© Brookside Press LLC, 2017. Unauthorized use and/or duplication of this material without express and written permission from the author is strictly prohibited. Excerpts and links may be used, provided that full and clear credit is given to Brookside Press LLC with appropriate and specific direction to the original content.
If you are looking for ADS FOR PIANO-RELATED PRODUCTS AND SERVICES and don't see any here, please turn off the AD-BLOCKING FEATURE in your web browser, or list www.pianobuyer.com as an EXCEPTION.
When we began work on the four-octave keyboards for John Eaton, we at once became aware of how pampered we had been by the Yamaha. Purchased from Pratt-Read in the late 1970s, these were no-frills, generic organ keyboards designed to be adapted by each manufacturer for its own use. (They were actually five-octave keyboards from which we removed the lowest octave to make room for a large touch plate used for making global changes in sound, sort of like a pitch-bend wheel on a synthesizer, but operating in the same manner as a touch-sensitive keytop sensor.) In contrast to the precision and uniformity of the Yamaha, however, the Pratt-Read keyboards were a mess. Warpage of the keys and misalignment of the keyframe pins made even removal and replacement of the keys difficult. Whereas throughout most of the project I took my directions from Moog, here he made it clear that I was on my own — he could be of no help to me. I really felt put to the test. Miraculously, after several hours of furious pin bending and key sanding, I was able to create four fully functioning keyboards, but it had been like taming a team of wild horses. On my side was the fact that, among the four keyboards, there were four octaves of unused keys that could be scavenged for replacements. Unlike piano keys, organ keys are straight, so any natural key in good condition could serve as a replacement for any other natural of the same note name, and sharps were completely interchangeable.
Closeup of hand-shaped keytops made from circuit boards, on four-octave MTS keyboard.Photo Credit: William LaVista
Once the keyboards were functioning, I proceeded to perform much the same operations on them as I had on the Yamaha, but with some differences. Since I could easily remove the plastic keytops without removing the fronts, I did so (actually, these keys didn’t have fronts per se, but rather beveled undersides). I also removed the plastic sharp tops and replaced them with ones of a better size and shape for our purposes. Then I glued the sensor keytops directly to the wooden keys of the naturals or to the plastic sharps in the usual fashion, with ribbon cables attached, and filed and shaped the nearly 200 of them to match the shape of the surfaces on which they were mounted. Since these keyboards did not need to resemble pianos, the naturals were left the black color of the resistive paint rather than being painted white. The vertical-position sensors and circuit boards were installed in much the same manner as on the Yamaha.
Completed four-octave MTS keyboard in its walnut cabinet.Photo Credit: William LaVista
Moog had already built attractive walnut cabinets to house the four-octave keyboards, leaving room for the circuitry at the rear. Therefore no major surgery was necessary other than to cut off the keys a few inches behind the balance point, the rear of the keys serving no purpose. The ribbon cables were arched over the balance point to the scanning circuitry at the back, leaving the key bushings available for servicing should that ever be necessary.
Four-octave MTS keyboard with cabinet uncovered.Photo Credit: William LaVista
Since there was no action, as on a piano, to hold down what little there was of the rear of the keys and, therefore, to hold up the front, it was necessary to devise a spring-loaded mechanism. The springs had to be strong because they would be acting on an extremely short key-lever arm, but they also had to be very small to fit in the space between the key ends and the circuitry. After a few tries, Moog found some short, stiff springs that fit the bill, and I attached them to the keys with screw eyes. The springs were actually a little too strong, but some experimentation taught me just how much I had to deform them to get them about right. Thereafter, just turning the screw eyes provided the fine adjustment. Frankly, this spring system didn’t seem particularly elegant to me, but it did the job reasonably well.
Closeup of spring-loaded key-return system for four-octave MTS keyboard.Photo Credit: William LaVista
I balanced the spring-loaded keys with standard piano technician’s gram weights to a weight of about 64 grams, rather than the usual 50 or so grams found on a piano with the dampers disengaged. When balanced to 50 grams, these MTS keys felt terribly insubstantial, probably due to the extremely low inertia caused by the absence of a hammer action. A weight of 64 grams was similar to that of some other electronic keyboards I measured, and to that of piano keys with the dampers engaged. Since we had no idea just how these keyboards were going to be used, our principal aim was to make the touch uniform from note to note, and reasonably consistent with the touch a player of electronic keyboards would expect.
