OK long story short I heard a pretty bad flutter in my rear speakers while listening to music the other day. I can tell you right now that my cabling runs are not ideal. I have a 7.1 system. Will running the left\right rears\back rears together cause some interference? Also will running the wire in metal\pvc conduit help or hurt rejecting outside interference or will that make much difference from running it bare? My left side currently runs just under the base board heaters, does this introduce interference. My apt. setup is def not going to net great acoustics but I'm sure I could gain some by running the wiring a little better.

On the actual wiring side, will running twisted wire help with the noise interference? I am not trying to spend a boat load on an overhaul, but wouldnt mind spending $100-200 on new rerouting.

OK long story short I heard a pretty bad flutter in my rear speakers while listening to music the other day. I can tell you right now that my cabling runs are not ideal. I have a 7.1 system. Will running the left\right rears\back rears together cause some interference? Nope.

Also will running the wire in metal\pvc conduit help or hurt rejecting outside interference or will that make much difference from running it bare? Nope.
My left side currently runs just under the base board heaters, does this introduce interference. Nope.
On the actual wiring side, will running twisted wire help with the noise interference? Nope. Not for a speaker.

Flutter is likely an acoustic issue.

Thank you very much. I think I am going to run proper lenth wires and do a little cable management which should help a little. I was streaming MP3's off my laptop through my PS3 so that might have something to do with it. I will try various sources next.

Nope. Not for a speaker.

Is this due to the low impedance of the circuit?

Nope. Not for a speaker.

Most of your post I agree with. but.... While I agree that the slight increase in recieved ambient noise energy incident on the speaker terminals with the slightly more efficient antenna, that non twisted speaker cables represent as compared to twisted speaker cables, should be small enough not to drive most speakers enough to be noticeable due to the low impedance of the speakers. It is the electrical path that speaker cables provide that leads this incident noise back into the electronic equipment that worries me. Depending on the design, once the noise gets conducted inside near to the circuit, there is the opportunity to couple with higher impedance amplifier inputs and the like.

In other words, since the noise immunity gained by twisting speaker cables is improved and speaker cables are a potential path for noise into the amplification circuitry, I say twist.

since the noise immunity gained by twisting speaker cables

Twisted conductors are not inherently resistant to induced noise. Differential receivers connected to twisted pairs can nullify noise that has been equally induced into both conductors. This equal induction is a result of twisting the conductors.

Twisted pairs are slightly more resistant to differentially recieving ambient noise fields than even tightly run parallel pairs. As you probably know it is the differential noise which will be most damaging to our differential signal based sound systems...ground referneced single ended or not. It has to do with the area of the current loops (antennas) as seen by incident ambient noise fields. Twisted pair cables make slightly less efficient antennas resulting in less differential noise that gets seen by the system.

It is more the physics of the conductive paths 3D shape in whatever incident fields that may be present which makes the difference regarding how effficeient the antenna is and how much noise is then on the line and less the electronics that makes the difference. Now, I'm not disagreeing that twisting also has it's benefits to differential signaling in it's ability, when carefully manufactured, to help impedance match. Just that there is ALSO the extra crappy (in a good way) noise reception antenna property it also exhibits.

Is this due to the low impedance of the circuit? Yes, the amplifier output as well as the speaker are both low impedance devices.

It is the electrical path that speaker cables provide that leads this incident noise back into the electronic equipment that worries me. Depending on the design, once the noise gets conducted inside near to the circuit, there is the opportunity to couple with higher impedance amplifier inputs and the like. It's rare to find a power amp output stage that does not have a series choke. That pretty much kills the path for any RFI. (It's main purpose is to prevent the amplifier from becoming an oscillator with certain loads, but it takes the RFI issue along for the ride).

If there were even a hint that such noise interference were a real problem, you can bet that the speaker wire industry would be all over it with solutions like twisting or even shielding. Unless you live under a broadcast tower I wouldn't be concerned.

It's rare to find a power amp output stage that does not have a series choke. That pretty much kills the path for any RFI. (It's main purpose is to prevent the amplifier from becoming an oscillator with certain loads, but it takes the RFI issue along for the ride).

Yes, point well taken. The problem is: no filter is perfect and will block all of the noise so any possible EMI hardening on the cables is still useful in helping keeping out noise. Besides, I'm guessing these inline filters would be one of the first thing that gets eyeballed for removal by tweakers considering they be right in the signal path. The cost to twist being what it is, I just can't see any downside but can see several potential upsides to the EMI hardening that twisted wiring offers. Even if it is overkill for most cases, I certainly won't be unravelling my cables any time soon.

