[pianotech] Hammer Cant Angle, was D Hammers

[pmc033 at earthlink.net]

George:
	This discussion is very helpful.  I haven't heard it explained before, but
sort of discovered it while trying to mate some bass strings.  I realized
that unless I filed the hammers on the piano at an angle (!) the level
strings weren't going to mate with them well.  
	You could probably illustrate the phenomenon by mounting a level (vial) on
the end of a hammer.  If the hammer weren't canted, as you raise it the
level wouldn't change.  Once you cant it, your vial is going to show the
level changing as you raise it.  Verrry interesting.
	Paul McCloud
	San Diego


> [Original Message]
> From: George F Emerson <pianoguru at cox.net>
> To: <pianotech at ptg.org>
> Date: 04/22/2010 10:54:02 PM
> Subject: Re: [pianotech] Hammer Cant Angle, was  D Hammers
>
> ----- Original Message ----- 
> From: "David Ilvedson" <ilvey at sbcglobal.net>
> To: <pianoguru at cox.net>
> Sent: Wednesday, April 21, 2010 1:16 PM
> Subject: Re: [pianotech] D Hammers
>
> >The first thing we did was checking traveling of the shanks and made a
few 
> >adjustments then it was
> >checking hammer movement between neighbors as you are talking about.
>
> Sounds like the same thing by a different means.  I would still what to
know 
> if the piano I was working on was originally set up with a strike point 
> scale different from the action scale.  You can determine this as I 
> previously described with measurements at the bass-tenor break.  I would 
> have some concern about adding this clearance cant on a piano that was
not 
> designed and built for it.  It would result in having to rotate hammer 
> flanges to center on the strings, bend the backcheck wires left to right, 
> and compromising wippen alignment.
>
> >We burned in by heating with an electric heat gun and bending towards
the 
> >direction the hammer was
> >moving...which seemed counter-intuitive but worked... '-]
>
> Yes, it works, but all of the hammers in the section have to have the
cant 
> for it to work.  It's pretty much the same throughout the bass, gradually 
> changing through the tenor, and none in the treble sections.
>
> >The hammers all had the cant you are describing.   So with this hammer
cant 
> >the hammers do come up
> >to "about" level with the strings..right?
>
> Not exactly level, but the gang sanding you described pretty much takes
care 
> of it.
>
> >> In the factory, we have a fixture that holds the hammer
> >>molding where the shank engages it, at the spacing of the "action
scale,"
> >>but also holds the crown of the hammers at the spacing of the "strike
> >>scale."   This predetermines a slight mounting angle in the bass and 
> >>tenor,
> >>and perfectly vertical in the treble sections.
> >PLEASE describe more...diagram/photo or whatever if this jig.
>
> It's not something that would be useful in the field, since it requires 
> interchangeable components, specific to each make and model.  Imagine two 
> giant combs, one to secure each hammer where  it engages the shank, and
the 
> other to secure the hammers near the crown.  The hammers are held above
the 
> fixture for gluing, then rotated down into the two slots, between the
teeth 
> of the combs, for drying.
>
> >I'm still having a hard time understanding the Physics of the
phenomenon. 
> >This has nothing to do shank
> >traveling but with the angle of the hammers in the bass and tenor.
>
> David, your questions caused me to rethink some of what I said earlier. 
> While it is true that the hammer, at the point of engagement with the
shank 
> move toward the rear as it moves upward, that does not fully account for
the 
> clearance problem.  I modified my grand action 3-D computer model to 
> illustrate and to do some motion studies.  I have saved some JPG files
from 
> the CAD software.  I could have saved the motion studies in video format. 
> It is fun to watch in dynamic motion, but that would be too much band
width 
> for the list.  In a computer model there is cleaner definition of the
shape 
> of the felt, and there are not variations in hammer pinning, etc.  With
CAD 
> software, it's easier to rule out possibilities, and to be less likely to 
> wrongly attribute what you see to something else.
>
> (See: No Cant All Up Back 15 View.jpg)
> The hammers are all mount perfectly perpendicular to their shanks.   
Each 
> hammer is identical, as they are all brought into the assembly from the
same 
> subassembly model.    This view is at a 15° angle from the back looking 
> straight down the hammers.  You can see they are evenly spaced, at the
same 
> bore angle, and perpendicular to the shanks with no cant angle.
>
>
> (See: No Cant 2Up Back 15 View.jpg)
> The outside hammers appear to be mounted at a cant, while the center
hammer 
> appears not to be canted.  The outside hammers appear to be at an angle, 
> because they are, not because of the way they are mounted on the shank,
but 
> because they are rotated downward on the hammer center, compounding the 
> angle of the hammer's centerline.  The hammer is tilted on one axis by
its 
> bore, and once again on another axis by rotating downward on its center.
>
> (See: No Cant 2Up Top 15 View.jpg)
>  The angle difference is even clearer from a top view.  Looking straight 
> down, you cannot see the sides of the raised hammer, but you clearly see
one 
> side of each outside hammers, at rest.
>
> (See: Cant 2Up Back 15 View.jpg)
> These hammers are canted at 2°.  You can still see the differences in the 
> angles, but they are less severe on the outside hammers.  The spacing
where 
> the hammers come closest is slightly greater.
>
>
> (See: Cant 2Up Top View.jpg)
> >From the top, you can see that the angles of the outside hammers are
less 
> severe, but there is an angle in the opposite direct of the raised hammer
in 
> the center.  The discrepancy in the angles is fairly evenly shared by the 
> raised vs. lowered hammers.  It is hard to see from these images, but the 
> computer measures about 3mm more clearance with the cant, than without
it. 
> I had expected more, but that is significant, especially in a tightly
spaced 
> action.  Tapering the tails would help too, but that was not a component
of 
> this study.
>
> Thanks, David, for prompting me to add to my body of knowledge and 
> understanding.
>
> Frank Emerson
>
> P.S.  Sorry, David.  I did not realize until I started to send this email 
> that I was replying to an email you sent directly to me, rather than the 
> list.  I hope you do not mind my sharing it on list.