Key Ratio

[stanwood]

Dennis Johnson:

>I must also clarify that I understood "place temporary weights on
>the back of the key so that the balance weight of key and wippen is zero"
>to mean that key should balance at rest in a half-way position.

Two ways to check that the balance weight of the key/wippen is zero:

1. Bounce the key gently down and see that it bounces up to halfway
position, then bounce the key up and see that it bounces down to a halfway
position.

2. Take a gram guage to the front of the key and move the key in both
directions with the quage.  The force readings should be the same.


>This is what I got:

>Note            27              49              Average
____________________________________________________________

>Old Parts       5.64            5.86            5.75  (w/old hammer 6.9g)

>Renner N.Y.     6.0             6.06            6.03  (w/light hammer 6.0g)

>Renner Hamburg  5.2             5.26            5.23  (w/light hammer 6.0g)

>Renner Hamburg  5.2             5.2             5.20  (w/heavy hammer 8.4g)

>Imagine my surprise at the same numbers on Hamburg parts with a
>2.4g heavier hammer!

Does strike balance weight increase when strike weight increases?
Of course it does!  In this case it didn't seem to.  I've had this kind of
result in some cases and its an area that deserves the kind of study that I
can only do on the fly, in the process of doing work.  But surely we all
notice that there tend to be more key leads in the bass end of pianos where
hammers are heavier.  Part of the equation is the up/down weight measurement
which is crude and governed by the vagaries of friction.  Also the balance
weight value is based on the assumption that friction is equal in either
direction.  I'm sure there are variations in this assumption.

I will say this: From a statistical standpoint you don't have near enough
data to draw any kind of conclusions in this particular case.  When we (my
group of associates) look at touchweight data we never draw conclusions from
less than 6 data points for preliminary conclusions and the strongest
conclusions are drawn from 88.

>Is Strike ratio more relevant to leverage
>than to hammer weight?

Of course!

I can say this very strongly:

There is no doubt in my mind, and my experience bears this out: when actions
are made to have similar strikeWt, Strike Ratio, balance weight, and
friction weight, they all feel the same regardless of the key ratio.

By the way, I find that, with post 84 Hamburg parts, if the key ratio
(weight ratio) is from .51-.53 the strike ratio will be close to 5.5.  No
need to alter the capstan and heel.

I've moved capstan lines as much as 1/2 inch to correct for high key ratio
and the action feels the same as actions with the right key ratio.


>Please correct any misunderstanding but this statement lead me to
>believe that you would not automatically change underleveraged key ratios
>if the strike ratio came out acceptable anyway, presumably by using Hamburg
>dimension parts.

I do move the capstan if the strike ratio isn't where I want it, regardless
of the type of shank I'm using.  Remember that moving the capstan to change
key ratio also changes the wip ratio.  What counts in the end is the product
of all three ratios (shank, wip, and key).


David C. Stanwood


P.S. - For clarity I'm posting here below the text of my previous post
defining key ratio:


I recommend the following concise definition of standard
key ratio:

The shortest horizontal distance between the capstan line
and the balance line divided by the shortest horizontal
distance between the balance line and the touchweight
measuring point.

Where, with the keyframe sitting on the horizontal plane,
the capstan line is defined the line of intersection
between the horizontal plane and a vertical plane which
intersects the capstan/wippen contact point.  The balance
line is defined as the line of intersection between the
horizontal plane and a vertical plane intersecting the key
balance point (center of the balance hole at the bottom of
the key).  The touchweight measuring point is defined as
the line of intersection between the horizontal plane and
the vertical plane intersecting the touchweight measuring
point.  The touchweight measuring point is defined as the
point on the front top surface which intersects a vertical
line through the center of gravity of touchweights placed
on the front of the key for the purpose of measuring
touchweight.

For the measuring point, we use a point 13mm in from the
front edge of the keys.

Key ratios by this definition usually range from .48 - .56

Sorry about the long winded definition.  Just imagine
your reading a patent.


David C. Stanwood