Helmholtz and Steinway

[Ron Nossaman]

>Well Stephen has suggested a simple way of showing this and it seems 
>to me pretty obvious that compression waves travelling the length of 
>the wire will not be stopped by impediments to transverse movement. 
>Imagine a thick steel rod six feet long passing through a one foot 
>cube of concrete in the middle of its length which prevents any 
>sideways movement.  You put your ear to one end and I'll tap the 
>other with a small hammer.  You are suggesting that you will hear 
>nothing that is due to waves travelling along the rod.  I think that 
>unlikely.

Maybe. Stephen didn't specify how his "bridge" was attached. If it was
merely hung on the suspended wire, it wasn't remotely the termination that
a bridge attached to a soundboard would be. I don't know that this is all
that good an illustration of principal. Given your example, if I were to
hit the other end of your embedded rod, where Richard's ear recently was,
after relocating him out of harm's way, would you hear the sound from the
end you originally hit? If he heard the sound you produced, I would
rationally expect that you would hear the sound I produced from his end of
the rod. If it works one way, it will work the other way with the same
parameters. It's the LAW. So if the rear duplex excitation is the result of
longitudinal vibrations passing across the bridge, then exciting the rear
duplex should result in the SPECIFIC excitation of the speaking length of
the same string. That's not a GENERAL excitation of anything we would
expect under the proposed circumstances, but a SPECIFIC excitation of the
speaking length of the same string that shares the duplex that received the
excitation. Furthermore, if the rear duplex works on the same principal as
the front duplex, and vice versa (according to the, rather THE patent),
ostensibly by the transmission of longitudinal vibrations across the
appropriate "termination" from the speaking length, then the excitation of
either the front or rear duplex should result in the excitation of the
speaking length of THAT PARTICULAR STRING with similar effect. Symmetry (If
it's demonstrable on one end, it had better be demonstrable on the other or
the premise doesn't apply). It doesn't, by actual real world "anyone can
verify it by getting off their ass and giving it three minutes without
undue expenditure or inconvenience" trials. Why not? I say it's because the
premise is false. Why else?


>I feel that more experimental data is needed.  Conklin is concerned 
>primarily with the audible longitudinal waves in covered strings and 
>his demonstrations are borne out to a degree by my own experience as 
>a bass string maker.  To that extent I find his thesis interesting, 
>particularly in view of the actual proofs he uses.  On the other hand 
>he fails to mention other aspects of string design which, in my 
>experience, have a greater effect on tonal quality in covered 
>strings.  All Conklin's demonstrations are done with low bass notes.

He isn't addressing tonal quality resulting from other aspects of string
design here, and there's no indicated reason to drag the discussion off
line in that direction. He is quite specifically addressing the
ramifications of scale design decisions relating to longitudinal
vibrations. These would, presumably, be the same longitudinal vibrations
mentioned in the patent filed by Theodore Steinway regarding the action of
the front and rear duplex scales, which we are currently discussing. 


>We then make the jump to Theodore Steinway's longitudinal waves, 
>which are right at the other end of the scale, from A4 (440c/s) 
>upwards and not dealt with at all by Conklin.  According to Conklin 
>the frequency of these waves would be roughly 7000 c/s rising to over 
>60,000 at C (4186 c/s).

The are indeed dealt with by Conklin in the text I quoted, as I quoted.
What "other end of the scale" are you referring to here? I don't understand.


>You wrote, quoting from Conklin:
>
>>  "The longitudinal frequency of a plain steel string in a piano can 
>>be changed only by altering its speaking length..."
>
>Since he produces no demonstrations relating to plain wire strings, 
>it is quite possible that he would need to prove that this would make 
>a difference if it did not alter the TOTAL length of the wire.  Nor 
>are we told how these longitudinal waves are initiated, how they are 
>affected by the relative tension of the wire etc.  He leaves more 
>questions unanswered than he solves, though what he does show is 
>interesting.
>
>JD

Again, Harold has proven to my satisfaction (though I realize that pulls no
weight whatsoever) that he is adequately thorough in his presentations.
Since he does not even hint that anything but the speaking length is
relevant here, I am willing to accept it. If you are not, the we are
finished with this discussion, pending proof to the contrary. He doesn't
address how these waves are initiated, assuming that he knows, which may
not be the case. Since Theodore Steinway didn't, or couldn't enlighten us
on the same subject, we may presumably conclude that neither of them knew
at the time, or chose not to say. In either case, that has no bearing
whatsoever on the fact that the front duplex and rear duplex aren't
apparently driven by the same mechanism. How these longitudinal vibrations
originate isn't reagent at this point. The premise that the front and rear
duplexes are driven by the same mechanism, yet react very differently is.

Ron N