Long interconnects or speaker cable?

[Steve Eddy]

Now that we seem to be on the same page, let's take this full circle.

Going back to the "statement in question":

Noise induced by varying magnetic fields will be differential and not rejected by a balanced circuit, which rejects common mode noise.

This seems to be quite correct. The noise induced into the loop will be differential.

Can anyone explain how you could possibly have a common-mode voltage induced?

se

 

[sully]

1..if the time varying flux goes between the twisted pair and it's shield

2..if the IC shield forms a loop with the equipment and the line cords of the equipment, any shield currents caused by that loop picking up delta flux can cause IR drops in the shield..so the receiver shield can be moving relative to the pair. Even though the pair and shield are seeing the same delta flux and emf generation, the shield has a low impedance path for the voltage, giving higher currents..the signal pair see high impedance, so current is much lower.

Cheers, John

PS..that is why I prefer using short ic's and long speaker runs..

 

[Steve Eddy]

1..if the time varying flux goes between the twisted pair and it's shield

2..if the IC shield forms a loop with the equipment and the line cords of the equipment, any shield currents caused by that loop picking up delta flux can cause IR drops in the shield..so the receiver shield can be moving relative to the pair. Even though the pair and shield are seeing the same delta flux and emf generation, the shield has a low impedance path for the voltage, giving higher currents..the signal pair see high impedance, so current is much lower.

Yes, I can see how you can get a common-mode voltage when you have TWO loops. But I don't see how that could be the case with one loop.

se

 

[mrfeedback]

If It's Not a Personal Question...............

There is a nulling circuit that can compensate for this in a large degree with a very small and simple network that fits easily inside the male RCA.

Do tell.

Eric.

 

[JOE DIRT®]

Steve....you know how how common mode noise is picked up

these threads confound me as the the original question gets twisted and the person who started it is never around

 

[Steve Eddy]

Steve....you know how how common mode noise is picked up

Well I thought I did when I entered this thread.

But now it seems apparent that there is no common-mode component with regard to magnetically induced noise in a single loop. That in such a case the induced noise voltage is differential. That contrary to at least one reference cited, it has nothing to do with the proximity of each conductor to the source relative to the other, but rather how much flux is changing within the loop and that all you can do to reduce it (seeing as a differential input will not reject differential voltages) is to keep the loop area as small as possible by keeping the conductors as closely spaced as possible and/or make the loop area effectively smaller by way of twisting and local cancellation within the loop itself rather than common-mode rejection at the end of the loop.

My understanding of the issue has changed dramatically since I entered this thread thanks to those who choose to engage in actual dialogue rather than just throw up a bunch of URLs which in this case served only to confuse the issue further as some were at odds with others.

se

 

[Fred Dieckmann]

Don't believe every thing you read.......

"My understanding of the issue has changed dramatically since I entered this thread thanks to those who choose to engage in actual dialogue rather than just throw up a bunch of URLs which in this case served only to confuse the issue further as some were at odds with others."

You're welcome...... Engaging in dialogue only helps if it is dialogue if you actually intend to listen. Don't take one persons word for something even if you think that they are your guru. Alternately don't assume something is wrong because you don't like the person telling you. There is still some misunderstand about magnetic fields but I dare not reference any URLs. I am thinking about taking up an easier and more relaxing hobby such as deprogramming people who have been brain washed by religious cults.

 

[Steve Eddy]

Re: Don't believe every thing you read.......

Engaging in dialogue only helps if it is dialogue if you actually intend to listen.

I always intend to listen. But when I brought up the contradiction between the Whitlock references and the Capgo reference along with the "statement in question" with regard to noise being induced differentially and no common-mode, nothing followed to explain the contradiction, either by explaining how both are correct or which one was wrong.

Don't take one persons word for something even if you think that they are your guru.

