Bridge as a 2nd class lever

[Gene Nelson]

Ric,
As I recall from Ron's class the concept was that the treble end of the 
bridge - under load - acts as a fulcrum - it transfers  bearing load to the 
curved part of the long bridge in the killer octave area. The case being 
made that more bearing load happens there than you would think.
The only formula on the handout was for sizing ribs as support beams and 
deflection under load.
I can scan it into Word if you like - I do not think that Ron would mind.
Gene
----- Original Message ----- 
From: "Ric Brekne" <ricbrek at broadpark.no>
To: "pianotech" <pianotech at ptg.org>
Sent: Tuesday, August 08, 2006 3:54 PM
Subject: Bridge as a 2nd class lever


> Hi list
>
> Those of you who were at Rochester and attended Ron N's class might 
> remember he spent a bit of time talking about bridge leverage and had some 
> printouts of the results of a mathmatical model he and Phil Ford (I 
> believe) worked out. If anyone has copies of these printouts and could get 
> them to me I would be very greatfull.
>
> As I understood this, any increase in downward pressure on just about any 
> place on the long bridge also exerted some significant pressure on the 
> area roughly equivalant to the killer octave.
>
> I've been trying to see if there should be any consequences for tuning. 
> One would think that if the model is correct, then any significant change 
> in string tension over a large region other then this killer octave area 
> would result in a predictable change in string tension there as well. This 
> because string tension directly affects how strong the downbearing on the 
> bridge is.  I've been looking at this for several days now and today I 
> lowered tension on a Bechstein E that I had tuned to 442 the other day.  I 
> lowered the tension from C5 to A0 and then measured C5 upwards. With the 
> lesser tension from A0 to C5 I would have expected an area above C5 to 
> rise slightly in pitch if the model was correct. No change at all in this 
> test, at least nothing I could measure with Tunelab useing two and a half 
> hours to do the deed.
> I'd like a much closer gander at the printouts and the assumptions made. 
> As far as I understand there has been no attempt (outside of mine today) 
> to actually measure in any way whether the model holds true ??  If someone 
> knows otherwise please send along whatever info you have.
>
> Thanks for any and all comments
> Cheers
> RicB
>
>