Skip to main content ↑ ↓

Notice

Please note that most of the software linked on this forum is likely to be safe to use. If you are unsure, feel free to ask in the relevant topics, or send a private message to an administrator or moderator. To help curb the problems of false positives, or in the event that you do find actual malware, you can contribute through the article linked here.
Topic: Meier Corda Amps (Read 18434 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

Meier Corda Amps

Does anyone have a clue what Dr. Meier is talking about with his "active balanced ground driving" concept?  What sort of audible effect might it create, if any?

Meier Corda Amps

Reply #1 –
Quote
Does anyone have a clue what Dr. Meier is talking about with his "active balanced ground driving" concept?
No idea...  Do you have a link so we don't have to search?

It could be a virtual ground design or a bridge-output design.*

Quote
What sort of audible effect might it create, if any?
Hopefully none!!!    It's been possible to build audibly transparent amplifiers since the advent of solid state electronics** and with modern integrated circuits and low-cost electronics it's super-easy and cheap.





* These approaches can be helpful if you have a single-ended (positive only) power supply and/or a low voltage supply, such as with a battery operated device.


** It was possible with vacuum tubes & transformers, but not cheap or easy.

Meier Corda Amps

Reply #2 –
That would be my guess as well. Rather than go to the expense of implementing a bipolar supply, create a virtual ground at half the supply voltage. You can only get away with this if you are driving headphones, and the virtual ground had better be very good or else you can get crosstalk between the channels.

It works OK, but I would not have chosen that route.

Meier Corda Amps

Reply #3 –
Quote from: theriverlethe on
Does anyone have a clue what Dr. Meier is talking about with his "active balanced ground driving" concept?  What sort of audible effect might it create, if any?


If it is what I think, this is the methodology used to produce output-capacitor free headphone amplifiers by many portable digital players including most of the later iPods, and the Sansa Clip/Fuse etc.

It gets the headphone amp source impedance way down, all the way down to DC, with no physically large parts.

Since the lowest cost Meier headphone amp is €215 I think I'll just enjoy this feature on one of my Fuses ($< $40).

Meier Corda Amps

Reply #4 –
Quote from: DVDdoug on
Quote
Does anyone have a clue what Dr. Meier is talking about with his "active balanced ground driving" concept?
No idea...  Do you have a link so we don't have to search?

It could be a virtual ground design or a bridge-output design.*

Quote
What sort of audible effect might it create, if any?
Hopefully none!!!    It's been possible to build audibly transparent amplifiers since the advent of solid state electronics** and with modern integrated circuits and low-cost electronics it's super-easy and cheap.





* These approaches can be helpful if you have a single-ended (positive only) power supply and/or a low voltage supply, such as with a battery operated device.


** It was possible with vacuum tubes & transformers, but not cheap or easy.


Here is a link to his website, but you have to click on "ground driving:"

http://www.meier-audio.homepage.t-online.de

I found a comment on it by nwavguy, but he appeared to be stumped by it as well.

Meier Corda Amps

Reply #5 –
This appears to be the relevant page:

http://www.meier-audio.homepage.t-online.de/grounds.htm

I'm not the technical whiz, but it smells like BS to me: Do we really need to concern ourselves with electric fields radiated by headphone cables, given the sorts of voltages and currents they carry? And for all of the bother and expense in solving a non-problem, I wonder if crosstalk performance suffers.

Meier Corda Amps

Reply #6 –
Quote from: 4season on
This appears to be the relevant page:

http://www.meier-audio.homepage.t-online.de/grounds.htm

I'm not the technical whiz, but it smells like BS to me: Do we really need to concern ourselves with electric fields radiated by headphone cables, given the sorts of voltages and currents they carry? And for all of the bother and expense in solving a non-problem, I wonder if crosstalk performance suffers.



I did the math and it is not BS. It is exactly what I said: This is the methodology used to produce output-capacitor free headphone amplifiers by many portable digital players including most of the later iPods, and the Sansa Clip/Fuse etc.

The L & R ground wires are part of the same active ground and algebraically, it matches the math shown below:



which is titled: "Active Balanced Ground Driving".

I have to admit that left to my preferences I would make G = -(L+R)/2  instead of G = -(L+R)/4  but as long as the math is right, and G is properly added to L & R,  G will cancel and it will be all the same.

Meier Corda Amps

Reply #7 –
Quote from: Arnold B. Krueger on
Quote from: 4season on
This appears to be the relevant page:

http://www.meier-audio.homepage.t-online.de/grounds.htm

I'm not the technical whiz, but it smells like BS to me: Do we really need to concern ourselves with electric fields radiated by headphone cables, given the sorts of voltages and currents they carry? And for all of the bother and expense in solving a non-problem, I wonder if crosstalk performance suffers.



