[TheSonicTruth]

That's why in my car I don't drive, don't turn on the engine and only listen to pure tones, because otherwise I put extra strain on the transducers.
That's a philosodiophile argument IMO. How do you estimate, if it's broad noise, when the sum of music+NC causes extra amplitude and effort, and when it leads to lower amplitude and facilitates the movement? Plus if the movement is to counter air pressure fluctuations from ambient noises, don't we end up with increased fidelity?
It's without a doubt, challenging to get things right in a car and I can think of a few reasons why the sound with NC could feel worst(or really be worst), but I honestly doubt that your hypothesis is the explanation for that.

I heard the intermittent diminished bass myself, in my 2010 Honda Accord with the ANC system. After disconnecting it from the back of the Audio Unit(Honda's term for 'radio' or deck), I no longer experienced intermittent low-end performance from the speakers.

And before you suggest the ANC was 'malfunctioning' in my specific case, I'll suggest that no noise cancellation system is perfect, particularly when it deploys through the same amp & speakers as the entertainment system in a vehicle.

 

[pinnahertz]

Adaptive beam-forming over 20-20kHz, 5 to 6 seats, differing head heights, using existing speaker locations and drivers. Might want to give that one a little more than 30 seconds of thought.

 

[TheSonicTruth]

Adaptive beam-forming over 20-20kHz, 5 to 6 seats, differing head heights, using existing
speaker locations and drivers. Might want to give that one a little more than 30 seconds of thought.

My 'Active Noise Cancellation': Actively adding sound-deadening foam to the doors, firewall, and floor so that the music can be enjoyed at any volume preferred, without electronic noise bogging down the speakers. That's how we rolled.. No fancy whiz-bang circuitry.

 

[GRUMPYOLDGUY]

Even if all of this is considered, and even implemented, it still puts additional load on the main speakers - if they are asked to produce this antinoise in addition to faithfully reproducing their first order of business - your favorite music or other programming. Since most of the ANC-created anti-noise is down in the low-to-lower-mid-range frequency realm, it means speakers, even if a sub is present, are being asked to simultaneously produce that anti-noise and whatever vocal or instrumental information also occupies that bandwidth.

End result: Both lower frequency program info and the appropriate(as determined by the total ANC system) anti-noise signal are compromised(weak reproduction, distortion, etc) to some degree. This is not a possibility, but rather a probability, unless, as I have proposed in automotive circles, the ANC system utilizes a separate amp & set of speakers, perhaps ones with a frequency response encompassing just enough bandwidth to cover efficiently just the range that is typically used by Active Noise Cancellation - from 50 up to 200 Hz, for instance.

It doesn't matter if the combiner is electronic behind a single (set of) speakers or at the ear.... The audio quality is still compromised. Nobody has ever claimed (and if they did, they are wrong) that ANC does not reduce audio fidelity.

 

[GRUMPYOLDGUY]

My 'Active Noise Cancellation': Actively adding sound-deadening foam to the doors, firewall, and floor so that the music can be enjoyed at any volume preferred, without electronic noise bogging down the speakers. That's how we rolled.. No fancy whiz-bang circuitry.

That's not active, it's passive. And it's the same philosophy companies like Shure take over companies like Bose - noise isolation vs noise cancellation. Despite what you may think, you are not covering any new ground here.

 

[GRUMPYOLDGUY]

Adaptive beam-forming over 20-20kHz, 5 to 6 seats, differing head heights, using existing speaker locations and drivers. Might want to give that one a little more than 30 seconds of thought.

Bandwidth is not a concern. There will be negligible squint over such a narrow BW. Yes, thought has to be taken in constructing a desired beam pattern, and some new technology needs to be developed. But I'm not here to solve the world's problems, chief. I'm just thinking about whether or not it is technically feasible.

 

[bigshot]

`

It doesn't matter if the combiner is electronic behind a single (set of) speakers or at the ear.... The audio quality is still compromised. Nobody has ever claimed (and if they did, they are wrong) that ANC does not reduce audio fidelity.

Signal processing is the place where reduced audio fidelity can mean improvements in audio quality. We tend to think they're one and the same, but they aren't necessarily.

 

[pinnahertz]

It doesn't matter if the combiner is electronic behind a single (set of) speakers or at the ear.... The audio quality is still compromised. Nobody has ever claimed (and if they did, they are wrong) that ANC does not reduce audio fidelity.

Um...well, I'm afraid I have to disagree here. I have a pair of the Parrot Zik 3 headphones. They start with respectable isolation, then apply their own ANC. The ANC is quite good, and does not impact fidelity at all. You can turn it on and off, nothing changes except the background noise. I'm not so sure this is true with the Bose product, but I don't own Bose because I can't tolerate the raw sound quality much less the ANC version. The Parrots are a different story. Not without issues, but the app includes a very adept parametric EQ, so you can "fix" the bigger problems. I've done it, it works. The ANC itself has no SQ degenerative impact. Nor, in theory, should it.

