MacBook Pro speakers physically damaged by playing sounds through them

[Zbig]

So, yeah, it's apparently possible to physically damage MacBook Pro's speakers by making them play noise I'm not exactly the greatest fan of Adobe and their software quality but FFS, one could reasonably assume no audio signal, however random, should be able to physically damage any part of a fully-integrated audio reproduction system. What's next, "Playing dubstep on this device will invalidate warranty" disclaimers?

https://www.dpreview.com/news/2610969677/some-macbook-pro-owners-report-speaker-damage-after-experiencing-adobe-premiere-pro-audio-bug

[m98]

Isn't that a product liability issue on Adobe's side? It's also not too unusual that you can mess up speakers and preamps by playing the wrong waveform at a too high volume.
Drop the speaker thing, its scary that this bug will also play that sound at maximum volume through headphones, which can cause serious injuries.

[bsudbrink]

If you have not already, you may be interested in reading "Godel, Escher, Bach - An Eternal Golden Braid" by Douglas R. Hofstadter.  The full text is available online here:

https://archive.org/stream/GEBen_201404/GEBen_djvu.txt

Hear is an excerpt that starts a discussion somewhat relevant to this thread:

Tortoise: A type of music which you are most unlikely to have heard of. call it "music to break phonographs by".

[amyk]

It appears Apple is not the only ones cheaping out with DC-blocking capacitors or otherwise underrated components...

https://www.dell.com/community/Laptops-General-Read-Only/VLC-Media-Player-WARNING/m-p/4082242

https://marcin.juszkiewicz.com.pl/2012/12/10/how-to-fry-speakers-in-your-chromebook/

https://bugs.launchpad.net/ubuntu/+source/linux/+bug/522982

[ciccio]

My humble opinion, after 40 years of designing, building and servicing audio equipment, is that there is a serious bug (or deliberate stinginess) in the design of the laptop audio system.
There is no reason for the integrated speakers to fail when driven by the integrated amplifier, regardless of the software used.
It i not Adobe's fault, nor user's fault. It is Apple's fault.

Best regards

[Zbig]

My humble opinion, after 40 years of designing, building and servicing audio equipment, is that there is a serious bug (or deliberate stinginess) in the design of the laptop audio system.
There is no reason for the integrated speakers to fail when driven by the integrated amplifier, regardless of the software used.
It i not Adobe's fault, nor user's fault. It is Apple's fault.

Best regards

This. Just imagine the laptop display breaking because a faulty piece of software displayed garbage image on it.

[glarsson]

Just imagine the laptop monitor breaking because a faulty piece of software displayed garbage image on it.

That happened to the IBM PC junior.

[bsudbrink]

Just imagine the laptop monitor breaking because a faulty piece of software displayed garbage image on it.

That happened to the IBM PC junior.

And the original IBM PC could be programmed to send signals to the monitor that would "burn it out" after a minute or two.

[T3sl4co1l]

Not as odd as it sounds.  Audio systems have been underrated for decades.  Anything rated "W RMS" is already trying to pull one over on you.  (I already forget if that's the peak wattage figure, or the best-case maximum figure (which would be a square wave output), or what.)  Amplifiers are rated more than they'll actually deliver under all combinations of conditions (e.g., producing a sine wave with < 0.001% THD or whatever that spec happens to be, while also delivering rated "W RMS"), and so are speakers (the rating usually assumes a modest duty cycle, due to the crest factor of typical music).  To say nothing of "PMPO", which is just a completely meaningless label.

Playing low crest-factor sounds, like CW sines or squares, or spectrally rich sounds like white noise, will easily exceed the ratings of systems designed for lower duty cycle material.

Is that good design?  Probably not.  Is everyone doing it?  Maybe.

Tim

[BrianHG]

It appears Apple is not the only ones cheaping out with DC-blocking capacitors or otherwise underrated components...

https://www.dell.com/community/Laptops-General-Read-Only/VLC-Media-Player-WARNING/m-p/4082242

This dell one is BULLSHIT...  I've seen modern POP music recordings with harder clipping naturally recorded that way compared to VLC's amplified maximum volume gain.