Because the spring-loaded keys exerted considerable upward pressure at the front, we used a key-stop rail to prevent the keys from jumping off their pins and to adjust the key height. Wooden rails turned out to not be strong enough to avoid bowing in the middle, so Moog turned up some lengths of heavy iron bar that, though clearly overkill, served nicely. We mounted each bar on threaded rod so that it could be adjusted, and padded it with felt where the keys would rest. Fine adjustments to the key height were made by affixing punchings (thin paper or cardboard spacers) to the keys where they contacted the felt. The key dip was regulated to be about the same as that of a piano.
The foregoing account probably gives the impression that, other than a few problems, the construction process went rather smoothly and in a rational sequence. Not so. Actually, we frequently switched our work from one keyboard to another, sometimes for reasons of efficiency and sometimes because Moog had not yet worked out certain design details or the solution to some problem. Sometimes we switched just so we could honestly assure each of Moog’s two anxious customers that his keyboard was receiving our undivided attention. This was especially important because, when we began working, the project was already several years behind schedule.
Each step of the way, too, required some experimentation and its attendant failures before we could proceed. For example, our earliest days were spent experimenting with adhesives. We were looking for a type with which we could glue the vinyl ribbon cables to the wooden keys so that the cables could be removed for servicing the keys, then just stuck back down without regluing. After many unsuccessful attempts, we ended up bending and routing the cables so that gluing was largely unnecessary.
Another problem was the clamping of glued surfaces. The keytop sensors, in particular, had to be lined up very precisely with the keys during the gluing process. The clamps we were using at the beginning applied pressure unevenly, causing the keytops to move. Attempts to correct the positioning while the epoxy was drying often resulted in bad glue joints that sometimes were not apparent until after I’d spent hours trimming and filing the keytops to shape. At such times, I tended to sulk, but Moog, ever inventive, rose to the challenge by building jigs that utilized clamps of the type found in industrial mass-production applications, largely solving our problem.
The most serious challenge we encountered was in the installation of a system for sensing aftertouch pressure. Moog had received a license to use a patented system of force-sensitive resistors consisting of a circuit board with a conductive pattern printed on it, loosely overlaid with a resistive film. When no pressure was applied to the film, the resistance between film and conductive pattern was high and no current would flow. When a piece of felt at the bottom of a key would press down on the film, however, the film was supposed to contact the conductive pattern on the circuit board over an area roughly proportional to the amount of force applied, and the current to flow in like proportion. Building this system required a tremendous amount of detailed, intricate wiring and soldering.
Unfortunately, after installing and testing this system in several keyboards, we came to the conclusion that it was highly unreliable for this application. Applying the same amount of pressure would not always yield the same or even a similar amount of current flow, and sudden, unpredictable current spikes were frequent. Moog decided to change the system from one based on resistance to one based on capacitance. He removed the resistive film entirely and covered the circuit boards with a thin Mylar tape to act as the insulating dielectric. The piece of felt at the bottom of each key was replaced with a piece of electrically conductive rubber, to act as the other plate of the capacitor. When the rubber was forced down on the Mylar-covered board, it would expand slightly in surface area and would electrically interact more strongly with the conductive pattern on the circuit board below. This difference would be picked up by the scanning circuitry we installed, and, like the other data being scanned, would be turned into musically usable information. The new system worked like a charm — but now, of course, I had to rip out all the old wiring and replace it with new.
During many of our Saturday-afternoon sessions, Moog and I worked together, he at one workbench and I at another. Over time we arrived at a satisfactory division of labor. Usually he operated the power tools, assisted me with testing, and performed general troubleshooting. I did the operations requiring good hand-eye coordination: soldering, gluing, and shaping of keytops. Occasionally we invaded each other's territory — I on the drill press and he with the soldering gun. Of course, Moog did all the electronics design and provided overall direction for the project; I did all the piano-technician work.
Moog taught me what I needed to know about electronic assembly: how to make good solder joints, how to identify resistor values, and so forth. He also tried a few times to teach me basic electronics, but despite my professed interest in the subject, I proved quite dense (I’d probably breathed in too many solder fumes). Moog did not ask to be taught how to regulate keyboards.
Hundreds of electronic components had to be hand-soldered to dozens of circuit boards.Photo Credit: William LaVista
Occasionally our work together was punctuated by other interesting projects. Once Moog took on a short-term consulting job for a company that made stenograph machines for court stenographers, and was researching the feasibility of producing a digital stenograph machine. For a few days, Moog put my key-making talents to use making typewriter keys. On another occasion, in his capacity as a Kurzweil executive, Moog hired me to critique the serviceability by piano technicians of a digital piano Kurzweil was developing. He also kindly took a few days out to read the completed manuscript of The Piano Book and give me valuable feedback.
To comment on any article, go to Larry's Blog.