I don't. But I'm not going to change my view simply because someone says I don't know what I'm talking about. I'm always open to changing my view in light of a cogent argument as to exactly WHY any currently held view of mine is in error, regardless of how I came to hold such a view.

Alternately don't assume something is wrong because you don't like the person telling you.

I don't. But I'm not going to change my view simply because someone says I don't know what I'm talking about. I'm always open to changing my view in light of a cogent argument as to eactly WHY any currently held view of mine is in error, regardless of who makes the argument.

It's true I don't like you as a person, but I don't judge the veracity of your words because of that. I'll give any argument you care to present as much consideration as I'd give an argument from anyone else. And if it's a convincing argument, I'd have no more problem admitting to it than a convincing argument from anyone else.

But if all someone does is claim that what I've said is incorrect, I don't care who they are or how much I like or dislike them, if their claim that what I've said is incorrect isn't followed by a cogent argument as to exactly WHY what I've said is incorrect, I will challenge them to do so. Just ask SY. I like him quite a lot but I've challenged him before to explain WHY.

In spite of what some seem to think, I don't look at this stuff in any sort of personal terms such as WHO is right or WHO is wrong. I look at it in terms of WHAT is right and WHAT is wrong. And what's right is right and what's wrong is wrong regardless of WHO may say it.

Instead of ridiculing the PERSON who may say something you believe is wrong, why don't you simply address WHAT was said? There's absolutely no reason to make any of the technical issues discussed here a personal issue. None of us here are pefect. None of us here know everything and understand everything. And just because someone's understanding of an issue may be incorrect, that makes them no more inferior a person than correctly understanding an issue makes one more superior a person.

So how 'bout putting the personal stuff aside and just stick to the issues by way of respectful dialogue and debate?

There is still some misunderstand about magnetic fields but I dare not reference any URLs.[/B]

And the page you post here seems to further reinforce the notion that the "statement in question" and the Capgo references are correct and the Whitlock reference is incorrect. That the noise induced into a single loop will be differential, not common-mode and that twisting doesn't help by making the wires equidistant from the source, but rather works by effectively reducing loop area by way of local cancellation.

se

 

[Steve Eddy]

By the way, is that page you posted from Ott?

se

 

[Christer]

An observation I have made, not only in this thread, is that many
controversies, not the least those that tend to lead to personal
insults, seem to have their origin in the fact that arguments are
presented only in plain text and are often somewhat sloppy or
ambiguous. I think many of these controversies could have been
avoided if someone had brought in some mathematics early in
the disussions. Unfortunately, it seems many threads have a
tendency to die out when somebody dares to do so. Personally
I applaud Freds initiative to post the formula above.

In Sweden we use to say that a picture says more than a
thousand words (I don't know if you say so in english too).
I think one could just as well say that a formula says more
than a thousand words.

 

[millwood]

I think many of these controversies could have been
avoided if someone had brought in some mathematics early in
the disussions.

I thought I was the only one suggesting that more math would be helpful in understanding the skin effect topic.

Unfortunately, not one soul listened, .

 

[Steve Eddy]

An observation I have made, not only in this thread, is that many
controversies, not the least those that tend to lead to personal
insults, seem to have their origin in the fact that arguments are
presented only in plain text and are often somewhat sloppy or
ambiguous.

But that's no excuse for making things personal. If something is sloppy or ambiguous, try and get it clarified.

I think many of these controversies could have been
avoided if someone had brought in some mathematics early in
the disussions. Unfortunately, it seems many threads have a
tendency to die out when somebody dares to do so. Personally
I applaud Freds initiative to post the formula above.

In Sweden we use to say that a picture says more than a
thousand words (I don't know if you say so in english too).
I think one could just as well say that a formula says more
than a thousand words.

But I don't know that the forumla actually resolves the issue.

