[pureAC]

Hi all,
Just wanted to start a discussion

<insert liability statement:"I announce that I know Diddly Squat about electronic theory, etc."  >

So I was listening to some music one day..  

Started to think about power conditioning..   

Then about capacitors across the line (active and neutral).

Point 1: Fact(Hopefully?). All electrical equipment is run parallel to the power buses (active and neutral). I think about it as a two long wires with all your equipment conneted to it.

eg:
+ve________________________________________________
                     |                  |                    |

            Fridge, PC, etc       Cap             CDplayer

                     |                  |                    |
-ve________________________________________________


If we attempt to use a capacitor across the line, to suppress high frequency noise, all noise on that bus will be shunted to neutral?

To put this in a real world perspective. Here you are saying, "Oh, auricap, cool, I'll install that on my CD player". If we assume the above statement is correct, then all your TV, Fridge, computer, etc. noise wil now be shunted through the one capacitor right next to your CD player?

This is only assuming you don't have other capacitors/ shunts draining noise. These can be in PC PSUs, other power conditionig products, etc.  Now you have further power degradation with the inclusion of these other capacitors/ shunts.

Now we are left with one noisy power source?!?!?

Now I politely ask a few questions for the gifted:
- Does the above hold any water?
- How can you stop other capacitors/ shunts polluting your audio/video power buses?
- If a single capacitor can change the sound of a system so profoundly, then what impact will all the "old/low quality" capacitors have connected throughout your house?
- Does the noise spread evenly across the multiple shunt capacitors throughout your home?

Hope we have a informative discussion.

pureac

[Occam]

If we attempt to use a capacitor across the line, to suppress high frequency noise, all noise on that bus will be shunted to neutral?

A capacitor may be viewed as a frequency dependant resistor, but rather than resistance, it has reactance that varies with frequency. [impedance is the square root of the resistance squared plus the reactance squared]
http://en.wikipedia.org/wiki/Reactance
http://tpub.com/neets/book2/4c.htm
and a calculator
http://www.kusashi.com/reactance-c.php
Now if we look at the reactance of a .47uf cap typically used as an 'X' cap across 'hot' and 'neutral' in the North American 60Hz power environment.

@60Hz Xc = 5,644Ohms
@600Hz    = 564 Ohms
@6,000     =56.4 Ohms
....
@6MHz     = .056 Ohms

So no, it obviously doesn't shunt all noise, but its efficacy increases with frequency. And the standard caveats apply - this is only addressing transverse mode noise, capacitors are not perfect, and they have inductance (increasing with size which compromise efficacy....)
The fact that a .47uf Auricap can often provide startling improvements, is a testament to how crappy our mains power can be.

Power filtering (as opposed to spike and surge protection) is no different from looking at a passive woofer crossover, and those same considerations apply, source impedance, load impedance, desired slope, power dissipation, etc....
Ultimately, good passive conditioning requires inductive elements in addition to capacitive ones. This is a very complex discussion that requires an understanding of filtering.

As to your specific questions - Yes, the capacitor is in parallel to all the components/appliances on a mains line. But the connections between have resistance, inductance, capacitance themselves which make the analysis less than straightforward.

[Daryl]

The fact that a .47uf Auricap can often provide startling improvements, is a testament to how crappy our mains power can be.

Auricap?

[ctviggen]

See:

http://www.vhaudio.com/acpowerconditioning.html#auricap

[Daryl]

See:

http://www.vhaudio.com/acpowerconditioning.html#auricap

LOL

Hi Bob,

I know what an Auricap is I was curious why Occam would mention them except to warn others against them.

Boutique capacitors have nothing to offer.

Extended foil construction on film and foil types will have lower inductance and resistance.

However physical size and lead inductance are much of the total inductance for the capacitor and no matter what the construction will not go away.

Lower resistance is not as big a benefit as you might think since the impedance curve for a capacitor is shaped lik a V.

The impedance starts very high at low frequency goes down in proportion to frequency until it reaches the capacitors self resonance.

After self resonance the capacitors impedance rises in proportion to frequency due to inductance.

Resistance has it's main effect near the capacitors self resonance where it determines the Q of the resonance which could be deep notch with low resistance or underdamped with electrolytics (specifics will vary with value).

Over most of the frequency range a capacitors impedance will be dominated by it's reactance.

If you want performance from capacitors stick to the garden varieties and use multiples of smaller units to achieve the value necessary.

Multiple smaller units in parallel will divide inductance by the number of units employed and be reduced further by the smaller physical dimensions of the smaller capacitors employed (especially if you consider that some boutique caps are huge for their value).

