Action Geometry

[Richard Brekne]

Farrell wrote:

> I gotta problem. Well, actually lots, but one immediate piano related
> problem.
>
> My  analysis shows that my action stack is set too low.

Could you expound a bit as to why you think the action is to low, or are the
symptoms you describe below your reasoning.

> My hammers appear to be over-centering, the
> capstan/heel cushion contact lies below the
> "magic line" at key rest,

It should, and at half blow it should intersect, and at full blow it should be
as far above as it was below at keyrest.

> and just barely touches it at full key depression
> (so it is arguably on the low side),

Grin... something wrong alright.... could indicate too narrow a spread perhaps..
but then that depends...

> the knuckle/rep lever contact lies
> about 3 mm below the line between wip center and hammer center and full key
> depression,

havent used this line yet so I dont know what to say about that.

> and the perpendicular from the knuckle wood to the jack center
> lies about 4 mm aft of the jack center (on the B+ side) - which is good that
> it is on the B+ side, but I wonder if this is too far.

Sure is too far.. this line should be perpendicular,, or just barely on the B+
side... prerequisites proper hammer to string distance and I find it helps to
adjust the jack itself to the knuckle core in looking at this line. This line is
checked with hammer at rest.

> Also, sighting down
> the action rails, one can clearly see that the middle portion of the stack
> sags lower than the far bass and hi treble ends.

That dont sound good at all.. would make things very difficult... grin.

> I placed hardwood shims under the stack feet to bring the hammer rail up to
> a elevation where the shanks would be parallel (or nearly so) to the
> strings. I put about 2mm at bass end, 3mm at bass break, 2mm at tenor/treble
> break, 1mm at treble/hi-treble break, and no shim at high treble end.
>
> Sight down rails now and all is very straight. Capstan/wippen heel contact
> point appears unchanged :-( , perpendicular from the knuckle wood to the
> jack center is improved and now lies about 1 or 2 mm aft of the jack center
> (on the B+ side). BUT, Downweight  in the most-raised section went up 2
> grams (from 45 to 47) and Upweight went down in same section 2 grams (from
> 26 to 24). Bass and treble had smaller changes, and Up and Down weight were
> the unchanged in hi-treble (where no elevation change occurred).

Are you making sure you are re-regulating when making each change before taking
measurements of any and all sorts... I made this mistake a bunch when I started
doing all this.. ended up scratching my head nearly hairless before I figured
out what I was doing wrong.

>
> Now what does all this say? I would think my geometry improved, although
> maybe I didn't change action geometry much, but rather only the angle at
> which the hammer strikes the string. Do I have to pay for better
> string-striking angle with action efficiency? My thoughts are now turning
> toward considering moving the capstan in toward the center pin to improve
> its relationship to the "magic line".

Before doing this...make darn sure your top action geometry is correct. Seems to
me that the top action can really be regulated and geometrically set independant
of the keys.. you need a few jigs to do this but it can be done easily enough.
When this is done then the capstan / whippen cushion positions in relationship
to the magic line can be dealt with at the same time as you decide key ratio.
This is basically what I did on the Duysen I am working on and it worked great.

> The bottom horizontal piece of wood of the wippen (what is it called?) to
> which the heel is attached is nice and horizontal on my Boston. The keys are
> nearly horizontal. The capstan is 90 degrees to the key. But my wippen heel
> cushion is not horizontal, but rather it slopes upward toward the pianist.
> As such, the heel cushion contacts only the back portion of the capstan and
> this relationship is exasterbated through the downstroke of the key. Is this
> a bad thing?

>From a friction standpoint, and I believe from a leverage standpoint.. the
capstan /cushion should act as nearly like to a pivot as is possible. For
example... if at half key stroke you have capstan at 90 degrees from both the
key and the whippen cushion horizontal line.

> I know that if I decide to move the capstan line I may need to
> do something with the wippens. If my heel are all funky because of their
> wierd angle, maybe that will help me decide whether to move the capstans if
> other info suggests that would be good to do. How important is it for the
> capstan/wip heel to cross the magic line at half-keystroke?

I have begun to ask this question myself... I believe I understand that it can
be a friction issue... but I have a hard time understanding this as a probem
otherwise. Seems far more important to assure the hammershank perpendicular to
jack center at key rest. The key must move the whippen assembly a certain
distance for any given key dip, and this can be achieved several places along
the key itself.... its just a matter of assuring enough total length of the
whippen cushion and capstan height combined. Still I cant help thinking  there
must be an optimal for this placement..So on my Duysen I placed this such that
the contact point between the two at half key dip was on the magic line I
established by the spread, and that I had a total whippen balance weight (ala
Stanwood) of just over 9.0. (key ratio ended up at .56)

>
> Eyegh, yie, yie. What a nightmare - but I sure am learning - I think. ;-)  I
> started this crap at 4 am this morning. Maybe I need to get a life.

yea... but its fun aint it ?? grin.. Hope you get some responses from some folks
who know more then I, as I am still very much an egg in these regards as well.

>
>
> Terry Farrell
> Piano Tuning & Service
> Tampa, Florida
> mfarrel2@tampabay.rr.com

--
Richard Brekne
RPT, N.P.T.F.
Bergen, Norway