For example, look what happens when theta equals 90. The cosine of 90 is 0. Which will give a V_n of 0. Does that mean that V_n is zero because there's no induced voltage anywhere in the loop or because the voltages across each half of the loop are of the same magnitude and polarity, i.e. common-mode?

se

 

[fdegrove]

Hi,

An observation I have made, not only in this thread, is that many controversies, not the least those that tend to lead to personal insults, seem to have their origin in the fact that arguments are presented only in plain text and are often somewhat sloppy or ambiguous.

How very true...

Unfortunately, it seems many threads have a
tendency to die out when somebody dares to do so.

Not every formula will be understood unless carefully explained IMO.

In Sweden we use to say that a picture says more than a thousand words (I don't know if you say so in english too).

Probably in many languages but than you need to find the right pic for the job...not always an easy task.

I'd still like to see the appropriate Whitlock reference (AN#?) as it does seem open for interpretation.
That's not good scientific practice if that's the case but when things are torn out of context they often don't make a lot of sense.

The topic at hand wasn't even all that hard to figure out as it's a simple logical deduction. I somehow have a hard time understanding how a publicized paper by someone as reputable as Whitlock could have sat there being just plain wrong for such a long time??

Anyway, glad to see the storm in a glass is over...

Cheers,

 

[Christer]

But that's no excuse for making things personal. If something is sloppy or ambiguous, try and get it clarified.

Absolutely. However, I think it is often the case that those
who write something do not realize it is ambiguous. That's
elementary pedagogics really. Often people are so biased
by their own understanding of something that they can't see
that what they write is unclear to someone not already knowing
what it is supposed to mean.

But I don't know that the forumla actually resolves the issue.

For example, look what happens when theta equals 90. The cosine of 90 is 0. Which will give a V_n of 0. Does that mean that V_n is zero because there's no induced voltage anywhere in the loop or because the voltages across each half of the loop are of the same magnitude and polarity, i.e. common-mode?

The posting of a formula doesn't necessarily solve an issue
immediately, but a formula is usually a much more precise
statement than text is, so it is easier to discuss the issue
when referring to the formula. Your question above is very
clear becuase it directly refers to the formula. As for the
answer, well that's for somebody else to give, I am afraid.

Not every formula will be understood unless carefully explained IMO.

Quite so. Usually one needs a combination of text and formulae.
Of course all variables and assumptions must be explained
unless obvious. My point, however, is that a formula is more
precise than text usually is, so even it further discussion is
required to clarify what it means, such a discussion rests on
a firmer ground.

The topic at hand wasn't even all that hard to figure out as it's a simple logical deduction. I somehow have a hard time understanding how a publicized paper by someone as reputable as Whitlock could have sat there being just plain wrong for such a long time??

You'd be surprised what errors sometimes slips through even
in scintific journals, where peer reviewing is supposed to
prevent that from happening. For better or worse, I have made
myself a bit of a reputation for publishing papers pointing
out errors in other publications within my professinal area.

 

[Steve Eddy]

Not every formula will be understood unless carefully explained IMO.

Sure. But this one's just simple algebra.

I'd still like to see the appropriate Whitlock reference (AN#?) as it does seem open for interpretation.
That's not good scientific practice if that's the case but when things are torn out of context they often don't make a lot of sense.

The one I was referring to was one of the two cited by Fred, from the two part article in Sound & Video Contractor. Specifically:

CMRR in Balanced Interfaces: Part 2

More specifically, where Whitlock says:

AC magnetic fields induce voltages in wires. The magnitude of the voltage depends on the strength of the field, which depends on the distance between wire and field source. Therefore, equal voltages will be induced in the two wires of a balanced cable only if they are equidistant from the field source. Any voltage difference is added to the signal as noise.

Twisting of the two wires is a first-order technique to make induced voltages identical by averaging the physical positions of the wires.

The "statement in question" however says that magnetic fields induce voltages differentially, not common-mode and therefore cannot be rejected in a balanced interface. To repeat:

Noise induced by varying magnetic fields will be differential and not rejected by a balanced circuit, which rejects common mode noise.

