<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN"> <html> <head> <meta content="text/html;charset=ISO-8859-1" http-equiv="Content-Type"> <title></title> </head> <body bgcolor="#ffffff" text="#000000"> Hi Cy For the moment... I'm just insuring it has a bit more tension then the speaking length... and that the <<unisons>> are tuned to the same frequency... and that that frequency is one in which I hear nothing (or as little as possible) of anything resembling a single string beat when any of the three strings back lengths are plucked. Interestingly... I found right from the get go that factory installed aliquots (or whatever you call them) are far from consistantly placed in just about any make. That is to say, I havent measured the actual distance from the back bridge pin to the contact point of the aliquot on pianos... but I have noticed that the highest obtainable pitch for any given back length varies greatly from piano to piano of same make and design. Easy example... take the top few notes of Steinways.  The speaking length and back length should be able to yield the same frequency with the string up to pitch.  But I've found several that no matter what you do... you cant get the back length up to the same pitch as the speaking length. Perhaps the speaking length varies from piano to piano and the back length remains constant... I dont know...  But clearly... if the idea is that these lengths should be able to have clearly defined consonant frequencies to harmonics of the speaking lengths... the placement of the back length termination is not consistent enough to do the job. I dont really know that this matters much... for the moment I just get the feeling that as long as the back length has a higher tension then the front length, and that these are tuned to as close a state of beatlessness as possible... that the result is a cleaner sounding piano. Cheers RicB <blockquote>Do you tune the rear duplex by its proportional length, geometrically -- i.e. move the little Monopoly house so that the length of the rear duplex is exactly half of the speaking length -- or do you account for inharmonicity, and tune the rear duplex to the exact frequency of the second partial of the speaking length? --Cy--</blockquote> </body> </html>