<!doctype html public "-//W3C//DTD W3 HTML//EN"> <html><head><style type="text/css"></style><title>Re: model B wire gauges & string calculations</title></head><body> <div>Hello All,</div> <div><br></div> <div>Thank you so much for the many replies to my request for the standard plain wire gauges of the early Steinway model B. I haven't rebuilt any early Bs, so I didn't have a scale on file. With the wire gauges filled in, here's a jpg below of the 85-note model B scaling graphs.</div> <div><br></div> <div align="center"><br></div> <div align="center"><br></div> <div align="center"><br></div> <div align="center"><img src="cid:a06110408c1487ce948b6@[192.168.0.4].1.0"></div> <div align="center"><br></div> <div align="center"><i>Steinway circ 1890 model B scaling graphs</i></div> <div align="center"><br></div> <div align="center">Just in case the above jpg doesn't get through the Pianotech sever, I have uploaded a copy to;</div> <div align="center">http://members.optusnet.com.au/ronovers/1890B.scale.jpg</div > <div><br></div> <div>The scale is not surprisingly quite similar to the 'modern' B. There's a gaping tension hole from F21 to C#29, and for the lowest singles. No surprises why Ron Nossaman designed a nine note transition bridge from F21 to C#29 for the remanufactured model B he exhibited at the recent Rochester national PTG convention. I'll be doing the same for the 1890 piano if it gets to be rebuilt in our workshop<font color="#0000FF"> (as an aside, several years ago we rebuilt a modern model B. Before removing the strings at teardown, I pulled note F21 up to B flat #26 - to 41% of breaking strain - just to have a listen. The tone was so much improved, and so too would have been the tuning stability, had the piano been originally designed with the break on B flat #26 at the original #F21's 145 cm speaking length)</font>. Look also at the Z (impedance) jump between note F21 and E20. This jump can be plainly heard in the piano as well. Just makes me wonder how folks who claim to be at the cutting edge, and who also claim to be making continual improvements, can sit on such scaling problems for over a century and pretend that there's not an issue. While it is excusable in a 1890 piano, it is now 2006 and our understanding has totally changed. Of course, the contemporary industry is plagued with similarly poor scales amongst a variety of manufacturers.</div> <div><br></div> <div>I'll include below a small-piano scale graph, taken from one of my own scales, for those of you who may be interested, since there has been quite a lot of scaling discussion on the Pianotech list recently. Without cutting the back-scale short, this scale can be squeezed into an upright of just over 130 cm (the "% breaking strain" graph is the uppermost one).</div> <div><br></div> <div align="center"><br></div> <div align="center"><br></div> <div align="center"><img src="cid:a06110408c1487ce948b6@[192.168.0.4].1.1"></div> <div align="center"><i>An Overs design, Short-piano scale graphs</i></div> <div align="center"><i><br></i></div> <div align="center">Just in case the above jpg doesn't get through the Pianotech sever, I have uploaded a copy to;</div> <div align="center" >http://members.optusnet.com.au/ronovers/upright.scale2.jpg</div> <div><br></div> <div>As some of you have mentioned on the list (David Love may have been one who mentioned it), you can't design with smooth curves on all three parameters simultaneously. We will therefore have to compromise on one or more parameters depending on our priorities. Over the past several years I have chosen to smooth inharmonicity and Z, while juggling the speaking length and wire gauges to minimise the deviation of percentage of breaking strain at the transitions between tri-chord and bi-chord strings, and bi-chord and singles. I regard a 15% tension change between sections to be unacceptable. It is possible to get the deviation down to around 10% without compromising on inhamonicity and impedance. In the upright scale shown above, the tension falls to around 41% at the lowest treble note, #F33, with the first bichord covered note in the bass at around 51%. This will result in quite good tuning stability at the cross over. By comparison, the Steinway model D has a tension in the treble which falls to around 36% at F21 (the lowest treble note), with the first trichord covered bass note (E20) at around 51%. So the upright scale shown above has quite a bit less tension deviation than the industry-standard model D scale. Therefore, all other things being equal, one would expect the tuning stability of the above short-piano scale to be improved, when compared to that of the industry-standard model D. I look forward to the possibility that within 20 years a majority of pianos might use scales with deviations similar to the upright scale shown above, or better. With longer pianos it is possible to design a scale with even less tension deviation than the upright scale shown above.</div> <div><br></div> <div>Richard Brekne asked recently about the reliability of calculated inharmocity verses the real world measured figures. Yes there are differences. In the wound string section in particular, the deviation between calculated inharmonicity and real world figures get worse as the wrap diameter increases. However, the deviation when the windings are small is quite insignificant, so the cross over calculations will still be quite close. And this is where it matters most. I don't own a Cybertuner, but these devices can measure the real inharmonicity. As usual, in uncle Joe G's recent post, he was carrying on with all the usual commonsense that he has become known for. Well done uncle Joe.</div> <div><br></div> <div>I see Terry F is getting all geared up to do a transition bridge which includes moving the agraffe line - good idea Terry, but it is a lot of work. I've done a transition in a Yamaha G2 where I moved the hammer line to correct for the strike point problems created by the inclusion of a transition bridge - the result was good but it also is too much work, creating some action geometry issues which have to be worked around. When you check enough scales from existing manufacturers, including some who claim the high ground, for many the strike ratio is already quite varied across the break between bass and treble.</div> <div><br></div> <div>By the way Terry, I suspect that the dust extractor you've got might be a better unit than many of the other commercially available units. Bill's design uses an impeller with backward curved blades. Many of the commercially available units use impellers with straight blades. Both impellers will work fine while the filters are unblocked. But once the filters start to get restricted the backward curved impeller will keep pumping relatively well, while the straight bladed version will stall. Good design is usable for a lifetime. Trash is trash.</div> <div><br></div> <div>We need to believe that we can make a difference in our own lifetime, by not putting our collective heads in the sand. The current Pianotech list scaling-discussion would seem to be indicating that serious head-from-sand extrication is really happening. Bravo!</div> <div><br></div> <div>Ron O.</div> <x-sigsep><pre>-- </pre></x-sigsep> <div><font face="Times New Roman" size="+1" color="#0000FF"><b>OVERS PIANOS - SYDNEY</b></font></div> <div><font color="#000000"><i>   </i></font><font color="#0000FF"><i>Grand Piano Manufacturers</i></font><font color="#000000"><br> _______________________</font></div> <div><font size="-2" color="#000000"><br> </font><font color="#000000">Web http://overspianos.com.au</font></div> <div><font color="#000000">mailto:ron@overspianos.com.au<br> _______________________</font></div> </body> </html>