So the question to me was, which is it? Are voltages induced by magnetic fields always induced differentially as the "statement in question" states, or can they also be induced common-mode as the Whitlock reference states?

If one is correct, the other must be incorrect. If the induced voltage is always differential, then there's no common-mode component.

The Whitlock reference says that the purpose of twisting is to assure that the induced voltages are as common-mode as possible so that they can be rejected at the input. The Capgo reference as well as the page Fred just posted say that twisting works by alternately flipping the loop so that the voltages induced in the loop cancel such that ideally, there's no noise voltage to deal with at all, differential or common-mode.

The topic at hand wasn't even all that hard to figure out as it's a simple logical deduction.

Well no one has denied that twisted pairs result in less noise than parallel pairs. Question remains however as to exactly WHY. And I think that's an important question as depending on which is correct, there are two different implications here.

If the Whitlock reference is correct, then balance matters. You need to assure balance in order to maintain common-mode voltages which are then rejected at the receiver.

If the "statement in question" and the other references are correct, then balance doesn't matter as the induced noise voltage will always be differential in nature and the reduction of such induced noise is a function of the cable, not the rejection of common-mode noise at the receiver.

I somehow have a hard time understanding how a publicized paper by someone as reputable as Whitlock could have sat there being just plain wrong for such a long time??

Well, if sully's arguments have any veracity, then Hawksford's Essex Echo has been sitting there just plain wrong for quite a long time.

se

 

[SY]

Invoking SY's Second Law again

In Sweden we use to say that a picture says more than a thousand words (I don't know if you say so in english too).
I think one could just as well say that a formula says more
than a thousand words.

And one can say even better that 5 minutes spent with a couple pieces of wire, a resistor, a scope, and a source of EM field (like a transformer) can clear up the issue more certainly and inarguably than 10,000 pages of equations and "logical" argument.

I've got company coming tonight and Mrs. Y has me vacuuming, scrubbing, and minding Red Chief, so it's up to someone else this time.

 

[Christer]

Re: Invoking SY's Second Law again

And one can say even better that 5 minutes spent with a couple pieces of wire, a resistor, a scope, and a source of EM field (like a transformer) can clear up the issue more certainly and inarguably than 10,000 pages of equations and "logical" argument.

Sorry, I beg to disagree. Experiments and simulations are often
important, but they can only tell you that, not why.
For the latter, you need a theoretical basis to interpret the
results of the experiment/simulation. As an example, yesterday
I was wondering about the PSRR of a diamond buffer. I was
lazy, so instead of sitting down with pen and paper, I ran some
Spice simulations. Since I hadn't tried to analyze it theoretically,
I got a dissapointing and surprising result. Thinking a little
about, I realized what might be wrong with the simulation and
tried to change that. This gave a much more satisfying, and
probably more correct result. However, it didn't tell me why the
two simulations gave different results. IN this case it turned out
the explanation was rather obvious when thinking a little about
what happens. So, yes, the simulations helped me to understand,
but they did not per se explain anything. Ironically, the fact that
I made the simulation in the wrong way in the first place actually
helped me to understand the issue theoretically.

 

[Steve Eddy]

Re: Invoking SY's Second Law again

And one can say even better that 5 minutes spent with a couple pieces of wire, a resistor, a scope, and a source of EM field (like a transformer) can clear up the issue more certainly and inarguably than 10,000 pages of equations and "logical" argument.

True.

I've been mulling over how best to work up a jig to do just that.

Mark's heading over with some parts for a rush order of circuit boards I gotta knock out but might be able to get to it Monday or Tuesday.

se

 

[Steve Eddy]

Re: Re: Invoking SY's Second Law again

Sorry, I beg to disagree. Experiments and simulations are often
important, but they can only tell you that, not why.
For the latter, you need a theoretical basis to interpret the
results of the experiment/simulation.

