Touch weight Metrology Question

[David Love]

The different SWR calculated by Terry indicates a different KR between
sharps and naturals.  I have not actually had to put up an altered capstan
line.  The one action that I encountered recently where it was an option, I
based the solution on the naturals and let the sharps fall wherever.  I kept
the sharps close to maximum FW and it played just fine.  Moreover, it wasn't
in the scope of the job, i.e., they didn't want to pay for it.

David Love


----- Original Message -----
From: "Isaac OLEG" <oleg-i@wanadoo.fr>
To: <pianotech@ptg.org>
Sent: May 17, 2002 1:31 AM
Subject: RE: Touch weight Metrology Question


> David, all,
>
> Almost any (somewhat short) piano I measure the KR with the weight method
> show a 0.2 difference between whites and sharps.
>
> I had the same result the first time I measured a grand S&S C model, then
I
> measure again and every note was right (no more difference showing).
>
> This difference puzzle me and I am looking to understand where does it
come.
>
> Indeed the capstan are often misaligned, but that will not lend to a
> difference between sharps & whites I guess.
> Misplaced balance rail ? I believe that the way the keys are designed ,
> there is a flaw in the computation that lend to this difference, or may be
> the way the keys are measured on the jig can add a little error margin
too.
>
> On short keys, the force is acting at the perpendicular of the lever when
> playing, I don't believe this is reproducible with weights, or may be the
> place to weight may be different for short keyboards than for long ones,
> (just trying to understand).
>
> I've seen a keyboard where the key arms where computed by measuring under
> the key, and the good lever in theory transform in large (0.4) difference
> between whites & sharps.
>
> But David , did you ever modify an action so the sharps have a different
> capstan placement ? I believe you don't.
>
> Why not compromise the data on the basis that we play more the white keys
> (and they are more) so give some "weight" (for example 30%) for the sharps
> SWR and the remaining for the whites.
>
> On Terry's action, may be the bolts are not well bedded when the action is
> on the bench, and the SWR is too high because of that too and may be the
> whippen rail is not very straight.
>
> Anyway before taking the samples, seems very important to me to check the
> pinning and regulate the notes. It is amazing how with my little
> understanding of the method, I could yet see some facts in the Terry's
data.
> This touch weight analysis is a must !
>
>
> Regards.
>
> Isaac OLEG
>
>
>
>
> > -----Message d'origine-----
> > De : owner-pianotech@ptg.org [mailto:owner-pianotech@ptg.org]De la part
> > de David Love
> > Envoyé : vendredi 17 mai 2002 04:57
> > À : pianotech@ptg.org
> > Objet : Re: Touchweight Metrology Question
> >
> >
> > Terry:
> >
> > Well they didn't exactly get the balance rail right did they?  The
sharps
> > have a different KR than the naturals and thus the SWR varies
> > considerably.
> > You will have to decide where to compromise or to have a varied
> > capstan line
> > to get even SWR.  A move of the capstan of 2-3 mm toward the
> > balance rail on
> > the sharps will drop the SWR by about the right amount if you want to go
> > that route ( you should check this by experimentation).
> >
> > It appears that you generally have quite a bit of room to add lead.  For
> > example:  Remember that 1 g added to the FW reduces the BW by 1 g.  So
on
> > note 28, with a BW of 48.5 adding 10 g to the FW would still
> > leave you under
> > the FWC.   On note 52 adding 18 g to the FW would still leave you
> > below your
> > FWT.
> >
> > If the hammers are untapered, then you could easily remove 1 g +  from
the
> > SW by simple tapering and tailing.  I don't think that would have
> > a negative
> > effect on tone.
> >
> > The biggest problem is what to do with the different SWR between
> > sharps and
> > naturals.  Given the limited budget, I would probably use the
> > naturals as a
> > guide and let the sharps fall where they fall.  For example, on
> > note 53, if
> > you reduce the SW by .5 g, you will reduce the BW to approx 54g.
> > To achieve
> > a 38 g BW you would need to ad 16 g to the FW which would put you at 24,
> > just below your FWC.  Assuming that is a representative sample, that
> > naturals around there would be well under the FWC.  Moving the
> > capstan line
> > on the sharps is not that big a deal.  Thirty-two capstans wouldn't take
> > that long assuming there is room under the heal to accommodate the move
> > without refelting.
> >
> > If you are not going to move the capstans, I would follow this
