McIntosh Transformer causing noise in speakers while disconnected.

[Doodski]

Regarding the metering of the AC EM field around the McIntosh I am staying in my lane because I know zero about what this metering means... I will observe now.

 

[mhardy6647]

I find the technicality of this textual communication to lend me very strong spoken word organizational skills. When I speak now I always minimizing my words, skipping all those extra words that are not required. I also use the proper words and of course substitute generalizations the word, "Stuff." Sometimes peeps I speak with really don't need all the details and we all know what the stuff is. LoL... I really need to get the <;> function use understood. I never seem to be able to use properly and I fudge it always.

I am now officially rolling on the floor.

 

[Doodski]

Apples to apples - I ripped the whole second video with Audacity. In other words the time axis is true to the original video.

View attachment 357796

The resolution of the time scale is wayyy too low. The resolution needs to be @ the most ~mili seconds to determine the duration of a full 360 degrees of the distorted sine wave.

 

[Zapper]

... and just for (as I like to say) shoots and googles -- 150 Hz

Assuming @ClassG33 is in the US and not Europe, that does not support the idea it's 60Hz hum. However, this is one of the weirdest things I've heard of. It is not an ordinary occurrence, with ordinary answers! So we should keep open minds and not jump to conclusions.

TLDR;
Hypothesis: one or more channel is defective and oscillating at high power, radiating RF at > 100kHz. This RF is inducing current in the inductors, voicecoils, and wiring in the speaker. Internal current loops allow current to flow through voicecoil. 2nd order nonlinearities in speaker demodulate the RF into audible noise.

I suggested the radiated energy may be at a much higher frequency than we hear, and the audible tone is a demodulation of the high frequency carrier wave, like an AM radio. So let's explore that hypothesis a bit.

We have transfer of energy over a distance of several feet, enough to make an audible sound in the speakers. It is hard to transfer energy at 60 Hz magnetically. 60Hz transformers need steel or iron cores to give them the large magnetizing inductance they need to transfer power at 60Hz. Air core transformers are very inefficient at 60 Hz. And the buzzing speakers are separated from the amp by feet of air.

Wireless charging can be performed over an air gap. To address the inductance problem, the frequencies are much higher - on the order of 100's of kHz. At these frequencies the required inductance is much lower and air core transformers can be used. Typically these are formed into resonating tank circuits with a capacitor. So it seems possible that the energy is being transferred from the amp to the speakers by a high frequency EM field rather than a low frequency one, as appreciable energy can be transferred through relatively small inductors, such as those in the LC networks of a speaker crossover.

An amp can generate high frequencies if it is oscillating. The joke in Analog Design 101 classes is that if you want to design an amplifier you get an oscillator; if you want to design an oscillator you get an amplifier. High gain amplifiers will oscillate without careful control of the phase and gain in the feedback loop. Certain defects can cause a loss of feedback control, resulting in oscillation. For example, many amps rely on a single capacitor to establish the "dominant pole", which attenuates the gain with increasing frequency, so the amp runs out of gain before it reaches a frequency that it becomes unstable. A bad solder joint on that capacitor, and the amp can oscillate. Oscillation can be expected to occur around the unity gain frequency (the frequency where the amp gain = 1), which is often in the 100kHz range or higher for high performing amplifiers. So there is the theoretical possibility that the amp is oscillating and transmitting RF power in the longwave or maybe even AM band.

How can that produce sound in the speakers? The coils in the speaker, both the inductors in the crossover and the voicecoils, can convert the alternating magnetic field into AC current. There are several current loops internal to a multi-driver speaker with crossovers that the current can flow through, through voicecoils, inductors, and capacitors. So no Kelvin's laws need be violated in the production of the buzz. @mhardy6647 gives an example of this phenomenon in a previous comment.

Last for the source of the 50Hz, or 60Hz, or whatever frequencies show up in the spectrum. If the oscillator were producing a steady sine wave, there would be no sound in the speaker, even if the HF was received, converted to current, and demodulated. But unintended oscillations can be messy. They can pulse on and off, as the large amplitude signals shift the bias points, which change gain and capacitance in the devices. They can oscillate chaotically, which will produce a broad spectrum of frequencies. They can oscillate at some mix of frequencies, with intermodulation components. The most likely modulating frequency would be 60Hz from the power supply, but others are possible.

