Notes on balanced power, isolation, etc

JohnR · « **on:** 10 Jul 2003, 09:02 pm »

Here's an interesting (long) thread on AA:

http://db.audioasylum.com/cgi/m.pl?forum=tweaks&n=25697

Look interesting... (haven't read them yet):

http://www.soundstage.com/articles/pete01.htm
http://www.soundstage.com/articles/pete03.htm

About isolation transformers:

http://www.blackmagic.com/ses/bruceg/EMC/isolatrans.html

Equi=tech make balanced power systems, their FAQs and articles might be of interest:

http://www.equitech.com/

JohnR · « **Reply #1 on:** 10 Jul 2003, 09:12 pm »

Jon Risch's DIY balanced power note:

http://www.geocities.com/jonrisch/catch2.htm

MaxCast · « **Reply #2 on:** 11 Jul 2003, 12:45 am »

How many different types of devises are there in this category?
We have isolation devises that do.......
We have filtration devises that do........
We have balanced power devises that do.......
We have surge suppressors that do......
any more?

DVV · « **Reply #3 on:** 11 Jul 2003, 09:14 pm »

Quote from: MaxCast

How many different types of devises are there in this category?
We have isolation devises that do.......
We have filtration devises that do........
We have balanced power devises that do.......
We have surge suppressors that do......
any more?

Just as we have:

* bicycles,
* motorbikes,
* automobiles,
* vans,
* busses,
* etc

to carry us and/or loads over roads. Nothing worng in that Max, it's a young filed still searching for many answers, still in its infancy (well, for most, anyway).

Ideally, the same device should filter, protect from surges, decouple (isolate) and be in balanced mode all at the same time. To the best of my knowledge, the DeZorel devices come close to this ideal, but are by no means perfect, best seen from the fact that over the last two years or so, they have introduced new models which are better and better. This is an admission that we still have much to learn in the firled.

The fact that others use different technologies is telling us that there are different approaches to the same problem, and in my experience, this means that no single approach is good enough yet to claim absolute advantage. Which will make the industry work harder. And in the end, we, the users, will benefit.

Cheers,
DVV

MaxCast · « **Reply #4 on:** 12 Jul 2003, 03:26 am »

That's what I about figured, DVV. I have heard nothing good about surge supression other than saving components from spikes and bolts. I get miffed on post talking about filtration and issolation. What is the difference? Are they inclusive? Sometimes? I guess I need to read some web sites. Better yet, how long is the waiting list for the Dezorel?

DVV · « **Reply #5 on:** 12 Jul 2003, 09:45 pm »

Quote from: MaxCast

That's what I about figured, DVV. I have heard nothing good about surge supression other than saving components from spikes and bolts. I get miffed on post talking about filtration and issolation. What is the difference? Are they inclusive? Sometimes? I guess I need to read some web sites. Better yet, how long is the waiting list for the Dezorel?

Filtration is removing of unwanted additions to what should be total silence, in an attempt to go back to as much silence as is possible. Isolation is separating the power consumer from the source, but does not necessarily mean filtration as well, although in many instances it does include one form of filtration or another. The general idea of isolation is to provide a low or lower source impedance, which makes the transfer function of the power towards the consuming equipment smoother.

Surge suppression on its own is usually using one surge arresting diode, retail price in Europe around $2, to pull down whatever comes your way above a certain preset limit. It usually does this by shunting it to the ground. In this lies its greatest weakness, because it assumes the ground is at a certain potential - which it never, really never is (at least, not once have I seen or measured it, nor have I ever heard anybody claim to have it as it should be, at 0.8 V and less than 100 ohms). In the process, if the surge is large and long enough, that diode will die and will need to be replaced, and you can bet your puppa it will cost a lot more than $2.

Since a spike, any spike, will also have a frequency it resides on, otherwise it will be DC and that's not permissible in an AC grid, a well designed line filter will also by default act as a surge suppressor. How much depends on the filter's efficiency and the frequency the spike is on, as you will have gathered.

OK, Max? If there's anything else you want to know and I can answer, just ask - much better that way, else I'll start rambling. Here or in private, as you please.

Cheers,
DVV

MaxCast · « **Reply #6 on:** 12 Jul 2003, 10:35 pm »

Thanks, DVV. I'll be sure to let you know of any more questions as I do some more reading.

JohnR · « **Reply #7 on:** 13 Jul 2003, 02:35 am »

Actually, a good isolation transformer does filtering as well -- it has two electrostatic shields between the primary and secondary to stop noise coupling between the windings. Isolation transformers also provide the benefit of being able to isolate the (possibly noisy) mains ground from your equipment, which you can do IF you have a technical ground installed. (Unlikely for most of us.) A "balanced" transformer, to work properly, should provide the same shielding as an isolation transformer, but provides the output as +/- 60V phases. The balancing, as far as I understand at present, prevents certain types of noise current from getting into the ground and thus showing up as noise from the low-level signal stages.

One reason why one would want more transformers when there are already transformers in your equipment is that transformers in audio equipment are not of the same spec as a good isolation transformer (two shields, etc). It is also not necessary to have a "balanced transformer" in your equipment to obtain the benefit.

A filter reduces noise by using inductors and capacitors. Poorly done, they can actually increase the amount of noise in the ground circuit. It's important to distinguish the technology from a particular unit. IIRC BPT includes this kind of filtering as well as the transformer.

