Probably a silly question - Does an aluminum speaker cabinet attract more noise?

[cjf]

Hello, I wonder what folks thoughts are on this topic.

Is it possible for an aluminum cabinet to attract EMI/RFI/EMF noise compared to say a wood or other non conductive enclosure material? If so, is there such a thing as someone "grounding" a speaker cabinet to earth to eliminate such things or would that not do anything at all for such an issue?

I've recently aquired new S3 towers and immediatly after their installation I noticed some HF interference from the tweeter/mid driver that I've never encountered before in my room with the same equipment and my previous wooden speakers.

I've narrowed down that the issue is not related to the Input signal as I can remove the XLR connectors from the amps and still hear the same noise. I've also eliminated the speaker cables and power cords as being possible culprits as well by switching those out for different ones with no change either.

Lastly, I believe I've eliminated the amp as being a culprit by swapping in a different set of amps and the noise still occurred. But, I should also mention that even though I've swapped amps both amps are of the same type/flavor (Class D Hypex) but both are known for their very low noise floor and neither pair of amps made this noise hooked up to previous speakers (B&W 803D2).

I've captured a sound clip of the noise which comes from both speakers mainly during the first 10mins of initial power up of the amps afterwhich it dies down significantly. After 10mins or so the only remaining noise I hear is a very high pitched (16khz or so) constant whistle when my ear is right up on the tweeters. During the first 10mins the noise I hear is loud enough to hear across the room and that sound is in the 8khz range of the audio spectrum (according to an APP used on my I-Phone).

The things I find strange here are that the noise comes from both speakers which are powered by a set of mono blocks so it's very unlikely to be related to any one amp having some kind of issue.

Here's the kicker and the main reason for my question in this post. Today while probing around the listening area with my trusty AC/Current "No Touch" sensor to look for possible sources of the noise I looked up and noticed my ceiling fan staring at me. The fan is probably 3-4ft away from the top of the right speakers cabinet. As I raised the voltage sensor up towards the ceiling fan it started going wild with alerts. Now bare in mind that my hand was still at least two feet away from the fan itself when the device started alerting me of voltage being present. I should also mention that the light and fan were NOT turned on.

One last point to mention here and that is my house is very old (120yrs) and still has some old school knob and tube wiring in place. I suspect that this ceiling fan may not be grounded and is probably wired up using the old wiring.

If there is any interest I can post up the sound clip I captured of the noise I hear but I'm curious to hear people's thoughts on this. Could my beautiful aluminum speaker cabinets be picking up some voltage spray from this fan or other source I've yet to identify due to their build material?

Thanks for your thoughts on this topic

 

[DonH50]

I'd get someone to check the amp's outputs with a wideband oscilloscope ASAP. I doubt the cabinets are doing anything, but the load of the new speakers may be making the amps unstable.

 

[Folsom]

Unstable? Not likely... unless they're broken.

It's possible for the cabinet to capacitively couple to the wiring. But grounding the speaker will only work if the path is much lower impedance. On top of that we don't know that it's not a bad connection inside one speaker that's affecting the feedback in the amp, giving an amplified signal back to the speaker. Your speakers are less efficient so you're not hearing something that was already there.

What happens if you touch the cabinet?

If you can't reach your equipment, listen to the tweeter and have someone touch your amps and anything they are plugged into.

The answer to which of those makes a difference can tell use where capacitive coupling is occurring, or if it is something else.

However my experience with old wiring like that is that you'll never get things correct. It's not possible. If you can afford S3 speakers you can afford an electrician to run a separate new line to the area.

Also the most obvious question, what does it sound like when the fan is off? Fans can actually play RF to be audible. It sounds nuts but in high RF area you can't hear radio stations in fans if you got good hearing. It's very subtle, and no one I know besides myself has been able to hear it, but it's 100% consistent and not depending on which fan! Plus they're playing the local radio stations...

