Basic electrical questions.

Jea48 -- That was an excellent paper; thanks!

However, it does not really address my question, which as you appear to realize was about having MULTIPLE dedicated lines powering different parts of the same audio system. The author describes his own setup, which has a single dedicated line running to a receptacle into which he plugs a diy power strip + switch. He does provide some good information as to connection order, such as connecting the power amp to the end of the strip closest to the incoming ac.

I understand your point about decoupling the power supplies of each component from the supplies of the other components, particularly to isolate digital components from analog components. But I'm uncertain how to reconcile that with the concerns I expressed, about ground loops and the possibility of having ac ground points that are offset from each other at least at high frequencies.

When I get a chance I'll try to take a look at the AA archives you mentioned.

Thanks again,
-- Al

I'm mostly with Al on this one. The paper that Jea48 refers to is interesting and has some valid information, but there are also some typos and mistakes . . . and I think that some of his conclusions are a little questionable. With ANY of the Audio Express, Audio Amateur, Glass Audio, etc. articles, I HIGHLY recommend also reading follow-ups in the "letters" section of the following issue(s), as there is frequently a bit of ongoing dialogue. I definately would NOT just take them as gospel, simply because they're published in an audio magazine.

But there are cases, esp. with multiple high-powered amplifiers, that I can see some benefit to having multiple circuits run. But if all of the issues that Al mentions aren't taken into account when the circuits are wired, then it will likely make things worse.

In professional sound, lighting, and video, multiple circuits and distros are usually required simply to provide anywhere near enough current . . . not to mention a combination of house and generator power. This is also usually the main portal a grizzly parallel dimension of horrific shocks, sparks, and humm that knows no ends . . . so for a domestic system of modest power requirements, a single dedicated circuit avoids a LOT of pitfalls.

I have read on many occasions various AGON members stating that in effect "separate dedicated circuits alone help decouple components from one another".

From an electical engineering perspective, separate dedicated circuits all connected to the same subpanel would all be in parallel with each other. This means that any spurious artifacts from a component fed back into its dedicated circuit would be superimposed on all other dedicated circuits. Other components power supplies would have to deal with these artifacts, e.g., filter them out. In this regard, this approach is not any better than a single dedicated circuit.

In my own system, I have installed an isolation transformer feeding the subpanel, with separate dedicated lines to each component. To address the issue above, I have put a second isolation transformer on the CD dedicated line to keep its digital hash from contaminating the other components.

This has proved very effective for me, as there is now no difference in the background noise level between the CD player while off, and the CD player playing while muted. If you are contemplating multiple dedicated lines off a single subpanel, I suggest you consider this approach.

12-10-08: Kirkus
I'm mostly with Al on this one. The paper that Jea48 refers to is interesting and has some valid information, but there are also some typos and mistakes . . . and I think that some of his conclusions are a little questionable.

Kirkus,
Maybe you are not familiar with Charles Hansen

Charles Hansen - Manufacturer, Ayre Acoustics, Inc.
============================

12-10-08: Zargon
From an electical engineering perspective, separate dedicated circuits all connected to the same subpanel would all be in parallel with each other. This means that any spurious artifacts from a component fed back into its dedicated circuit would be superimposed on all other dedicated circuits. Other components power supplies would have to deal with these artifacts, e.g., filter them out. In this regard, this approach is not any better than a single dedicated circuit.

Many would disagree with you.

In the instance of a sub panel where the sub panel might be installed in a close proximity of the audio equipment where the length of the dedicated branch circuits were very short, then I would agree with you.

Quote from Shunyata Research:

There is up to a hundred feet of wire in the walls, so the last 6 feet of power cord can’t possibly make any difference.

Answer: The PC is NOT the last 6 feet as stated in #1 and the local current and EM effects directly affect the sonic performance of the component. The power cord is not the last 6 feet, it is the first 6 feet from the perspective of the component. The further a noise source is from a component, the less of an impact it will have on the components power supply. The high-frequency noise sources that have the greatest impact on audio and video performance are the system components themselves -- which are usually all in close proximity of one another and all emit radiated fields of high-frequency noise. A well designed power cord can act as a noise-isolated extension of the primary winding of a component’s power supply and will help isolate the power supply from the fields of radiated RF and EM noise energy that is ever present in all electronics systems.

The further a noise source is from a component, the less of an impact it will have on the components power supply.

If you reread Hansen's article you will find he is basically saying the same thing
.

From an electical engineering perspective, separate dedicated circuits all connected to the same subpanel would all be in parallel with each other. This means that any spurious artifacts from a component fed back into its dedicated circuit would be superimposed on all other dedicated circuits. Other components power supplies would have to deal with these artifacts, e.g., filter them out. In this regard, this approach is not any better than a single dedicated circuit.

I would think that the inductance of the one or two hundred feet of round-trip wiring that is involved is likely to be significant at noise frequencies.

In my own system, I have installed an isolation transformer feeding the subpanel, with separate dedicated lines to each component. To address the issue above, I have put a second isolation transformer on the CD dedicated line to keep its digital hash from contaminating the other components.

Interesting approach! The combination of isolation and bandwidth limiting that the transformers provide would overcome the concerns I expressed about ground loop noise, while still achieving the decoupling benefits of multiple dedicated lines. Excellent! Do you tie the system to a (single) earth ground at some point, though, to keep it from floating relative to earth?

Regards,
-- Al