Just to keep balance, the Parrots are far from perfect, and their idea of where the EQ is controlled from, is bollox. The Bluetooth vs corded option is also bollox. The touch control is half-baked most of the time. But for SQ, they're my preference for ANC, they are not my preferred headphones for wired non-ANC applications.

 

[Davesrose]

Humm....when it comes to the basis of this thread, I find it pretty absurd (at least any technologies in the horizon). I've always considered the given that a car is not an acoustical environment for the best audio quality. Now better quality speakers and such do have some improvements for a feeling of better bass or more enveloping sound, but there's no way you're going to over-compensate the environment. The OP mentions ANC for engine noise. I drive a hybrid: hardly has any engine noise. I do hear road noise (which can be important for driving), and also, more importantly AC noise. Now that we're in summer months, my AC can start with high fan noise. While that's going I just accept it and not try to drown it out with too loud music.

 

[pinnahertz]

Bandwidth is not a concern. There will be negligible squint over such a narrow BW. Yes, thought has to be taken in constructing a desired beam pattern, and some new technology needs to be developed. But I'm not here to solve the world's problems, chief. I'm just thinking about whether or not it is technically feasible.

Oh, bandwidth is an concern alright. Because at the higher frequencies the distances to a null are tiny. Just take 10kHz. A perfect null (not necessary for ANC, but it makes the point) is achieved when the cancellation signal and the noise signal are perfectly matched in phase (time) and amplitude. If the amplitude match is off by 1dB, that perfect null drops a 20dB null. OK, still usable? But now, you've moved the listening ear by .33", which is 90 degrees at 10kHz. Now that null is nonexistent, completely gone. This isn't a big problem with headphones because the distances are all fixed. But in an enclosed space with free position seating?

OK, but that's 10kHz. How about something more reasonable, like 3kHz? The 90 degree shift distance is now 1.1" to blow the null completely, and only .25" just to keep a reasonable null of 9dB. What are you going to do, put the listeners head in a vice? Track their movement with a worn motion tracker? This would only be practical for low frequency ANC, otherwise it's just not possible, I don't care how you form your beams. Even at 250Hz, which might be the start of engine noise, the listener can't move 3" or he'll be out of the null.

 

[TheSonicTruth]

That's not active, it's passive. And it's the same philosophy companies like Shure
take over companies like Bose - noise isolation vs noise cancellation. Despite what
you may think, you are not covering any new ground here.

Never thought I was. Just suggesting tried-&-true.

 

[TheSonicTruth]

Oh, bandwidth is an concern alright. Because at the higher frequencies the distances to a null
are tiny. Just take 10kHz. A perfect null (not necessary for ANC, but it makes the point) is achieved
when the cancellation signal and the noise signal are perfectly matched in phase (time) and amplitude.
If the amplitude match is off by 1dB, that perfect null drops a 20dB null. OK, still usable? But now,
you've moved the listening ear by .33", which is 90 degrees at 10kHz. Now that null is nonexistent,
completely gone. This isn't a big problem with headphones because the distances are all fixed.
But in an enclosed space with free position seating?

OK, but that's 10kHz. How about something more reasonable, like 3kHz? The 90 degree shift
distance is now 1.1" to blow the null completely, and only .25" just to keep a reasonable null of 9dB.
What are you going to do, put the listeners head in a vice? Track their movement with a worn motion
tracker? This would only be practical for low frequency ANC, otherwise it's just not possible, I don't care
how you form your beams. Even at 250Hz, which might be the start of engine noise, the listener can't
move 3" or he'll be out of the null.

In a car, or a 767 cabin for that matter, most of the noise you want to cancel is below 1kHz. So you'd really be overtaxing the noise cancellation system, not just the speakers(!) by attempting to cover most of the human hearing range.

 

[pinnahertz]

In a car, or a 767 cabin for that matter, most of the noise you want to cancel is below 1kHz. So you'd really be overtaxing the noise cancellation system, not just the speakers(!) by attempting to cover most of the human hearing range.

I cited an example of 250Hz. Still doesn't work in a situation with variable ear location.

 

[TheSonicTruth]

I cited an example of 250Hz. Still doesn't work in a situation with variable ear location.

Another reason why vehicular ANC will never be perfect - IE: completely cancel out all engine hum or road rumble at all seated positions in the cabin.

I never noticed much of such noises in my comparable size '81 Buick Century - it had padding beneath the carpeting, and that padding extended above and beyond the front footwells, up the firewall behind the dashboard.

Best noise cancellation around!

 

[pinnahertz]

Another reason why vehicular ANC will never be perfect - IE: completely cancel out all engine hum or road rumble at all seated positions in the cabin.

I never noticed much of such noises in my comparable size '81 Buick Century - it had padding beneath the carpeting, and that padding extended above and beyond the front footwells, up the firewall behind the dashboard.

Best noise cancellation around!

That's isolation, non cancellation. Isolation works by preventing the problem from occurring in the first place.