If VLC is a problem, as an example, just play in a normal music player Daft Punk's RAM album's track Loose Yourself to Dance or any heavy metal at full volume, and you'll destroy those speakers.

[Blocco]

Not as odd as it sounds.  Audio systems have been underrated for decades.  Anything rated "W RMS" is already trying to pull one over on you.  (I already forget if that's the peak wattage figure, or the best-case maximum figure (which would be a square wave output), or what.)  Amplifiers are rated more than they'll actually deliver under all combinations of conditions (e.g., producing a sine wave with < 0.001% THD or whatever that spec happens to be, while also delivering rated "W RMS"), and so are speakers (the rating usually assumes a modest duty cycle, due to the crest factor of typical music).  To say nothing of "PMPO", which is just a completely meaningless label.

Playing low crest-factor sounds, like CW sines or squares, or spectrally rich sounds like white noise, will easily exceed the ratings of systems designed for lower duty cycle material.

Is that good design?  Probably not.  Is everyone doing it?  Maybe.

Tim

Well, yes it is good design, since most people buy an audio system to listen to music and not sine waves at full power.

[bd139]

I can believe this and it's Apple's fault. The MBP and MBA speakers are seriously loud. Room filling loud. I reckon they're pushed way close to the line.

$600 repair bill is bollocks though. Should be $80 or so. The speakers come right out when you crack the case open. Takes about 2 minutes.

[Ice-Tea]

Just imagine the laptop display breaking because a faulty piece of software displayed garbage image on it.

If memory serves, you could do just that in a not-so-distant past: you could latch your screen by applying nonsense timings.

[BillyD]

Many years ago I wrote some software to print graphics on a Star dot matrix printer for use by machine operators on the factory floor.
After a period of months in which several printers were destroyed, we discovered that a particular layout of dots would cause the print head to try to move too far to one side where it jammed and would eventually burn out the the stepper motor.
We analyzed the print data with a fine tooth comb but concluded it was correct, and in fact it wasn't even possible to officially command the print head out of range like this, so it must have been a firmware bug.
Though that didn't stop me getting hauled over the coals a few times by the boss.
Pity because those printers were almost bulletproof besides.

[rrinker]

 TRS-80 Color Computer was one of the first to have a method of physically damaging hardware via software - there was a POKE you could do in BASIC on those things that doubled the clock speed. Perfect. Up until it was running a while and the CPU fried.

[Zbig]

Well, yes it is good design, since most people buy an audio system to listen to music and not sine waves at full power.

All kinds of music or should I be careful not to go too hardcore? Is there an Apple-curated list of approved music I'm allowed to listen on a MacBook so I won't strain this precious good design? What about action movies with loud explosions? Or sci-fi? I heard aliens making some pretty sick noises in some of these. Also, sine waves are pretty easy on the amps and speakers - it's the square waves and other harmonics-rich content that's hard so you kind of failed trying to make your point. And no, it doesn't matter what most people listen to on their Macbooks: if you designed a complete, integrated system and sold it to me for a big bucks, I'd better be able to play whatever I damn well please on it - whether it's Beethoven's 5th or white noise with jackhammer on top. If you let your amplifier destroy your speakers in the closed system of your design, you've failed.

[amyk]

And no, it doesn't matter what most people listen to on their Macbooks: if you designed a complete, integrated system and sold it to me for a big bucks, I'd better be able to play whatever I damn well please on it - whether it's Beethoven's 5th or white noise with jackhammer on top. If you let your amplifier destroy your speakers in the closed system of your design, you've failed.

Put another way, if the system is designed well then the ears of the listener should be the weak point...

[Blocco]

Well, yes it is good design, since most people buy an audio system to listen to music and not sine waves at full power.