I did the math and it is not BS. It is exactly what I said: This is the methodology used to produce output-capacitor free headphone amplifiers by many portable digital players including most of the later iPods, and the Sansa Clip/Fuse etc.

The L & R ground wires are part of the same active ground and algebraically, it matches the math shown below:



which is titled: "Active Balanced Ground Driving".

I have to admit that left to my preferences I would make G = -(L+R)/2  instead of G = -(L+R)/4  but as long as the math is right, and G is properly added to L & R,  G will cancel and it will be all the same.


So, it's a really complicated way of saying, "low output impedance?"

Meier Corda Amps

Reply #8 –
Quote
So, it's a really complicated way of saying, "low output impedance?"
It's a way of making a DC coupled output (no capacitor).

A capacitor does have capacitive reactance, which is an impedance component.  It's not the only  impedance component and it's possible to have high impedance without a capacitor.  If the capacitance value is too low, you can loose bass.  At DC (zero Hz) a capacitor has infinite impedance and no current flows. 

With a large enough capacitor (high uF value) you can get "flat" frequency response below 20Hz.    So... 

A capacitor-less design is not always  better...  It's one engineering solution...  It's OK to be curious, but as a consumer/user you don't need to worry about engineering choices & trade-offs or what's inside...  Just look at the performance, features, appearance, cost, etc.

Meier Corda Amps

Reply #9 –
Quote from: theriverlethe on
Quote from: Arnold B. Krueger on
Quote from: 4season on
This appears to be the relevant page:

http://www.meier-audio.homepage.t-online.de/grounds.htm

I'm not the technical whiz, but it smells like BS to me: Do we really need to concern ourselves with electric fields radiated by headphone cables, given the sorts of voltages and currents they carry? And for all of the bother and expense in solving a non-problem, I wonder if crosstalk performance suffers.



I did the math and it is not BS. It is exactly what I said: This is the methodology used to produce output-capacitor free headphone amplifiers by many portable digital players including most of the later iPods, and the Sansa Clip/Fuse etc.

The L & R ground wires are part of the same active ground and algebraically, it matches the math shown below:



which is titled: "Active Balanced Ground Driving".

I have to admit that left to my preferences I would make G = -(L+R)/2  instead of G = -(L+R)/4  but as long as the math is right, and G is properly added to L & R,  G will cancel and it will be all the same.


So, it's a really complicated way of saying, "low output impedance?"


Hopefully. The typical means for achieving low output impedance are facilitated, but not guaranteed.

Meier Corda Amps

Reply #10 –
To see how such a design can perform badly see NwAvGuy's mini3 article. It also links to a discussion about the problems with active grounds iirc.

The "active balanced ground" subtracts a bit of a mix of both channels from each channel, as well as from ground. If this doesn't cancel perfectly then crosstalk figures will suffer.
"I hear it when I see it."

Meier Corda Amps

Reply #11 –
Quote from: xnor on
To see how such a design can perform badly see NwAvGuy's mini3 article. It also links to a discussion about the problems with active grounds iirc.


Oh yeah, I built Mini 3 and he's right it's pretty bad: Compared to O2, it sounds almost like mono what with all the crosstalk!

Meier Corda Amps

Reply #12 –
Quote from: Arnold B. Krueger on
I did the math and it is not BS. It is exactly what I said: This is the methodology used to produce output-capacitor free headphone amplifiers by many portable digital players including most of the later iPods, and the Sansa Clip/Fuse etc.

It seems you are confusing things here. For that you just need a normal virtual ground.

You can mess that up by making it active as in the mini3.

Since the ground is now driven like an audio channel, you can make it follow whatever signal you want (virtual ground i.e. the output of the rail splitter which is 1/2 the supply voltage like in the mini3; -(L+R)/4 on that Meier page, which he calls "active balanced").

---

The headphone doesn't care what voltages are applied, it just cares about the difference in voltage.
"I hear it when I see it."

Meier Corda Amps

Reply #13 –
I assure you, the Corda doesn't sound anything like mono, even with the crossfeed engaged.  I saw nwavguy' article and someone mentioned the Corda Concerto in the comments.  Nwavguy didn't really have a conclusive answer, but said it was something "different."

Meier Corda Amps

Reply #14 –
Quote from: 4season on
Quote from: xnor on
To see how such a design can perform badly see NwAvGuy's mini3 article. It also links to a discussion about the problems with active grounds iirc.


Oh yeah, I built Mini 3 and he's right it's pretty bad: Compared to O2, it sounds almost like mono what with all the crosstalk!


