Sound-upon-sound question

Okay, imagine this scenario:

You're watching the news on TV at a normal volume, and your roommate suddenly turns on the vacuum cleaner, which is very loud. The noise from the vacuum cleaner totally drowns out the noise from the TV, so it looks like the TV commentator's mouths are moving but no sound is coming out.

QUESTION: what is the total noise level in the room? Is it vacuum cleaner + TV? Or is it just vacuum cleaner? After all, you can't hear the TV; but is the noise still 'there?' Or does the noise from the vacuum cleaner sort of 'envelop' the noise from the TV?

This has been bugging me for a long time, so please offer a suggestion instead of just a one-word answer.

 

[brentkiosk]

Multiple sources of sound just add. What actually adds is the sound waves, the total wave amplitude at any time and location being the sum of all the waves. Usually, such as with the vacuum cleaner and TV, the sources have no defininite phase relationship, and you can just add the intensities (NOT the dB levels) to get the total. So, the vacuum cleaner just drowns the TV out. The TV sound is still there.

 

[CycloWizard]

Sound is nothing but pressure waves travelling through a medium. So, if your ear is at a location where the two sources are producing a reinforced peak, then you'll hear a louder sound. Think of it as sine waves traveling from various directions. Wherever they overlap, there will be crests, troughs, and nothings (where the trough of one wave cancels out the crest of another). Then, consider that these sine waves all have different frequencies and amplitudes. The answer isn't so simple. As with any good engineering question, the answer is "It depends."

 

[The Boston Dangler]

A wave can amplify or attenuate another wave, depending on phase, timing, frequency, etc. Active noise cancellation works by measuring a sound, and outputing the opposite sound.

The TV's sound is added to the vaccuum's sound to give the room's SPL, measured in dB. This may be A-weighted for human hearing.

http://www.ptme.com/et/audio/reference/sound/dB-SPL.htm

http://www.jimprice.com/prosound/db.htm

http://www.neurophys.wisc.edu/comp/docs/notes/not006.html

 

[TuxDave]

In the simpliest terms, it's general additive and you can picture the vacuum cleaner as a noise and the TV as your signal. If the amount of energy from the TV sound falls below this noise floor, all you hear is vacuum cleaner.

 

[glorygunk]

The vaccuum is a lot louder than the TV, probably a difference of at least 20dB. 20dB corresponds to a multiple of 100x so you shouldn't be able to hear the relative 1% sound that the TV is making. The sound of the TV is still there, but it essentially becomes negligible.

Well picking the vaccuum cleaner is a tough choice because it tends to produce a broad range of frequencies (maybe the newer models are quieter, but my Windtunnel isn't). Sort of like the static you hear on your TV when there is no signal. Thus any sound you make that is in the frequency band of the static will be destructively interfered, i.e. you won't hear it. Perhaps you should try this...make your TV produce static, turn the TV to normal conversational volume. See if you can talk to your roommate Just my idea anyway.

 

[Calin]

Well, let's move a bit up in frequency (and change the wave category).
If we would use radio waves, the situation is as follows:
We live in a place where radio waves (from the lowest common frequency like 50 and 60 Hz of power lines to the highest frequency like strong X-Rays from outside the Earth) are present.
However, if you use just a narrow band to transmit, you can overcome the noise on that band, and if you have a sufficiently sensitive receiver, you can receive only the band you want. However, if your receiver is not capable to receive on a very narrow band, the signal will "get drowned" in noise received on a larger band, and the "signal to noise ratio" increase (you have just as much signal, but the noise is added not on the signal band but on a much larger band).
Now, if the noise is concentrated on a small band, and the signal is concentrated on a larger band, you could just cut out that noise band, and get the signal from the rest of the listening band.

Coming back to vacuum cleaner... the noise is generated mostly by airflow over the turbine (or vacuum pump), and this kind of turbulence covers a larger band (it depends on the airspeed I think). Also, the rotational motion will generate a base harmonic at the same frequency as the rpm of the pump, and at multiples of that (this kind of sound is on a pretty narrow band at each of its frequencies).
So, to hear the TV you would need to drown the sound of the vacuum cleaner. However, if your TV would be pinging (like in submarine movies) and the vacuum cleaner would be low, you could hear the "pings" over the low frequency sound of the cleaner (even if the pings would have much lower power, their frequency localization would help you hear them). You might also hear drums on the TV with the vacuum cleaner at max power and max rpm. However, voice (with its pretty large range of frequencies) over a vacuum cleaner with its also large range of frequencies and random distribution or power on those frequencies? No chance

 

[KoolAidKid]

As others have said, you can consider the intensities of complex sounds (such as your vacuum or the TV) to be additive. So a sound level meter would register a higher reading in the TV+vacuum condition than in the vacuum condition.

However, you may not be able to perceive this increase because the vacuum is masking the output of the TV. In simplified terms, your inner ear is a mechanical fourier transformer: it breaks sounds up into their component frequencies. If you are presented with two tones played at the same time, 100 Hz and 5 kHz, for example, the basilar membrane located in your inner ear will vibrate at two distinct locations, one associated with 100 Hz and the other with 5 kHz. Each of these places on the basilar membrane has a sensor associated with it so that any vibration in close proximity will cause the sensor to trip.

Masking happens when two or more sounds cause the same sensor to trip. The listener can't tell the difference between the sensor's response to just the vacuum and the sensor's response to vacuum+TV.

Audio compression techniques such as MP3 take advantage of this by discarding portions of the music that are likely to be masked: if the listener can't hear it anyway there is no point in including it in the compressed signal.