 

[DonH50]

I have seen some very good amps go unstable to the point of destruction when the load impedance hits a "bad" point. I am sure a few on here remember the Infinity class D amp demo and Cobra Cable fiasco at that long-ago Chicago CES. I was there for that one, ugh. And there have been plenty since.

If the cabinets are coupled to the input jacks then they could act like an antenna that would send signals back to the amplifier. In that case it implies the output stage of the amplifier is susceptible to EMI/RFI. You could try getting some of those big ferrite cores and wrapping the speaker wires through them to see if that makes the noise go away. It's also possible the fan is coupling EMI/RFI someplace else in the system and you either didn't notice before or the new speakers are emphasizing the noise.

You could try grounding the fan, speaker cabinets, and/or the rest of your system, but it is best to have an electrician do that. Do you even have a good safety ground? Most tube and knob wiring jobs do not include a ground line. That may be the root cause of everything, lack of a decent system ground...

 

[cjf]

Unstable? Not likely... unless they're broken.

It's possible for the cabinet to capacitively couple to the wiring. But grounding the speaker will only work if the path is much lower impedance. On top of that we don't know that it's not a bad connection inside one speaker that's affecting the feedback in the amp, giving an amplified signal back to the speaker. Your speakers are less efficient so you're not hearing something that was already there.

What happens if you touch the cabinet?

If you can't reach your equipment, listen to the tweeter and have someone touch your amps and anything they are plugged into.

The answer to which of those makes a difference can tell use where capacitive coupling is occurring, or if it is something else.

However my experience with old wiring like that is that you'll never get things correct. It's not possible. If you can afford S3 speakers you can afford an electrician to run a separate new line to the area.

Also the most obvious question, what does it sound like when the fan is off? Fans can actually play RF to be audible. It sounds nuts but in high RF area you can't hear radio stations in fans if you got good hearing. It's very subtle, and no one I know besides myself has been able to hear it, but it's 100% consistent and not depending on which fan! Plus they're playing the local radio stations...

Hello and thanks for your response,

The noise occurs in both speakers but is slightly more pronounced in the Right Channel than the Left. Since both channels are affected I would think its less likely to be a bad connection inside the speaker.

If I touch the cabinet or the Amp there is no change to the noise.

The stereo system itself is run from X 4 dedicated, grounded 20Amp circuits so whatever cabling is being used by the ceiling fan is separate and not anywhere nearby.

Also, the fan itself was and is turned off at all times. I don't use it at all. But, despite the power on state of the fan it does still seem to be "Live" at all times based on my test with the probe. Here's the scary part, if I hold the AC/Voltage probe in my left hand with my arm extended towards the floor and the device point down and then simultaneously reach up and touch the fan with my right hand the probe starts to beep

 

[cjf]

I'd get someone to check the amp's outputs with a wideband oscilloscope ASAP. I doubt the cabinets are doing anything, but the load of the new speakers may be making the amps unstable.

Thanks for your reply. Its interesting you mention this because I've recently read a review of the S3 done by Hi-Fi News found here: http://www.absolutesounds.com/pdf/main/press/S3_HFN_NOV2014.pdf

In that article the following was mentioned:

"Magico claims a nominal figure of 4ohm but we measured a minimum modulus of 1.6ohm, more in keeping with a 2ohm nominal rating. Moreover, at high frequecies low modulus is combined with high impedance phase angle, the figures at 20kHz being 1.6ohm and -54deg respectively, sufficeint to drop the EDPR (equivilent peak dissipation resistance) to a scary 0.5ohm. Only at around 3kHz does the EPDR drop below the frequency peak of 1.7ohm at 67Hz"

So assuming what they say is accurate in the above measurements it sounds like the speaker is pretty challenging in the upper Freq range in terms of load. This is interesting considering that the noise I hear and measured with my i-Phone APP shows that the noise occurs at 8kHz (during the first 10min of Amp warmup) and then after it settles the constant high pitched noise present in the tweeter is at around 16kHz. This is with no music playing of course so I dont know what to make of being able to hear this sound without any Input present at the Amp end.

