Farrell wrote: > > Remember... if you have an Even FW curve AND and even SW curve... > > then any uneveness is simply got to be found in either friction or leverage. > > I thought one would want to target an even SW curve and a constant BW. Differences in FW would then be due to any differences in leverage? Think about it for a second Terry. You have basically two sides to balance against one another... on the front side of the balance rail pin you have the lead in the key (FW) and whatever force your finger provides (BW). On the other side you have leverage, friction and dead weights of parts... most notably the SW. Indeed one way of looking at the formula is BW + FW = the rest of it. So if you know FW is on an even curve and you know SW is also such, and you "assign" a constant value to BW, then the only thing that can keep the equation from working out are the left over variables... and they are basically friction and leverage. So yes... you install a SW curve based on a BW and WBW spec. Any unevenness in FW's needed at this point to achieve that constant BW are due to either friction and or leverage. So go ahead and even out the FW to fit the equation.... then whatever uneveness you find in BW are due to leverage and / or friction problems. Makes problems easy to find. R = ((BW + FW) - (KR x WW))/SW.... is the same thing as saying BW + FW = (R * SW) + (KR * WW) R can be calculated ahead of time actually... its the overall action ratio... which is the result of the length ratios of the three levers. Stanwood doesnt do it this way as far as I know... but there is no reason not to as far as I can tell. You want a 5.7 ratio... install one. Just make sure the KR component fits into your scheme. Jon Page has a neat way of going about this leverage question also. It is simply to find the optimum position for the capstan relative to regulating any given action. You set the spread right first, then find the best spot for the capstan by seeing what works best using a template capstan. This always results in a useable ratio that you can work with. Stanwood doesnt do it this way either... but again nothing says you cant use the basic Balance equation in other ways. Technically, I suppose one could be accused of installing a FW curve based on writting the formula as FW = (R * SW) - BW + WBW.... which is a breach of Stanwoods patent. But since his maximums are published charts provided in his kits,,,, it seems ok enough to use these to me. And if you need / want to taper them as a last step in your action job to accommodate whippen assist springs for example.. you could at that point very easily and successfully use the old weigh off style of determining FW's.... as by that point all other issues of eveness are resolved. This is at least how I see it all, and after quite a few jobs now I havent seen any problem with going forward in this fashion. > > Now, if someone could reduce all this to only one variable, I could get a job done AND get some sleep! I guess 47 variables and the combinations possible make for lots of fun/frustration............ well, interesting anyway! :-( > > Terry Farrell > > - Actually.... this is sorta what i do.... reduce it to one variable... friction. It does have the limitations of being less then creative designwise tho, as your range for choice of SW curves is somewhat limited to those appropriate for a 5.3 to 5.7 Ratio for any given BW spec. Grin.. NOW I bet you are REALLY confused... or maybe not :) -- Richard Brekne RPT, N.P.T.F. Bergen, Norway mailto:rbrekne@broadpark.no http://home.broadpark.no/~rbrekne/ricmain.html
This PTG archive page provided courtesy of Moy Piano Service, LLC