procedure.
> > Calculate the SW curve based on a BW of 38 - 40 g (your choice),
> > an R of 6.0
> > and use the FWC in the equation.  That will give you the maximum
> > SW for the
> > sharps to keep the FW at maximum.  Then take some samples SWs and see if
> > that is achievable in terms of tapering and tailing without
> > having to remove
> > too much weight from the hammers.  If that requires too much reduction
in
> > the SW, then either allow the FW to go over max for the sharps, or push
up
> > the BW until you get the compromise you want.  Take off each hammer and
> > shape to get the SW curve, check the pinning and replace.  Then use the
BW
> > method of setting the leads in the keys.  I think that will give
> > you a good
> > result.
> >
> > If your average SWR is 5.5 on the naturals, you should not have a
problem
> > with regulation.
> >
> > David Love
> >
> > ----- Original Message -----
> > From: "Farrell" <mfarrel2@tampabay.rr.com>
> > To: <pianotech@ptg.org>
> > Sent: May 16, 2002 5:43 PM
> > Subject: Touchweight Metrology Question
> >
> >
> > > Well, I'm trying to plug along here with my Yamaha G5 heavy
> > action. Below
> > is a table of my data for ten sample notes (I hope it comes through
> > un-screwed up). I think all is fairly self explanatory perhaps except
for
> > the last four columns. FWC is Front Weight Ceilings (from Stanwood data
> > tables). FWT is Front Weight Targets, which is the FWC, less 10% (safety
> > factor). BWC is Balance Weight achievable with using the ceiling
> > FWs. BWT is
> > Balance Weight achievable with using the target FWs.
> > >
> > > #    DW  UW   F      BW     FW     KR    WW    KC     SW    R
> >  FWT  FWC
> > BWT BWC
> > > 16   61   37  12.0   49.0   29.0   0.49   18.4   16.5   11.9
> > 5.8   33.8
> > 37.5   44   41
> > > 17   67   39  14.0   53.0   30.3   0.49   19.4   16.5   11.9
> > 6.2   33.5
> > 37.2   50   46
> > > 28   59   38  10.5   48.5   23.1   0.48   18.9   16.5   11.3
> > 5.5   30.6
> > 34.0   41   38
> > > 29   73   48  12.5   60.5   17.7   0.50   19.1   16.5   11.2
> > 6.1   30.3
> > 33.7   48   45
> > > 40   67   40  13.5   53.5   14.1   0.48   19.7   16.5   10.5
> > 5.5   27.0
> > 30.0   41   38
> > > 41   72   46  13.0   59.0   13.5   0.50   18.5   16.5   10.4
> > 6.1   26.6
> > 29.6   46   43
> > > 52   66   47    9.5   56.5     3.6    0.48   18.6   16.5     9.6   5.3
> > 22.6   25.1   38   35
> > > 53   68   46  11.0   57.0     8.0    0.50   18.4   16.5     9.5   5.9
> > 22.2   24.7   43   40
> > > 64   62   41  10.5   51.5     2.6    0.48   19.3   16.5     8.3   5.4
> > 17.2   19.1   37   35
> > > 65   63   41  11.0   52.0     6.7    0.50   19.0   16.5     8.3   5.9
> > 16.7   18.6   42   40
> > > 76   59   41    9.0   50.0    -1.7    0.48   18.3   16.5     7.3   5.4
> > 10.4   11.6   38   37
> > > 77   62   42  10.0   52.0    -0.3    0.50   18.8   16.5     7.1   6.0
> > 9.9   11.0   42   41
> > >
> > > Action spread is right on the money at 112.5 mm.
> > >
> > > My SWs are right along the divide between medium and heavy
> > hammers on the
> > smart chart. I should think this is a reasonable weight set of hammers.
I
> > see someone forgot to put lead in the keys along the line somewhere! But
I
> > also see that even if I go right up to the FW ceiling, I will be
> > a bit over
> > ideal DWs.
> > >
> > > Not having seen all this data before, it seems rather erratic to me.
How
> > in the world can the BWs range from 49.0 to 60.5? And that is with a
very
> > smooth SW. I assume this is mostly erratic leading? Leading is
> > two in bass,
> > one in tenor, small one near center rail in treble, and one in rear of
key
> > in high treble. I just played around a little with weighting key
> > #16, and I
> > would have to add two big leads and one small one to get near the
> > FW Target
> > or Ceiling. But perhaps that is OK - basically three big leads in the
bass
> > keys.
> > >
> > > Recommendations?
> > >
> > > I should point out here that this piano is in a small church,
> > and it seems
> > that the piano is not a high priority for spending money - so I'm
probably
> > not going to get to do everything I might want.
> > >
> > > I could trim all hammers, but I hate to have hammers too light for
tonal
> > reasons. Geez, I look at this mess and start to think that I
> > simply need to
> > take a saw and cut off everything above the keyframe and start all over.
> > Help!
> > >
> > > I know I can trim hammers a bit, and add lead until I get about
> > to where I
> > want to go, but that seems so........ pre-cambrian. Waddaya think?
> > >
> > > I think I also need to regulate out a few notes to be sure that the
> > current geometry is not requiring excessive dip.
> > >
> >
> >
>
>