I guess the good news here is that after the first 10mins passes the remaining noise is not loud enough to hear under normal listening conditions or at the listening chair so it doesn't really affect my listening enjoyment of the speaker. But, I am still fascinated by this strange noise and how its entering the system. I do plan to have the ceiling fan removed from the listening room and "capped" off to officially kill it since it does still seem to be spraying all sorts of voltage out of itself despite not being turned on.

 

[amirm]

Can you short out the input to the amp and see if the problem goes away? Just get a cheap RCA cable, cut one end, strip the center wire and wrap to the shield. Just do this with one amp with the other shut off.

 

[amirm]

If I were to guess what is going on, I would say it is a simple case of the speakers being very sensitive in high frequencies and simply playing what is coming out of the amps. It has nothing to do with the speaker cabinet and such. I had a class D amp that did the same thing. You could hear the high pitched buzz and tones from 6 feet or so and that is with my degraded ear .

If so, your solution would be to get an amp that is more quiet. BTW, don't go by the sensitive specs of the speaker. They are measured at a single (usually unspecified) frequency, much lower than what you have measured.

 

[cjf]

I have seen some very good amps go unstable to the point of destruction when the load impedance hits a "bad" point. I am sure a few on here remember the Infinity class D amp demo and Cobra Cable fiasco at that long-ago Chicago CES. I was there for that one, ugh. And there have been plenty since.

If the cabinets are coupled to the input jacks then they could act like an antenna that would send signals back to the amplifier. In that case it implies the output stage of the amplifier is susceptible to EMI/RFI. You could try getting some of those big ferrite cores and wrapping the speaker wires through them to see if that makes the noise go away. It's also possible the fan is coupling EMI/RFI someplace else in the system and you either didn't notice before or the new speakers are emphasizing the noise.

You could try grounding the fan, speaker cabinets, and/or the rest of your system, but it is best to have an electrician do that. Do you even have a good safety ground? Most tube and knob wiring jobs do not include a ground line. That may be the root cause of everything, lack of a decent system ground...

I would love to hear or read more about what occurred during that Demo. Is there somewhere on the web I can find out more?

Would it be considered common for a speakers Input jacks to be coupled to the cabinet? I think I may try playing around with those ferrite cores you mention. Is there any brand or type in particular you can recommend I try? I'm not sure how I would get the speaker wire to wrap around them since the majority of the jacket covering my Cardas Neutral Ref speaker cables is fairly fat except on the very ends where the Spades attach. At the ends, the inner conductors do extend out past the fat jacket into individual conductor runs but I think I would lucky to have enough length there to even get one wrap around anything related to a ferrite core. Maybe one that attaches itself to the wire could work though?

Thanks

 

[cjf]

If I were to guess what is going on, I would say it is a simple case of the speakers being very sensitive in high frequencies and simply playing what is coming out of the amps. It has nothing to do with the speaker cabinet and such. I had a class D amp that did the same thing. You could hear the high pitched buzz and tones from 6 feet or so and that is with my degraded ear .

If so, your solution would be to get an amp that is more quiet. BTW, don't go by the sensitive specs of the speaker. They are measured at a single (usually unspecified) frequency, much lower than what you have measured.

Thanks for your response Amirm,

One thing to mention is that when I put the Amp into standby mode (still powered on from an AC receptacle standpoint but the Output is "Muted") the noise goes away completely. Not sure if this would be an equivalent test to the one you mention with the RCA cable or not? But in my case, I only have XLR Balanced Inputs on the Amp so maybe the test parameters can still be the same but substituting in the use of a balanced cable instead?

In terms of possibly needing to use a more quiet amp, that one leaves me scratching my head a bit. All of the measurement data and Specs I have read about the Hypex modules shows that they are probably considered one of the quietest amps available today in terms of noise floor (somewhere in the -124db down range). So I can't help but wonder how a different amp whose specs are most likely worse would fair under the same conditions. It seems unlikely that this noise is related to the switching frequency of the Class D module but I don't know. I believe I've read something in the past that the switching Freq occurs somewhere in the 48kHz range for the Hypex module which would be well above what I measured.