Instead of the 0.47uf Auricap for $19.00 you might get 100 of  the 0.68uf G.E. polypropylenes that Madisound has for $15.00 per 100 units (68uf total, same money but you would have a lot of soldering to do).

Of course getting coils involved and isolation transformers would be even better.

Daryl

[Occam]

I know what an Auricap is I was curious why Occam would mention them except to warn others against them.
Boutique capacitors have nothing to offer.

I mentioned them because I, and others, have yet to find a single capacitor which subjectively works as well in that role. There is a reason that the Good Lord, in Her infinite wisdom, has given us dpdt switches, so that we can make INFORMED EMPIRICAL judgements.

Extended film construction on film and foil types will have lower inductance and resistance.

While this is true, sadly foil construction does not have the advantage of 'self healing' metalized construction, as found in 'X' , Auricap, GE 4xL, etc... capacitors. There is nothing like that sinking feeling that you get when you watch that really expensive [what you thought was over rated] film and foil CDE or RelCap literally go up in a puff of smoke when subjected to the surges/spikes typical of an AC line.

However physical size and lead inductance are much of the total inductance for the capacitor and no matter what the construction will not go away.

For the DIYer this is true, but for manufacturing/development environment, this can be addressed. Look at the recent papers from Perrault -
IEEE TRANSACTIONS ON POWER ELECTRONICS, VOL. 19, NO. 3, MAY 2004 p591
Filters With Inductance Cancellation Using Printed Circuit Board Transformers
IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS, VOL. 40, NO. 2, MARCH/APRIL 2004 p483
Filters and Components With Inductance Cancellation

Multiple smaller units in parallel will divide inductance by the number of units employed and be reduced further by the smaller physical dimensions of the smaller capacitors employed (especially if you consider that some boutique caps are huge for their value).

I believe you oversimplified the matter, a bit. The inductances of the paralleled caps are not really in parallel, but rather, the paralleled caps should be viewed, at minimum, with each cap a series R-L-C, which is then paralleled. A very simple Mat-Lab or Spice simulation will illustrate this. Better yet, a sweep with a QuadTech 7600c with the actual components will confirm it. [and yes, I've done this] Given the great price and source for those GE caps that you've so generously shared with us, paralleling them does make sense, but you are simply rescaling that impedance 'V', in a good way. Typical industry practice is to use multiple different value capacitors to spread the attenuation spectrum. I've found that a triplet of .47, .1 and .01uf low cost X2 rated caps (used in concert with inductive components) can equal the performance of a single .47uf Auricap in the same circuit. While that makes tremendous sense from a manufacturer's perspective, for a DIYer, doing a oneoff, dunno.

Instead of the 0.47uf Auricap for $19.00 you might get 100 of  the 0.68uf G.E. polypropylenes that Madisound has for $15.00 per 100 units (68uf total, same money but you would have a lot of soldering to do).

Ya pays yer money, and you take yer choices. You can get them at the Partsconnexion.com for 17.50 and they often run 15% off sales. And dang!, they are expensive.

As a general comment - Anytime one uses a non 'X2' rated cap (UL1414) for across the line use, such as Auricaps or the GE 41Ls, one should take extra precautions, or not use them at all. Even with UL1414 caps, about 50 fires a year are caused by them in the US. I've never heard tell of a fire caused by that use of an Auricap, but due to their limited use in that role, such an assurance is anecdotal, at best.
Put them in a non flamable environment like a Hammond diecast aluminum enclose or fuse them where practical -
http://www.audiocircle.com/index.php?topic=24939.msg221401#msg221401

FWIW

[ctviggen]

I may have to rejoin IEEE to get me some of those articles. 

[Scott F.]

Now I politely ask a few questions for the gifted:
- Does the above hold any water?
- How can you stop other capacitors/ shunts polluting your audio/video power buses?
- If a single capacitor can change the sound of a system so profoundly, then what impact will all the "old/low quality" capacitors have connected throughout your house?
- Does the noise spread evenly across the multiple shunt capacitors throughout your home?

Hi pureAC,

I definitely don't consider myself one of the gifted but maybe I can point you to a couple of papers that might help.

This first one is on common mode filter analysis,
http://www.coilcraft.com/pdfs/doc200_CMFiltAnalysis.pdf

and then this one is on common mode filter design,
http://www.coilcraft.com/pdfs/doc191_CMFiltDesign.pdf

Hope that helps.

[Daryl]

Hi Occam,

I miss our chats (and the PM's especially).

I mentioned them because I, and others, have yet to find a single capacitor which subjectively works as well in that role. There is a reason that the Good Lord, in Her infinite wisdom, has given us dpdt switches, so that we can make INFORMED EMPIRICAL judgements.