I recall something earlier in the thread about component matching.

Meier Corda Amps

Reply #15 –
Quote from: theriverlethe on
I saw nwavguy' article and someone mentioned the Corda Concerto in the comments.  Nwavguy didn't really have a conclusive answer, but said it was something "different."

I disagree.
It is fundamentally the same. It's simply an active ground. The "difference" is that it doesn't just float at half the supply voltage, but floats at that plus an attenuated inverted mono mix. The same needs to be added to both channels, so that it cancels at the headphone drivers.

Sadly, on that linked page there is also a confusion between balanced and bridged operation. Balanced operation does not need an inverted part of the signal. What matter is impedances.


After a quick search I found this comment by NwAvGuy:
Quote
The Corda amps are less clearly specified. And, to be honest, some of the Corda amps use designs that simply don't make any sense from an engineering perspective. My opinion is Corda tries to designs amps that are different just for the sake of being different.



Nevertheless, Meier's products probably sound just fine.
"I hear it when I see it."

Meier Corda Amps

Reply #16 –
Quote from: xnor on
Quote from: Arnold B. Krueger on
I did the math and it is not BS. It is exactly what I said: This is the methodology used to produce output-capacitor free headphone amplifiers by many portable digital players including most of the later iPods, and the Sansa Clip/Fuse etc.

It seems you are confusing things here. For that you just need a normal virtual ground.


A normal virtual ground works, but it isn't the only thing that works.

Quote
You can mess that up by making it active as in the mini3.


A designer can avoid messing things up if he gets the algebra right.  Just because someone got it wrong doesn't mean that everybody has to get it wrong.

But, it turns out that the Mini3 schematic shows a plain-vanilla active ground design, so what are you complaining about?  I guess circuit analysis is not one of your strengths!

BTW, this schematic for the Active 3 shows a straighforward active ground: Mini3 Active ground schematic

The NWAVGuy  review :
NWAVguy Mini3 Review

highlights three major failings - high THD at high frequencies, low output into 150 ohm loads, and poor crosstalk. The crosstalk issue could come from other areas than the circuit design, including the active ground. 

Cross talk can be influenced by circuit elements that aren't on the schematic, and all I have is a schematic.  So here's the point: The inclusion of the Mini3 was a deflection or an error on your part because it uses the straightforward active ground circuit that you seem to favor, based on the schematic.

I recognize that issues like channel separation can be influenced by design features such as the circuit board layout that don't show up on the circuit schematic and whose analysis can be more complex than can be easily discussed here.

Quote
Since the ground is now driven like an audio channel, you can make it follow whatever signal you want (virtual ground i.e. the output of the rail splitter which is 1/2 the supply voltage like in the mini3; -(L+R)/4 on that Meier page, which he calls "active balanced").

---

The headphone doesn't care what voltages are applied, it just cares about the difference in voltage.


Exactly. Why is it so confusing for you when I rely on the same math?

Gee, you want to accuse me of being confused, and then make the same points? And bring in a device that uses the design features you seem to want to favor?

There's someone here who is very badly confused, and it ain't me!

Meier Corda Amps

Reply #17 –
Ok Arny, if you want to play:

Quote from: Arnold B. Krueger on
A normal virtual ground works, but it isn't the only thing that works.

You talked about capacitor free outputs design like in portable devices. An active ground is not required for that, especially not one that floats with the input signal*.

Here's what you said specifically:
Quote
This is the methodology used to produce output-capacitor free headphone amplifiers by many portable digital players including most of the later iPods, and the Sansa Clip/Fuse etc.

Which is simply factually wrong.
See *, which is not commonly used at all. In typical portable devices it would just increase cost and decrease performance.


Quote from: Arnold B. Krueger on
A designer can avoid messing things up if he gets the algebra right.  Just because someone got it wrong doesn't mean that everybody has to get it wrong.

That wasn't my point.
My point was that an active ground is an addition e.g. to a basic virtual ground design. You seemed to confuse the different designs, hence my attempt at clarification.


Quote from: Arnold B. Krueger on
But, it turns out that the Mini3 schematic shows a plain-vanilla active ground design, so what are you complaining about?  I guess circuit analysis is not one of your strengths!

You are answering your own questions:
Quote
three major failings - high THD at high frequencies, low output into 150 ohm loads, and poor crosstalk
.
and:
Quote from: Arnold B. Krueger on
The inclusion of the Mini3 was a deflection or an error on your part because it uses the straightforward active ground circuit that you seem to favor, based on the schematic.

Eh... what?
Where do I favor active grounds? I don't.

