I have an input to a system that requires a 10mVpp max signal. I have to use my computer to input it due to constraints out of my control. I am well aware an audio card cannot go down that low. I have several attunators, one active and one passive (just a potentiometer and some resistors).

no matter what I do I cannot get the signal to go that low. I have also tried introducing a digital amplifier (just some high quality digital speakers with an audio out) into the system between the computer and the attenuators but had no success.

Where can I go next? what other information can I provide to help.

I am running audacity on a dell optiplex 760 (my computer) and it has the manufacturers audio card in it. I did some quick research and it seems most built in audio cards have about 46Ohm impedence. I couldn't find info on my exact card. It maxes out its output at about 3Vpp and starts to struggle keeping its output stable below 150mVpp.

I am also using a probe master 20:1/200:1 active attenuator (model 4231) and a homemade potentiometer attenuator.. about as simple as it sounds.

ANY help is appriciated.. this is stumping several of us and we are unsure of where to go next.

thanks in advance!

  • \$\begingroup\$ Is Ohm's law not working for you? A simple resistor divider should be able to provide an arbitrarily low signal. If the frequency is very high you may get capacitive feed through from poor layout/wiring but you don't say what kind of signal you are trying to attenuate. \$\endgroup\$
    – John D
    Commented May 2, 2016 at 19:43
  • \$\begingroup\$ This should be simple. If your voltage divider does not work, there's an underlying issue. Please draw a systems diagram so we can see what you're trying to do, and where you're attaching the probe, and such things. \$\endgroup\$
    – pipe
    Commented May 2, 2016 at 19:50
  • \$\begingroup\$ What's wrong with your sound card? A 16-bit sound card set to 1Vrms output at full scale (2.828 Vpp) should have a resolution of about 43 uV. a 10 mVpp signal would be generated by numbers in the range of -116 to +116. \$\endgroup\$
    – Dave Tweed
    Commented May 2, 2016 at 19:52
  • \$\begingroup\$ so it turns out it must have been a quality issue.. although I am still unsure why I was getting absolutely nothing through the potentiometer (yes its wired correctly and has a large bandwidth.. its a unit that has has passed through production and has been used in other applications so I would hope it works! haha). I finally found someone with a high precision handmade attenuator used for ham radios.. and it worked flawlessly. Lesson learned would be when in doubt about integrity, get high quality parts ;) thanks! \$\endgroup\$
    – scarlso9
    Commented May 2, 2016 at 21:10
  • \$\begingroup\$ The attenuator that finally worked was similar to this JFW industries inc model 50TA-007 BNC \$\endgroup\$
    – scarlso9
    Commented May 2, 2016 at 21:10

1 Answer 1



simulate this circuit – Schematic created using CircuitLab

Figure 1. Audio attenuator. This gives a signal output from zero to 1% of the input, depending on wiper position.

The circuit above is what you require. Make sure you identify the wiper correctly. It's normally the centre pin of three.

I did some quick research and it seems most built in audio cards have about 46 Ohm impedance.

That's fine. This circuit won't load that to any significant extent.


Your Answer

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

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