Does air reflect sound?

[duronboy]

How would a completely still and quiet open field with sound deadening panels on the ground compare to an anechoic chamber with the same panels covering 360 degrees?

[dcibel]

If you find a day with zero wind, far from any civilization and birds, trees, crickets, etc, then it can be comparable to an anechoic chamber. This situation doesn't really exist however, there will always be a slight breeze, rustle of leaves and grass, etc, so you can compare to an anechoic chamber with a higher noise floor. Some people dig a hole in the ground and place a box in there, speakers pointing up to measure a 2pi (infinite baffle) response.

[duronboy]

More of a pointless theoretical question than anything. I know a chamber is the best, practically.

I just wondered if there's any measurable reflection from just air. Or does there have to be something to reflect the sound.

[dcibel]

There has to be a change in media for a reflection to occur. Not specifically a reflection, but at night sound will travel farther in air due to the change in density of air at ground level vs air at altitude. The change in density due to temperature causes the waves to bend back towards the ground again.

[billfitzmaurice]

So long as the background noise is at least 30dB down from the measured level outdoors away from barriers is for all intents and purposes anechoic. In the suburbs background ambient noise tends to run around 50-55dB, sans passing cars, barking dogs and kids playing, so if you're measuring at even 90dB the ambient noise is moot.

[duronboy]

There has to be a change in media for a reflection to occur. Not specifically a reflection, but at night sound will travel farther in air due to the change in density of air at ground level vs air at altitude. The change in density due to temperature causes the waves to bend back towards the ground again.

So if there were a perfectly 2-dimensional freezer with a door which was parallel to the source of traveling sound waves, and it were dumping a wall of freezing air into the path of sound, could the colder air then reflect some of the sound back to the source?

So long as the background noise is at least 30dB down from the measured level outdoors away from barriers is for all intents and purposes anechoic. In the suburbs background ambient noise tends to run around 50-55dB, sans passing cars, barking dogs and kids playing, so if you're measuring at even 90dB the ambient noise is moot.

So how far from fences/structures does one need to be? How do I create a sound and make sure the reflections are at least 30dB down?

[devnull]

So if there were a perfectly 2-dimensional freezer with a door which was parallel to the source of traveling sound waves, and it were dumping a wall of freezing air into the path of sound, could the colder air then reflect some of the sound back to the source?



So how far from fences/structures does one need to be? How do I create a sound and make sure the reflections are at least 30dB down?

No the sound wave will speed up in the cold air because the wave is seeing a density change. A simple example of a wave encountering a density change is a drinking straw in a glass of water. The straw looks like it's bent at the spot it enters the water.

Depends on the frequency you're trying to measure to. The lower the frequency the longer the wave, the further from reflective objects you need to be. The easiest way to look for reflections is to look at the impulse response. You'll see a spike at the distance from the reflective object, about 1ms per foot. Then just gate the measurement. The impulse response shows a reflection at 5.5ms so you'd set the gate to 5ms.

[andy19191]

No the sound wave will speed up in the cold air because the wave is seeing a density change. A simple example of a wave encountering a density change is a drinking straw in a glass of water. The straw looks like it's bent at the spot it enters the water.

At a sharp change in impedance in the air due to, for example, a temperature inversion in the atmosphere sound waves hitting the interface between the two states will be partly transmitted and partly reflected. The latter is likely to be fairly modest due to practically achievable impedance changes in air being small. Large reflections follow from large impedance changes such as that between air and a solid like wood.

[billfitzmaurice]

So how far from fences/structures does one need to be? How do I create a sound and make sure the reflections are at least 30dB down?

That's a bit of a complicated question, as it's not only distance that matters, it's also size. From the standpoint of distance the inverse square rule applies. If the boundary is eight times as far from the mic as the mic is from the speaker a reflected wave would be down 36dB, if it was fully reflected. However, if the boundary is less than a wavelength wide and high the full wave won't be reflected, if it's reflected at all. It may diffract around the obstacle completely.

[Dwight]

A little off topic but try listening to a radio or speaker with a campfire in between. Unbelievable how turbulent air affects the sound.

[kevintomb]

A little off topic but try listening to a radio or speaker with a campfire in between. Unbelievable how turbulent air affects the sound.

I just went to post something similar~!!

it is almost as if sound moves through air and is air vibrating (sarcasm)

[billfitzmaurice]

At a sharp change in impedance in the air due to, for example, a temperature inversion in the atmosphere sound waves hitting the interface between the two states will be partly transmitted and partly reflected.

Cool damp air can have a major effect. I used to do sound monitoring for a major outdoor concert venue. One cool damp night my off-site readings anywhere from 500 yards to 2 miles from the stage were reading a 6 to 9dB higher than normal relative to 150 feet from the stage. I later confirmed that dry air attenuates sound more than humid air does.

[Dwight]

They will not raze (implode) buildings if it's foggy because the shock wave travels further and with more intensity causing more damage to surrounding structures.

[duronboy]

No the sound wave will speed up in the cold air because the wave is seeing a density change. A simple example of a wave encountering a density change is a drinking straw in a glass of water. The straw looks like it's bent at the spot it enters the water.

So one would have to purposefully create a series of strategically placed higher density airflows in order to have a hope of making the sound do a 180 with just air.

That's a bit of a complicated question, as it's not only distance that matters, it's also size. From the standpoint of distance the inverse square rule applies. If the boundary is eight times as far from the mic as the mic is from the speaker a reflected wave would be down 36dB, if it was fully reflected. However, if the boundary is less than a wavelength wide and high the full wave won't be reflected, if it's reflected at all. It may diffract around the obstacle completely.

For higher frequencies and smaller subs, an urban environment is ok, but big ol' subs need to be measured out in the wide, open spaces. Not just for neighbor comfort purposes. I keep coming across a post or two that recommends doing the 100 watt, 10 meter measurement for large subs because you're effectively going from half to almost quarter space. Although I can't find those posts, now. But with that in mind with what you just said, buildings would need to be 80 meters away if the mic was 10 meters from the speakers.

[billfitzmaurice]

Danley Sound Labs measure at 10 meters/100 watts outside their Gainesville, Georgia factory. Judging by the view on Google Maps it's not exactly lonesome prairie. They probably use gating to filter out whatever bounce may occur. There's residential housing close by as well, so they must use a short pulse.