Please read what I wrote. I wrote you can mess it up. Also, I talked about the mini3, not the theoretical performance of some schematic. I could respond to your strawman more in depth, but I don't see the point.



Quote from: Arnold B. Krueger on
Exactly. Why is it so confusing for you when I rely on the same math?

Gee, you want to accuse me of being confused, and then make the same points? And bring in a device that uses the design features you seem to want to favor?

There's someone here who is very badly confused, and it ain't me!

You completely missed my point. I differentiate between plain (virtual) ground, active ground and the uncommon "balanced active" ground.
Also see the first part of my response that deals with the confusion and straw man of me favoring something.


edit:
Maybe you should also read virtual-grounds-3-channel-amps (or anyone else interested in the topic).
"I hear it when I see it."

Meier Corda Amps

Reply #18 –
Quote from: xnor on
Quote from: Arnold B. Krueger on
A normal virtual ground works, but it isn't the only thing that works.


You talked about capacitor free outputs design like in portable devices. An active ground is not required for that, especially not one that floats with the input signal*.


Case in point: The AS AS3525 microchip which centerpieces the Sansa Clip and Fuse:

AS 3525 (Sansa Clip & Fuze) detailed tech specs

P 131:

Figure 50 (right)  shows capacitor-free "phantom ground" or if you will virtual ground.

The way that the 02 and similar products avoid output coupling capacitors is via a split supply which is less practical in portable devices.

Quote
Here's what you said specifically:
Quote
This is the methodology used to produce output-capacitor free headphone amplifiers by many portable digital players including most of the later iPods, and the Sansa Clip/Fuse etc.


Which is simply factually wrong.

See *, which is not commonly used at all. In typical portable devices it would just increase cost and decrease performance.


My usual play: Refutation via real-world example, above.

However I am incorrect in referencing iPads, etc as they use Maxim's DirectDrive and sequel technologies that create a local negative supply that allows its output to be referenced to chassis ground and avoid output coupling capacitors.

Maxim MAX 13331 data sheet

Meier Corda Amps

Reply #19 –
Is muting the left channel and playing music an adequate test of crosstalk?  This thing does not play nicely with an unbalanced line input.  I think I need a stereo to dual mono cable into something like a Scarlet 2i2.

Meier Corda Amps

Reply #20 –
Quote from: theriverlethe on
Quote from: 4season on
Quote from: xnor on
To see how such a design can perform badly see NwAvGuy's mini3 article. It also links to a discussion about the problems with active grounds iirc.


Oh yeah, I built Mini 3 and he's right it's pretty bad: Compared to O2, it sounds almost like mono what with all the crosstalk!


I recall something earlier in the thread about component matching.


Also, I'll bet that the evaluations were no better than casual sighted. :-(

When you start working with matrixing schemes like the ones implied by some of these funny grounding circuits, parts tolerance becomes very important.

With common 1% resistors, crosstalk may be no better than 40 dB. 

NWAVguy seems to be making a big stink about how bad 40 dB channel separation is, but I don't know of any psychoacoustic tests that back up his apparent anxiety.

IME as little as 20 dB channel separation can still give a fine sound stage. However, its been a long time since I did any relevant listening tests.

Meier Corda Amps

Reply #21 –
Quote from: Arnold B. Krueger on
Case in point: The AS AS3525 microchip which centerpieces the Sansa Clip and Fuse:

Let me guess, you didn't even read what I wrote?
You don't need an active ground that floats with the input signal, nor is it common.


Quote from: Arnold B. Krueger on
Quote
Here's what you said specifically:
Quote
This is the methodology used to produce output-capacitor free headphone amplifiers by many portable digital players including most of the later iPods, and the Sansa Clip/Fuse etc.


Which is simply factually wrong.

See *, which is not commonly used at all. In typical portable devices it would just increase cost and decrease performance.


My usual play: Refutation via real-world example, above.

I am not sure who you are refuting here. Yourself?


Quote from: Arnold B. Krueger on
However I am incorrect in referencing iPads, etc as they use Maxim's DirectDrive and sequel technologies that create a local negative supply that allows its output to be referenced to chassis ground and avoid output coupling capacitors.

Sure, charge pumps are another (next to virtual grounds) possibility to prevent the need for output caps.
"I hear it when I see it."

Meier Corda Amps

Reply #22 –
Quote from: xnor on
Quote from: Arnold B. Krueger on
Case in point: The AS AS3525 microchip which centerpieces the Sansa Clip and Fuse:

Let me guess, you didn't even read what I wrote?


No, I could have written everything you wrote and saved you the trouble of adding your own errors and misrepresentations.

Quote
You don't need an active ground that floats with the input signal, nor is it common.