How would one film capacitor be any different from another in the audio range (ignoring the fact that we are speaking of shunting powerline noise and not handling the signal)?

They measure near perfection in both the linear and non-linear domains and you are trying to discern the difference with your ears.

[Daryl]

I believe you oversimplified the matter, a bit.  The inductances of the paralleled caps are not really in parallel, but rather, the paralleled caps should be viewed, at minimum, with each cap a series R-L-C, which is then paralleled. A very simple Mat-Lab or Spice simulation will illustrate this. Better yet, a sweep with a QuadTech 7600c with the actual components will confirm it. [and yes, I've done this] Given the great price and source for those GE caps that you've so generously shared with us, paralleling them does make sense, but you are simply rescaling that impedance 'V', in a good way. Typical industry practice is to use multiple different value capacitors to spread the attenuation spectrum. I've found that a triplet of .47, .1 and .01uf low cost X2 rated caps (used in concert with inductive components) can equal the performance of a single .47uf Auricap in the same circuit. While that makes tremendous sense from a manufacturer's perspective, for a DIYer, doing a oneoff, dunno.

No Occam I didn't oversimplify the matter (honest).

I was speaking of just the inductance the other circuit elements exist but are not related to the point that I was making.

That is paralleling capacitors divides the total inductance by the number of units employed.

The admittance and thus the ability of a capacitor to shunt noise continues to improve with increasing frequency until the units self resonant frequency is reached and then it falls.

If you replace one 10uf capacitor with four 2.5uf capacitors you can expect inductance to drop to 1/4 that of the 10uf unit, the self resonant frequency to go up an octave and the effective bandwidth to go up two octaves since the impedance must climb the other side of the 'V' to reach the level where it is considered ineffective.

The 100 x 0.68uf example I suggested would have a self resonant frequency 10 times higher than that of a similar 68uf capacitor and an effective bandwidth 100 times higher.

Of course connecting the individual units could cause it's own inductance so your mechanical design must be suitable.

It is a simple concept which needs no simplification.

[pureAC]

Hi all,

So basically all capacitors accross the line interact with other electrical elements. How can we separate the individual caps from others caps.

I have noticed a power conditioner with auricaps, with CMC between them. Will this do the trick? Would a hammond common mode choke do this also.

I have noticed that when I place the Hammond chokes between components, it seems to isolate the two components.

pureac

[Occam]

.....
So basically all capacitors across the line interact with other electrical elements. How can we separate the individual caps from others caps.

I have noticed a power conditioner with auricaps, with CMC between them. Will this do the trick? Would a Hammond common mode choke do this also.

Audience's Adept conditioners use their own Auricaps flanking CMCs. They've recently released smaller versions (fewer than their original 12 cmc isolated outlets) -
http://www.audiocircle.com/index.php?topic=37804.0
Basically, the same topology is implemented by PS Audio in their Soloist and Ultra Outlet and Monster, Belkin in their conditioners, as well as Schurter, Corcom, etc... in their inlet and sealed versions with various degrees of efficacy, all using X2 certified capacitors.

I have noticed that when I place the Hammond chokes between components, it seems to isolate the two components.

Referring to the first quote, I'm not aware of any common mode chokes produced by Hammond, only differential mode chokes. As to the Robert Gray type use of Hammond 193 chokes, they're across the line, from 'hot' to 'neutral', not part of a conventional L-C type filter. I frankly can't figure out what the heck they're doing, nor do I have any personal experience with them. I'm not claiming they don't work, just that after reading the patent, I still don't understand what is going on.

FWIW,
Paul

[Dan Banquer]

For Occam and DAJ: A most common technique of reducing the effect of inductance in a capacitor is to bypass the cap with a smaller non/much lower inductive cap. Why pay big bucks for Auricap when you can bypass?
              d.b.

[BobM]

There is nothing like that sinking feeling that you get when you watch that really expensive [what you thought was over rated] film and foil CDE or RelCap literally go up in a puff of smoke when subjected to the surges/spikes typical of an AC line.

It is usually preceeded by a loud "POP" and a spark, immediately followed by an exclaimation (probably "OH SHIT") and a frantic grab at the power cord to yank it out of the outlet. I recently experienced this with a Jensen 4 pole cap (too expensive) when I forgot to check for continuity across all 4 poles before plugging the unit in. Gratefully, nothin else was harmed in this experiment.

Enjoy,
Bob

[daj]

For Occam and DAJ: A most common technique of reducing the effect of inductance in a capacitor is to bypass the cap with a smaller non/much lower inductive cap. Why pay big bucks for Auricap when you can bypass?
              d.b.

Also helps with ESR at high frequencies.