Is it possible that another amp can have more rejection in the upper Freq range to EMI/RFI but still have worst measurements and specs everywhere else?

Thanks for any additional info you can provide

 

[amirm]

Thanks for your response Amirm,

One thing to mention is that when I put the Amp into standby mode (still powered on from an AC receptacle standpoint but the Output is "Muted") the noise goes away completely. Not sure if this would be an equivalent test to the one you mention with the RCA cable or not? But in my case, I only have XLR Balanced Inputs on the Amp so maybe the test parameters can still be the same but substituting in the use of a balanced cable instead?

Yes, you can substitute the same with balanced cable. Just want to make sure the noise is not being picked up at the input of the amp and then getting amplified. Instead of cutting the cable another thing you can do is to short out the connection on the cable with a piece of wire. Assuming your amp has female connections, then the other end of the balanced cable will have the pins. Simply short those out with a screwdriver, blade of a scissor, alligator clip, etc. Do this when the amp is turned off as touching those pins may cause a pretty loud hum through your speakers!

In terms of possibly needing to use a more quiet amp, that one leaves me scratching my head a bit. All of the measurement data and Specs I have read about the Hypex modules shows that they are probably considered one of the quietest amps available today in terms of noise floor (somewhere in the -124db down range). So I can't help but wonder how a different amp whose specs are most likely worse would fair under the same conditions. It seems unlikely that this noise is related to the switching frequency of the Class D module but I don't know. I believe I've read something in the past that the switching Freq occurs somewhere in the 48kHz range for the Hypex module which would be well above what I measured.

The switching frequency is in hundreds of Kilohertz (usually in 300 KHz range) so it is not that you are hearing. The amp I had was based on ICE modules and it too had good specs but here it was, with clearly audible high frequency noise. As Don mentioned, this may be an oscillation issue inside the amp due to poor layout and supporting components that is not normally heard but happens to be the case in your setup. Class D amps are impacted by wiring and speaker loads and so act differently on every system (to some level).

Is it possible that another amp can have more rejection in the upper Freq range to EMI/RFI but still have worst measurements and specs everywhere else?

I don't think it is an EMI/RFI problem. Certainly not one coming from the speaker. Shorting out the input of the amp will take care of anything coming into the amp.

Do you have a cheap mass market amplifier like an Audio/Video receiver that you could test with? If not you can pick one up for $300 these days and may be an important diagnostic tool here. If it doesn't generate the noise, then we know the arrow points at the amp.

 

[Folsom]

Ncores are stable for 2ohm speakers, so 1.6ohm would be expected to occur as a dip for them. But the ohm will be stable at nominal without music playing and you still get the noise.

If they are single ended output it's perfectly fine to ground the cabinets from the negative terminal of the speaker with a wire. However if they are not singled output amplifiers then this will not work. That's a question for the manufacturer.

Your ceiling fan sounds frightful. Maybe it should be looked at for wiring to it and the switch. Perhaps the hot/neutral were swapped and they tied the ground of it to the 'neutral' wire.

But overall I'll point out Ncores have a faint hiss sound when your ear to speaker, and that's normal. Without anything attached one should expect it to be loud. Your source's output impedance will also control how loud the sound is! The higher the output impedance the more noise you'll get.

You could open the amplifiers and disconnect the wires from the inputs, put the case top back on, and then see if you get noise without anything connected since the inputs won't be vulnerable to RF like when your source isn't connected. That would be useful information.

So far I doubt it's the speakers. Capacitive coupling almost always changes when you touch something.

 

[Folsom]

Shorting the inputs is not necessarily a shield for RFI, and may exhibit the same results since there isn't a return path away from the amplifier.

