[Steve Eddy]

Quote:

Originally Posted by Aynjell /img/forum/go_quote.gif Let's say I have two op amps in a sequence, is the capacitor that serves as an output buffer required for the first op amp required, or can I just setup the buffer on the second op amp? I was thinking about it last night, you could make 2 cmoy boards and use the same dual op amp twice, once for each channel, or if you wanted to you could send the output of the left side of left, to the right side of right and then send that to the jack, and vice versa. is this a terrible idea or could it work? Also, let's say I did such a setup, for example, if I did a gain of 1, and then a gain of 2, that'd give me a total gain of 3, or 4? I wasn't sure if I was multiplying or not.

I'm sorry, but I'm not quite following just how you're looking to hook everything up by your description here. And I don't know what you mean by "the capacitor that serves as an output buffer."

Quote:

Also, the way i understand it is the amps with 3 channels pull the ground through the third (or if fully balanced, both grounds) through the third and/or fourth board. Does the ground signal pass through the eletrical part of the amplifier or the signal part of the amplifier?

Generally speaking, "signal ground" and "power supply ground" are one and the same. Though if you're using a single supply with a level shifter, the signal ground reference is taken from a point midway between the power supply rather than from the power supply ground.

se

 

[Aynjell]

Quote:

Originally Posted by Koyaan I. Sqatsi /img/forum/go_quote.gif I'm sorry, but I'm not quite following just how you're looking to hook everything up by your description here. And I don't know what you mean by "the capacitor that serves as an output buffer." Generally speaking, "signal ground" and "power supply ground" are one and the same. Though if you're using a single supply with a level shifter, the signal ground reference is taken from a point midway between the power supply rather than from the power supply ground. se

I misread the circuit, the capacitor is the input buffer and I do believe it's required. As for the other question, I've decided not to follow that path.

Edit, this should be a good replacement if I understand tangent's site, for the standard virtual ground... and the way I understand it is both channels should have a better split and should be getting more power than what the default design allows... is this correct? Edit, here is the amp circuit I'm planning to build, the IC is the same IC, just going through the other side with a higher gain.

In essence what I'm thinking of doing is using 1 side of 1 op amp to pre-amp, and then send to the other side for further amplification. What I'm hoping this might accomplish is getting a similar or decent gain, while getting less noise (1st pass, gain 2, 2nd pass gain 3). I'd need two of each to complete it, and I'm thinking I might try the treads psu to feed it. In reality, this simply involves building 2 cmoy's, but matching the parts acrossed both boards symetrically, if you want to match them. The only real deviation is, you're bridging output from the one side to the input of the other, and having an entire cmoy handle either single channel.

 

[NelsonVandal]

Quote:

Originally Posted by Koyaan I. Sqatsi /img/forum/go_quote.gif But it doesn't entirely work as advertised. It's advertised as bypassing ground. It doesn't. That's a myth.

I've never seen it advertised as "bypassing ground". I don't even know what that means.

Quote:

A regulated low impedance PSU can be as simple as a pair of three terminal regulators.

Yes, but does L78xx/79xx or LM317/337 sound as good as eg AD825, LM6171 or LME49713? That's the question.

Quote:

And the ground channel doesn't provide any regulation of the supply rails. So unless you're powering your circuit from batteries, you'll get all the ripple voltage on the supply rails that you'd get with any unregulated AC power supply.

You're right, but like you said, a regulated PSU is as simple as an LM317 and a couple of resistors and capacitors.


Quote:

A perfect blend of what?

There are no perfect sounding opamps, are there? Many of them are good, but I haven't heard any without coloration. Take eg LME49720. It has an annoying dryness and "hollow" mid. With LME49710 as ground amp, it's twice the annoyance. If you use LM6171 with it's more blooming mid as ground amp, the blend gives a more neutral and enjoyable presentation.

Quote:

Ideally the ground channel behaves as a low impedance path to ground. What "tweaking" is there to do on a low impedance path to ground?

What tweaking is there not to do? The influence is major. Are you saying that the path to ground sounds the same no matter how it's done, as long as it's low impedance?


Quote:

That's incorrect. All of the current emanates from and returns to the power supply caps. You'd need just as large capacitors with a ground channel as without.

I was refering to passive vs. active ground. I regard regulators, active rail splitters and ground channel as being active and capacitors, batteries etc being passive - like CMOY or RA1.

Quote:

As I said in another thread on this issue, the return currents don't just spill out onto the floor or evaporate into the air.

I never said that. This is just bitching. This is the whole point - how the return currents are handled.

 

[Steve Eddy]

Quote:

Originally Posted by NelsonVandal /img/forum/go_quote.gif I've never seen it advertised as "bypassing ground". I don't even know what that means.

From AMB's website:

The ground channel amplifier sources or sinks the return current from the transducers, which would otherwise have been dumped into signal ground or power supply ground.

From AMB in another thread:

In an active ground amp, the load current doesn't go to ground.

From Morsel, who claims credit for the "active ground" idea:

Cmoy, I came up with the idea for using a headphone amplifier ground channel to shift responsibility for the high current reactive load of the headphones from signal ground to the supply rails, thus removing the primary source of signal ground contamination in 3 wire headphone systems.

All of these claims claim that the load currents magically go someplace other than ground, thus effectively bypassing ground.

Quote:

Yes, but does L78xx/79xx or LM317/337 sound as good as eg AD825, LM6171 or LME49713? That's the question.

Dunno. I haven't seen it demonstrated that either produces any actual audible differences over the other.

Quote:

There are no perfect sounding opamps, are there?

Guess that depends on how you're wanting to define "perfect."

There is nothing that's perfect from a purely technical point of view. But if an opamp does not produce any actual audible difference, it may well be considered "perfect" for all intents and purposes.

Quote:

Many of them are good, but I haven't heard any without coloration. Take eg LME49720. It has an annoying dryness and "hollow" mid. With LME49710 as ground amp, it's twice the annoyance. If you use LM6171 with it's more blooming mid as ground amp, the blend gives a more neutral and enjoyable presentation.

Well, this is all purely subjective and transcends any issue of actual audible differences so there's not much point arguing over it.

Quote:

What tweaking is there not to do? The influence is major. Are you saying that the path to ground sounds the same no matter how it's done, as long as it's low impedance?

No. But it's a rather simple matter to just do it right in the first place.

Quote:

I was refering to passive vs. active ground. I regard regulators, active rail splitters and ground channel as being active and capacitors, batteries etc being passive - like CMOY or RA1.

I know that you were referring to passive versus active ground.

And I'm saying that in either instance, all of the current emanates from and returns to the power supply caps. So you'll need just as much power supply capacity with an active ground as you would with a passive ground.

Quote:

I never said that. This is just bitching. This is the whole point - how the return currents are handled.

No, you didn't say that. However that's how "active ground" has been promoted and how most I've come across understand it to work. I'm just trying to set the record straight on that matter and reiterating it.

se