Can a faulty amplifier blow a tweeter or a woofer just by being turned on?

[Veganfloyd]

Hi all,

I'm a long time lurker here. Many thanks to Amir and others running this place and other contributors to audio science.

I recently bought a class D amplifier from one of the highly-regarded companies in this forum for a desktop setup. I will not disclose the make and model for now as I'm not sure if the amplifier is at fault, so I would rather avoid bad publicity.

I went ahead with Dali Menuets as my first choice of speakers, but it was immediately apparent that they were damaged with soundstage skewed to the left. Playing test tones with known frequencies confirmed that the tweeter in the right speakers was dead. In the process of figuring out what was wrong, I switched the channels between the speakers multiple times and played some mono test songs with Dali Mentor 1s using the amplifier in question. I assumed that the pre-owned Dali Menuets were sent to me damaged and I sent them back for a refund.

While keeping an eye out for a good deal on a second pair of Menuets, I came across a good deal for used Dali Rubicon 2s which I went ahead with. I used them for a couple of hours in total across short listening sessions. No unpleasant surprises there, but I realised that they were too big for my listening distance and I had to keep them too close to the rear wall, resulting in a boomy bass.

I went ahead with another pair of pre-owned Dali Menuets, only to find out that the woofer in the right speaker was blown. I did not notice this during my first 1-2 minutes listen, which I carried out only for a quick check to see if the bass boom was gone. Having started my proper listening session a few hours later, I quickly noticed that the sound was asymmetrical, though not as much as previously with the blown tweeter. Playing test tones with ascending frequency again, I realised that the right speaker was not producing any sound up to the crossover.

So, I am either extremely unlucky with buying used Dali Menuets or the amplifier keeps damaging them merely by being turned on. Can you please advise if the latter is even hypothetically possible for an otherwise seemingly well-functioning class D amplifier? Just in case, the amplifier turns on with a quiet 'pop' from the speakers, possibly louder on the right side, and louder with the Menuets than the Rubicons. That said, I get a much louder pop from my Nord One MP NC502 via Dali Mentor 6s in my main rig.

Factors against the scenario that the amplifier is faulty include no damage to other speaker models after brief or moderate use, and no damage to the second Menuet from the first pair when switching them back and forth between channels for testing. However, I am worried by the fact that I did not identify the imbalance caused by the blown woofer immediately on the first short listen, so I cannot rule out that it briefly played fine before getting blown whilst in my posession. The memory of these initial impressions is too vague for me to make an educated guess either way. I would appreciate any advice on whether the amplifier could be at fault and how it could be tested for safety, without which I would feel uneasy about hooking it up with any speakers in my posession.

 

[MaxwellsEq]

Wow, this sounds like an unfortunate sequence of events.

In theory, yes, an amplifier fault could damage a speaker at the point it is powered on. For example: if it generated a large DC offset; if it generated a very powerful impulse on switch on; if it generated a very powerful ultrasonic oscillation when switched on. However, many of these would be noticeable if you were in the room. The ultrasonic burst would be above your hearing and the DC offset would be below your hearing, but the latter would be visibly obvious because the cones would be sucked in or pushed out.

What source are you feeding the amplifier with? Could it be generating a random DC offset?

The fact that your issues only seem appear occasionally and when trialling used speakers suggests to me that you might just have been very unlucky. There's also a possibility that the Dali speakers have a reliability issue.

 

[Veganfloyd]

Thank you for taking the time to respond. The amplifier was connected to my laptop via the old-style mechanical Thinkpad docking station using a USB to USB-C cable. There were no extraordinary sonic events prior to noticing the blown tweeter/woofer. The popping sound that I described was there, but as far as I understand this is a feature of class D amplifiers.

 

[somebodyelse]

That sounds a bit like a friend who blew 3 speakers when an output transistor on a bridged bass guitar amp failed. Each one apparently seemed ok for a few minutes before they died. An output transistor failed short, so one speaker terminal was at rail voltage. The other terminal still had audio output, so the result was audio signal with a big DC offset. I would expect this to have a major pop at turn on though. Pops on power on are NOT normal for any of the Class D amps I've used.

If you have a multimeter you can use the DC voltage range to check for DC on the amp output when it's not connected to any speakers.

 

[EdW]

Thank you for taking the time to respond. The amplifier was connected to my laptop via the old-style mechanical Thinkpad docking station using a USB to USB-C cable. There were no extraordinary sonic events prior to noticing the blown tweeter/woofer. The popping sound that I described was there, but as far as I understand this is a feature of class D amplifiers.

A correctly working class D amplifier should produce no pops (or almost inaudible pops) at switch on or off.

