steps, given the state of technology back then. I should warn you that I speak as someone who knows not a stitch of player actions, old or new. Sorry about the 1700 words, gang. Heck, it's Sunday. Stick this right in with the sports section. I repeat my offer to fax out illustrative charts. Bill Ballard RPT "I'll play it, 'n tell you what it is later" NH Chapter Miles Davis..........
This PTG archive page provided courtesy of Moy Piano Service, LLC