Shanks parallel to strings

[Mark Davidson]

>>When hitting the string, we could consider two points : the point of first
>>contact between hammer and string, and the point of largest excursion of
>>the hammer accompanying the string at it's largest deflexion from rest
>>position before rebounding, at loudest playing power (this will be very
>>different in the bass than in the trebble).  I believe that for maximum
>>power transfer between the hammer and the string, the movement of the
>>hammer should be not really perpendicular at contact time to the plane of
>>the resting string, but to a plane a bit higher (higher of something
>>approaching half the distance between the two points defined
>>earlier).  This, because of the cosinus coefficient in the vectorial
>>force.  Let us call this third very point the optimal quasi tangential
>>point of intersection between hammer arc and quasi perpendiculat to
>>largest portion of string plane when at approximately 1/2 of it's  largest
>>deflection from rest position at loudest power playing
>>(OQTPIBHAQPLPSPA1/2LDRPLPP).
>
>You leave me speechless.
>
>Phil Ford

I think I understand what Stéphane is trying to say.  Imagine the piano
hammer is a bowling ball, and you drop it on a trampoline (get
some other sports in here besides baseball).  There is a point where
the ball first touches (his first contact), and a point where the ball
is stopped before reversing direction (maximum excursion - hammer
compressed and string deflected).

Halfway in between these two points is where the hammer should
be perpendicular to the string, since that is the vertical "center"
of the impact. This makes the average hammer to string angle
be 90 degrees.

But do you optimize for hard blows or soft...
and what happens if the floor's not level?

-Mark