Best transistor to use for audio amplifier

Question

This term, we will be designing an audio amplifier. So far in our lecture, we are still at BJT and based from what I've heard, FETs will just be partly discussed unlike the thorough one on BJT. Anyway, I would like to have an idea this early so I can plan on what transistor to use for best audio amplification. I have read some threads how the other transistor (BJT/FET) is better, but other forums say that the performance relies not on the component but on how the transistor is properly biased and how the circuit is properly designed.

In designing an audio amplifier, which of the four subtypes of transistors is the most efficient? (NPN/PNP/JFET/MOSFET)

By the way, the requirement of my professor is just this: impress me. Right now my group haven't decided yet on the specifics of the circuit (wattage, impedance, etc).

Answer 1

You could successfully build a audio amp from many different types of BJTs. It will be the circuit, not the transistor, that makes the amp work well. I'd pick jellybean parts like the 2N4401 (NPN) and 2N4403 (PNP) and stick with them for everything except for the final power output transistors. Lots of parts could fill that role. If you have your own favorite jellybean small signal transistors, use them if you prefer. The ones I mentioned have reasonable gain and can handle up to 40 V, which should be plenty good enough to allow for a amp to impress your profesor with.

There are lots of possible power transistors to use as the final output. If you are aiming for a few Watts, I'd probably go with basic parts like the TIP41 (NPN) and TIP42 (PNP).

Again though, it's not the choice of transistor that will make or break this project. You can certainly create a impressive audio amp with the transistors I mention, but you can also make a mess. It's really up to the design. In audio, overall noise and harmonic distortion are high priorities. Those come from careful circuit design and attention to these parameters at every step along the way.

You can also use other types of transistors, like JFETs or MOSFETs. Those would require a different circuit topology to utilize properly, but can be used to make a good amp too. Since you will be going over BJT details more thoroughly, I'd stick to them for now. This will be a great learning exercise. Designing a amp with very low noise and very low distortion is not trivial.

Answer 2

You will probably make a more effective power output stage using BJTs for the same number of components compared to MOSFETs. I use the word effective to mean that your output voltage will swing higher/bigger for the same power supply with BJTs used in a simple push-pull circuit. This is because, to turn-on a BJT, you only need about 0.6 to 0.7V whereas to get a MOSFET supplying several hundred milliamps you might need to drive its gate with 3 or 4 volts.

Again, this will be a simple emitter-follower push-pull class AB output stage. You can only drive the output transistors with a signal that is restricted to the power rails and if this is (say) 24V dc - you should be able to drive a signal that is 22Vp-p to the power transistors. Given that each BJT would "lose" 0.7 volts (because of the base emitter junction), the maximum output voltage will be about 20.6 volts peak to peak. If you were using mosfets, it would be more like 14 volts peak to peak into a decent load.

There is a bit of hand-waving in my answer so far but, just do your homework on mosfets connected as a source follower and pick one with the smalled Vgs(threshold) and examine the data sheet to see how much gate drive voltage is needed to get a few hundred milliamps flowing through it.

There are more complex designs that are quite difficult to get working where the output transistors are collector-connected or drain-connected but, for a beginner I'd stay away from these because they will be unstable if not carefully designed and, require more silicon to get working effectively.

So, given that you haven't specified power output, speaker load or voltage rails I'd say a BJT power output stage is probably the best choice. As for the other transistors I'd stick with BJTs - they've been used in tens of thousands of good commercial designs. You could of course consider a class A output stage using an output transformer - this is probably worth considering but the down-side is the loss in efficiency due to the final transistor biasing.

I've just had a look around for a fairly simple output stage that shows the biasing arrangement you'll likely need for a decent amplifier and came across this one: -

It came from this site. I'm recommending it because it seems to have a decent spec and the site also recommends a cut-down version without the diodes/biasing. I personally think it would be a good start for a beginner. The site discusses several things about what is needed to make a good output stage.

You can take the basic design and add gain to it and swap the op-amp out for individual transistors if you do a bit more research.

Answer 3

This is a bit of a late answer, but I'm hoping it may help someone asking the same questions.

I prefer BJTs, but MOSFETs are super easy to use and can outperform BJTs in terms of fidelity. Both can give excellent results, just use what you find you prefer. MOSFETs can generally handle higher supply voltages (higher max Vds). So design with what you feel most comfortable with (calculation wise) and if you feel equally comfortable with both, use random.org.

To add to what Andy aka said, just know that you will need to have a very complex design to get 0.7V under each rail as your output swing. This is because the amplifier stage of a BJT amplifier also needs the signal to drive it which normally cuts one of the rails' voltage down by roughly 10% (don't quote me on that number, it's just a general rule of thumb I use). And I don't think an op-amp amplifier will impress a professor. At least where I studied I would have failed outright had I used an op-amp. And besides, the maximum you can get out of one (with a carefully designed driver stage) is 18 W into 8 ohms - this using an NE5532 if I remember correctly. In general you're only looking at 10-15 W with an op amp. Firstly, an op amp requires 5 minutes to design and secondly the power is dismal.

And to add, using two diodes to bias a BJT output stage isn't particularly the best idea, unless you match your diodes and transistors perfectly and thermally connect the diodes and the output transistors. BJT amplifiers are very susceptible to thermal runaway. You'll probably find in practice that you end up with very high bias current if you use normal signal diodes. Use rectifier diodes if you're going to be using diodes - 1N4001.

Answer 4

Define "performance". Why are you interested in "efficiencey"? Transistors are used in audio amplifiers in different ways. You have discrete class-a circuits that overdrive well like the infamous Neve console mic pre. On paper an op amp designs will have the best performance (actually putting separate transistors in front of a conventional op amp is probably going to get close to the theoretical limit of performance). But more generally you have input transistors, gain transistors and output transistors.

Input transistors should be low noise. BJT tend to be lower noise if the correct source impedance (for op amps you can look this up in the datasheet by looking at voltage noise / current noise which for NE5534A at 30Hz is ~5.5/0.0015 = 3k7). JFET has super low current noise so they will tend to have better noise performance with high Z inputs.

Gain transistors should be low noise and high gain. I'm not sure about what makes a good output transistor. Bandwidth or thermal characteristics maybe.