<html> <head> <META HTTP-EQUIV="Content-Type" CONTENT="text/html; charset=us-ascii"> <meta name=Generator content="Microsoft Word 10 (filtered)"> <style>  </style> </head> <body lang=EN-US link=blue vlink=purple> <div class=Section1> The scales we are using tend to come with increments to the .10 grams.  It makes sense to divide the SWs that way because each 1 gram difference in the SW produce a difference at the key as a function of the action ratio, usually 5-6 grams.  A .10 difference will only produce a BW change of .5 or .6 grams, something most of us can live with.  Going to .01 is overkill and the scale reading will tend to float around from the air currents in the room or if you belch too loud from the Chinese food you had at lunch.  For those who want to produce very smooth actions with smooth leading patterns and reasonably even inertia, .10 grams is a good place to start.  If you aren’t concerned with those things then it doesn’t much matter.  Of course, the actions are not really balanced accurately to the .10 grams since the margin of error when you do the actual weigh off is greater than .10 grams, especially since the grams weights we use to measure up and down weights are calibrated only to the 1.0 grams.    Key dip and blow, btw, should change whether you change the action ratio at the knuckle or the capstan though the amount of movement required to make an equal change will be different--the capstan will have to move farther in absolute terms than the knuckle to produce an equal change in the action ratio.  It’s just a different way of skinning a cat.       The number of leads in the key is not really the deciding factor as to whether there is too much lead.  It’s location, location, location—just like real estate.  It’s the overall front weight of the key that matters.  If the leads are located near the balance rail, you will have more leads for the same BW than if they are located out toward the front of the key.  An argument can be made for more leads closer to the balance rail in which case having 4 or 5 in the bass or even higher up is not necessarily a problem.    <div> David Love davidlovepianos@comcast.net www.davidlovepianos.com </div> -----Original Message----- From: caut-bounces@ptg.org [mailto:caut-bounces@ptg.org] On Behalf Of Keith Roberts Sent: Sunday, September 23, 2007 4:54 PM To: College and University Technicians Subject: Re: [CAUT] Key ratio measurement (was Re: moving capstans question)   <div> Sometimes it cracks me up. Correct me if I'm wrong but are you guys using scales that have an accuracy of a tenth of a gram? I suggested to a friend we get the one that has an accuracy of a hundreth of a gram. Your hammers range from 10 grams down to 5 grams. (give or take) So with 5 grams divided into 88 increments that yeilds a weight change of slightly more than half a tenth per hammer. It is senseless to use a strike weight curve unless I can post hammer weights after the taper like, #1, 9.64 #2, 9.60  #3, 9.55 #4, 9.45  #5, 9.49  #6, 9.35 </div> <div>   </div> <div> Now that could read 9.6, 9.6, 9.6, 9.4, 9.5, 9.4 or 9.6, 9.6, 9.5, 9.5, 9.5, 9.3. However with the scale we got it makes it possible to really dial in the weight of the parts. In the run above I can see that i can switch 4 and 5 and being the bass it won't show. A tenth of a gram scale doesn't cut it in my opinion for the proclaimed smoothness of curve. Essentially it all has to come out in the wash when the action is reweighed and leaded. </div> <div>   </div> <div> Really we don't want to redesign the action, we want to make it work right. We are taking the measurement of an existing action and it doesn't matter what anybody else gets as the action ratio. 40 people can measure it and have 40 different answers for their computer/math models but nothing has changed in the action in question. It still has the same ratio it had before the measurements as afterwards. We have several options. The consensus is that 4 to 5 leads in a key is two much. I took out 4 pounds of lead. We can lighten the hammer, we can move the knuckle, we can move the capstan and even the heel. Sometimes we can get different whippen configs. Each one has it's own criteria that has to be considered before using. Key dip and blow distance really didn't change much with this capstan move but moving the knuckle, it sure does. </div> <div>   </div> <div> Keith   </div> </div> </body> </html>