godfreyl
Advanced Member level 5
To design a power amplifier, all we are really trying to do is to make an opamp that can deliver high output currents so that we can use a circuit like the one below.
The very easy way to do that is to start with a normal opamp, and add an output buffer to give the extra output current needed. As you say, opamps like the TL074 are limited to supplies of +-18V, but there are opamps that work with higher voltage supplies.
If we do that, then there isn't much to learn though. It would be better to design our own high power opamp using discrete transistors. This can be made up from the three gain stages I explained earlier.
Before we get into those details, there are some important considerations even in the simplified circuit below.
Firstly, there are the obvious points:
A) Electrolytic capacitors are nasty.
They have lots of problems but the most important for audio circuits is that they create distortion. We want to have low distortion, so it is best not to pass the audio signal through electrolytic capacitors unless we really have to. Plastic capacitors e.g. polyester/mylar or polypropylene are much better.
C1 is only 1uF so there is no problem - we can use a plastic capacitor, but if we make the value of C1 too much higher, we will need a very large and expensive plastic capacitor, or an electrolytic. This also means we can't make R1 too small, because then C1 would need to be bigger.
C2 is a problem. The value has to be high (at least about 22uF), so we have to use an electrolytic. The answer here is to make the value much higher, say 220uF. This high value means that the signal voltage across the capacitor will be very small, so the distortion it causes will be very low.
B) DC offset.
We do not want any DC voltage across the loudspeaker. If we use a perfect opamp, then the capacitors in the circuit below would make sure there is never any DC voltage across the speaker, even if there is DC at the amplifier input. However opamps are not perfect (even ones we design ourselves).
The problem is that in our circuit, the inputs will be connected to the bases of the transistors in the first gain stage, and they will draw a small current. These currents flow through R1 and R3, causing a small voltage drop across them. If the voltage across those two resistors is the same, then they will cancel and there should be no DC voltage across the speaker.
There are four things we can do to make the voltages as close as possible:
So in summary, there are four guidelines:
The very easy way to do that is to start with a normal opamp, and add an output buffer to give the extra output current needed. As you say, opamps like the TL074 are limited to supplies of +-18V, but there are opamps that work with higher voltage supplies.
If we do that, then there isn't much to learn though. It would be better to design our own high power opamp using discrete transistors. This can be made up from the three gain stages I explained earlier.
Before we get into those details, there are some important considerations even in the simplified circuit below.
Firstly, there are the obvious points:
- R1 sets the input impedance to 22K.
- The ratio of R2 and R3 set the gain to 23, so a little less than 1V RMS input will give full power.
- The combination of R1 and C1 roll off the low frequency response at about 8Hz.
- The combination of R2 and C2 roll off the low frequency response again but at a much lower frequency, about 0.8Hz.
A) Electrolytic capacitors are nasty.
They have lots of problems but the most important for audio circuits is that they create distortion. We want to have low distortion, so it is best not to pass the audio signal through electrolytic capacitors unless we really have to. Plastic capacitors e.g. polyester/mylar or polypropylene are much better.
C1 is only 1uF so there is no problem - we can use a plastic capacitor, but if we make the value of C1 too much higher, we will need a very large and expensive plastic capacitor, or an electrolytic. This also means we can't make R1 too small, because then C1 would need to be bigger.
C2 is a problem. The value has to be high (at least about 22uF), so we have to use an electrolytic. The answer here is to make the value much higher, say 220uF. This high value means that the signal voltage across the capacitor will be very small, so the distortion it causes will be very low.
B) DC offset.
We do not want any DC voltage across the loudspeaker. If we use a perfect opamp, then the capacitors in the circuit below would make sure there is never any DC voltage across the speaker, even if there is DC at the amplifier input. However opamps are not perfect (even ones we design ourselves).
The problem is that in our circuit, the inputs will be connected to the bases of the transistors in the first gain stage, and they will draw a small current. These currents flow through R1 and R3, causing a small voltage drop across them. If the voltage across those two resistors is the same, then they will cancel and there should be no DC voltage across the speaker.
There are four things we can do to make the voltages as close as possible:
- Make the currents as small as possible by running the input transistors at a low collector current and choosing transistors with high current gain.
- Make the currents as similar as possible by running both input transistors at the same collector current and using transistors with similar current gain.
- Make R1 and R3 the same value.
- Keep the value of R2 and R3 as small as possible.
So in summary, there are four guidelines:
- Make R1 and R3 the same value.
- If their value is too low, there will be a problem with the capacitors.
- If their value is too high, there will be a problem with DC offset.
- Make C2 a very high value to avoid distortion.
