Simple, Low-Power Audio Amplifier
While browse the internet, especially on electronics-related websites, I came across a simple audio amplifier circuit or schematic that uses only three transistors. It seems the designer intended it for low-power speakers, probably just 2 to 3 watts, similar to an extension speaker to boost the audio signal from a laptop, phone, or other audio devices.
I took a brief look, and my intuition told me that while the amplifier would likely produce sound at its specified wattage, the distortion would be quite high. I noticed the circuit was a bit unconventional because it used a bias for the power stage that would inevitably fluctuate, even though it was said to be used for limiting overcurrent. In my experience, if that was the goal, it wouldn't work as intended.
However, out of curiosity to see if my intuition was correct, I ran a simulation. And I was right—the circuit's amplification was not linear or symmetrical. For simple **playback**, this might not be an issue, but for audio enthusiasts, it would be a significant consideration.
Therefore, I modified the circuit (based on my knowledge) to achieve more linear amplification. The simulation results from LTspice (which I've used for over a decade and trust) showed an accuracy level of over 96%. Any deviation in a real-world build would be due to the fact that physical components are never ideal. The improved circuit is as follows:
My Improvements
- I used a constant current source of about 5 mA for the bias of the power transistors Q3 and Q4 (or the active load of the front amp Q1). The voltage reference uses two standard signal diodes in series, with a bias of about 0.4 mA, which is set by the chosen value of R4.
- The feedback resistor R2 remains 100 K$\Omega$
- Because this amplifier configuration is inverting, it's necessary to add an input resistor (R1), with a value that can be adjusted from about 2.5 to 3 K$\Omega$ to achieve a voltage gain of approximately 28 to 30 dB.
- To improve DC stability over temperature change, a feedback resistor R3 of 33 to 100 ohms is installed on the emitter of the input stage (Q1), a degenerative resistor. This resistor also improves the dynamic range of the input stage.
- For a slightly higher AC gain, it is traditional to parallel a capacitor of around 100 ~ 200 uF across the emitter resistor. This is optional.
- To control the size of the zero-voltage crossover at the output, which means adjusting linearity and THD (total harmonic distortion), the bases of Q3 and Q4 are forced to be spaced apart by 2 x 0.7V. This is achieved by placing two signal diodes, such as 1N914 or 1N4148, in series, so the two bases are about 1.4 volts apart. This places the amplifier in Class B mode.
- If you want to achieve lower distortion or shift towards Class AB, the spacing between the two bases can be increased, with the consequence that transistors Q3 and Q4 will run hotter.
- To do this, you can use a series of three or four diodes or even an LED. Alternatively, you can use a classic VBE multiplier, which consists of a transistor and two resistors.
- Since the application is for low power, perhaps a maximum of around 2 watts, the speaker impedance cannot be too low. The optimal impedance is around 16 ohms.
The second image shows the LTspice simulation results. The output is very linear and quite symmetrical between the positive and negative signals.
The third image shows a -3dB Bandwidth that reaches 120 KHz, which is quite good for a simple amplifier.
Please feel free to try it. However, it's important to remember that in practice, there are always technical factors that can cause a circuit's performance to differ slightly from the simulation results. It could be a little better or a little worse. That's where the importance of practice and precision comes in. That's the essence of the Art of electronics.
I hope this is useful, and TABIK !
(This post is parallel to the status on the FaceBookGroup The Art of Electronics with the same topic)
Comments
Post a Comment