I'm guessing these inline filters would be one of the first thing that gets eyeballed for removal by tweakers considering they be right in the signal path. That would be a serious mistake. Could cost some their tweeters.

Not necessarily a mistake, only if it introduces an instability while speakers are attached. Potentially a huge benefit to sonics if there is no instability.

I would certainly be thoroughly testing for oscillations were I so inclined to pull off a move like that. Then again IF there is a problem it is usually easily remedied by start small and keep bumping up series R at the output until stability is achieved. As you point out though it may be wise to tune it while driving a dummy load. The people running the estat speakers and/or long capacitive cables that would be who really need to worry most. I half way wonder if my own amps could drive my speaker cable combo right now filterless without instability...may have to find out one day in the name of science.

Not necessarily a mistake, only if it introduces an instability while speakers are attached. Potentially a huge benefit to sonics if there is no instability. By what theory or mechanism would there be a huge benefit in sonics?
I would certainly be thoroughly testing for oscillations were I so inclined to pull off a move like that. How would you know if there was an instability or not until it was too late? You cannot simulate the load.

Then again IF there is a problem it is usually easily remedied by start small and keep bumping up series R at the output until stability is achieved. Series R? Are you kidding? Now that would cause a serious degradation in sonics. Poor damping factor!

As you point out though it may be wise to tune it while driving a dummy load. The people running the estat speakers and/or long capacitive cables that would be who really need to worry most. I half way wonder if my own amps could drive my speaker cable combo right now filterless without instability...may have to find out one day in the name of science. I never suggested tuning with a dummy load. It would be impossible for someone at home to simulate the exact load of their actual speaker. A 4 or 8-ohm resistor will not tell you anything abouty instability.

In the name of science, I suggest you do your investigations via electronics textbooks. It will save you lots of money. ;)

By what theory or mechanism would there be a huge benefit in sonics?

Mostly the ones dealing with filters being imperfect and the fact they will alter passband frequencies as well as the desired filter frequencies with the result the ideal goal of transparency is sacrificed. There are also the theories regarding the addition of any components adding thermal, contact etc. noise. I won't mention that each bit of length added to the signal path introduce opportunities for coupled/recieved ambient noise energy along with the unpredictable nature of coils being subjected to vibrations and other ambient environement changes.

How would you know if there was an instability or not until it was too late? You cannot simulate the load.

You can be sure that someone with experience in this area such as myself will easily be able to identify the sysmptoms of an amplifier operating on the edge of stability. Obviously if your testing showed these signs you'd be proceeding with caution at that point.

You can come close enough experimenting/simulating with lumped elements to get a good feeling about whether you can expect stability with a real world load.

Series R? Are you kidding? Now that would cause a serious degradation in sonics. Poor damping factor!

Not kidding at all. Series R right at the amp output is one tool designers have available to stabilize amps driving capacitive loads and this method is accepted industry wide as being standard practice. In other words, this IS how you do it.

You want to add just enough of this series R to create stability with a given loading. The idea is that the amp will be much happier, aka more likely to stay stable, if it sees on its output a resistance rather than the short circuit that a grossly capacitive cable appears to be during high frequency transients. Sure, it will sacrifice a small amount of damping but this is usually not a big deal with modern solid state amplifiers. Most of which seem to have plenty of this particular parameter to spare. Keep in mind that the series R is kept small as possible to achive stability but not any higher than is necessary in order to keep damping factor reduction to a minimum. One can even remove the the associated voltage drop by including this series R in the feedback loop.

The fact you didn't know about this industry standard amp design prctice raises questions about YOUR own experience with the matter. This is one of the basics of amp design that is outlined in beginning amp design guide books. ;)

I never suggested tuning with a dummy load. It would be impossible for someone at home to simulate the exact load of their actual speaker. A 4 or 8-ohm resistor will not tell you anything abouty instability.

I know. You recomended care, advice I likely would heed by either using some lumped element dummy load or a sacrificial speaker. BTW it may be wise to try not to make assumptions about me or put words into my mouth. A "dummy load" is not necessarily just a resistor by itself. I neither implied that it is or that a lone resistor was the dummy load I would plan to use in the event I were inclined to play with this further.

In the name of science, I suggest you do your investigations via electronics textbooks.

