1
\$\begingroup\$

I am wondering if there is any way for me to eliminate the DC offset present in this Class B amplifier circuit.

Proposed Amplifier Circuit

Waveform Exemplifying DC Offset Issue

The circuit is to be used to amplify music for a subwoofer and, as I understand it, a DC voltage across the speaker is potentially harmful. I built and tried a physical circuit that was almost identical (I think the resistors were 10k) that did amplify sound well but shifted the speaker's cone a noticeable amount immediately (even when no sound was playing), indicating a DC offset. The problem might be easily solved with a dual-rail supply as the output would then be centered around 0V, however, that is not possible for me to add to this design; I intend to have this circuit run off of a 12V battery. I assume there is a clever way to use some basic component to traverse this issue, but I am unsure what it might be.

My first idea was a coupling capacitor, as is used in many amplifier output stages. However, the coupling capacitor only allows the AC signal past it, making the voltage across the speaker somewhere in the millivolt range meaning that the current will be in the milliamp range. I assume this is a result of this circuit being more of a current amplifier/impdeance converter. So, I'm pretty sure that using a coupling capacitor is not at all the correct solution here, but please let me know if I'm wrong in this respect.

At this point, as far as I can see, I must maintain the same current while not having that unfortunate DC offset. Does this require an isolating output transformer? Though I was previously unfamiliar with the concept, I think I understand how a transformer can be used to make the relative voltages across the speaker terminals favorable while not eliminating any of the actual power amplification that is desired.

Additionally, is this even an optimized class B amplifier? I've tried to play around with the values and this seems to give a pretty optimal amplification, but I am not too knowledgable when it comes to the specifics of this amplifier class.

\$\endgroup\$
2
  • 1
    \$\begingroup\$ How does a 1mF output coupling capacitor fair in simulation? \$\endgroup\$
    – Michael
    Aug 13 at 5:11
  • \$\begingroup\$ Three letters: NFB. \$\endgroup\$
    – jonk
    Aug 13 at 5:12
3
\$\begingroup\$

My first idea was a coupling capacitor, as is used in many amplifier output stages. However, the coupling capacitor only allows the AC signal past it, making the voltage across the speaker somewhere in the millivolt range meaning that the current will be in the milliamp range.

You need to use the right value capacitor such as 4,700 μF. It has an impedance at 30 Hz of about 1 Ω and will pass current to your speaker pretty well.

Or, you can make two back-to-back push pull stages forming what is known as a bridge amplifier: -

enter image description here

The two push pull stages need to be biased the same but whereas one is fed an AC signal as per the original circuit, the other is fed an inverted AC signal. It also produces the potential for four times the power into your speaker.

\$\endgroup\$
1
  • \$\begingroup\$ Thank you; this was very helpful. I had considered a bridge arrangement at one point as I have previously built a H-Bridge inverter, but dismissed the idea before looking into it any more. \$\endgroup\$
    – SctKwl
    Aug 13 at 17:45

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.