So maybe there is a mechanism to produce sound in a disconnected speaker at a distance from a high power oscillation. A lot of things would have to happen for this to occur, and it is not a likely scenario. But @ClassG33 has speakers buzzing with no electrical connections! Something very unlikely has already happened here, so anything goes.

How to test this hypothesis? Easy if @ClassG33 had an oscilloscope, which I presume he does not. Do you have a portable AM radio, @ClassG33 ? If so, does it exhibit any strange behavior around the haunted amp? Try sweeping the dial and see if it picks up any static or buzzing.

 

[Zapper]

This is a totally new thing for me to see.

You and me both. This is one of the most interesting and bizarre things I've seen in a long time.

 

[Doodski]

Assuming @ClassG33 is in the US and not Europe, that does not support the idea it's 60Hz hum. However, this is one of the weirdest things I've heard of. It is not an ordinary occurrence, with ordinary answers! So we should keep open minds and not jump to conclusions.

I suggested the radiated energy may be at a much higher frequency than we hear, and the audible tone is a demodulation of the high frequency carrier wave, like an AM radio. So let's explore that hypothesis a bit.

We have transfer of energy over a distance of several feet, enough to make an audible sound in the speakers. It is hard to transfer energy at 60 Hz magnetically. 60Hz transformers need steel or iron cores to give them the large magnetizing inductance they need to transfer power at 60Hz. Air core transformers are very inefficient at 60 Hz. And the buzzing speakers are separated from the amp by feet of air.

Wireless charging can be performed over an air gap. To address the inductance problem, the frequencies are much higher - on the order of 100's of kHz. At these frequencies the required inductance is much lower and air core transformers can be used. Typically these are formed into resonating tank circuits with a capacitor. So it seems possible that the energy is being transferred from the amp to the speakers by a high frequency EM field rather than a low frequency one, as appreciable energy can be transferred through relatively small inductors, such as those in the LC networks of a speaker crossover.

An amp can generate high frequencies if it is oscillating. The joke in Analog Design 101 classes is that if you want to design an amplifier you get an oscillator; if you want to design an oscillator you get an amplifier. High gain amplifiers will oscillate without careful control of the phase and gain in the feedback loop. Certain defects can cause a loss of feedback control, resulting in oscillation. For example, many amps rely on a single capacitor to establish the "dominant pole", which attenuates the gain with increasing frequency, so the amp runs out of gain before it reaches a frequency that it becomes unstable. A bad solder joint on that capacitor and the amp can oscillate. Oscillation can be expected to occur around the unity gain frequency (the frequency where the amp gain = 1), which is often in the 100kHz range or higher for high performing amplifiers. So there is the theoretical possibility that the amp is oscillating and transmitting RF power in the longwave or maybe even AM band.

How can that produce sound in the speakers? The coils in the speaker, both the inductors in the crossover and the voicecoils, can convert the alternating magnetic field into AC current. There are several current loops internal to a multi-driver speaker with crossovers that the current can flow through, through voicecoils, inductors, and capacitors. So no Kelvin's laws need be violated in the production of the buzz.

To produce an audio frequency from a carrier, a demodulating mechanism is required. But there are several 2nd order distortion mechanisms in speakers - and those involving eddy currents are likely to be very strong at high frequencies. Second order nonlinearities can demodulate the carrier.

Last for the source of the 50Hz, or 60Hz, or whatever frequencies show up in the spectrum. If the oscillator were producing a steady sine wave, there would be no sound in the speaker, even if the HF was received, converted to current, and demodulated. But unintended oscillations can be messy. They can pulse on and off, as the large amplitude signals shift the bias points, which change gain and capacitance in the devices. They can oscillate chaotically, which will produce a broad spectrum of frequencies. They can oscillate at some mix of frequencies, with intermodulation components.

So maybe there is a mechanism to produce sound in a disconnected speaker at a distance from a high power oscillation. A lot of things would have to happen for this to occur, and it is not a likely scenario. But @ClassG33 has speakers buzzing with no electrical connections! Something very unlikely has already happened here, so anything goes.

How to test this hypothesis? Easy if @ClassG33 had an oscilloscope, which I presume he does not. Do you have a portable AM radio, @ClassG33 ? If so, does it exhibit any strange behavior around the haunted amp? Try sweeping the dial and see if it picks up any static or buzzing.

