1
\$\begingroup\$

I have a DC motor speed controller (PWM) that uses a 100 kΩ potentiometer to adjust motor speed. I also have an electronic throttle pedal that used a 49E Hall-Effect sensor (+5 V, 0.8-4.2 V signal, 1 K KΩ). I want to use this pedal with this speed controller.

I measured the voltage drop across the potentiometer and it is 5 V. I confirmed that the pedal's signal changes from 0.8 V (released) to 4.2 V (fully pressed) when given 5 V.

What would be a right way to replace the potentiometer in the speed control unit with a Hall-effect sensor? Note that I would prefer to preserve full speed control range (from 0% to 100%).

If you're curious about my application: replacing binary on/off throttle in kids ride on car.

UPDATE 1

PWM controller built-in potentiometer works like this:

  • Max speed: GND to wiper = 0 V, wiper to V+ = 5 V
  • Min speed: GND to wiper = 5 V, wiper to V+ = 0 V

And Hall-effect pedal is opposite:

  • Max speed (fully pressed): GND to signal = 4.2 V
  • Min speed: GND to signal = 0.8 V

So as a drop-in replacement pedal it will behave more like a break pedal :)

UPDATE 2

On further examination, it appears that the potentiometer's GND (I mean, one of its legs that I referred above as GND) is connected directly to input GND of the PWM controller (where the negative terminal from battery goes). So, from a practical standpoint, it's probably safe to assume that I also have GND - Xv at my disposal (Xv however is variable, since it's coming from an unregulated battery).

But I think this extra voltage source makes this question less general and less useful for other people.

\$\endgroup\$
13
  • \$\begingroup\$ What voltage do you measure between the two far ends of the pot? It will probably be constant no matter what the pot is adjusted to. Then you need to run your hall-sensor through an opamp circuit that offsets and amplifies so the 0.8 and 4.2V gets remapped of 0 and Vpotmax. \$\endgroup\$
    – DKNguyen
    Aug 6, 2020 at 3:21
  • \$\begingroup\$ @DKNguyen As I mentioned, voltage drop on the pot is 5V. Yes, the range difference is what bothers me. Can you be more specific on op-amp approach? AFAIK, they don't really do rail-to-rail, so I still will have a gap at least on one end of the spectrum (will not go to 5V, will be same ~4V upper range). I have 12V input of course, but I have no idea what kind of PWM circuit interference I can cause by powering op-amp from it... \$\endgroup\$ Aug 6, 2020 at 3:28
  • \$\begingroup\$ I will post circuit. Just don't have time at the moment. You're saying the pot only adjusts from 0-4V? You could, but do not need to, power the opamp from the 12V if you add some decoupling caps and stuff. But best to just power straight off the two end terminals of the pot you measured the 5V between after you cut it out and inject the output into where the center terminal of the pot was. \$\endgroup\$
    – DKNguyen
    Aug 6, 2020 at 3:29
  • \$\begingroup\$ @DKNguyen pot on PWM controller adjusts from 0 to 100K or from 0 to 5V. \$\endgroup\$ Aug 6, 2020 at 3:31
  • \$\begingroup\$ So what did you mean by "they don't really do rail-to-rail, so I still will have a gap at least on one end of the spectrum (will not go to 5V, will be same ~4V upper range"? Rail to rail opamps exist and are not uncommon. \$\endgroup\$
    – DKNguyen
    Aug 6, 2020 at 3:32

1 Answer 1

0
\$\begingroup\$

schematic

simulate this circuit – Schematic created using CircuitLab

This won't reverse it. Will continue to look at it tommorrow. I'm not sure I can do it without a negative supply.

https://www.falstad.com/circuit/circuitjs.html?ctz=CQAgjCAMB0l3BWEA2AzNA7ATi8gHACxhaKTJYgIaWWQBQAhiAEx57hbMgGpfFfswSALRDw8eFGi5myBGDxyCyMs1QUYM5GhwZlYDAlbiJdAO4s24MF1aCMyKOe68OXHnxtOATpcFePay5IcAR4Ol8bf1srMC8QsFNfOxZIdhTmBEcEsLg6AHM-Fiyi1EgCHyLmNLdU9gTTAHkXPk5uXFqQ+gAHWurBNszsqCdC-mLHcbKK+gA3cGMhhfSakIqCBMwWaCQunYLlicPmDGCDqPAVQ4Ns5wv+vtW6VBnoPAA1AB0AZwAvAFMAHb-XybDBgHhGPD8aHwZj-YTUELBN50X7XK4XOK2KQfH4A4HeCLXLwXAj1EDqPIAJUO5IxIzWIV2jP2AHsQBRkDNuJAsOwkbB4FgHFlMiMwHQOexuSMNvzxFIJMwCEQNshxRBJUA

\$\endgroup\$
4
  • \$\begingroup\$ Thanks for the efforts, really appreciate it. I tried to simulate your circuit, but DC swipe produces several megavolts on output :) I came up with simplified version that simulates well, but still needs to be reversed, which turned out to be tricky: circuitlab.com/circuit/9n5k5gsjrsvu/throttle \$\endgroup\$ Aug 6, 2020 at 15:20
  • \$\begingroup\$ I built it for visual purposes so opamp power pins might be swapped, and labels are for human eyes, not parsers, etc. Use Falstad instead \$\endgroup\$
    – DKNguyen
    Aug 6, 2020 at 15:39
  • \$\begingroup\$ After some pondering I was able to reverse with a second op-amp: circuitlab.com/circuit/9n5k5gsjrsvu/throttle, but I'm sure it's possible to do better than that. I also need to read more about Zener diodes and capacitors, since I don't really understand why you put them there. \$\endgroup\$ Aug 6, 2020 at 17:23
  • 1
    \$\begingroup\$ The zener is providng a 0.8V reference which gets subtracted from your input signal which shifts your 0.8-4.2V signal to 0-3.4V signal, prior to being amplified 1.5x relative to GND to become 0-5V. The capacitor is there because if the opamp needs a quick burst of current, the power supply isn't fast enough and too far away which will cause the opamp to misbehave. \$\endgroup\$
    – DKNguyen
    Aug 6, 2020 at 18:31

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.