Rob,

For this sort of problem I'd set up the equations in excell & let it do the
maths. You could also use the "optimise" function (based on maximising the
minimum voltage differences) for the variuos options. It's a bit of an issue
to set up, but will give you the best possible results in the end. Then,
plug in the nearest available resistor values and see what sort of margin
you get.

RP


On 29/11/2007, Rob Robson <rob@silk.net> wrote:
>
> > Think binary :)
>
> I'm trying!  Do you mean for me to binary weight the four switch resistor
> values (R4=2xR3;  R3=2xR2; R2=2xR1)?  I'm okay with that, but how does one
> optimize the value of the common pull-up resistor to minimize bunching of
> the resulting _voltages_, keeping in mind that V= 5 x (Rx / (Rx+Rpu))?
> Is
> some sort of scaling factor required for the four
> otherwise-binary-weighted
> resistors to offset the bunching effect?   The system needs to reliably
> distinguish between the case where all four buttons are pressed
> simultaneously, and any one of the possible 3-button combinations.  My
> back-of-an-envelope scenarios yield some voltage values that are less than
> 1% apart, which is too tight for comfort.  The archived solutions some
> folks
> have kindly identified seem to be invalid for multi-button combinations,
> which doesn't work for this situation.  Any more hints?
>
> Thanks,
> RR
>
>
> --
> http://www.piclist.com PIC/SX FAQ & list archive
> View/change your membership options at
> http://mailman.mit.edu/mailman/listinfo/piclist
>
-- 
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist