Aside from not needing the complexity of a DC offset servo circuit, is there any inherent advantage technically or especially sonically to using a capacitor coupled output? As I recall, the JLH Class A amplifier is capacitor coupled, for example. I've been researching this over the weekend after becoming intrigued with the Ion (now Heed) Obelisk circuit. Heed claims an output capacitor acts as a "buffer" or "translator" to more naturally couple the amp to the speakers akin to the output transformers of a tube amp or how Macintosh uses them. Is this claim simple marketing hype or is there something possibly advantageous here?
I believe to remind an article Papa wrote about it, where he put a cap across the wire at the speaker-input to perfectly match something (capacitance? Whatever?) and he was able to measure an improvement of something‘s waveforms to some amount. All that was way beyond 20 khz, so completely inaudible but still an improvement.
Sorry for not being able to answer your question in a „more sensible“ way (that is, at all), but there’s something to it. (Which, since inaudible, can be considered as hype
)
Sorry for not being able to answer your question in a „more sensible“ way (that is, at all), but there’s something to it. (Which, since inaudible, can be considered as hype
)
Most SS amps are capacitor coupled on the output. DC servos are less common, and there are even some transformer coupled SS amps.
As far as capacitors being more natural........yeah no. That is complete marketing nonsense.
If you don't want the complexity of a DC servo, I would actually look at using a transformer coupled output. A transformer can simply block DC like a capacitor would, or you can use one to make some pretty unusual topologies.
As far as capacitors being more natural........yeah no. That is complete marketing nonsense.
If you don't want the complexity of a DC servo, I would actually look at using a transformer coupled output. A transformer can simply block DC like a capacitor would, or you can use one to make some pretty unusual topologies.
Very few ss amplifiers have output capacitor coupling. Most ss amps are direct coupled,
with a large capacitor in the nfb loop to reduce the LF gain to unity, to minimize DC offset.
Last edited:
Shrug.
When I look for SS schematics, or work on SS amplifiers, about 60-70% are capacitor coupled.
Even when I go to audio shows where I would expect to see a majority of DC coupled amps, I still see about a 50/50 split between capacitor coupled and DC coupled amps.
Obviously my experiences probably are completely anecdotal, but my observations make some amount of logical sense too. It is a lot easier to chuck in some 3000uf cap at the output and call it done than it is to make a dc servo.
When I look for SS schematics, or work on SS amplifiers, about 60-70% are capacitor coupled.
Even when I go to audio shows where I would expect to see a majority of DC coupled amps, I still see about a 50/50 split between capacitor coupled and DC coupled amps.
Obviously my experiences probably are completely anecdotal, but my observations make some amount of logical sense too. It is a lot easier to chuck in some 3000uf cap at the output and call it done than it is to make a dc servo.
Capacitor coupled output is as rare as hen's teeth today. And by today i mean any time between the early 70s and now. Perhaps 1% of show-worthy amps are cap coupled.
DC coupled means something entirely different and has no bearing upon the speaker interface.
Actually, that is much more expensive to do. One of the last commercially successful solid state amplifiers
(selling 100s of thousands of units) having capacitor coupled outputs was the Dynaco Stereo 120, around 1966.
Last edited:
The majority of SS amps are direct coupled to the speakers without the need for a DC servo. Just a nfb decoupling cap. There is an old argument whether a servo is indeed the lesser evil sound wise.
Not sure if i buy into the theory about output coupling caps improving anything. IME they always place some limit on the subjective bass performance and need to be pretty good not to screw up the entire landscape too.
A friend swears by the effect of autoformers and uses them with all his SS amps regardless of transformer ratio. I guess he just likes the sonic effect of that big coil at the output. Or the sonic effect of hysteresis
> more expensive to do
Just count.
A mono single-supply amp needs 2 big caps. A mono bipolar supply amp needs 2 big caps.
A stereo single-supply amp needs 3 big caps. A stereo bipolar supply amp needs 2 big caps.
Yes there are differences in cost because value and voltage. The cost may be like 3/4 rather than 2/3. Depending also on bass expectations. The Dynaco Stereo 80 had smaller caps than the ST120, and marginally lesser bass specs.
Many many exceptions. The first big amp I worked on was mono bipolar, just designer's choice. And many stereo single-supply, just being conservative.
A explicit "servo" sure is not needed. Only a differential pair. Even poorly balanced it has <50mV offset, and that won't hurt any speaker.
Just count.
A mono single-supply amp needs 2 big caps. A mono bipolar supply amp needs 2 big caps.
A stereo single-supply amp needs 3 big caps. A stereo bipolar supply amp needs 2 big caps.
Yes there are differences in cost because value and voltage. The cost may be like 3/4 rather than 2/3. Depending also on bass expectations. The Dynaco Stereo 80 had smaller caps than the ST120, and marginally lesser bass specs.
Many many exceptions. The first big amp I worked on was mono bipolar, just designer's choice. And many stereo single-supply, just being conservative.
A explicit "servo" sure is not needed. Only a differential pair. Even poorly balanced it has <50mV offset, and that won't hurt any speaker.
A friend swears by the effect of autoformers and uses them with all his SS amps regardless of transformer ratio.
I guess he just likes the sonic effect of that big coil at the output. Or the sonic effect of hysteresis
They certainly have a certain "special" sound. An advantage is that shorting the outputs has no effect.
I always thought the output capacitor simplified the power supply / power transformer. Someone has to wire in that center tap for +/- and that adds cost. Didnt the Stereo 120 have a series pass regulated supply? Two of those - or an output cap. Its easy to see which way would be cheaper to manufacture...
The Dynaco Stereo 120 was a single supply circuit, with a capacitor coupled output.
It also had a transistor/Zener series regulated 72VDC power supply, shared between channels.
The three heat sinks were smallish, flat rectangular pieces of black aluminum,
bolted to the chassis bottom. The outout inductors were wound by the builder around the
output coupling capacitors.
Dynaco Stereo 120 Stereo Power Amplifier Manual | HiFi Engine
It also had a transistor/Zener series regulated 72VDC power supply, shared between channels.
The three heat sinks were smallish, flat rectangular pieces of black aluminum,
bolted to the chassis bottom. The outout inductors were wound by the builder around the
output coupling capacitors.
Dynaco Stereo 120 Stereo Power Amplifier Manual | HiFi Engine
Last edited:
