## You are here

Log in or register to post comments
bifcake
Offline
Last seen: Never ago
Joined: Nov 27 2005 - 2:27am
Please edumacate me

In the September issue, Fred Kaplan reviews the Krell Evolution 505 CD player.

Can someone explain to me in layman's terms how it is that a CD player can operate in a "current domain" vs a "voltage domain"?

My understanding that as per Ohm's law, Current is the function of Voltage divided by resistance. So, what's within the chain of the components that provides sufficient resistance as to make the signal flowing from one component to another a function of the "current" rather than typical voltage?

If anything, I would expect the speakers' interaction with the amplifier to be in the "current domain" since the speakers provide resistance and require the amplifier to provide sufficient current to drive them. How does that translate vis-a-vis electronic components?

Furthermore, provided that there is something to the components being driven in the "current domain", as I understand it, the benefits of that would be that the interconnecting cables become irrelevant or close to it. Yet, Fred Kaplan states that when he hooked up Nordost CAST cables (in the current domain), he heard a pronounced difference vs. the Krell cables he used.

That statement leads me to believe that either:

a) The reviewer is deaf and he simply responds to more expensive shiny objects or

b) Current domain is a marketing ploy and there's nothing to it.

Either this circuit implementation makes a difference in terms of making interconnecting cables irrelevant or it doesn't.

Any thoughts are greatly appreciated.

KBK
Offline
Last seen: Never ago
Joined: Sep 30 2007 - 12:30pm
Re: Please edumacate me

'voltage domain' is like that of a tube device. PRIMARILY or largely VOLTAGE, and low current, strictly as a numbers/value calculation.

Now, when that ratio of Volts vs amps changes drastically..then the device is called a CURRENT type device. The signal has more mass, and is less affected by external vibration and interference issues. This is just one aspect of the differences between current type devices (Solid state) and voltage type devices (Tubes, for example)

Tube circuit (not unusual values): might be , oh, 500volts, and 100mA of current, as a ratio, it would be...500:0.1, or a ratio of "5000 to 1" 500Vx0.1A=50 watts

A analog audio SS type circuit, for lets say, driving headphones or small speakers might be 25Volts and 2 amp. that's a ratio of 25:2, or "25 to 2", or 12.5:1 as a ratio. 25Vx2A=50 watts. same value, achieved differently, via different devices.

These are simply a names for two general types of signal transmission that are used in the audio world.

Now that I've read your post a ..er...a bit more thoroughly, it leads me to believe that you still have issues with your psychological makeup, and I can't fix that. Only you can.

You seem to insist that Fred (or something connected to the situation) is full of it as it does not fit your interpretation of reality.

Too bad. Fred heard a difference. There is a difference between current drive and voltage drive, it's just too long to get into here. And there is no reason why one change in a 'current drive system' might make it sound different than another.

Basically, if one is transmitting analog signals via cables, then there will be some sort of audible difference when the cables are swapped out. Otherwise, in the case of loudspeakers, which could easily be construed as being 'current drive'...we would not hear a difference in speaker cables.

But we most emphatically - do.

bifcake
Offline
Last seen: Never ago
Joined: Nov 27 2005 - 2:27am
Re: Please edumacate me

Quote:
'voltage domain' is like that of a tube device. PRIMARILY or largely VOLTAGE, and low current, strictly as a numbers/value calculation.

Now, when that ratio of Volts vs amps changes drastically..then the device is called a CURRENT type device. The signal has more mass, and is less affected by external vibration and interference issues. This is just one aspect of the differences between current type devices (Solid state) and voltage type devices (Tubes, for example)

Tube circuit (not unusual values): might be , oh, 500volts, and 100mA of current, as a ratio, it would be...500:0.1, or a ratio of "5000 to 1" 500Vx0.1A=50 watts

A analog audio SS type circuit, for lets say, driving headphones or small speakers might be 25Volts and 2 amp. that's a ratio of 25:2, or "25 to 2", or 12.5:1 as a ratio. 25Vx2A=50 watts. same value, achieved differently, via different devices.

That makes sense, but we're not talking about driving speakers here, we're talking about connecting one electronic component such as a CD player to another, such as a preamp. In this scenario, I don't see anything in the chain that would present a load, thereby increasing the amperage within the circuit.

dbowker
Offline
Last seen: 2 years 10 months ago
Joined: May 8 2007 - 6:37am
Re: Please edumacate me

I believe there is a pretty thorough explanation on the Krell website, as well as a former Krell review in Stereophile (I may have been their big integrated amp).

Jan Vigne
Offline
Last seen: Never ago
Joined: Mar 18 2006 - 12:57pm
Re: Please edumacate me

Quote:
If anything, I would expect the speakers' interaction with the amplifier to be in the "current domain" since the speakers provide resistance and require the amplifier to provide sufficient current to drive them. How does that translate vis-a-vis electronic components?

The speaker load represents an impedance, not a resistance. If the load were purely resistive in nature, the amplifier would only have to produce voltage with minimal current. However, the speaker load is reactive rather than resistive and the reactive component of the load demands "x" amount of current in response to the impedance and electrical phase angle of the load at any specific frequency.

Quote:
In this scenario, I don't see anything in the chain that would present a load, thereby increasing the amperage within the circuit.

Alex, how can you mention Ohm's Law and not recognize the load component at the end? Without a load on the circuit, no current will flow. Everything works into the load present at the next link in the chain, Alex. Right? Output impedance/input impedance. Load.

It is present in each stage of amplification or signal processing just as it is from component to component through the chain of components. A designer's challenge is often how to maintain a proper output/input impedance without adding significantly to the complexity of the circuit and adding more noise and distortion to the signal path. The ability to do this sets the good designer apart from the excellent designer.

http://www.firstwatt.com/downloads/cs-amps-speakers.pdf

Place "first watt/nelson pass" in a seach engine and read.

bifcake
Offline
Last seen: Never ago
Joined: Nov 27 2005 - 2:27am
Re: Please edumacate me

Quote:

The speaker load represents an impedance, not a resistance. If the load were purely resistive in nature, the amplifier would only have to produce voltage with minimal current. However, the speaker load is reactive rather than resistive and the reactive component of the load demands "x" amount of current in response to the impedance and electrical phase angle of the load at any specific frequency.

Yes, so we're really talking about varying resistance over a wide frequency range, but that's speakers. I'm trying to understand how it relates to source components

Quote:
In this scenario, I don't see anything in the chain that would present a load, thereby increasing the amperage within the circuit.

Alex, how can you mention Ohm's Law and not recognize the load component at the end? Without a load on the circuit, no current will flow. Everything works into the load present at the next link in the chain, Alex. Right? Output impedance/input impedance. Load.

Actually, without a load in a circuit, you'll get a short. But, yes, there has to be a load, but that load can be minimal since there doesn't seem to be a requirement for high current among components. Therefore, it's a fairly simple (as far as I can tell) transfer of low current/low voltage signal from one electronic component to another.

Quote:

It is present in each stage of amplification or signal processing just as it is from component to component through the chain of components. A designer's challenge is often how to maintain a proper output/input impedance without adding significantly to the complexity of the circuit and adding more noise and distortion to the signal path. The ability to do this sets the good designer apart from the excellent designer.

http://www.firstwatt.com/downloads/cs-amps-speakers.pdf

Place "first watt/nelson pass" in a seach engine and read.

That's all well and good, but what does that have to do with "current domain"? I don't get that.

ethanwiner
Offline
Last seen: Never ago
Joined: Sep 1 2005 - 2:26pm
Re: Please edumacate me

Quote:
My understanding that as per Ohm's law, Current is the function of Voltage divided by resistance.

Your understanding is correct and any "alternative" explanations are bogus. There are two issues:

1) All electric flow has current and voltage. To use the common "water pipe" analogy, voltage is water pressure and current is gallons per minute. You can get the same amount of water from here to there using very different ratios - a narrow pipe with lots of pressure, or a wide pipe with only a little pressure. Likewise for moving electrons around - skinny wire with high voltage and low current, or a fat wire with lots of current and low voltage.

Another analogy is that a transformer is identical (and I do mean identical) to an automobile transmission. An engine having 100 horsepower can use that power to go slowly up a steep hill, or faster on a level road. Either way you have 100 horsepower. Likewise, 100 watts can be sent down a wire with 100 volts at 1 amp, or 1 volt at 100 amps.

2) There are devices that are considered current sources and others that are voltage sources. You still have both volts and amps! But with a current source the device outputs whatever voltage is needed to maintain the desired current, and a voltage source will supply as much current as needed to maintain the desired voltage.

No sensible audio device uses a current source output because by definition that means the output impedance is very high. Which in turns means the frequency response will roll off with even short runs of signal wire.

In audio, efficient power transfer matters only with power amps, and is irrelevant with line-level gear. The accepted convention is for all devices to be voltage sources having as low an output impedance as possible. But inputs are always high impedance. So little actual power is ever transferred.

--Ethan

bifcake
Offline
Last seen: Never ago
Joined: Nov 27 2005 - 2:27am
Re: Please edumacate me

Thanks Ethan. That makes sense. Seems like CAST is just another marketing ploy.

Jan Vigne
Offline
Last seen: Never ago
Joined: Mar 18 2006 - 12:57pm
Re: Please edumacate me

Quote:
Actually, without a load in a circuit, you'll get a short. But, yes, there has to be a load, but that load can be minimal since there doesn't seem to be a requirement for high current among components. Therefore, it's a fairly simple (as far as I can tell) transfer of low current/low voltage signal from one electronic component to another.

No, if there is no load, you have an open circuit. A shorted circuit is quite a different animal. If the load is minimal a voltage source will have to produce higher current to have a resulting signal flow and there is the likelyhood of damage to the signal in the process. It's not very simple at all once you get off the piece of paper.

Quote:
That's all well and good, but what does that have to do with "current domain"? I don't get that.

Did you read the article? Did you do some research on first watt?

Jan Vigne
Offline
Last seen: Never ago
Joined: Mar 18 2006 - 12:57pm
Re: Please edumacate me

Quote:
Another analogy is that a transformer is identical (and I do mean identical) to an automobile transmission.