All kinds of music or should I be careful not to go too hardcore? Is there an Apple-curated list of approved music I'm allowed to listen on a MacBook so I won't strain this precious good design? What about action movies with loud explosions? Or sci-fi? I heard aliens making some pretty sick noises in some of these. Also, sine waves are pretty easy on the amps and speakers - it's the square waves and other harmonics-rich content that's hard so you kind of failed trying to make your point. And no, it doesn't matter what most people listen to on their Macbooks: if you designed a complete, integrated system and sold it to me for a big bucks, I'd better be able to play whatever I damn well please on it - whether it's Beethoven's 5th or white noise with jackhammer on top. If you let your amplifier destroy your speakers in the closed system of your design, you've failed.

First, note that I was not replying to comments regarding the Macbook Pro but to a more general statement about consumer audio equipment.

Products are designed to meet the requirements of their intended use and this is a good thing. If everything was designed to operate at 100% of its capability 100% of the time then very few people would be able to afford anything. A production road car would soon begin to fail if it were used for racing, does this mean that it's a bad design?

Incidentally, sine waves are not "easy on amps" since they keep the output devices in their linear region where they dissipate the most heat. Can you point me to a loudspeaker that can withstand a sine wave at full power continuously?

[Zbig]

First, note that I was not replying to comments regarding the Macbook Pro but to a more general statement about consumer audio equipment.

Fair enough, although the topic I started is about a specific case of a specific laptop. I kept reiterating my opinion that a failure mode like this is unacceptable in an integrated design because I fully agree it's a whole different story when you buy separate pieces of equipment and undersize your speakers - in that case you only have yourself to blame.

Products are designed to meet the requirements of their intended use and this is a good thing. If everything was designed to operate at 100% of its capability 100% of the time then very few people would be able to afford anything. A production road car would soon begin to fail if it were used for racing, does this mean that it's a bad design?

Oh come on, a car analogy, really? Car analogies rarely work with anything other than cars yet people still love using them so much. Please, no.

Incidentally, sine waves are not "easy on amps" since they keep the output devices in their linear region where they dissipate the most heat.

AFAIK, most of the laptopey, portable, highly integrated consumer type of devices use class D amps almost exclusively nowadays, so not much of a linear region to speak of. I say speaker diaphragm slamming hard back and forth between its two extremes is much worse for the speaker than a simple sine wave.

Can you point me to a loudspeaker that can withstand a sine wave at full power continuously?

Not really as I'm not going to sit in a room with my speakers at full blast just to prove you something. But isn't that exactly how a speaker power handling figure is defined? I.e. speaker being able to dissipate the power specified while being driven by a sine wave signal indefinitely? Can you point me to a speaker (of reputable brand) that got damaged after it was driven with a sine wave signal within its specified power?

[dmills]

The real screwup here is not using an amp capable of way more power then the speakers can handle steady state, that is actually an entirely reasonable, rational design decision, and one I sometimes make.

Music has a significant peak to average ratio (In pro audio we usually figure 8:1 when sizing things like power feeds), and fitting an amp that can manage say 10dB more then the speakers cont. power rating is perfectly sane and useful as it lets you play real program audio 10dB louder then you otherwise could.

The cockup was in not configuring the amp (or the driver, but many class D amp chips can do it internally) to limit average power measured over some suitable timeconstant to a level that will not damage the voice coil (And the power at low frequency to avoid exceeding Xmax), do it right and this limit is vanishingly unlikely to ever be hit in the absence of a bug, and you still get that 10dB of louder.

Regards, Dan.

[Kleinstein]

It is rather common that speakers can be damaged by to much power of an extended time. A 100 W speaker usually can not stand those rated 100 W for a longer time, but will likely burn out, from thermal overload. The speaker nominal power is usually defined by the point were it reaches to mechanical limits or produces way to much distortion.

Ideally there would be some kind of protection and usually this would be way to much noise to stand. However with very low or high frequency one may not near the overload anymore. In normal audio systems a protection is not very common - one sees burned speakers more often.
An amplifiers starting to oscillate at not too high a frequency (so more likely in older designs) could easily destroy the tweeter - these are often the weakest ones anyway. So the maximum power is a function of frequency.