Nah been there done that. In fact I would guess that is the reason I've been employed as an electronic designer (non audio) for the last ten years doing this type of design work on a regular basis. Luckily for me and them my employers have given me the opportunity to prove my hair brained ideas by supplying them with many viable products instead of attempting to convince me to not even try in the first place as it would seem you are doing. The truth is that the majority consumer/pro amps are designed to be extremely flexible in what types of loading it can tolerate. This flexibility comes at the expense of what could be extra performance available to a more optimised and less general purpose tweak offshoot of the design. It's not as if these designs are stamped approved by God and can't be optimised for a more specific set of conditions.

It will save you lots of money.

Reading and staying fresh is good no doubt. It's just that this particular subject is such old hat routine for me by now I don't even need to check the books anymore for this stuff.

Yes, the amplifier output as well as the speaker are both low impedance devices.

interesting, I'm only an EE, but the hobbyists on this board have schooled me on this, and apparently it's all about signal level...mention impedance and you hear the sound of crickets. Thanks for the answer.

You can come close enough experimenting/simulating with lumped elements to get a good feeling about whether you can expect stability with a real world load. Not unless you accurately model your loudspeaker/cable. That's not easy. How would you do that?

Not kidding at all. Series R right at the amp output is one tool designers have available to stabilize amps driving capacitive loads and this method is accepted industry wide as being standard practice. In other words, this IS how you do it. Maybe that's how an amateur does it. And Bob Carver for his "tube simulation" mode (even there is is defeatable). But not me, and not Nelson Pass, David Reich, George Kaye...

Keep in mind that the series R is kept small as possible to achive stability but not any higher than is necessary in order to keep damping factor reduction to a minimum. One can even remove the the associated voltage drop by including this series R in the feedback loop. What would be a typical value resistance you have seen or used? BTW, if it's inside the feedback loop, it does not solve the stability problem the same as when it's outside. Didn't you know that?

If you want to put resistors in series with your speakers and claim it sounds better than a choke, knock yourself out. But don't try to kid me. I've been in audio electronics longer than you (BSEE 1973). The only time I've found resistors in the output stages of audiophile power amps (aside from the Carver mentioned earlier which was not for stability but sonic effect) were the emitter degeneration resistors in a parallel output structure to prevent thermal runaway and current hogging. But these are inside the feedback loop, where it does not impact output impedance/damping factor.

Do you know some amps with output series resistors? Might be good to list them here, so we can all avoid them. :D

Not unless you accurately model your loudspeaker/cable. That's not easy. How would you do that?

That IS easy. To guarantee stability with my speaker cable combo would simply require I guarantee stability with a worst case example that would exceed the requirements of remaining stable with the actual load. For example If I slap a ridiculously larger, low esr capacitance than my speaker/cable combo present across the output termianls of my amp and it remains stable then it will be safe to assume it will be fine w/ my speakers. Once I guarantee stability then I use my speakers as the load for remaining testing.

Maybe that's how an amateur does it.

This is how professionals educated recently are taught to do it in engineering school.

The first related book I grab from my office "IC op-amp Cookbook 3rd Ed." By Walt Jung published 1986 talks about this on page 160-161 and says exacty what I did. I think Walt gets at least a little respect among amp designers still. This is also what was taught in my electronics course at my school, BSEE Washington State University 1997.

Please no offense here but could it be that the 1973 BSEE is outdated? I'm asking since I truly am confused since I did truly believe that every EE was familiar with this technique.

In case you really are interested I've included a link that covers the general ideas fairly well.stability in the presence of capacitive loads (http://www.analog.com/Analog_Root/static/techSupport/designTools/interactiveTools/stability/stability.html)

not Nelson Pass, David Reich, George Kaye...

Not to doubt your word but I would be interested in digging deeper if you'd be kind enough to cite source material info.

What would be a typical value resistance you have seen or used? BTW, if it's inside the feedback loop, it does not solve the stability problem the same as when it's outside. Didn't you know that?

I'm (wild ass) guessing my own amps would need very little if any at all with my speakers and cables. That remains to be seen IMO but I'd guess that it's much less than 10 ohms. However I will say that back in my industrial motion controller designer days where control cables that are expected to be strewn potentially hundreds of feet across factory floors we used the ridiculously huge value of 50 ohms if memory serves just to guarantee there was never a problem with our off the shelf products. And yes as is pointed out in the link I provided putting it in the feedback loop works quite well and still allows it to address the stability issues it was added for in the first place. So you are incorrect on that point.

