keyboard modifications

[WeAreAs1 at aol.com]

In a message dated 12/12/00 12:13:03 AM, efm3 at mediaone.net writes:

<< Michael, I have used a lot of different (key-cap) keyboards mostly Casios

and they all work fine and have close to zero resistance when closed. As 

long as the resistance is a constant it won't matter if it is there. You 
could 

build the chain with any "realistic" resistor value as long as they were all 
the 

same. Think of the opamp as a positive voltage supply. The voltage is 1/2 

the supply if you use 2 100K resistors as a divider....and the voltage is 
still 

1/2 the supply using 1K resistors.....so 200 ohms works just as well as 100.  

Got it? Somebody (? sorry) built one using 1.3K 1%  resistors and it 

worked just fine. >>

If this is the case, then it might be best to use a set of larger-valued 
resistors for the string.  Say, 1K or larger, rather than 100 ohms.  This 
way, the variations in resistance between the individual rubber switches 
would be less significant to the effective resistance at each point in the 
string.  For example, if you used 100 ohm resistors, and the rubber switches 
have an average resistance of 100 ohms, then a 10 ohm variation beetween 
switches would represent a 5% error for that key.  If you use 1K resistors, 
the same 10 ohm variation would  cause only 0.5% (approx.) error in effective 
resistance.  Using 10K resistors would further reduce this error to just 
0.05%, which would hardly be measurable in terms of effect on pitch accuracy.

Michael B.