 

[cjf]

Yes, you can substitute the same with balanced cable. Just want to make sure the noise is not being picked up at the input of the amp and then getting amplified. Instead of cutting the cable another thing you can do is to short out the connection on the cable with a piece of wire. Assuming your amp has female connections, then the other end of the balanced cable will have the pins. Simply short those out with a screwdriver, blade of a scissor, alligator clip, etc. Do this when the amp is turned off as touching those pins may cause a pretty loud hum through your speakers!


The switching frequency is in hundreds of Kilohertz (usually in 300 KHz range) so it is not that you are hearing. The amp I had was based on ICE modules and it too had good specs but here it was, with clearly audible high frequency noise. As Don mentioned, this may be an oscillation issue inside the amp due to poor layout and supporting components that is not normally heard but happens to be the case in your setup. Class D amps are impacted by wiring and speaker loads and so act differently on every system (to some level).


I don't think it is an EMI/RFI problem. Certainly not one coming from the speaker. Shorting out the input of the amp will take care of anything coming into the amp.

Do you have a cheap mass market amplifier like an Audio/Video receiver that you could test with? If not you can pick one up for $300 these days and may be an important diagnostic tool here. If it doesn't generate the noise, then we know the arrow points at the amp.

Thanks again for the info.

Is it safe to assume that one of these jumpers would also do the trick for testing the Input on the amp strapped across Pins 1 & 3 assuming Pin 2 is Hot? I'm thinking of this option at least initially before opening up the case (which I cant do for warranty reasons) and in substitute for messing with the XLR cable itself. I guess I could also just make one of these jumpers with a very short length of solid core wire of the appropriate size to fit inside the holes of the female XLR Input?



I may try to lug downstairs my old Adcom GFA 535 II amp and hook it up to the speakers as a test like you mention. My god, its probably a crime somewhere to hook that amp up to these speakers but I suppose I can live with that at least for testing purposes

 

[cjf]

Ncores are stable for 2ohm speakers, so 1.6ohm would be expected to occur as a dip for them. But the ohm will be stable at nominal without music playing and you still get the noise.

If they are single ended output it's perfectly fine to ground the cabinets from the negative terminal of the speaker with a wire. However if they are not singled output amplifiers then this will not work. That's a question for the manufacturer.

Your ceiling fan sounds frightful. Maybe it should be looked at for wiring to it and the switch. Perhaps the hot/neutral were swapped and they tied the ground of it to the 'neutral' wire.

But overall I'll point out Ncores have a faint hiss sound when your ear to speaker, and that's normal. Without anything attached one should expect it to be loud. Your source's output impedance will also control how loud the sound is! The higher the output impedance the more noise you'll get.

You could open the amplifiers and disconnect the wires from the inputs, put the case top back on, and then see if you get noise without anything connected since the inputs won't be vulnerable to RF like when your source isn't connected. That would be useful information.

So far I doubt it's the speakers. Capacitive coupling almost always changes when you touch something.

Yes, I consider myself lucky right now that the remaining high freq noise I hear at the tweeter is very faint and only audible when my ear is basically resting on it. Its the initial noise I hear at bootup of the amp that is the most concerning to me right now. If that noise remained at all times I would be very salty.

Just for kicks, here is the noise I captured coming from the speaker that occurs just after initial bootup of the amps and remains cycling thru at various levels for about 10mins. I broke out my Audacity Cowboy skills to plot the spectrogram of the sound so the various Freq can be seen

[video]https://flic.kr/p/E8SRAn[/video]

 

[amirm]

Thanks again for the info.

Is it safe to assume that one of these jumpers would also do the trick for testing the Input on the amp strapped across Pins 1 & 3 assuming Pin 2 is Hot? I'm thinking of this option at least initially before opening up the case (which I cant do for warranty reasons) and in substitute for messing with the XLR cable itself. I guess I could also just make one of these jumpers with a very short length of solid core wire of the appropriate size to fit inside the holes of the female XLR Input?

View attachment 25779

Sure. Any way you can short them out with a short path would work.