Ideally the computer should have some protection. But as it is relatively complicated it may be realized in software and may thus fail or be ignored by direct HW access.

[Yansi]

It is rather common that speakers can be damaged by to much power of an extended time. A 100 W speaker usually can not stand those rated 100 W for a longer time, but will likely burn out, from thermal overload. The speaker nominal power is usually defined by the point were it reaches to mechanical limits or produces way to much distortion.

Not true.

Nominal power of speakers (at least those professional ones) is indeed the one that can be applied continuously; typically specified as a continuous pink noise signal within the frequency range* of the speaker. Power is calculated on rated nominal impedance. Loudspeaker in free air.

But, speaker power is NOT defined as from mechanical limits! The cone excursion is highly dependent on the acoustical load the  speaker is working with (i.e. the design of the enclosure).

The excursion is dependent on the inverse of frequency squared.   Half the frequency means typically four times the excursion. (However, this can be heavily affected by the acoustic circuit the speaker is loaded with, ie., the cone excursion can become a very complex function. That's where the simulators come in handy.

Exceeding the Xmax does NOT damage a speaker. Xmax is the maximum linear excursion, defined by the voice coil height and magnetic gap height. What definitely will damage a speaker, is going above the Xdmg excursion - that one is however typically a couple times larger than Xmax.

*typicaly a range defined by a multiple (10x) of the resonant frequency of the speaker.

[dmills]

Lost a few drivers in horn loaded subs over the years when some clown (Who me?) forgot to change the HPF setting on the LMS when going from an 12 box to a 4 box array. The low frequency cutoff in horn speakers is strongly a function of mouth area, and effective mouth area scales with the number of boxes in the array.

Get it wrong and the acoustic impedance goes very low once below horn resonance and you end up shredding the cones, which is annoying given how much of a pain they can be to get to, never mind the expense of the recone kits.

Not that the failures seen in big PA should be any kind of issue in pony computer audio domestic tat.

Regards, Dan.

[Blocco]

First, note that I was not replying to comments regarding the Macbook Pro but to a more general statement about consumer audio equipment.

Fair enough, although the topic I started is about a specific case of a specific laptop. I kept reiterating my opinion that a failure mode like this is unacceptable in an integrated design because I fully agree it's a whole different story when you buy separate pieces of equipment and undersize your speakers - in that case you only have yourself to blame.

Products are designed to meet the requirements of their intended use and this is a good thing. If everything was designed to operate at 100% of its capability 100% of the time then very few people would be able to afford anything. A production road car would soon begin to fail if it were used for racing, does this mean that it's a bad design?

Oh come on, a car analogy, really? Car analogies rarely work with anything other than cars yet people still love using them so much. Please, no.

Incidentally, sine waves are not "easy on amps" since they keep the output devices in their linear region where they dissipate the most heat.

AFAIK, most of the laptopey, portable, highly integrated consumer type of devices use class D amps almost exclusively nowadays, so not much of a linear region to speak of. I say speaker diaphragm slamming hard back and forth between its two extremes is much worse for the speaker than a simple sine wave.

Can you point me to a loudspeaker that can withstand a sine wave at full power continuously?

Not really as I'm not going to sit in a room with my speakers at full blast just to prove you something. But isn't that exactly how a speaker power handling figure is defined? I.e. speaker being able to dissipate the power specified while being driven by a sine wave signal indefinitely? Can you point me to a speaker (of reputable brand) that got damaged after it was driven with a sine wave signal within its specified power?

The only point, the only point!!! I was making is that "good design" takes into account the intended usage of a device. This is why we have consumer/light-duty/short duty-cycle products and commercial/ heavy-duty, continuous duty-cycle products, they both do essentially the same thing but the commercial product is typically larger, heavier and more expensive. I hope that's clear.

[IanMacdonald]

This has been a common problem for many years where underpowered audio equipment is used at parties, discos, clubs etc. If the amp is driven into distortion to get loud enough sound, it generates an excess of treble, and this burns out the tweeters. It's the square wave, rich in harmonics, that does it.