I have a piece of electronic that controls a motor. It takes power from AC power.

Now there is a 10KOhm Potentiometer. The data sheet of this controller says that the control signal is 10Vcc.

Here's the connections diagram (in Italian):

enter image description here

I'd like to emulate the Potentiometer with Arduino. I'm not sure if PWM will do the job.

Any hints?

  • 1
    I'm not sure your question is clear enough. Can you tell us more about what you want to accomplish? What will be receiving input from this emulated pot? – Ricardo Jun 5 '14 at 13:10
  • all right.. is it better? – nkint Jun 5 '14 at 13:43
  • It's better, thanks! But it's still not enough I suppose. Do you have more info on the board you want to modify? Maybe a picture of the board. Or even better, can you share the datasheet you have? – Ricardo Jun 5 '14 at 13:52
  • 1
    Usually a pot is part of a voltage divider. I'm assuming that 10Vcc is the voltage at one of the far terminals of the pot. The other end is usually grounded and the middle pin goes from 0 to 10Vcc. But there's no way of knowing if that's what's going on. If it is, you would need a DAC (digital to analog converter) to simulate it... – Ricardo Jun 5 '14 at 13:56
  • 1
    Hi ricardo, the data sheet is in italian sorry. It is here: we.tl/Kk89gm2PRL – nkint Jun 5 '14 at 14:07

Adding to Ricardo's answer:

PWM can do the job, sort of. You need a low pass filter (LPF) to convert the PWM to an analog voltage and an amplifier to take it from 0-5 to 0-10V. Or you can get a dedicated digital-to-analog (DAC) chip but that will obviously be more expensive than a resistor and capacitor for the LPF.

Here's some tutorials on converting your PWM to an analog voltage:

Analog-Output: How-To Arduino’s AnalogWrite – Converting PWM to a Voltage

Then you need to basically double the voltage. You do that with an operational amplifier. The resistors you pick for the op amp set its "gain", in this case you want it to be 2 so you can go from ~5v (max output) to 10v.

Here's an example of that: My First OP-AMP Circuit

  • You can avoid the op-amp by using a PNP transistor to switch the 10v, and pulling down to ground with a resistor, and then filtering that, but the op amp way might make more sense. – EternityForest Jun 12 '14 at 11:49

You can take an analog (PWM) output from the Arduino and turn it into 0 to 10 V DC using something like this:

Op-amp buffer


The op-amp shown is hooked up as a non-inverting amplifier, where the gain is given by:

VOut / VIn = 1 + (R2 / R1)

Since R2 and R1 in this case are both 1 k then the output will be the input multiplied by:

1 + (1000/1000) = 2

Thus, if we have 5 V coming from D3 we will have 10 V at the output of the op-amp. We need to supply a bit more than 10 V to the op-amp supply pin (pin 8) because this is not a rail-to-rail op-amp, that is, it cannot output all the way up to its supply voltage.

Low-pass filter

The resistor R3 and the 10 µF capacitor form a low-pass filter that filters the PWM signal from the Arduino into a more-or-less smooth DC voltage.

This shows the difference between the Arduino PWM output, and the output of the op-amp:

PWM and op-amp output

You can see from the scope traces that the yellow line (the PWM output) has a 50% duty cycle, and the blue line (the op-amp output) is 5 V and has a slight ripple.

With a 100% duty cycle, the output is 10 V.


Switching the scope to AC coupling we can see how much ripple we have in the op-amp output:

Ripple on PWM output

The scale for the blue channel has changed in this image, and the cursor shows that we have around 140 mV of ripple.

We can reduce the ripple by using a larger resistor for R3. For example, making it 10 k reduces the ripple to 86 mV.

The larger resistance make the output less responsive to changes at the input (it takes time to charge or discharge the capacitor) but in testing it still seems pretty responsive.


Test code:

void setup () 

void loop () 
  analogWrite (3, analogRead (A0) / 4);

I had a potentiometer connected to A0 and adjusted it to alter the duty cycle of the PWM output on pin D3.


