xen keyboards 2

I wrote: "Do you know if the force required to move the keys increases
towards the back rows?"

For those of you not familiar with this, the Janko keyboard, being
originally designed for acoustic instruments in 12tET, has keys sharing the
same internal lever and strings (as Manuel explained). This causes the key
travel to be longer and easier in the rows nearer to the performer, and
shorter and harder in the rows farther from him. I think it can be agreed
that this is undesirable. The makers of Janko pianos never solved the
problem.

Norman Henry has suggested simply making all the levers very long compared
to the difference in key contact points, so that the difference between
rows will be minimal. That this can work without requiring unreasonably
deep key travel for all the rows remains to be seen.

I have suggested the following arrangement...

............string
key key ( )
----- ----- ( | )
| | (|)
---------------- | <--hammer
| |
--------------------------------------------------------- key lever
^ ^
branch point fulcrum


But the question is, even assuming that we can make this branching bit
absolutely rigid (which would be difficult in the least), will the force on
each key be...

(a) the same, as if each were attached at the branch point (it works).
(b) different, as if each were attached at the point directly beneath it.
(c) half and half -- the force is triangulated, or something.

Norman Henry and Brian McLaren both think (a), and this seems to me the
intuitive choice. But I have been skeptical of intuition in such matters
ever since I learnt that the torque on a screw can be changed by changing
the length of the screwdriver (anyone who would care to explain that,
please mail me at clumma@nni.com).

Either way, can anyone come up with a different solution? It would be
worth a great deal, I think. Might even get one in the history books (I
think we're all already there anyway, but... :~)


Carl

> * As far as acronyms go=85 (E)quidistant (D)ivision of the (O)ctave=92s=
=97 EDO, is
> certainly seems to me a rather colorless creation (immanently indispose=
d as
> it is to the oddly resonant charm of (T)one (E)qual (T)emperament=92s=97=
TET),

My immediate take would be that the two are not exactly synonymous, by=
which
I mean "equidistant divisions of an octave" and "equal temperament", the
distinguishing case being nonoctave equal temperaments. Carlos' Alpha, f=
or
example, would be an equal temperament, but no particular subdivision of =
the
octave. Well, not an integer division anyway, and if not integer, then t=
hat
nomenclature becomes inconvenient.