Optimizing repetition

[FSSturm@aol.com]

A couple weeks ago, I tuned for Alicia Dellarocha (did I spell that right?)
when she was here in Albuquerque, on tour with the Cincinnati symphony. This
was one of those no frills appearances - load in at 3:00, pm, Dellarocha came
to try out the piano at about 6:15, the orchestra bus appeared at 6:45,
curtain at 7:30. I was there at 6:30 to deal with any piano problems, touch
up tune, or whatever.

Dellarocha's only complaint was that repetition in the octave 5-6 area was
not rapid and reliable enough. She demonstrated to me what the problem was.
The concerto was the Ravel two-hand, which has some rapid triple repetitions.
The technique she used was a smooth drop of the wrist, with a 4-3-2 finger
action, faster than any repetition I have ever seen. Each finger slid off the
front of the key to allow it to rise slightly before the next finger struck.
The third blow was held down.

This is a technique which requires extremely precise timing of the finger
actions, and is certainly at the upper limits of what is possible
mechanically on the best designed and regulated action. I am a fairly
accomplished pianist, and am able to duplicate what she was doing, but it is
very difficult to do this consistently and repeatedly. I don't doubt
Dellarocha's ability to be consistent, at least almost all the time. She has
performed this concerto many, many times on many pianos, and has recorded it
several times, so she can be relied on to know what a piano should be capable
of.

What she seemed to be finding was that on the third repetition, which was
held down, the key was not in contact with the front rail punching (as she
showed me more than explained). My guess was that the hammer was at that
point lower than its normal checking position, so the key was "hung up". Also
there seemed to be a complaint that the second repetition wasn't always
reliable/didn't sound. Time was, as mentioned, very short, and my examination
of the action found nothing out of the way (this is an instrument which gets
light use, and I maintain it regularly). All I was able to come up with was
to raise the check distance slightly, maybe as much as 1/8" closer to the
string. This seemed to me to produce a slight improvement, but there was
unfortunately no opportunity for feedback from her, and I heard no comment
later.

At any rate, as is usual in similar circumstances, I came to feel my
limitations when faced with the demands of truly top-notch artists (not
enough opportunities out here in our otherwise wonderful isolation). So I
thought I'd ask what approaches others might have to achieve best possible
repetition.

So as not to begin from absolute scratch, let's assume fairly standard
condition and regulation. Letoff closer to 1/32" than 1/16". Drop about
3/32". Check near as close as it can be without tail rubbing on backcheck
when the hammer movement is resisted by one hand while the other presses down
the key. Hammertail rough enough to check except on the softest pianissimo
blow (that perennial Steinway struggle between repetition spring tension and
tail/check distance). Repetition spring strong enough to raise the hammer
smoothly and rapidly (without hesitation) when released from check, though
not enough to jerk or bounce. Repetition lever set so that when you trip the
jack tender lightly with a finger, you can just feel the resistance of the
knuckle (friction), but the jack returns reliably all the way without hanging
up when released slowly by the finger. Jack position set to line up with
knuckle molding (and doesn't trip when you impede the movement of the
hammer). Blow distance, key dip, aftertouch set to fairly standard
measurements and relationships to one another - in the range of 1 7/8",
3/8"+, .050". Action centers free (no sluggishness). Knuckles brushed, Teflon
powdered, firm and round. Key bushings almost friction free.

With this much given, what changes and refinements might be made to optimize
repetition? What areas should be given particular attention? Is it possible
too little friction might adversely affect repetition? There are plenty of
other questions, but these will do to start. Any brilliant ideas out there?
(or simple, obvious solutions for that matter)

Many thanks
Fred Sturm
Albuquerque, NM