[fishski13]

Quote:

Originally Posted by Koyaan I. Sqatsi /img/forum/go_quote.gif And with regard to the one thing the active ground does do, i.e. keeping supply current draw constant, that won't even work properly in the situation at hand, which is a four board amp with dual power supplies. It might provide some marginal benefit in an unbalanced amp with a single power supply, but that's not what's under consideration here. k

really? what is under consideration here? here's what you wrote in post #6:
Yes.

"And as I said, I see no particular advantage or benefit of the active ground channel.

k"

i think your confused, or rather just like to start circular arguments.

 

[Lifthanger]

the mosfets provide current gain. the gained current comes from the V+- rails.
due to the high input impedance of the jfets the current to ground should be in the microamperes, while the current sourced from the rails provide the rest.
edit: most of the time the rails are not connected to signal ground, too.

 

[Fitz]

Quote:

Originally Posted by Koyaan I. Sqatsi /img/forum/go_quote.gif Any "attitude" on my part has been in response to amb's increasing defensiveness in response to my having done nothing more than ask some simple questions. Case in point, where does the load current go if it doesn't go to ground? I asked it quite simply and as politely as anyone could ask. In response, I got non-answers, wrong answers and dismissals, which eventually culminated in amb's sticking his fingers in his ears and shouting "BLAH! BLAH! BLAH! I CAN'T HEAR YOU! BLAH! BLAH! BLAH!" If he can't substantiate his claim that the current "does NOT go to ground," then he should just say so. There's no reason to get so childishly defensive over it. k

Cool story, bro.

 

[Steve Eddy]

Quote:

Originally Posted by fishski13 /img/forum/go_quote.gif really? what is under consideration here? here's what you wrote in post #6: Yes. "And as I said, I see no particular advantage or benefit of the active ground channel. k" i think your confused, or rather just like to start circular arguments.

What's under consideration here is what would at minimum be a four board balanced B22 with dual, left/right power supplies, and the ability to power unbalanced headphones as well.

It's within that context that the question as to the benefit of an active ground channel was raised.

And in that context, I see no particular advantage to it.

First, the claim that the active ground results in load currents not going to ground is simply incorrect.

Second, the one thing that the active ground can do, i.e. keep power supply current constant, won't work in a dual power supply situation.

So what advantage is there that I'm missing here, and within the context of that which is under consideration?

k

 

[Lifthanger]

Quote:

Originally Posted by Koyaan I. Sqatsi /img/forum/go_quote.gif First, the claim that the active ground results in load currents not going to ground is simply incorrect.

nope

 

[Steve Eddy]

Quote:

Originally Posted by Lifthanger /img/forum/go_quote.gif the mosfets provide current gain. the gained current comes from the V+- rails.

Yes.

Quote:

due to the high input impedance of the jfets the current to ground should be in the microamperes, while the current sourced from the rails provide the rest.

What you're neglecting here is the fact that the current sourced from the rails goes to ground. See the illustration I provided above.

Quote:

edit: most of the time the rails are not connected to signal ground, too.

But all of the current flows through ground. And "signal ground" and "power supply ground" are one and the same.

k

 

[Steve Eddy]

Quote:

Originally Posted by Lifthanger /img/forum/go_quote.gif nope

No?

Ok, here's a blank slate for you.

Draw the path of the load current and explain how it doesn't go to ground.

k

 

[luvdunhill]

Steve:

needs MOAR transformers

 

[Steve Eddy]

Quote:

Originally Posted by luvdunhill /img/forum/go_quote.gif Steve: needs MOAR transformers

k

 

[Lifthanger]

I'm far from being an expert, but seeing that V- and V+ are DC, there won't be any current through the caps.
Say it would make it through the caps: the output of the sigma22 has very very low impedance. which means the current will leave/enter this circuit through the voltage rails. I don't know what happens to the current in the ps. but at that point it won't matter any more because of the regulation.
idk but this should be what happens. I'm also pretty sure, there's a difference between going into ground and passing through ground which is what happens in your schematic.

 

[fishski13]

Quote:

Originally Posted by Koyaan I. Sqatsi /img/forum/go_quote.gif So what advantage is there that I'm missing here, and within the context of that which is under consideration? k

i have no idea in this contex and nothing further to add. grounding topologies aside, 5-channels and 2 PS seems a bit OTT to drive a pair of HPs.

regarding passive 2-channel vs. 3-channel active ground. i'm specifically building a 2-channel first before a 3-channel to personally assess what sonic impact there is.

 

[TzeYang]

Quote:

Originally Posted by Koyaan I. Sqatsi /img/forum/go_quote.gif No? Ok, here's a blank slate for you. Draw the path of the load current and explain how it doesn't go to ground. k

urm, one sees a ground with a variable low impedance (following the frequency) and one sees a stiffer more constant low impedance?

 

[Steve Eddy]

Quote:

Originally Posted by Lifthanger /img/forum/go_quote.gif I'm far from being an expert, but seeing that V- and V+ are DC, there won't be any current through the caps.

With no current through the caps, there won't be any current at the output, and all you'll get is silence.

Quote:

Say it would make it through the caps: the output of the sigma22 has very very low impedance. which means the current will leave/enter this circuit through the voltage rails.

The output impedance is irrelevant. The current still goes to ground.

Quote:

I don't know what happens to the current in the ps. but at that point it won't matter any more because of the regulation.

The regulation doesn't control the current. It only keeps the rail voltage constant. The same current flowing through the rails flows through the regulators.

Quote:

I'm also pretty sure, there's a difference between going into ground and passing through ground which is what happens in your schematic.

The current can't simply go "into ground," it MUST pass through ground. The amount of current entering a node must equal the amount of current leaving the node.

See Kirchhoff's Current Law.

k

 

[Steve Eddy]

Quote:

Originally Posted by fishski13 /img/forum/go_quote.gif i have no idea in this contex and nothing further to add. grounding topologies aside, 5-channels and 2 PS seems a bit OTT to drive a pair of HPs.

Well, four of the channels is to do a bridged (i.e. "balanced") amp. The two power supplies are for powering the left and right channels separately. The odd man out is the fifth channel for essentially allowing for a three channel system for driving unbalanced headphones. But an active ground simply won't work in a two supply system that I can see.

Quote:

regarding passive 2-channel vs. 3-channel active ground. i'm specifically building a 2-channel first before a 3-channel to personally assess what sonic impact there is.

Coolness! Hope you're happy with whichever arrangement you end up with.

There's no arguing one's personal tastes and preferences. But when it comes to making claims like the load current not going to ground, that's a whole other matter that transcends personal tastes and preferences.

k

 

[Steve Eddy]

Quote:

Originally Posted by TzeYang /img/forum/go_quote.gif urm, one sees a ground with a variable low impedance (following the frequency) and one sees a stiffer more constant low impedance?

I'm not seeing that, unless by "a stiffer more constant low impedance" you're referring to the short length of wire in the passive ground configuration.

k