But . . . What I see here are attempts at explaining the basics, one at a time, which is no substitute for study. If I can make any contribution to this forum, it is to suggest study. I know most of the available books are bad, so please read on.

1) Firstly, it could help if you can make it to an extension course at your local college or Devry facsimile, making sure it covers analog and includes lab time. A course titled "Laboratory Electronics" for example would fit the bill. I'll bet a lot of you with so much interest will find your fascination and excitement increases, and a new world opens up to you.

2) In addition or as an alternative, there is an especially fine book that is a wonderful resource, especially for novices, but also practicing engineers: The Art of Electronics by Horowitz and Hill. (If you've ever tried to deal with an introductory text on EE, you probably haven't found one that isn't just awful.) This is an excellent book on how to actually "do" electronics, and one place to get it is:

http://www.amazon.com/Art-Electronics-Paul-Horowitz/dp/0521370957/ref=pd_bbs_2/102-6936263-4459344?ie=UTF8&s=books&qid=1173189170&sr=1-2

It's pricey but really very well worth it.

Here's the description from Amazon: 

"This is the thoroughly revised and updated second edition of the hugely successful The Art of Electronics. Widely accepted as the authoritative text and reference on electronic circuit design, both analog and digital, this book revolutionized the teaching of electronics by emphasizing the methods actually used by circuit designers -- a combination of some basic laws, rules of thumb, and a large bag of tricks. The result is a largely nonmathematical treatment that encourages circuit intuition, brainstorming, and simplified calculations of circuit values and performance. The new Art of Electronics retains the feeling of informality and easy access that helped make the first edition so successful and popular. It is an ideal first textbook on electronics for scientists and engineers and an indispensable reference for anyone, professional or amateur, who works with electronic circuits."

3) There are also some decent little handbooks of formulas. The one I like is Handbook of Electronic Formulas, Symbols and Definitions, by Brand. It's ridiculously expensive new, but reasonable used. You might even find it in the used section of your local college bookstore, where it should be in the $10 to $20 range. Amazon link, $15 used at the moment:

http://www.amazon.com/s/ref=nb_ss_b/104-8541173-2299151?url=search-alias%3Dstripbooks&field-keywords=Handbook+of+Electronic+Formulas%2C+Symbols+and+Definitions&Go.x=8&Go.y=11&Go=Go

4) There are a couple of books on acoustics that are worth trying to find, though many decades out of print.

The Elements of Acoustical Engineering by Olson
Acoustics by Beranek

A bit more is known now than then, but these are the foundational tomes.

5) Lastly, a foundational text on speakers and speaker-like gizmos is Electro-mechanical Transducers and Wave Filters by Mason.

[Dan Banquer]

Thanks for the recommendation, but I already have a degree in electronic technology from Wentworth, and over twenty years experience. My texts and on the job experience have already covered most of what you recommended.
           d.b.

[Daryl]

There is nothing like that sinking feeling that you get when you watch that really expensive [what you thought was over rated] film and foil CDE or RelCap literally go up in a puff of smoke when subjected to the surges/spikes typical of an AC line.

I would recommend proper design rather than crossing your fingers.

Now if we look at the reactance of a .47uf cap typically used as an 'X' cap across 'hot' and 'neutral' in the North American 60Hz power environment.

@60Hz Xc = 5,644Ohms
@600Hz    = 564 Ohms
@6,000     =56.4 Ohms

Yeah looks like a 0.47uf capacitor will be next to useless unless the noise sources effecting the line have pretty high impedance (if you spend at least $100 for that capacitor and display it where you and everyone can see it then it probably will work better than anything I might suggest).

Also not much energy will be needed to run such a small capacitor beyond it's voltage rating when a surge occurs.

68uf of total capacitance would attenuate noise much more aggressively and require much more energy to push over the top when a surge comes along.

Make up your 68uf up of capacitances small enough to provide the bandwidth your looking for (if you like soldering you could even go with my 100 x 0.68uf idea which will give extreme bandwidth).

Now preceed you capacitors with a bunch of MOV's, preceed those with chokes and preceed the chokes with fuses.

Viola, surge proof.

Let the other guys cross their fingers.

[Dan Banquer]

Not a bad idea to put a choke in front of the MOV's. This will help take the energy out from underneath that spike and the MOV's will last longer.
                 d.b.

[daj]

Thanks for the recommendation, but I already have a degree in electronic technology from Wentworth, and over twenty years experience. My texts and on the job experience have already covered most of what you recommended.
           d.b.

Right, of course. I am speaking to those who are turning here for tidbits of advice from knowledgeable people like you.

Oh, and I think becoming cliche free is the wave of the future.

[Dan Banquer]

"Oh, and I think becoming cliche free is the wave of the future. "
 
Good One!
          d.b.