 

[Sokel]

Please get a multimeter and measure it on DC scale.
Even a cheap one will do.

 

[MaxwellsEq]

The popping sound that I described was there, but as far as I understand this is a feature of class D amplifiers.

That may be a problem. I have a Class D amplifier which is almost silent when switching on or off - there's an almost inaudible tick. My previous Class A/B had soft thud on switch on and a slightly louder thud on switch off, but both were louder than my current Purifi.

 

[Veganfloyd]

Thank you all for the advice, including, surprisingly, a fellow Cantabrigian. I will get this multimeter and report later today: https://www.screwfix.com/p/lap-dc-digital-multimeter-600v/793rt#product_additional_details_container Hopefully, the manual will make it clear which settings to use as this will be my first time using a multimeter. Otherwise, can you please let me know which voltage/current/resistance selection would work best for testing the amp?

 

[Sokel]

Thank you all for the advice, including, surprisingly, a fellow Cantabrigian. I will get this multimeter and report later today: https://www.screwfix.com/p/lap-dc-digital-multimeter-600v/793rt#product_additional_details_container Hopefully, the manual will make it clear which settings to use as this will be my first time using a multimeter. Otherwise, can you please let me know which voltage/current/resistance selection would work best for testing the amp?

You will put the pointer dial to the yellow area I marked (start with the lowest one 200 mV and go up if it's not enough)

Then place each of the two probes on each amp speaker binding post (one on + and one on - ) WITHOUT ONE TOUCHING THE OTHER and see what the screen shows.
Then repeat with the other channel.
If you don't understand what it says just take a picture and post it here.


Edit:
Everyone should do that with any amp,no matter cheap or expensive straight after unboxing it,before plugging anything to it.
You never know if it's faulty,if some damage occurred during transport,etc.

 

[DVDdoug]

So short answer, yes, it's possible.

It could be a short-term high-power "pop" or DC could blow the woofer. It would be less likely to blow the tweeter. If you can see the woofer, you'd likely see it move in or out. (Some slight, slow movement, isn't always dangerous.)

Constant DC will also move the woofer in or out, and hold it there. And, you'll hear a click or pop when the DC is connected or disconnected. A 1.5V battery is enough to demonstrate that, and it's less than 1W so the battery won't hurt anything.

A preamp or other source can also generate a loud-dangerous pop if you turn-on the power amp first or if you turn them both on at the same time. (Some power amps have delay to let everything stabilize.)

An ultrasonic oscillation could burn out the tweeter at a frequency you can't hear. I once built a preamp that turned-out to be an RF oscillator. One channel of my power amp burned-out before I realized anything was wrong. The tweeters survived, possibly because the frequency was so high that the inductance of the tweeters limited the current.

 

[Veganfloyd]

You will put the pointer dial to the yellow area I marked (start with the lowest one 200 mV and go up if it's not enough)

View attachment 365802


Then place each of the two probes on each amp speaker binding post (one on + and one on - ) WITHOUT ONE TOUCHING THE OTHER and see what the screen shows.
Then repeat with the other channel.
If you don't understand what it says just take a picture and post it here.


Edit:
Everyone should do that with any amp,no matter cheap or expensive straight after unboxing it,before plugging anything to it.
You never know if it's faulty,if some damage occurred during transport,etc.

Thank you very much. I tested about 10 power on-off cycles in each channel. There was one very brief read of 174.8 mV in the left channel upon turning on. I'm glad that I was recording as it was too fast to read in real time and I had to replay the video slowed-down to 50% to catch it. Other than that, both channels kicked off quite variably in the range of 5-64 mV. The right channel tended to hover more consistently over the 20-30 mV range. Both went down within 10-15 seconds to ~2.4-2.8 mV in the left channel and 1.3 mV in the right.

EDIT: Turning the amp off typically resulted in spikes up to about 5-10 mV, but there were few 50ish readings too.

How does that sound?

 

[somebodyelse]

None of that would be enough to cause speaker damage - it would be a fraction of a watt. The failed output transistor I mentioned would give many volts of offset.

 

[Joe Smith]

This is weird, for sure. But having two sets of speakers affected...seems like an unlikely coincidence.

 

[Sokel]

Thank you very much. I tested about 10 power on-off cycles in each channel. There was one very brief read of 174.8 mV in the left channel upon turning on. I'm glad that I was recording as it was too fast to read in real time and I had to replay the video slowed-down to 50% to catch it. Other than that, both channels kicked off quite variably in the range of 5-64 mV. The right channel tended to hover more consistently over the 20-30 mV range. Both went down within 10-15 seconds to ~2.4-2.8 mV in the left channel and 1.3 mV in the right.