If you want to put resistors in series with your speakers and claim it sounds better than a choke, knock yourself out. But don't try to kid me. I've been in audio electronics longer than you (BSEE 1973).

It's very confusing to sit here and see someone who I believe probably has survived as an EE all these years say that putting a resistor sized small enough to insignificantly affect amp output impedance(ie somewhere on the order of the parasitic resistance in the choke filter you love but without winding capacitance lol) and is inside the feedback loop to compensate for the series voltage drop is somehow more damaging to signal integrity than a filter containing chokes.

Might be good to list them here, so we can all avoid them.

Now there is an open mind. :rolleyes:

Now Roger please don't misunderstand me here, as I mean no disrespect. I would certainly yield to hands on experience with these things but that isn't what you claimed. You acted as this thing I thought is industry standard is some alien tech. Certainly it's worth the time to discuss theory and debate it I hope you agree. I just want to make clear that I don't know how my experiments would end only that I wasn't satisfied by the counterarguments you provided. It's definitely nothing personal. I'd be man enough to admit in the face of good evidence that I'm wrong. In fact that's a big part of why I hang out at places like this... I llike learning new things and seeing what works in other industries (audio in this case) often helps me to see the bigger picture in my own little corner of the universe. I just haven't seen strong evidence and unfortunately I don't know you so taking your word without question is hard.

That IS easy. To guarantee stability with my speaker cable combo would simply require I guarantee stability with a worst case example that would exceed the requirements of remaining stable with the actual load. For example If I slap a ridiculously larger, low esr capacitance than my speaker/cable combo present across the output termianls of my amp and it remains stable then it will be safe to assume it will be fine w/ my speakers. Once I guarantee stability then I use my speakers as the load for remaining testing. Glad to hear it. What about the various inductances distributed in the load?

This is how professionals educated recently are taught to do it in engineering school.

The first related book I grab from my office "IC op-amp Cookbook 3rd Ed." By Walt Jung published 1986 talks about this on page 160-161 and says exacty what I did. I think Walt gets at least a little respect among amp designers still. This is also what was taught in my electronics course at my school, BSEE Washington State University 1997.

Please no offense here but could it be that the 1973 BSEE is outdated? I'm asking since I truly am confused since I did truly believe that every EE was familiar with this technique.

In case you really are interested I've included a link that covers the general ideas fairly well.stability in the presence of capacitive loads (http://www.analog.com/Analog_Root/static/techSupport/designTools/interactiveTools/stability/stability.html) I'm quite familiar with the technique. Used it myself. Works great.

That link to Analog Devices is talking about opamps. I suspect the same for the OpAmp Cookbook. I'm talking about power amps where one desires a high damping factor and minimal loss in the output stage.

Not to doubt your word but I would be interested in digging deeper if you'd be kind enough to cite source material info. Look at the schematic for a power amplifier. I happen to have one for a Phase Linear 400, as attached. It's not recent, but it's valid and relevant. My Threshold and Classe amps have no such resistors either--they just don't publish their schematics.

I'm (wild ass) guessing my own amps would need very little if any at all with my speakers and cables. That remains to be seen IMO but I'd guess that it's much less than 10 ohms. 10 ohms would be a dampling factor of around 1. Even just 1 ohm would be a damping factor on the order of 10. Both unacceptable in a high fidelity power amp. QSC quotes theirs at >500. Couple more links on the subject: http://en.wikipedia.org/wiki/Electronic_amplifier
http://www.st-andrews.ac.uk/~www_pa/Scots_Guide/audio/part2/page2.html

And yes as is pointed out in the link I provided putting it in the feedback loop works quite well and still allows it to address the stability issues it was added for in the first place. So you are incorrect on that point. I did not say you could not design an amplifier with a series output R in the feedback loop. I said the result would be different than when the R is outside the loop. In your earlier post you made it sound like you could choose to put the output resistor inside the feedback loop and be done. But that's not the case. For proof, just have a look at your own reference, where it says: >>In order to assure cancellation of both pole/ zero combinations, the above equations must be solved accurately.<< This means that the amplifier's newly added internal feeback compensation capacitor (Cf) has to be chosen based on the capacitance of the load. That's east to know when the following stages are part of the overall design, as in a multistage opamp case, but it's not easy to know when it's a power amp feeding the outside world. Do you see the snubber in the Phase Linear circuit? That works better in this case.

