0
\$\begingroup\$

I have a LMC555 which generates a audio signal between 440-880 Hz. The signal is DC between 0-3V after the chip. To use it as input for an amplifier. The amplifier, a LM386N-1, starts clipping the signal quite early, somewhere around 20mV. Currently I use this solution:

schematic

simulate this circuit – Schematic created using CircuitLab

I lack the really deep theoretical understanding. As I know, each resistor in series will increase the noise of the signal. Higher resistance will cause more Johnson–Nyquist noise.

Here I have a 1MΩ resistor in front of the amp input, so I wonder...

...is there an alternative solution to this which causes less noise?

\$\endgroup\$
  • 1
    \$\begingroup\$ Why are you attenuating it and then feeding it into an amplifier? Have you measured the actual noise? A quick online calculator suggests it's four microvolts .. \$\endgroup\$ – pjc50 Aug 16 '16 at 20:27
  • \$\begingroup\$ @pjc50 I am attenuating it, because the amplifier starts clipping the signal at 20mV. \$\endgroup\$ – Flovdis Aug 16 '16 at 20:58
  • \$\begingroup\$ It seems to me that the problem is in the amplifier, is it a microphone amplifier? \$\endgroup\$ – pipe Aug 16 '16 at 21:07
  • \$\begingroup\$ Reduce the gain of the amplifier... esp if 20mV clips the output. The gain that high is asking for trouble if you don't need it... \$\endgroup\$ – Spoon Aug 16 '16 at 22:46
2
\$\begingroup\$
  1. Lower resistance will give you less Johnson noise. I seriously doubt your 555 output is going to be affected noticeably by Johnson-Nyquist noise. It is 'white' noise which sounds like old time TV static.

  2. You may be able to avoid some wiper noise (when the pot is turned) by capacitively coupling to the divider since the 555 output is unipolar and thus effectively has an offset. That would be a scratchy noise when the pot is adjusted.

  3. Since you are drawing the signal from a digital circuit, any noise on the power supply or ground can cause noise in the output. If you are hearing hum that is mains-related (50Hz/60Hz/100Hz/120Hz) that is probably the source.

  4. Regarding your title- you can substantially reduce the noise in a 'DC Audio Signal' by shorting the signal to ground, since we can't hear DC.

\$\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.