EDIT: Turning the amp off typically resulted in spikes up to about 5-10 mV, but there were few 50ish readings too.

How does that sound?

Nope,that's not enough to cause so much damage.
Is that the amp alone,with nothing connected?

 

[EdW]

Thank you very much. I tested about 10 power on-off cycles in each channel. There was one very brief read of 174.8 mV in the left channel upon turning on. I'm glad that I was recording as it was too fast to read in real time and I had to replay the video slowed-down to 50% to catch it. Other than that, both channels kicked off quite variably in the range of 5-64 mV. The right channel tended to hover more consistently over the 20-30 mV range. Both went down within 10-15 seconds to ~2.4-2.8 mV in the left channel and 1.3 mV in the right.

EDIT: Turning the amp off typically resulted in spikes up to about 5-10 mV, but there were few 50ish readings too.

How does that sound?

An important aspect is that the amplifier settling to just a few mV dc offset after a very short time. This is what I’d expect. The fluctuations before settling suggests to me capacitors charging (if there is ac coupling in the signal path) or the positive and negative power rails, as they come up from zero, causing temporary imbalances in the amplifier with the momentarily out of spec rail voltages.

 

[Veganfloyd]

Nope,that's not enough to cause so much damage.
Is that the amp alone,with nothing connected?

Thank you. Yes, this was the amp alone with nothing else connected. But after writing the post it struck me that I should check if connecting the USB cable to my PC would make any difference. So I checked this and it did not. Other than the USB cable, I have nothing else connected in this setup.

So, can the amp be exonerated beyond reasonable doubt?

 

[Veganfloyd]

An important aspect is that the amplifier settling to just a few mV dc offset after a very short time. This is what I’d expect. The fluctuations before settling suggests to me capacitors charging (if there is ac coupling in the signal path) or the positive and negative power rails, as they come up from zero, causing temporary imbalances in the amplifier with the momentarily out of spec rail voltages.

Thank you for the follow-up. Apologies, as I know next to nothing about the design of amplifiers - how should I interpret the final sentence in layman's terms? Is it a technical explanation for why the fluctuations occur? Or is there a qualitative judgment attached to it with implications for how likely/unlikely the amplifier is to have tweeter and woofer blowing capabilities?

 

[EdW]

Thank you for the follow-up. Apologies, as I know next to nothing about the design of amplifiers - how should I interpret the final sentence in layman's terms? Is it a technical explanation for why the fluctuations occur? Or is there a qualitative judgment attached to it with implications for how likely/unlikely the amplifier is to have tweeter and woofer blowing capabilities?

So a dc offset of several volts (10 or more) could damage a woofer type loudspeaker if left there for longish time (i.e. not startup pops!). Most likely the tweeter is ac coupled with capacitors in the crossover so that should be unaffected by any dc although tweeters are generally more easy to damage under other circumstances. Really the only things which are likely to damage a tweeter is ultrasonic instability in the amplifier or severe overdrive of the amplifier into hard clipping.
A few 100’s of mV dc offset will not damage a loudspeaker but will move the cone of the woofer speaker away from its central position - you might notice the cone move in or out, and remain there, whilst the amp is powered and return when de-powered if the offset were this large. This de-centring is undesirable since it means the speaker is forced to operate away from its most linear condition and so the speaker might sound more distorted.

 

[MaxwellsEq]

You can never completely exonerate the amplifier, but you haven't seen it do anything sufficiently naughty in the DC world to blame it for your speaker issues.

The next obvious question is does it run hot either after it's been on for a while but not playing music; or, if it's been playing music for a while. For example, too hot often means you can't hold your hand on it for more than a few seconds. Very cool means you know it's on, but only just.

 

[Veganfloyd]

So a dc offset of several volts (10 or more) could damage a woofer type loudspeaker if left there for longish time (i.e. not startup pops!). Most likely the tweeter is ac coupled with capacitors in the crossover so that should be unaffected by any dc although tweeters are generally more easy to damage under other circumstances. Really the only things which are likely to damage a tweeter is ultrasonic instability in the amplifier or severe overdrive of the amplifier into hard clipping.
A few 100’s of mV dc offset will not damage a loudspeaker but will move the cone of the woofer speaker away from its central position - you might notice the cone move in or out, and remain there, whilst the amp is powered and return when de-powered if the offset were this large. This de-centring is undesirable since it means the speaker is forced to operate away from its most linear condition and so the speaker might sound more distorted.

Cheers for that! This is really good to know that the mechanisms of tweeter and woofer damage are very different. Seems the more so unlikely that the amp could randomly choose two different victims on the two occasions.