Ulas · « **Reply #8 on:** 13 Jul 2003, 03:16 am »

There are at least two more major classes: Power correction (Exact Power) and power regeneration (PS Audio).

markC · « **Reply #9 on:** 13 Jul 2003, 02:36 pm »

A friend of mine just gave me an iso trans that they were throwing out at his work. It's an IBM device that was designed to be a clean source of power for test equipment in computer rooms. He figures it's from the early '80's. This thing looks like it's never been used. It's a 240v-120v rated 5A@240 and 9A@120v. That should make it just over 1KvA on the 120v side. I was thinking of using it for my amp which I believe has 2 500vA trans. in it. Will this be cutting it too close in terms of the 1KvA power available?

DVV · « **Reply #10 on:** 13 Jul 2003, 08:20 pm »

Quote from: markC

A friend of mine just gave me an iso trans that they were throwing out at his work. It's an IBM device that was designed to be a clean source of power for test equipment in computer rooms. He figures it's from the early '80's. This thing looks like it's never been used. It's a 240v-120v rated 5A@240 and 9A@120v. That should make it just over 1KvA on the 120v side. I was thinking of using it for my amp which I believe has 2 500vA trans. in it. Will this be cutting it too close in terms of the 1KvA power available?

Probably will, but in real life, I expect you may have some trouble only when switching on. At such times, the power transformers tend to suck up a short term initial surge which is 2...7 times greater than their actual rating (depends on transformer type, quality of make, etc).

In normal use, I would imagine a 2.5 kVA power transformer to be pushed hard would mean you are well into the ear splitting levels, unless you have a pure class A amp - in which case the 1 kVA rating is definitely below par.

Beware of hard pushed transformers - they tend to distort madly and overheat.

Cheers,
DVV

markC · « **Reply #11 on:** 14 Jul 2003, 04:57 am »

Thanks for the relpy DVV; I should have been a little more clear in my description. My power amp has 2x 500 VA transformers-it's a dual mono design. I guess a test would be in order to determine that the voltage drop is not too great on the iso trans while powering the amp?

JohnR · « **Reply #12 on:** 14 Jul 2003, 05:13 am »

I'd be inclined to try it and see how it sounds. The rating seems fine. I'm a bit puzzled by the 240V though. This is the primary voltage? Is it intended to be connected across both phases of the mains?

Dan Banquer · « **Reply #13 on:** 14 Jul 2003, 02:10 pm »

I see no one bothered to pick up on the hint I dropped in the earlier part of this thread. In all probability this is due to the fact that most of consumer audio is still done with a two prong plug and not using the three prong plug with it's added earth ground.
The additional earth ground has some real advantages. Common Mode filters can now be designed to "dump" high frequency components onto the separate earth ground line instead of on the neutral line where it has the chance to recirculate. Surge suppressors are designed to do the same thing. The above is why all computers, lab equipment etc.etc. are now wired this way. The electrical code has also been this way for a number of years here in the U.S. Can anyone remember the last time you could go to a place like Home Depot and get a two prong wall outlet? 10 years ago?, or is it closer to 15 years ago?
Isolation Transformers as well as power supply transformers can be constructed with "narrow band" technology as Plitron calls it. The problem here is that as far as I can tell the HF components are still dumped back onto the neutral line. The balanced transformer will only work in this application if the common mode noise is equal on both the legs. Common mode fliters such as Corcom, Delta, Schurter etc.etc. don't care if the HF components are equally balanced they will still work.
On a more personal note; this entire thread has shown me once again that audio is still stuck in the past and has not adapted to already well proven and readily available technology. In talking with another engineer about about a similar subject pertaining to audio his conclusion was that audio is the "bastard child" of technology. I have to agree.
The following is a quote from an article I wrote with Gene Pitts, the former editor of Audio Magazine.

"The most important part of getting rid of conducted interference is to establish that you have a good, solid earth ground. Earth ground is usually tied to a large water pipe in the basement of your house or apartment building. Having a local electrician check this should be the first item on your list."

MaxCast · « **Reply #14 on:** 14 Jul 2003, 04:47 pm »

"The most important part of getting rid of conducted interference is to establish that you have a good, solid earth ground. Earth ground is usually tied to a large water pipe in the basement of your house or apartment building. Having a local electrician check this should be the first item on your list."

Can one do a visual on this or use a MMeter?

My house is wired the old two wire way. I did add a 20A line which I see is grounded on a pipe.

Is this correct? All filtration/issolation devises will work only on three wire service.

Dan Banquer · « **Reply #15 on:** 14 Jul 2003, 06:52 pm »

Max;
1. Any equipment that comes with a three prong plug needs to be plugged in to a three prong outlet.
2. Get the electrician.

markC · « **Reply #16 on:** 14 Jul 2003, 08:00 pm »

Quote from: JohnR

I'd be inclined to try it and see how it sounds. The rating seems fine. I'm a bit puzzled by the 240V though. This is the primary voltage? Is it intended to be connected across both phases of the mains?

Yes, the 240v is the primary side of the trans. I guess in essance it is a step down trans. According to the wiring diagram on the unit, it is indeed intended to be hooked to both phases. It has a funky looking plug on the end of it, which I assume was intended to be plugged into a specific receptacle in a computer room.