It was/is a project with a budget vanishingly close to $0, living with a
LOG pot is probably preferable to spending $1 on this.
Junk box parts, beefy IGBT, leftover power supply, junkyard automotive
ignition coil =3D big fat sparks for the kids to play with.
Pots are controlling frequency and duty cycle of the IGBT.
Logic is provided by $3 arduino nano clone. (Not a fanboy, but $3 is tough
to ignore)

LOG pot works, but the area of interest is about 20 degrees of knob travel.
Good enough for the purpose at hand, but it's one of those things my mind
chews on.

This morning I created an inverse LOG lookup table based on
y=3D147.6*ln(AD_reading), where AD_reading is 0 to 1023 and 147.6 was a
number I played with till the output values also scaled to 0 to 1023.
Yeah, 1k lookup table is a bit excessive, but simple and as John mentioned,
fast. Neither are too critical, as this entire program is about 10 lines of
C (well, arduino brand C flavored code substitute)

Based on completely subjective feel, the area of interest is now ~150
degrees but it does not appear that the center of travel produces mid-range
values. I'm guessing this is the reality that this is not an actual LOG
taper, but a piece-wise linear LOG approximation. In a more critical
application, I could see measuring the physical position of the pot while
reading the A/D value and creating a better lookup table. More work than
this is worth I think.


-Denny
--=20
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist
.