You might ask "Why 138 mV of ripple? Why not 50 mV? Why not 200 mV?".

Let's look at the theory. First, the image posted above looks a bit noisy, so let's take an average of the input to the op-amp:

Ripple averaged

Now we can see that the ripple is about 57 mV (the output of the op-amp would be double that). Indeed, the averaged output is 57 * 2 = 114 mV

Ripple on op-amp output

The general formula for calculating the time taken by a RC network to reach a certain voltage is:

t = -log((V-Vc)/V)R*C

Where "exp" is the natural exponent.

See Ladyada - RC Delay Calculator.

Also Mathematical treatment of charging and discharging a capacitor.

V  = supply voltage   (initial voltage)
Vc = output voltage   (target voltage)
R  = resistance in ohms
C  = capacitance in farads
t  = time in seconds

These figures will be on the RC discharge graph at roughly the point indicated:

RC discharge graph

See Wikipedia - RC circuit

Introducing our values for the resistor and capacitor (using Lua) we get:

VS = 5.0       -- supply voltage
VMED = VS / 2  -- 50% duty cycle
VL = VMED - 27.2e-3   -- low voltage as measured
VH = VMED + 29.6e-3   -- high voltage as measured
RES = 4700            -- resistance in ohms
CAP = 10e-6           -- capacitance in farads

T1 = -math.log ( (VS - VH) /VS) * RES * CAP 
T2 = -math.log ( (VS - VL) /VS) * RES * CAP 

print ("T1 =", T1)
print ("T2 =", T2)
print ("diff = ", T1 - T2, "seconds")


T1 = 0.0331
T2 = 0.0321
diff =  0.0010 seconds

The time difference is 1 ms which agrees with the observations.

  • Great solution, but a small error in the circuit diagram ... R2 should be between pin 1 and 2, not 4. Pin 4 is ground. – user23709 Jul 8 '16 at 7:47
  • Quite right! How come no-one noticed that for a year? I've converted your "answer" to a comment. With a bit more reputation, you will be able to post comments. Thanks for pointing it out. And, welcome to Arduino Stack Exchange. :) – Nick Gammon Jul 9 '16 at 7:42

You can use any typical motor control chip like a ULN2003 with PWM and maybe a smoothing capacitor, Google "ULN2003 Arduino" for many examples.

Based on the pinouts from Ricardo's answer, you'd get the 10V from pin 1 and ground from pin 3, and feed the PWM'd output from the ULN2003 back into pin 2.

  • that ia darlington transistor right? I'm still a student in electronics.. does it means that a darlington transistor can amplify current right? but what for pwm VS analog signal? – nkint Jun 6 '14 at 17:52
  • @nkint Yes, you're using a Darlington like a fast relay switch, using the Arduino's 5V output to control the device's 10V. The switching speed makes it suitable for use with PWM from the Arduino. You could add a capacitor to help smooth the output to make it more like analog. – sburlappp Jun 7 '14 at 12:43

I won't be able to answer your question completely, but hopefully I'll be able to make things a bit clearer for someone more knowledgeable to complete the answer.

From what I could understand from the datasheet, inputs 1, 2 and 3 are used to connect a pot that will control the motor speed. It's the POTENZIOMETRO DI COMANDO (control pot) in the figure linked in the question.

The datasheet also specifies (in the figure below) that you can connect terminal 3 to ground and feed terminal 2 with a voltage from 0 to 10Vcc, that will control the motor. That's how you control the motor using a MCU, I suppose.

enter image description here

Here's where my knowledge is getting short. I'm guessing that a PWM signal won't do the trick. You'll need to feed an actual voltage in terminal 2 to control the motor properly. To do that you have a few alternatives:

  1. Use an ADC IC
  2. Use an R2R resistor ladder network using digital outputs and some means to amplify its voltage to 10Vcc.

I hope this helps a bit.


my solution is simple get a servo (pwm input) and connect the shaft of the servo to the pot with a coupler then just use a simple servo control program

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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