Tubes, Bipolars and MOS
Having listened to some old amplifiers I have my doubts – a lot of it has to do with the design. Compare a quasi-complementary amplifier to the more expensive complementary unit and then one which is capacitively coupled to a direct coupled amplifier. The more expensive it gets the better the quality. Then your class A compared to class AB (or as near as dammit to class B). Reducing cross-over distortion is obviously of prime concern but then what about the linearity of the output stage? Is there phase shift?
There are so many parameters that one must take into acount when building and spec’cing an amplifier that it can make the head spin. I remember reading many years ago (in a British hobbyist magazine, I think it was ETI) – if the electronic specs are good you can be rest assured that what you hear is going to be good. This is a fact but still no substitute for your own hearing. Yes, your own hearing – not the salesman’s. It’s also peculiar how a gamer will spend 2 000U$ on all his hardware but the least on the power supply. Good amplifier design is so important but yet when the constructor gets down to the nitty gritty short changes himself by not spending on the power supply. I’d appreciate readers of this column to give me some feedback on their own experiences with power supplies and audio amplifiers. Expensive RF transmitters use regulated power supplies throughout – not only in the synthesiser. Likewise spending a bit of money on the output transistors can make a lot of difference when taking headroom/ceiling, frequency response, slew rate, etc into account. But I digress – the audio amplifier has improved in leaps and bounds over the years. The biggest flaw of the bipolar transistor, I think, is the temperature issue. They do not like to run hot and easily self destruct. High power amplifiers (200W RMS per channel or more) often have very sophisticated power supplies, or should have and exhaustive measures taken to reduce thermal runaway. Bipolar transistors also have some very strange quircks when looking at the load line – specifically when being driven by an a.c. source – look at the gain and current graphs, ummm, a lot of design needed here. Look up transconductance on Wiki – should give you a very good idea).  Another issue is that bipolar device amplifiers have many stages to get to the final result – the output stage and low impedance load.
MOS devices and tubes can run hot – this uniqueness makes them ideal for Class A amplifiers.  Their transconductance makes them ideal for class A amplifiers. And we all know that class A amplifiers are the best, right? With that sorted out now you know why I mentioned the power supply as been so important. Class A’s are juggernauts when it comes to power hogging and are not so green. But the quality of sound is unbelievable. Go there first, don’t stop, don’t pay a fine – try it, test it and don’t give it back.
Do try to get an old tube amplifier and make a comparison. I think an ideal starting point would be something like a 6W EL84 unit. It’s not going to break your walls down – but when listening to quality who cares.
I sometimes wonder if I’m not giving my age away…