Ed M< There are people who have heard before and after examples.
I'm going to move the "In-Situ" thread sideways.
I haven't heard one of Ed's piano's yet. However I have tested, first hand, at least the practical application of some of his ideas. Based on evidence I gathered from a preliminary implementation of portions of the FTDS on my personal experiment piano, I felt I had enough data, at least in my own shop, to explore his ideas further. My experiments were practical applications of the ideas, so, they, of course lack the rigor of scientific investigation. My goal, is practical: to create a clear, clean, well-sustained, musically useful, easy to tune high treble which has clear pitch content. This especially from #69 all the way to #88.
This experiment was performed on the 1st piano I ever rebuilt...rebuilt it a while ago... a Chickering 122 (5') c1901. It is currently my personal at-home piano...I play it daily.
As a first time rebuild, the high treble was virtually un-tunable. Ridiculous Delignit pin torque coupled with, to me, un-readable rendering. The front termination angle is 20-22 degrees, coming off of a Chickering half-agraffe. Shortest possible duplex length was in theory maybe 35mm, but given the shape of the casting ended up 40+mm, so it was irritatingly noisy to boot. The duplexes' 1st terminating contact with the plate was felt covered, and then, in order to get the strings off the felt, a brass counter-bearing bar inserted between the duplex termination and the pin to achieve appropriate string height to the pin. This counter-bearing bar was also felt covered. There was a very long segment between the counterbearing bar and pins, the pins being snugged up close to the stretcher end of the plate.
An excellent configuration for difficult tuning and duplex noise. Also, it was noisy in the falseness department, and high treble sustain was poor...in other words, a great candidate for a further experiment.
Keep in mind this post is about changes I made to notes #69-88.
During the process of the experiment, strings were installed and un-installed multiple times, so the final pin torque ended up quite nice, solving that problem collaterally.
I did not set out to test Ed Mcmorrow's ideas in this experiment. The decision, to go with elements of his Fully Tempered patent was, as I will describe, incremental, following unsatisfactory initial fixes.
The first attempt at improvement in this top section was mostly a falseness fix and a rendering fix. I removed the front pins, and epoxied in WNG #7's. Then I removed all front segment bearing felt, and inserted a counter-bearing bar at the closest possible point relative to the speaking termination...35mm in theory. However, at 35mm the bar was right on the edge of the sloping plate edge, and the half-round brass, under string tension, kept being pulled off the plate. So I had to move the brass back to a strong 40-ish mm duplex. Two brass half-rounds were used as counterbearing bars. This counterbearing setup was similar to the original unsatisfactory setup except all felt bearing was eliminated. As in the pre-experiment conditions, the 2nd half-round was required to achieve adequate string height/angle at the tuning pin.
This duplex retained all the pre-experiment front segment noise and rendered just as poorly as before, even with the removal of all felt. To me, it felt identical to the previous counterbearing setup...very hard to read the front segment.
So I decided to flex. Having been mentored in short duplexes to avoid duplex noises, and having come to this work with my own design biases which favored short front segments, my first inclination was to go the short duplex route. However, since the existing plate limited my options to strong 40-ishmm duplex lengths, I figured it made sense to go with choices that the plate had already made for me. I decided to go with longer duplex lengths and work with them as a feature, rather than try unsuccessfully to achieve a short duplex...hence playing with some of the ideas in Ed Mcmorrow's patent made sense to me.
I was initially drawn to the ideas by the co-polymer counterbearing bars which I had been contemplating on my own when I read that Ed had already done it successfully...hence the thread title.
Here's what I did in the second experiment:
-Using his strike point ratios, re-hung hammers altering the strike points from #69-88
-shaping the hammers, I brought the hammer apex to a point right at the actual empirically confirmed string contact point, rather than simply lining up a perpendicular line on the hammer molding to the theoretical strike point. (custom bored hammers)
-hammer mass was already quite low, but reduced it further
-shaped the ½ agraffes to a "V" with .5mm flat land
-installed co-polymer duplex terminations based on graduated ratios of speaking length to duplex length as stated in the patent application
-cut the co-polymer to try and hit non-harmonic frequencies (hit one harmonic ratio by mistake...oops)
-eliminated the 2nd counterbearing bar, since the width of the co-polymer effectively raised the string to a decent take-off angle for the tuning pin.
-Took the takeoff on the pin down to 1/8" or so off the plate (no bushings)
-doused the Bacon felt hammers with 1:7 lacquer/acetone #69-88...1 full hammer soaking(didn't need this down to 69...could have been more conservative)
First attempt at bringing the co-polymer up to tension resulted in the plastic slipping away from my target location, so I staked the co-polymer down with 3/32" split pins. The pinned co-polymer stayed put nicely.
Results:
-rendering excellent, despite 22 deg termination angle
-speaking lengths and front segments very easy to read, and tuning is stable. Ease of tuning has gone from one of the hardest high trebles I tune(make that try to tune), to one of the easiest high trebles I tune. High treble clarity greatly improved, but room for more improvement. Even so, it is way beyond anything I service in the clarity department.
-no duplex noise, with the exception of 1 note where my duplex ratio mistakenly hit a multi-octave frequency right on the nose (oops)...I will alter that unison's ratio when I'm not so darn busy...the fix is not a biggy.
-musically, the high treble (#69-88) sustain and pitch clarity significantly improved to such a degree that as I improvise, I spend much more time up in the high treble playing off the high treble sustained audible pitches, rather than just using the high treble to finish out full keyboard arpeggios with pitch-flavored hammer knocks. It is really fun to have this sustain up in that 69-88 section. No pianos I service have high trebles with this kind of musically useable sustain...none.
-falseness was improved, but not eliminated. Several of the false strings were replaced with no improvement. This falseness does not get in the way of tuning or musicality in #69-88, but it remains there in a greatly reduced fashion
Overall: A very informative and piano improving, ultimately musical exercise. On the strength of this experiment, I purchased a single piano FTDS license and am incorporating the ideas into a Chickering 123 re-manufacture which will be making music late September. I will post sound samples when the piano is done.
One final note: This experiment addressed the high treble only and not the low treble (low treble: #49-68 in this piano). My boards are working quite nicely in the low treble & killer octave, based on the impedance model soundboard design (aka Nossaman and Fandrich), so I didn't feel the need to mess with this particular area. My interest in FTDS ideas at this time, is inspired by improvements in the top capo section, but I am, for smoothness of transition, incorporating the ideas through the first capo section as well.
Jim Ialeggio
-------------------------------------------
Jim Ialeggio
grandpianosolutions.com
Shirley, MA
978 425-9026
-------------------------------------------