But in this case, we have the why. In fact, we have two of them. And that's the problem. Each seems to contradict the other.

For a given situation, each would predict a different result. It would seem therefore that the issue at hand can be resolved by establishing a given situation, make the measurements, and see which why is most consistent with the measured result.

se

 

[fdegrove]

ECHOES...WITH A TWIST.

Hi,

AC magnetic fields induce voltages in wires. The magnitude of the voltage depends on the strength of the field, which depends on the distance between wire and field source. Therefore, equal voltages will be induced in the two wires of a balanced cable only if they are equidistant from the field source. Any voltage difference is added to the signal as noise.

This takes a few rereadings to understand the train of thought behind it but it seems correct.
When equal voltages are run on both send and return of the balanced cable it will be seen by the balanced receiver as being common to both and will be rejected.

The question that arises is :When are the two wires equidistant from the source of radiation?

When the send and return are tightly twisted they would be equidistant, wouldn't they?

Twisting of the two wires is a first-order technique to make induced voltages identical by averaging the physical positions of the wires.

Et voila...There it is.

Noise induced by varying magnetic fields will be differential and not rejected by a balanced circuit, which rejects common mode noise.

And that's correct too provided:

1)EMI fields of varying strength are induced in the send and return lines which will create a differential voltage and be treated as a differential signal at the receiving end.

2) Our send and return lines aren't twisted in the first place.

Hence there will be no CMR and the noise is added to signal.

The "statement in question" however says that magnetic fields induce voltages differentially, not common-mode and therefore cannot be rejected in a balanced interface.

The qualifier here is: only if they're of varying magnetic field strength (field A is different form field B) and no twisting is used (our wires aren't equidistant).

If that's the case a differential voltage will develop (send will carry a different voltage than return or vice versa) and will not be rejected since there's no common signal strength.

So the question to me was, which is it? Are voltages induced by magnetic fields always induced differentially as the "statement in question" states, or can they also be induced common-mode as the Whitlock reference states?

No, they can well be induced common-mode, see above.

If the induced voltage is always differential, then there's no common-mode component.

If it were to be picked-up as differential then, yes.
And that would have to meet the conditions outlined above.

In real life and with good engineering, that just won't happen too often.

The Whitlock reference says that the purpose of twisting is to assure that the induced voltages are as common-mode as possible so that they can be rejected at the input.

Does it say that? That's not how I see it.
It's really badly worded and it's not what goes on in a twisted pair.

The Capgo reference as well as the page Fred just posted say that twisting works by alternately flipping the loop so that the voltages induced in the loop cancel such that ideally, there's no noise voltage to deal with at all, differential or common-mode

That's how I view it too and as I mentioned before, it works whether we use a balanced or unbalanced receiver which proves the point IMO.
Not only does this work for EMI, the same occurs with RFI although not exactly in the same way.

Well no one has denied that twisted pairs result in less noise than parallel pairs.

I wasn't referring to just that but nevermind.

If the Whitlock reference is correct, then balance matters. You need to assure balance in order to maintain common-mode voltages which are then rejected at the receiver.

Of course you do, regardless of Whitlock or anybody else's statement.

If the "statement in question" and the other references are correct, then balance doesn't matter as the induced noise voltage will always be differential in nature and the reduction of such induced noise is a function of the cable, not the rejection of common-mode noise at the receiver.

You need to meet several criteria for the noise to be picked up and result in a differential signal on the cables, when the cable is twisted this is rather hard to do but probably not impossible.

See, a balanced line only rejects common mode noise, NOT differentail noise, if it did reject that too the net result would be no signal at all.

Well, if sully's arguments have any veracity, then Hawksford's Essex Echo has been sitting there just plain wrong for quite a long time.

I wisely hold my breath on that thorny topic but my calculations show me that Hawksford's sim isn't too far off...
The conclusion as printed by HFN&RR is a little simplistic at best.

Cheers,