Changing taper of pot is not really working (graph included)

[iandy4]

Hey Guys,
I've been experimenting with changing the taper of a linear pot to behave like an audio pot.  You can read out it here:
http://www.jt30.com/jt30page/potfix/

The thing is, I tried this technique with a 10k linear pot using a 1k, 2.2k and 5k resistor wired as shown in the link above.  Those are the four nearly identical lines in the graph (one is without resistor).  I don't know why the online graph I used doesn't show a key for which is which but it doesn't really matter considering there was nearly no change at all to the taper of the linear pot.  The red line was an 10k audio taper which I graphed just to see how close I was getting to the real deal with the resistors. Evidently, not too close...

Is there something I'm not aware of? Perhaps it needs to see a load to change the taper or this doesn't work well with lower value pots?  I'd really like to get this working because I'm using a 25k b pot in a modified volume pedal and id like to be able to use it as an expression pedal as well as a volume pedal (I have it figured out how to use a 25k pot even with a passive signal)


I did this simply by measuring the resistance between lugs 1 and 2 while the resistors were put between lugs 2 and 3.  I didn't have the pot connected to anything.  Just testing the pot with the resistor alone.

[Hides-His-Eyes]

Aigh... Well, obviously the resistance between lugs one and two doesn't change. (Obvious to me and will be obvious to you if you draw a schematic of what you did and what you're measuring!)

What you're changing is the RATIO between 1-2 and 2-3. Which is what matters in a volume control, etc.

[iandy4]

Aigh... Well, obviously the resistance between lugs one and two doesn't change. (Obvious to me and will be obvious to you if you draw a schematic of what you did and what you're measuring!)

What you're changing is the RATIO between 1-2 and 2-3. Which is what matters in a volume control, etc.

Thanks makes sense.  Is there a way to use the multimeter to measure it though?  I imagine I'd have to measure the resistance between 1&2 and 2&3 in each knob position and compare the ratios, right?

[petemoore]

 Yep, more dots = higher resolution pot taper graph.
  ...a number dial circle for behind the pointer knob, indexed to 0.0 by the pot thread-lock-tab on the pot casing, shows 1-10.
---------------------
   Going with 'volume control' assignment:
  Say >100k is desired, and you want it to twist/sweep different.
  Start with larger value pot [variable resistive divider = potentiometer] so the tapering resistors sum-value reduction
still = ~100k after the reduction/tapering value-sweeps has been re-curved.
  If the pot introduces large resistance in signal path [say 1 meg is turned to center = 500k impeding signal], a small cap can be placed across this resistance so when volume is reduced=high R, the HF's have a low impedance bypass path. =Small capacitor across the signal path of the large value pot...if needed....a little bigger and it becomes a treble 'boost' as volume is reduced.

[iandy4]

For anyone curious... I see now,  I had to measure the resistance between 1&2 and 2&3 in several different positions of the pot, then  compare the difference in each position. What surprised me was that the total resistance between 1&2 and 2&3 doesn't equal the pots value any more when you add a resistor so you have use 1&2 and 2&3's combined resistance in that position of the knob as the new pot value if your goal is to figure out the curve.

ex:
position 5 on the knob with out the resistor added was showing 5k between 1&2 and 5k between 2&3.  50% of the total 10k value of the pot as you'd expect from the middle position of a linear 10k pot.  The I added a 2.2k resistor between 1&2 and measured the resistance at 1.5k between 1&2 and 4.66k between 2&3. The total of both together no longer equal the pots value so I combined their values to calculate the percentage of resistance between 1&2 (24.3% at 5 on the knob) and 2&3 (75.6% at 5 on the knob) for each position on the knob.  Then I could plot it in the chart and found the curve to be exactly as it should be.

[frequencycentral]

http://www.diystompboxes.com/smfforum/index.php?topic=70732.0

[iandy4]

http://www.diystompboxes.com/smfforum/index.php?topic=70732.0

Rick, you are the man!! That is an extremely helpful thread.  I wish I caught that before!

[PRR]

Connect the pot. _as_ a potentiometer, across a 1V (or 9V) battery. Measure the output voltage. Linear will measure half-voltage at half-rotation. Audio-taper 10%-20%. Loaded-linear around 25%.

Note that loaded-linear will not change the taper at the ends, only in the center.

Note that loaded-linear will have a very low input impedance at full-up.

Note that loaded-linear will NOT! work correctly as a rheostat! Only as potentiometer.