Action Elevations

[Phillip Ford]

At 07:52 AM 1/20/03 , you wrote:
>Bill, Dave Love, Ron... anyone else who might have insights and
>thoughts...
>
>I've been re-reading Hohfs article of a couple years back and am
>wondering about the problem he sketches relating to the need for the
>string plane to be paralell with the key bed.

What's the need?  The hammers don't 'see' a plane.  Each hammer 'sees' one point on the string.  The sum of these points define a line (assuming the strings all lie in a plane).  I can see that it would be advantageous for this line to be parallel to the keybed so that bore distances for the hammers would be the same, but the orientation of the string plane to the keybed plane seems irrelevant to me.

>  One of the points made is
>that "most actions in common use today are designed to swing the
>hammershank up to the point where it is parallel to the keybed" and a
>statement following about the decrease in action efficiency when the
>shank is not paralell to the key bed. In addition we want the hammer to
>strike the strings at 90 degrees.

I don't see that the position of the hammershank is of much relevance.  I can see that the direction of travel of the hammer contact point and the jack-knuckle contact point is important and these are dependent on the hammershank, but the location of these points and their direction of travel could be the same without the hammershank being parallel to the keybed.  I think the idea that there is a decrease in efficiency if the hammershank is not 90 degrees to the hammer and not parallel to the keybed comes from the mistaken notion that if the hammer is perpendicular to the string at contact then the contact point is moving perpendicular to the string. It's not.  It's moving perpendicular to a line between the hammer center and the contact point.  Perhaps it would be most efficient for the shank to be perpendicular to the direction of motion of the hammer contact point.  It seems to me that this would minimize shank bending.  This might sound good in theory but wouldn't work in practice.
I believe that the hammer needs to strike at 90 degrees to present a symmetrical shape to the string and to ensure that the striking is done on the part of the hammer that's designed to do the striking.  However, the act of striking causes the hammer and shank to flex and to slide along the string to some extent (or roll along the string if you prefer).  So, its angle at start of contact is probably not its angle at end of contact.  Perhaps we should be raking the hammer so that it ends up being at 90 degrees to the string when it's halfway through its time of contact, if we could figure out what that is.  I think this would result in hammers raked at something less than 90 degrees.



>So first... if you have to choose the lesser of two evils... what would
>you say is the right choice to make ? Keeping the hammer shank
>centerline parallel to the keybed at strike distance or hitting the
>strings at 90 degrees ?


If these were my only choices I would definitely opt for the 90 degrees, although as I said above 90 may not be the right number.


>Another question I am wondering about....
>
>If the center spread is to be veiwed as a fixed value, and bore distance
>is fixed..doesnt that sort of dictate the height of the hammer shank and
>whippen centers over the keybed ?
>
>
>Thanks for your thoughts
>
>RicB

If bore distance is fixed and we want the shank to be 90 degrees to the hammer and hammer to be 90 degrees to the string, then the position of the hammer shank or hammer center is fixed relative to the strike line.  The location of the keybed would be determined by the height of the action stack that you wanted to use.  Given a fixed spread then the whippen center has to lie on a circle about the hammer center of radius equal to the spread.  Its up and down position is indeterminate.  There are some other considerations.  If it was too high then it would interfere with the hammershanks or hammers in the rest position.  So there's some upper limit.  But there's no real lower limit (other than the lower limit of the 
circle described above).  Once again it would be determined by the height of the action stack that you chose to use and where exactly you wanted to locate the whippen center.  It seems to me that you would want the whippen center to be as high as possible.  The ideal from a power transfer standpoint would be to have the line from whippen center to hammer center parallel to the string so that the jack-knuckle contact point is moving perpendicular to the string at some point in its travel.  This would be impossible with an action of current design.  So, as high as practically possible might be the best location for the whippen center.

Phil F



Phillip Ford
Piano Service & Restoration
1777 Yosemite Ave - 215
San Francisco, CA  94124