I thought I should add that in all this discussion, aside from the side spring along the BPP tangent, most of the discussion was confined to looking at the some of the consequences of attributing seasonal pitch changes to a strictly vertical deflection of the string by the rise and fall of the soundboard. Ron N threw out an example which my own spreadsheet confirms. He threw out an A4 of with a speaking length of 406 mm, back length of 130 mm (1 mm Ø ?) and a deflection of 1 degree (a 1,7 mm vertical deflection) and in subsequent posts provided some interesting thoughts. One point was that a 0.1" (2.54 mm) strictly vertical rise because of a soundboard pushing upwards would result in a 23 cent rise in pitch.... but the cost was a jump to a 2.5 degree down bearing angle. I'd add to that such a push upwards would have to fight off a total of 7.7 lbs of down bearing per string. I think he's right in stating it seems highly unlikely a soundboard can do this. His other interesting thoughts go along the line of what happens if the bridge surface swells and pushes the string vertically. If one looks at the strictly vertical forces again he points out that more then likely a bridge swelling more then very slightly would in actuality just force the board downwards. Again... because a soundboard can only push upwards so much. Figure out what 7.7 pounds per string does for downbearing on the whole board... :) Then there was the sideways deflection the bridge pins cause. Any change in vertical position of a string in relation to the pins, changes the length of the string segment over the bridge... in addition to deflecting (or at least trying to) the string strictly vertically. He states that a 0.2 mm climb of the pins produces about a 0.018 mm change in length for the string segment. He doesn't give the offset angle, bridge width or pin angle... but if one takes a 12 degree offset and pin angle with a 18 mm wide bridge surface I get that result. There are two things here two think about. First... since this change in length affects only the segment on the bridge, its only significance to pitch change is the net increase in tension this change has on the entire length of the string... from hitch pin to tuning pin. That increase in tension gets applied to the speaking length... which isn't really changed much at all here. Second... the speaking length is altered if ever so slightly by a 0.2 mm vertical rise at the front bridge pin. That change in speaking length also carries with it a small increase in tension... and the same thing happens at the back length. The combined change in tension gets applied to this very slightly longer speaking length to arrive at the new pitch, which is about a 6.8 cent change in this example. Here the increase in down bearing is quite minimal... only about 0.4 lbs per string. All of this assumes that nothing else is going on... purposely... so as to shed light on what looks plausible or not. To me, I agree right off that it does not seem plausible that the soundboard can actually rise or fall enough strictly vertically to effect pitch directly in any significant way. I find the <<string climb>> up the pin interesting enough... as it is certainly looks doable from the perspective of what load the soundboard can bear... but if you apply the results to all strings... you end up with a steep climbing curve that is very much more related to speaking length then anything else. To be sure... lower BPP's will be affected more then higher ones... but if you run the numbers it ends up looking like the shorter the string the bigger the effect. Since we don't see pitch changes in that follow this pattern in real life pianos... this tells me something else must be a far more dominant factor. Many folks have made points as to the strengths of the forces the soundboard attempts to apply upwards against the strings. Many have made measurements of crown that they say change largely... and at the same time it seems to be common to hear that down bearing angles don't change much. If both these are true... then perhaps what we are looking at is an overall swelling of both bridge and soundboard as a whole... and seeing this unable to push the strings much upwards, but rather distort the assembly downwards. With these kinds of forces going on it is also likely to think along the lines that the bridge itself is not going to remain in the exact same horizontal position. If it tilts even slightly one way or the other... then all kinds of interesting things happen to the speaking length... along with the strictly vertical deflection results that in the end are present whatever those are. The fact that climate control efforts in all their variants have a stabilizing effect attests that wood is moving and changing dimensions. This is why I say I don't think we really have a good answer at present. We have only shed a little light on what can not happen... and what can not account for what we observe. Strikes me that the little math involved here, and the subject matter at large should be of interest to just about any tech. Its a fascinating subject and a journey down a constructive road to boot. Perhaps if we looked at this from a reverse perspective... we might find some enlightenment. If one measures the pitch of each string before tuning on a few pianos one tunes often enough to know where they were pitch-wise a season before... and calculate what kind of change in string tensions these changes in pitch imply, one might be able to guess better at what length changes would be needed to affect those tension changes. Just a thought. Cheers RicB
This PTG archive page provided courtesy of Moy Piano Service, LLC