It's very confusing to sit here and see someone who I believe probably has survived as an EE all these years say that putting a resistor sized small enough to insignificantly affect amp output impedance(ie somewhere on the order of the parasitic resistance in the choke filter you love but without winding capacitance lol) and is inside the feedback loop to compensate for the series voltage drop is somehow more damaging to signal integrity than a filter containing chokes. Sorry to disappoint you. By the way, what is parasitic resistance? Sounds like bug spray. If you mean the DC resistance of the wire used to make the choke, it's probably less than the wire connecting the PC board to the output terminal.

Now Roger please don't misunderstand me here, as I mean no disrespect. I would certainly yield to hands on experience with these things but that isn't what you claimed. You acted as this thing I thought is industry standard is some alien tech. Certainly it's worth the time to discuss theory and debate it I hope you agree. I just want to make clear that I don't know how my experiments would end only that I wasn't satisfied by the counterarguments you provided. It's definitely nothing personal. I'd be man enough to admit in the face of good evidence that I'm wrong. In fact that's a big part of why I hang out at places like this... I llike learning new things and seeing what works in other industries (audio in this case) often helps me to see the bigger picture in my own little corner of the universe. I just haven't seen strong evidence and unfortunately I don't know you so taking your word without question is hard. I agree with your sentiments. I'd still be keen to see some high end power amplifiers that use an output topology as you espouse. There's a gentleman on the forum from QSC. I'd be happy to defer to anything he has to say on the subject. The man eats and sleeps this stuff.

What about the various inductances distributed in the load?

I figure that once stability is guaranteed it would be OK to move to to testing with the actual speaker itself eliminating the need for further guesswork.

That link to Analog Devices is talking about opamps. I suspect the same for the OpAmp Cookbook. I'm talking about power amps where one desires a high damping factor and minimal loss in the output stage.

Yep opamps. As I mentioned earlier, my presumption is/was that my amps would be stable with my speakers with very little if any compensation necessary, ie much much less than 10 ohms but hopefully none at all. Of course this is just pure guessing until I could try it out.

Look at the schematic for a power amplifier. I happen to have one for a Phase Linear 400, as attached. It's not recent, but it's valid and relevant. My Threshold and Classe amps have no such resistors either--they just don't publish their schematics.

I like old Phase Linear gear just can't justify what it goes for these days. My amps use a filtersimilar to what you've described too.

That's nice but I still wonder how well my amps would drive my speakers with no filter at all.

Recall the series R was only brought in the discussion as one possibility of ways to regain stability IF it were a problem. Obviously I wouldn't add it if it weren't needed. I never meant to imply that I already know that if I had to resort to this type of measure to achive stability after removing stock filtering that it automatically sounds best and I choose to keep it that way.

I think at this point I should apologise if my arrogance has blinded me from seeing any tidbits of wisdom you've dangled in front of me but honestly I tend to get my strongest lessons in the lab. I believe I grossly underestimated the value of what you had to say on my first overly hasty assesment of it.

It appears my amps have an even lower damping factor than the QSC's at >200 for 20-1K Hz for 8 ohms. Schematic toward the end of the manual. (http://safemanuals.com/user-guide-instructions-owner-manual/JBL/JBL%206290-_E) I tried attaching it but the file is to big.

I did not say you could not design an amplifier with a series output R in the feedback loop.

I see this now. Please see apology above. I misunderstood what you said, not that what you said is even unclearly stated for someone reading carefully. I admit it was a pretty big error on my part. I was spending time trying to explain stuff, perhaps even poorly, you clearly already understood.

Do you see the snubber in the Phase Linear circuit?

Yes. Interestingly my amps are using snubbers and an output filter both. :eek:

By the way, what is parasitic resistance? Sounds like bug spray/QUOTE]

lol! you are correct, DC winding resistance is what I meant. Is that bad terminology to call it parasistic? I am in the habit, maybe bad habit that is, of calling element nonidealness parasitic, ie inductor winding resistance, interwinding capacitance etc.

[QUOTE]I'd still be keen to see some high end power amplifiers that use an output topology as you espouse.

I honestly couldn't say whether something like this exists at all. I would note however that my particular amps, while certainly not modern, uses the interesting hybrid approach of using a 10 ohm resistor in parallel with it's output choke. I'm having trouble imagining the value of that right now.