These details open a whole new world of ideas for me. All I can say is far freaking out. Thanks for hanging out here @Zapper.

 

[Zapper]

So all things considered now... is it unanimous that this is a defective amplifier and needs to be exchanged?

I vote yes.

 

[Doodski]

I vote yes too. Get rid of it and get a replacement unit.

 

[Philbo King]

If it is new gear, I'd return or exchange it.

 

[mhardy6647]

... and yet the sound is, to me, just like ground loop buzz/hum.

I was mis-applying Audacity's FT tool. Having now tested it with some pure sine waves, I think I have a better feel for it (see above in my earlier, low signal-to-noise fusillade of posts!).

So, one last thing from me tonight
I edited the file down to contain just some of the nice, hummy/buzzy audio after 1:10 or so and then did the FT on that snippet. Here, FWIW, is what I can 'see'.
I annotated some of the more salient peak species' frequencies using the cursor readout on Audacity's FT frequency analysis.

 

[GXAlan]

tagging @amirm and @restorer-john @John Atkinson @Kal Rubinson and @SIY in case they have not seen this thread.

OP has a B&W center channel with zero wires connected to it. When he moves it close to his McIntosh amp, there is enough induction from presumably a magnetic field to induce the voice coils of the speakers to move, generating audible buzz… and he has the videos to prove it.

Has *anyone* ever seen/heard this?

 

[mhardy6647]

I have a funny feeling that a second amp will manifest the same symptom...
I'm gonna lay the odds at 50/50.
OK, I've gotta go to bed -- it's been a long day...

 

[Extreme_Boky]

It could be the backlight switching driver circuit... for those VU meters... can the backlight be switched off??

 

[Extreme_Boky]

Do we know if the thing uses SMPS??

 

[Doodski]

I edited the file down to contain just some of the nice, hummy/buzzy audio after 1:10 or so and then did the FT on that snippet. Here, FWIW, is what I can 'see'.
I annotated some of the more salient peak species' frequencies using the cursor readout on Audacity's FT frequency analysis.

I think the sample is not from a absolute peak of the distorted sinusoidal waveform and are taken from a very distorted section and as result the proportions of the fundamental frequency versus the odd and even order harmonics are not as they would be with a non-distorted sinusoidal waveform. Maybe try a clean pure 60 Hz sine wave and then compare that to your FT analysis image here<?>

 

[Doodski]

Do we know if the thing uses SMPS??

It has 6 high value capacitance smoothing caps from a image that I saw.

 

[Zapper]

We were very surprised one day to hear, very faintly, a radio station playing through one of the two speakers! Nothing was turned on. The radio station was WFBR-AM in Baltimore. Their transmitter and antenna array in those days was not too far from our house (about 2 miles as the crow flies, in the so-called "Middle Branch" of the Patapsco River). My father's theory was that the long cable plus voice coil was acting as the antenna and resonant circuit, and (perhaps) a 'defect' in the voice coil winding was acting as the detctor (i.e., extracting the amplitude modulated signal from the carrier at (then) 1300 kHz. The driver's high sensitivity transduced the detected signal to produce audible program. Sort of an accidental implementation of a so-called "foxhole radio" receiver.

Cool. That's empirical evidence for the idea that second order nonlinearities in the speaker can demodulate AM. I think something similar is happening here.

 

[Zapper]

It has 6 high value capacitance smoothing caps from a image that I saw.

So no. Not a SMPS. And @Doodski , didn't you find a schematic of the power supply, a standard transformer and bridge rectifier?

 

[Zapper]

I have a funny feeling that a second amp will manifest the same symptom...

Well, I hope not, because this is really unacceptable behavior. I trust McIntosh has competent engineers who wouldn't do this on purpose.

 

[LesterNZ]

Earlier on there was an OP mention of an "older" 80 years, or so, house. Weren't the 60Hz standards for phase (active Volts) and neutral ('ground -return') or whatever they called them pretty questionable? As in: "if it hums, reverse the power plug at the wall", or "turn the plug 180 degs" etc? OLD wiring, as it is, trusted - OK?

Point being: BEFORE the nice components, what's happening from the power transformer at the street and into and through the house itself to create all this EMI - RFI commotion?