Some thoughts on setting touchweight.

[Delwin D Fandrich]

Ron Torrella wrote:

> . . . .
>
> Can't move capstans any more than 3/16" without the capstan contacting
> wood. 1/8" would be fine, except it means drilling out the capstan holes
> *oversized*, plugging and redrilling.  If I didn't oversize the old holes,
> drilling new ones would be straddling old and new wood.  If I used pine
> plugs (not dowels....plugs cut from scrapped keyboards), I *may* be able
> to get away with "normal"  sized plugs, but there's still a glue joint to
> deal with. (Have I started splitting hairs yet?)
>
> I'm contemplating the option of a combination of things; move the capstans
> no more than 3/16" *and* swap small keyleads for medium or large ones. The
> net result would be a lighter touch (improved leverage), but a slightly
> heavier keyboard (on the order of 150-200grams or 5-7 lbs, overall, is a
> guess). . . .

------------------------------------------------

Ron,

Although touchweight and action geometry are related, they are not the same
thing.  Touchweight is the result of some fixed action geometry (that is, a
predetermined set of levers working through predetermined and fixed ratios), the
accumulated mass of the various parts involved and the accumulated friction of
the various sliding and rotating centers.  Ideally, the action geometry and
specifications are established by some action designer somewhere and this design
is faithfully followed by the action manufacturer.  In this best of all possible
worlds, both static and dynamic touchweight are predetermined by the designer.
The action parts -- including the keys -- will all be built to the specification
laid down by the designer.  Action centers will have some predetermined amount
-- or range -- of friction. A predetermined number of key leads will be inserted
into predetermined locations in each key. Etc. Etc. Etc.

When actions are not built to these standards, there are problems.  Sometimes
those problems are lever related.  It will be impossible to establish the
'proper' relationship between the key travel and the hammer travel.  Sometimes
the action will appear to be functioning normally, but the piano will very
difficult to play.  The pianist will complain of a 'heavy' touch even though you
may be able to verify that the down weight is right at 52 to 54 grams.

Without getting too involved, let me just separate static touchweight from
dynamic touchweight.  Static touchweight involves only the weight of the parts
involved, dynamic touchweight involves the mass of the parts involved.  Static
touchweight is what is measured by balancing the key and action assembly with a
precise mass.  Dynamic touchweight is what the pianist feels when the piano is
actually being played.  Dynamic touchweight includes the inertia of all of the
moving parts involved.

The action you are working with could be given the 'proper' amount of 'static'
touchweight simply by adding more mass to the front of the keys.  And this was
generally the recommended solution in the 60's, 70's and 80's.  At the same
time, of course, this made an already problematic action even more difficult to
play.  What was happening -- at least for the most part -- was that the action
geometry had varied from the way it had been designed.  And the resulting static
touchweight problems were being 'corrected' by adding lead to the keys.

The first step with any of these action problems is to verify that the action
geometry is correct -- or at least is within a functionally workable range.  The
device I described is one I used to make the action geometry functional on new
pianos without replacing a bunch of parts.  At the time we did not have the
option of replacing hammershanks and/or wippens.  Nor did we have the luxury of
relocating knuckles.  The problem was generally one of being able to measuring
'proper' static touchweight, but having an action that was 'very' difficult to
play due to the large number of leads installed in the keys.  A secondary
problem was the relationship between hammer travel, key travel and aftertouch.
The simple solution was to relocate the capstans.

This was done by using the device I described earlier to establish a lever ratio
between the key and the hammer that would result in a hammer travel of 44 mm, a
key travel of 10 mm and an aftertouch of 1.25 to 1.5 mm.  Once this relationship
was established, the Teflon bushed (hammershank) action centers were reamed
and/or pinned for a friction of between 4 and 7 grams of friction.  Following
this the action was carefully regulated and then it was finally weighed off.
The usual result would be to remove quite a few leads from the keys.  Typically
-- depending on the model -- we would go from 7 or 8 leads in the low bass to 4
or 5.

This whole issue illustrates the problem of individually weighing off each set
of keys as determined by the needs of a specific action and keyset combination.
This practice uses lead to compensate for variations in action geometry and
action center friction.  When this is done, even though the action may meet the
static touchweight specifications of the manufacturer, it may be a very
difficult piano to actually play.  There can be -- and very often is -- a great
deal of variation in the touch and action response from one piano to the next.
A much better practice -- at least in my view -- is to lead the keys to an
engineered standard.  Then when it comes time to check static touchweight, it
will be just that: a check.  Now it becomes a verification that the action is
functioning correctly.  If the action is not functioning correctly, it will give
the operator clues as to what is wrong with the action and what needs to be
corrected before the piano is shipped.  Now, please, don't somebody try to tell
me that actions cannot be properly manufactured to this standard.  They can be,
and it is being done. In my view, this 'feature' -- the need to weigh off each
action individually -- is not one to be proud of.  It is an indication of a
problem, or a flaw, within the manufacturing system that needs to be designed
out.

Regards,

Del