[xnor]

Yeah.. There are many free DIY amp designs available online, I suggest you take a closer look at those.

 

[Pageygeeza]

Quote:

Yeah.. There are many free DIY amp designs available online, I suggest you take a closer look at those.

I've looked at loads, but they only pose more questions as to how they shunt the current about.
 
Maybe I'm looking into the whole thing far too deeply, but I'll try to elaborate.
 
Say the left channels current is flowing from the amp to ground and the right channels current is also flowing to ground, on a 3 amp setup, wouldn't the ground be working twice as hard?
 
And say maybe the left and right channels currents are flowing in opposite directions to each other, that would mean that the ground channel current would have to flow in both directions at the same time.  :S

 

[nikongod]

Quote:

And say maybe the left and right channels currents are flowing in opposite directions to each other, that would mean that the ground channel current would have to flow in both directions at the same time.  :S

In a 3 channel amp, if the voltages (and correspondingly currents) from the left and right channels are exactly opposite (WRT ground) the current in the ground channel is 0a.
 
As you noted, the ground channel current can (and more often than not will be) twice the left or right board current.
 
Now this brings up the painfully obvious point that the ground channel may drop class-A before the "audio" channels (which is an odd but practical way of describing it because the ground channel is certainly part of the audio path) To which one must first answer: How much class-A current do you NEED. Once you have accounted for ~20db of overhead, how much do you need?

 

[Pageygeeza]

I take it when ground is 0v then basically that one channel is flowing to the other?  Surely that would misshape the sound wave?
 
This is why I've been dreaming of a 4 channel design.  The left and right ground channels would be amped seperately.
 
Dang, do you know of any documentation that would explain oddities like this?

 

[nikongod]

Quote:

I take it when ground is 0v then basically that one channel is flowing to the other?  Surely that would misshape the sound wave?

 
Ground is a constant, and unless you have a special application 0V. Even in special applications the key with ground is that it is a stable reference for the rest of the circuit and does not vary.
 
The example you asked about is a VERY specific case, and as you say the signal travels purely from left to right. To answer this quoted question, it does not matter! There is no distortion of the signal because of this effect.
 
Quote:

This is why I've been dreaming of a 4 channel design.  The left and right ground channels would be amped seperately.

 
If you have a 4 channel design you may as well put a 4-pin XLR on the headphones and run your system balanced.
 
Quote:

Dang, do you know of any documentation that would explain oddities like this?

They are as you say, oddities. Probably the best thing to do if you want to understand them is draw up a block diagram and note where currents go.

 

[Pageygeeza]

Cheers for that Nikongod.  
 
The long and the short of this is, does it or could it make any difference to the output signal?  
 
I know people are still looking for audio excellence, but if the ground for each channel could be separated all the way through the complete audio circuit (from source to headphones), would it make any difference?

 

[Mehve]

Quote:

Cheers for that Nikongod.  
 
The long and the short of this is, does it or could it make any difference to the output signal?  
 
I know people are still looking for audio excellence, but if the ground for each channel could be separated all the way through the complete audio circuit (from source to headphones), would it make any difference?

That's generally touted as one of the advantages of running balanced/bridged amps - the channel isolation. And there are people who make a distinction as to whether or not the audio circuit is fully balanced from source to sound. In theory, it shouldn't matter, but there are always little details that don't show up in the theoretical figures, that make themselves known in the actual circuit, possibly even to the point of being audible. But that's more of an issue as to how well the circuit has (or hasn't) been implemented and tuned.
 

 

[xnor]

Quote:

I know people are still looking for audio excellence, [...]

And always will. Though that probably has less to do with the actual performance of the amp than you'd think. 

 

[Leny]

Quote:

 
Maybe I'm looking into the whole thing far too deeply...
 

 
Or just read the link and pages I recommended in post #3.
 
: )

 

 

[mesasone]

Quote:

Maybe he's talking about something like that:

see wiki: amplifier

---
 
I think you will want to read up on the properties of Class A, B, and AB amplifiers.

 

[Tyll Hertsens]

Quote:

 
I think you will want to read up on the properties of Class A, B, and AB amplifiers.

What's your point?