Hi Don I have to admit I hadnt gotten around to looking closely at this paper before you mentioned it yesterday. So I ran through it a couple times to be sure of what our friend Hans Velo has to say here and this is what I get... His measurement device and calculations rely on a single measurement for each sample key. That in turn is based on the 0.4 kg weight attached at about half the length of the measuring devices' main lever, so that works out to about 0.8 kg (800) grams at the key. The device lever is released and the 800 grams pushes the key down while simultaneously a pair of electric contacts are opened to register the start of the measurement. When the key has traveled 5 mm another pair of electrical contacts are tripped to register the end of the measurement. This provides him with the measured distance the key travels (5mm) and how fast it travels that distance in, which he plugs into a formula along with a factor representing the influence of the 0.4 kg weight. (the mass itself times the device levers ratio). The mass of the lever itself is also accounted for. Once he has this he simply solves for the amount of force needed to move the key at different speeds then the measurement speed by plugging in whatever speed he wants to calculate for. So... its rather a <<measurement>> by proxy if you will. He doesnt actually measure the force for these speeds directly. He could have actually done it the other way around, and solved for key speeds for any given force value... but thats not really the point. I am not able to confirm or deny the validity of this approach, so I'll let Mark or Sarah or maybe Phil to take care of that. But assuming his conclusions are correct, then his magnetic assist configuration does indeed influence the dynamic touch weight in way that needs a bit of explaining. There are at least two very interesting and direct consequences of his conclusions. The first is that the magnets influence the dynamics of the action in a similar way a change of the physical ratio does. (one of the claims I make for my own variant of the MBA btw). There are really only two ways I know of that this can happen really...either the ratio itself is indeed altered or the force that the magnets exert change significantly through the keystroke. Ideally, the magnets should exert as close to the same amount of force on the system throughout the whole keystroke as is possible, so I can only assume something about the presence of the magnets directly influence the ratio itself. On the other hand, you have an obvious inconsistency either way. You get neither a constant force equivalent of so and so many grams of lead, nor a consistent change in the ratio (if thats whats going on at all). In addition, there is a strange inconsistency in the way black keys are influenced as compared to white keys. My first inclination is that number one, such inconsistencies are not particularilly desirable from the pianists perspective. And number two... I am become a bit suspect about the validity of the <<measurement>> methodology. Remember... he only actually measures the speed and distance resulting from the device (and accompanying mass) itself moving the key... the rest is calculations based on the assumption that his formula is valid for both conditions... with and without magnets. Perhaps this functions fine when dealing with pure mass and leverage... but fails when magnetic fields or springs are employed. But like I say.. I dont know and will let others who have a better grasp of physics answer to that. That said... should his methodology be sound, one has to explain how exactly the magnets accomplish what they do, and one has to accept the inconsistencies they create in touch as reflected by these calculations as being acceptable from the pianists point of view. The former should be explainable (predictable) in terms of leverage math and magnetics physics I should think. The later is up to the players and we pretty much already know what they like IMHO. All of which brings me back to my original concern that any such assist mechanism should only be used to even out minimal amounts of BW discrepancies, and not compensate for significant degrees of mass. Hope this stayed reasonably clear and understandable... Cheers RicB Don wrote: >Hi Richard, > >Apparently the magnetic action does change more than just static weight. Or >so it is being claimed--if I understand what is on this website. You are >better equipped than I am to understand this. > >http://home.hetnet.nl/~velo68/internlinks/EPENG%20version2ws.htm > > >
This PTG archive page provided courtesy of Moy Piano Service, LLC