2
\$\begingroup\$

I'm a hobbyist so my theoretical knowledge surrounding electronics is pretty limited.

I'm working on modifying a synthesizer that uses a sine wave control signal whose frequency is in the sub-audio range. The signal peaks at -5v & 5v, and is used to control things like pitch and amplitude of a signal.

I would like to add a potentiometer or some other control to the circuit so that this signal can be attenuated by the person playing the instrument. I would like it so that the potentiometer could allow the signal's amplitude to be changed over time, depending on how the user turns the knob/moves the slider.

To further clarify I've attached an image.

enter image description here

the blue line represents 0v, and the red line is to illustrate the scaling function as it works over time.

My uncertainty is about whether or not a simple circuit can achieve this, and if it is as simple as using ohm's law how I apply it in this situation to pick the proper resistance of a potentiometer.

\$\endgroup\$
  • \$\begingroup\$ Scale with regards to what? \$\endgroup\$ – Ignacio Vazquez-Abrams Apr 27 '14 at 4:15
  • 4
    \$\begingroup\$ What do you mean by "0:0"? If you want no output signal, why not just disconnect the input? \$\endgroup\$ – The Photon Apr 27 '14 at 4:29
  • \$\begingroup\$ Just to be clear ... your control signal can be in the range -5v to +5v? and you want to reduce it's range to -0.05v to +0.05v? This is just a guess because unless you disconect the signal you will never get 0v and if I'm not given a requirement I work with 1% if it's easy. \$\endgroup\$ – Spoon Apr 27 '14 at 8:45
  • \$\begingroup\$ By "AC" do you mean that the "+5V" and "-5V" are 180 degreees out-of-phase with respect to a circuit common voltage? ( possibly ground? ) \$\endgroup\$ – FiddyOhm Apr 27 '14 at 13:04
  • \$\begingroup\$ In re-reading my original post I realized that it lacked clarity. I've updated my post so that it hopefully makes some semblance of sense now. My lack of the appropriate domain-specific language is somewhat limiting here, so I hope my edit has cleared things up. \$\endgroup\$ – pdx_interactives Apr 29 '14 at 2:31
4
\$\begingroup\$

Your low-frequency control signal is no different than any other audio signal. You can control it with a simple potentiometer in the conventional way:

schematic

simulate this circuit – Schematic created using CircuitLab

The chosen resistance of the potentiometer depends on the circuit you connect it to. Usually, the wave generator will have a low impedance output and so isn't much affected by the choice of resistance (at "input"). So use a pot that compares reasonably to the input impedance of the stage that follows (connected to "output"). Note the resistance decreases anyway as the control is adjusted to a minimum.

\$\endgroup\$
4
\$\begingroup\$

Assumption: The requirement is for a controlled, deterministic attenuation.

A complete attenuation to zero signal is impractical using a voltage divider. In essence, a zero signal would be the equivalent of disconnecting the input entirely, presenting a short-circuit at the output. The answer by gbarry covers this simply and excellently.

Other than this, an attenuator will typically attenuate the signal asymptotically - approaching zero but not quite getting there.

This arrangement would get you approximately 100:1 attenuation (-40 dB), which might serve the intended purpose.

schematic

simulate this circuit – Schematic created using CircuitLab

When the potentiometer is at the bottom position (zero Ohms) the output signal would be +/- 5 Volts at maximum input signal. When the pot is at the top position (100k Ohms) the output would be +/-0.05 Volts. Thus, an attenuation range of 100:1 is achieved.

A voltage follower at the output is recommended, if the load to be driven is low impedance.

For the same -40 dB attenuation, an alternative is a precision attenuator such as described in this note.

Precision attenuator

If a greater attenuation is required, higher values of the potentiometer could be used, but at some point the resistive noise from extremely high pot resistance would smother the actual signal below usability. At that point, it would make sense to use an op-amp based attenuation circuit.

Alternatively, for high precision high attenuation, specialized ICs such as the LM1972 (-78 db attenuation, 7700:1) might be preferable. These are essentially single-chip digital potentiometer devices specialized for audio type signals.

\$\endgroup\$

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.