Is this due to the low impedance of the circuit?


That and the fact that the voltages on speaker wires are among highest signal voltages in an audio system, sometimes exceeding line level signals by a factor of 40 or more.

That IS easy. To guarantee stability with my speaker cable combo would simply require I guarantee stability with a worst case example that would exceed the requirements of remaining stable with the actual load.


Loudspeakers can be very complex and even signficantly nonlinear loads. Exactly what a worst case load would be, is probably dependent on the amplifier, and at this point an untested hypothesis.

May I add that in 2009, unstable power amps are in general far from being the rule, and are so rare as to be a possibility that can often be simply dismissed.

I researched this issue at some length and eventually came up with a speaker simulator. However, my goal was not causing instabilty, but instead just stimulating audible differences due to the non-uniform source impedance of the typical power amplifier.

In the end my speaker simulator was based on taking a speaker simulator that a certain magazine was using for testing power amps, and making it about twice as nasty. This is what it takes to be a good simulation of the usual range of modern hi-fi speakers.


For example If I slap a ridiculously larger, low esr capacitance than my speaker/cable combo present across the output termianls of my amp and it remains stable then it will be safe to assume it will be fine w/ my speakers.


IME the largest possible capacitance may not be even the worst case pure capacitive load, let alone the worst case load in general. I have found many power amps that would handle very large, very low-ESR caps very well, but were driven ar closer to the brink with caps in the 0.1 to 0.01 uF range.


Once I guarantee stability then I use my speakers as the load for remaining testing.


Using actual speakers for amp testing is a hassle. They are fragile and noisy. ;-)

Also, most speakers are in several ways an easier load for a power amp than pumping sine waves into a dumb old 4 ohm NI resistor.

That and the fact that the voltages on speaker wires are among highest signal voltages in an audio system,

Signal level plays no part in this. If it did, you'd hear hum during quiet passages or when the volume is turned down.

Signal level plays no part in this. If it did, you'd hear hum during quiet passages or when the volume is turned down. I would not be quite so hasty. It's amazing what sorts of noise passes as squeaky clean when it's 60 dB quieter (as in comparing a mic line @ 1 mV to a speaker wire at 1V, for example).

I appreciate the entirety of your reply.

It appears my amps have an even lower damping factor than the QSC's at >200 for 20-1K Hz for 8 ohms. Schematic toward the end of the manual. (http://safemanuals.com/user-guide-instructions-owner-manual/JBL/JBL%206290-_E) I tried attaching it but the file is to big.

Yes. Interestingly my amps are using snubbers and an output filter both. :eek:

I would note however that my particular amps, while certainly not modern, uses the interesting hybrid approach of using a 10 ohm resistor in parallel with it's output choke. I'm having trouble imagining the value of that right now. Thanks for finding the schematic of the JBL amp. To save others the need to download the entire 3 MB file, I have extracted the schematic page and attached it below.

I suspect the 10 ohm resistor is there to bleed off any residual charge in the inductor to reduce arcing when the protection relay activates.

In pro amps it's common to have an output network that comprises a parallel L and R combination in series with the output, followed by some sort of R and C combination as a shunt (looking at it in the direction of the signal).

The idea is to have a very low series and high shunt impedance for audio signals from the amp circuitry, as well as high series and low shunt impedances for both RF coming in on the output terminals from without and any ultrasonic oscillations that arise from an amplifier circuit malfunction.

In the audio spectrum, the output Z should typically range from a few mΩ or less in the bass end to some tens of mΩ at the high-frequency end.

RF energy of several hundred kHz and higher should see the amp output as a shunt Z of much less than 1Ω, with a series Z behind it of several ohms effectively decoupling it from the amp's output stage circuitry.

I would not be quite so hasty. It's amazing what sorts of noise passes as squeaky clean when it's 60 dB quieter (as in comparing a mic line @ 1 mV to a speaker wire at 1V, for example).

So you can induce enough current into a speaker circuit to move the cone?

So you can induce enough current into a speaker circuit to move the cone? Yes, but not so's you'd notice or even be able to see if you tried. For example, just the ambient sound in the room is moving your speaker cones--but no one sees it.

I'm not disagreeing with you insofar as there's little probablility these induced noise signals would ever become audible. It's precisely because the signal levels are so low that this is the case, not because they cannot or do not exist. When you said >>signal level plays no part in this<< I took that to mean they are nonexistant. Maybe I misunderstood your meaning.