You seem to be treating this like a political question, not a technical question.

Active grounds are problem solvers. Of course if you don't have the problem (example: the O2 headphone amp) then you don't need any solution. That is known as belaboring the obvious.

OTOH, building an efficient power amplifier while avoiding large and expensive output coupling capacitors can be facilitated by using an active ground configuration.  The other reasonably practical alternative usually involves building a switchmode DC-DC converter to develop a second power source out of the same battery.  However the switchmode power supply is potentially a source of audible electrical noise.  I don't know of any other solutions, do you?

Therefore, the statement that "You don't need an active ground that floats with the input signal, nor is it common." ignores many relevant facts.

Meier Corda Amps

Reply #23 –
It doesn't look to me as if Meier's design was due to the unavailability of a negative supply voltage in his amp. The capless drivers mentioned above do solve a problem, i.e. how to drive a headphone by a small portable device that works off a single voltage, without needing large capacitors for which there's no room in the small enclosure.

But that's almost certainly not Meier's problem.

So I don't understand the discussion you are having here.

I assume Meier's headphone amps work with a symmetrical supply internally. In some of the descriptions of his amplifiers, he states that explicitly. The external DC connection is a simple wall-wart plug, but he uses a voltage doubler/inverter internally, as per his own technical description. This means that there is no technical reason for using virtual grounds, or similar techniques to those found in the capless driver chips that were mentioned.

Meier's technical argumentation is bullshit, of course, even if the math is technically correct.

Firstly, he appears to confuse balanced lines with balanced drive. As has been pointed out here already, a line is balanced when its impedances are balanced, irrespective of how the drive voltages pan out. The headphone is not grounded, and only sees the voltage difference between its two connections. It doesn't matter whether one of the wires is arbitrarily called "ground" or not. Remember: Ground is just a label that one sticks on a certain node in a circuit. It is a convention, no more. It often is convenient to do so, but it can also obfuscate the situation when people attach some magical property to "ground" which it doesn't have.

Because of the above, the left and right diagram in Meier's explanation ("single ended driving" and "active balanced ground driving"), are in fact the same. What's called G in one case is called 0 in the other. The difference is only in the names.

Another piece of bullshit is the assertion that the electric fields in and around the cable matter a lot to the sound. He's merely using complicated language to say that the cable has a capacitance. Of course the cable has a capacitance that depends on the dielectric, we all know that. The effect on the sound can be calculated using the capacitance value, and it turns out that it is much too small to have any significant effect in the common cases. As usual with such audiophile assertions, the language used inflates and obscures the problem, and no attempt is made to provide a practical estimate of the effect.

In short: Meier postulates a non-problem, to which he provides a non-solution, promoted by a non-explanation.

Meier Corda Amps

Reply #24 –
Quote from: pelmazo on
It doesn't look to me as if Meier's design was due to the unavailability of a negative supply voltage in his amp. The capless drivers mentioned above do solve a problem, i.e. how to drive a headphone by a small portable device that works off a single voltage, without needing large capacitors for which there's no room in the small enclosure.

But that's almost certainly not Meier's problem.

So I don't understand the discussion you are having here.

I assume Meier's headphone amps work with a symmetrical supply internally. In some of the descriptions of his amplifiers, he states that explicitly. The external DC connection is a simple wall-wart plug, but he uses a voltage doubler/inverter internally, as per his own technical description. This means that there is no technical reason for using virtual grounds, or similar techniques to those found in the capless driver chips that were mentioned.


Agreed. The ready availability of virtually unlimited power in any conceivable resaonble format needed favors a more conventional approach.

Quote
Meier's technical argumentation is bullshit, of course, even if the math is technically correct.


Agreed. The technical facts are correct as far as they go, but their use as a justification for the unecessary complexity seems unwarranted.

Quote
Another piece of bullshit is the assertion that the electric fields in and around the cable matter a lot to the sound. He's merely using complicated language to say that the cable has a capacitance. Of course the cable has a capacitance that depends on the dielectric, we all know that. The effect on the sound can be calculated using the capacitance value, and it turns out that it is much too small to have any significant effect in the common cases. As usual with such audiophile assertions, the language used inflates and obscures the problem, and no attempt is made to provide a practical estimate of the effect.

In short: Meier postulates a non-problem, to which he provides a non-solution, promoted by a non-explanation.


Agreed. making a standard sized, AC powered headphone amp should be one of the easiest design chores imaginable in audio, and result in a very simple, low cost and effective device.

The presence of a market for wildly overpriced headphone amps had led to a lot of weird stuff.

Meier must know about and be motivated by the value of creating the perception of a significant difference in a crowded marketplace.