Quote:
In audio, efficient power transfer matters only with power amps, and is irrelevant with line-level gear.

Those statements both seem rather broadbrush, Ethan. A car's transmission has the advantage of progressively stepping up the gear ratio to aid the transfer of power. Audio transformers in general do not have the ability to begin in "low gear" and move through the ratios until they get to overdrive.

Equally, saying line level components do not suffer from impedance mismatching might be fine when you are dealing with an integrated amplifier but not so when you get to a high output impedance pre amp feeding long cables to a relatively low input impedance power amplifier. This situation has been responsible for many buyers not hearing in their home what they expected from the store's audition.

My understanding is a designer of voltage sources must constantly move from a high impedance input to a low impedance output to feed the next stage of amplification or processing. This makes for more complex circuits than would be required by a current source device feeding a known input impedance. In theory this should make cables a less sensitive item in a system's overall operation. That doesn't discount the fact that better quality matterials and construction will still result in a better quality cable.

ethanwiner
Offline
Last seen: Never ago
Joined: Sep 1 2005 - 2:26pm
Re: Please edumacate me

Quote:
A car's transmission has the advantage of progressively stepping up the gear ratio to aid the transfer of power.

Well, there are transformers with switchable taps, but I didn't mean they're identical in that way. Just that a transformer is exactly analogous to a gear box.

Quote:
not so [fine] when you get to a high output impedance pre amp feeding long cables to a relatively low input impedance power amplifier.

But that was my point. A preamp with a high output impedance, or an input with a low input impedance, is an incompetent design. This excludes old-style gear that used 600 Ohm matching, back when telephone engineering standards reigned.

Quote:
My understanding is a designer of voltage sources must constantly move from a high impedance input to a low impedance output to feed the next stage of amplification or processing.

That happens naturally in a well designed circuit. A simple one-transistor circuit inherently has a high input impedance and a low output impedance. Op-amp type circuits can be inverting or non-inverting, and a typical input stage will be non-inverting which has an extremely high input impedance. All op-amp outputs have a very low output impedance.

--Ethan

CharlyD
Offline
Last seen: 5 years 9 months ago
Joined: Jul 20 2006 - 4:01pm
Re: Please edumacate me

Quote:
No sensible audio device uses a current source output because by definition that means the output impedance is very high. Which in turns means the frequency response will roll off with even short runs of signal wire.

Not so Ethan. With the "current mode" used by Krell, the source component is a current source driving an output current that is proportional to the signal (music). The receiving end receives that current, and, if it's an amplifier driving speakers, produces an output voltage that is proportional to the input current (a transimpedance amplifier). The impedance requirement for the input of that amplifier would be as low an impedance as possible. This would allow the source component to drive its maximum current witout hitting the voltage limits of its output stage.

As I described in the previous thread on this topic last June, properly designed "current mode" or "voltage mode" systems will exhibit minimal effects from interconnecting cables. Also, there is no inherent advantage of one mode over the other. The cynic in me would say that they're pushing this CAST architecture because it contrains their customers to sticking with a single vendor.

bifcake
Offline
Last seen: Never ago
Joined: Nov 27 2005 - 2:27am
Re: Please edumacate me

Quote:

Not so Ethan. With the "current mode" used by Krell, the source component is a current source driving an output current that is proportional to the signal (music). The receiving end receives that current, and, if it's an amplifier driving speakers, produces an output voltage that is proportional to the input current (a transimpedance amplifier). The impedance requirement for the input of that amplifier would be as low an impedance as possible. This would allow the source component to drive its maximum current witout hitting the voltage limits of its output stage.

Can you repeat that in English, please?

Thanks

Jan Vigne
Offline
Last seen: Never ago
Joined: Mar 18 2006 - 12:57pm
Re: Please edumacate me

Quote:
Well, there are transformers with switchable taps, but I didn't mean they're identical in that way.

So you meant identical in a way that wouldn't be identical?

Quote:
But that was my point. A preamp with a high output impedance, or an input with a low input impedance, is an incompetent design.

Oh, Ethan! Don't you mean they suffer from comb filtering?

CharlyD
Offline
Last seen: 5 years 9 months ago
Joined: Jul 20 2006 - 4:01pm
Re: Please edumacate me

Quote:
Can you repeat that in English, please?

Sorry about my speaking geek, but my brain feels like it's doing yoga when trying to map an electrical circuit into garden hose and transmissions. Somebody should put together something simple like and amplifier using these hydraulic and mechanical analogies. Is there still a Rube Goldberg competition?

KBK
Offline
Last seen: Never ago
Joined: Sep 30 2007 - 12:30pm
Re: Please edumacate me

Quote:

Oh, Ethan! Don't you mean they suffer from comb filtering?

I used to suffer from that until I went with the Bald look.

ethanwiner
Offline
Last seen: Never ago
Joined: Sep 1 2005 - 2:26pm
Re: Please edumacate me

Quote:
The cynic in me would say that they're pushing this CAST architecture because it contrains their customers to sticking with a single vendor.

Okay, fair enough.

But I would never do it that way, and apparently I'm in good company with 99.9 percent of all pro circuit designers.

--Ethan

• X