Hi all
Without getting into a "which is best" war over output caps or no output caps. I can try and clear up some of the pros and cons of each.
Most audio power amplifiers use DC coupling of the loudspeaker to the amplifier with the amplifier having both + and - power supplys.
The main advantage of this is that the output circuit and loudspeaker have a very low output impedance (AC resistance). This not only means a more efficient drive to the loudspeaker but also means that the amplifier has better control of the loudspeaker particularly when the amp has to stop the loudspeaker cone moving (Amplifier damping factor is the key term here - the lower the output resistance, the lower the damping factor and control of the loudspeaker).
The BIG downside of DC coupling (especially to people who short out fuses!) is that if the amplifier output devices fail or 1 supply fails the poor old loudspeaker gets a destructive DC current flowing through it and the life of the voice coil is usually measured in seconds!
Putting a fuse in the Loudspeaker lead helps to protect the loudspeaker but ruins the whole point of DC coupling due to the extra resistance of the fuse!
The alternative to this is to run the amplifier off a single supply which means that the loudspeaker feed will be at half supply with no signal so you need a large capacitor to block the DC to the loudspeaker. The advantage of this setup is that if theres an amp failure there is no danger of a destructive DC current damaging your precious loudspeaker.
The downside of this is that the output capacitor has impedence as well which is why it should be as big as possible (a 2200uF has an impedence of about 1.5ohm at 30Hz). This extra resistance can significantly affect the damping factor and AC coupled amps tend to have less control over the loudspeaker and can sound less "tight" on very low bass.
Its one of the reasons that AC coupled amps like the Nytech tend to work better with Infinite baffle (sealed box) loudspeakers as these speakers are heavely damped by the sealed box.
It also creates a whacking great thump when you swich on which wont damage the loudspeaker but it is a bit of a shock if you're not used to it
(The little add on boards on some of the pictures on the forum are "anti thump" circuits designed to reduce this thump - The later Ion oblisk amps had them built into the main circuit)
One of Richard Hay's (Nytechs original co founder) brilliant ideas was to include the output capacitor in the feedback loop. Therefore the gain of the amplifier increased slightly at very low frequencies as the feedback reduced due to the increasing impedance of the output capacitor (-ve feedback so less feedback, more gain). It also improved the damping factor (although not as much as a DC coupled amp could).
This technique certainly contributes to the charictaristic sound of a classic Nytech amp but does also mean that theyre not at their best driving a modern ported or transmission line speaker due to their higher damping factor.
In answer to NWRlers question - Because the output cap is included in the feedback path, the effect of a larger capacitor will just mean less feedback at VLF so as long as the feedback is able to compensate for the capacitor then theres theoretically no improvment in doing it. However a larger capacitor generally has a higher ripple current capability which will help improve the damping factor slightly. You will also get a much bigger switch on thump!
I tried 4700uF o/p caps instead of the original 2200uF ones in my CPA602 and could not detect any difference! - Going to 10000uF appeared to tighten the bass at VLF (couldnt hear any differnece on my 050's - Only noticed it on the 202's) but the caps were too big to get on the board
I'm not going to venture which setup is best - There will be as many different views on that as there are people on the forum
However I've lived with my Nytech setup for 32 years now and it still sees off a lot of the modern equivilents so "low fi" it certainly aint!
I've now got the CPA 602 power amp and EXO102 Crossover board circuit diagrams finished and I'll upload them to the forum (as soon as I've figured out how to do it