I may try to lug downstairs my old Adcom GFA 535 II amp and hook it up to the speakers as a test like you mention. My god, its probably a crime somewhere to hook that amp up to these speakers but I suppose I can live with that at least for testing purposes

Hehe .

 

[amirm]

Yes, I consider myself lucky right now that the remaining high freq noise I hear at the tweeter is very faint and only audible when my ear is basically resting on it. Its the initial noise I hear at bootup of the amp that is the most concerning to me right now. If that noise remained at all times I would be very salty.

Just for kicks, here is the noise I captured coming from the speaker that occurs just after initial bootup of the amps and remains cycling thru at various levels for about 10mins. I broke out my Audacity Cowboy skills to plot the spectrogram of the sound so the various Freq can be seen

[video]https://flic.kr/p/E8SRAn[/video]

Oh, that sounds like noise induced through power line that the amp is sensitive to. Let's see how the Adcom does.

 

[Folsom]

The startup noise is the only noise that sounds like it is unusual here, atm. I can see why you're concerned... such lovely noise.

 

[cjf]

OK...things are starting to get interesting now.

I've just returned from a few more tests. I lugged my old Adcom downstairs and wired it up with just its power cable and a speaker cable hooked up to my Right channel speaker. No Input cable connected. After crossing my fingers that the Adcom would not blow up my new speakers upon switch on I flipped the switch and listened for the HF noise that was present before.

No HF Noise Now!

I should mention that the Adcom uses a 2 prong power cord so no ground wire.

After the above test I did something else this time just to see the outcome. I hooked back up only one of my Hypex amps (Right Channel Only). I did this to test and see if maybe there was some kind of crosstalk happening between the power cables I use in the X 4 outlet duplex in the wall. The power cables normally sit directly across from each other in this outlet in fairly close proximity to each other but the X4 outlets are actually on two separate circuits each with its own run to the breaker box. In any case, the noise returned with only one plug in use so that can be ruled out.

So here is the other interesting part that I noticed. As I powered on both Hypex amps and while they were going thru their "normal" HF noise routine during startup I heard a "break" in the noise for a brief second. At the same time I noticed that the Refrig in the Kitchen had just kicked on! It seems that during the spike of the Refrig kicking on the noise was stopped from the speaker for a brief second then continued on as it had been.

This is interesting indeed considering all these items are on different circuits with the audio system itself on multiple dedicated circuits (with ground). So this makes me wonder despite these circuits being separate if there is a leak finding its way into my audio circuit perhaps causing the noise I am hearing?

So as was mentioned earlier in this thread by others their is something about my combo of equipment as it sits today compared to how it sat a month ago that is now bothered by this noise that I assume was always present but not audible before. So now the question becomes how do I fix it? Do I delve into the possibility of trying some form of power conditioner that I've never had a need for before OR do I just live with this HF noise that only occurs at bootup of the Amps and then all but disappears after 15min or so?

Power conditioners have long been considered taboo or a bad idea when combined with these Hypex amps. Even ones that say they dont limit current but still end up somehow doing so or at least darkening the sound.

Hrmmm

 

[Folsom]

It's possible you've got bad resonations occurring from being on the same power service. When the fridge turns on it dampens them by swamping whatever is causing the problem. The amps have less of it when they heat up, potentially running hotter than they should. (I can't confirm that, but it is possible).

I'd have to take a closer look at the SMPS power supply, but I could potentially dampen it and see if that helps. My experience with the CMC's like which are for sure used in the SMPS of all Hypex units, is that they can thoroughly be resonated to the point that the amount of current they are suppose to be putting out is greatly reduced. It can be so bad in a CD player the tray won't operate. You can even get audible noise from the CMC.

Have you tested your AC level from those sockets? Another thing is that the voltage spike from fridge may be making the SMPS work correctly. I might be wrong but depending on which SMPS you're using, some of them have PFC. When it turns the DC back into AC it might not be stable due to low voltages.