SV: Re: SV: Re: [sdiy] Expo converters, lin log etc.

[Dave Manley]

John L Marshall wrote:
> Trying to fit a curve? Try the least squares method. 

Excel spits out the following 'trendlines':

expo: y=250.4e^0.1722x (not a very good fit)

get rid of e using b^x = e^(x*ln(b)):

y=250*1.188^x
       ^^^^^^<-------this should be 2 to give typical expo VCO results.

poly 2nd order: y=22.673x^2 - 96.089x + 448.37
poly 3rd order: y=3.816x^3 - 28.839x^2 + 100.42x + 259.5
poly 4th order: y=0.756x^4 - 9.803x^3 + 53.41x^2 - 88.30x + 387.91

Does that help?  :-p

Whatever it is it's pretty useless except as a noise maker.

Karl, how about posting a link to the schematic?

-Dave