Similar topics
Recommended
John Caldwell

In this month’s post on our sister blog, the “Precision Hub” on Planet Analog, I share a recent question I got from a customer in the Precision Amplifier Forum here in the TI E2E™ Community.

His circuit used an op amp to amplify the output of a microphone at very low frequencies. He used a large (47uF) AC coupling capacitor and a high input resistance (100kΩ) to achieve a low corner frequency for his measurements.

Unfortunately, a significant amount of DC offset was appearing at the output of the op amp – almost a volt!

What could cause this?

Check out my full post, “Plug that leak: Look out for capacitor leakage!,” on Planet Analog to find out.

Anonymous

  • Hi Neil,

    All great points, the customer was already using a FET input amplifier (TL082) and reducing the capacitance value to allow for the use of different dielectric types was the eventual solution that I recommended.

    On the contrary, because the voltage drop across AC coupling capacitors is low within the desired bandwidth, the distortion electrolytic capacitors contribute is low when used in this way. I have also written about these effects in ceramic capacitors with high-k dielectrics (www.edn.com/.../Signal-distortion-from-high-K-ceramic-capacitors) In any case, capacitor distortion is proportional to the voltage across the capacitor. I agree that electrolytic capacitors would contribute distortion in active filter circuits, and at frequencies below the desired cutoff in coupling applications.

  • John;

    Wouldn't a more straightforward approach be to simply use a good FET op amp such as an OPA134 or a low noise CMOS type and take advantage of its very low input bias current to increase R2 by an order of magnitude or two? If R2 is 1M then C1 needs to be 4.7uF; if R2 is 10M, then C1 can be 470nF. These values can be achieved with a much smaller film capacitor or even a ceramic type.

    Since the voltage is low, a ceramic type is not out of the question even for a 4.7uF. This appears to be an electret microphone, which has a high output level so noise is not a serious concern. Electrolytic capacitors have poor distortion characteristics anyway-- see Walter Jung's work on characterizing distortion of capacitors for audio applications.

    Regards, Neil