How to find out when something will reflect sound

[Puncher]

I try to find some information about until which frequency (from high to low) a wood panel of a specific mass will reflect sound and when it will start vibrating itself and not reflecting anymore.

I hope i described this understandable - actually i'm not having much of a clue yet

[SAC]

Its not that easy.

Materials can indeed exhibit resonance based upon mass and compliancy, but there are many factors that determine when and if such a material will resonate sufficient to dissipate energy.

Likewise, a materials acoustical impedance will determine its reflectivity and absorptive qualities.

Perhaps the most basic way to evaluate whether an object will reflect sound is to evaluate it in terms of size. I realize that this may seem strange, as many do not think in terms of sound having size; but the relationship of an objects dimensions and the length of the wavelength is fundamental.

Where an objects dimension is equal to, or larger than the wavelength, the energy at and above that frequency (where the wavelengths are still shorter) will be reflected. Where the wavelength is larger than the objects dimension, the acoustical energy will effectively flow around the object as if it is not there.

But I am not aware of any 'table' or 'tables' that list a specified frequency above which an object is reflective. There are too many variables to reduce a materials behavior in this manner. And that is also why we must employ such esoteric designs to create tuned mechanical filters and absorbers - it isn't as easy as simply going to the store and buying "x" type of material as it acts as say, a high pass filter with a corner frequency of 'y' with a slope of 'z' dB per octave!

[Ethan Winer]

Originally Posted by Puncher

I try to find some information about until which frequency (from high to low) a wood panel of a specific mass will reflect sound and when it will start vibrating itself and not reflecting anymore.

You can do this with room measuring software, a microphone, and a reasonably large sample of the material to be tested. The key is to set up the microphone and speaker on opposite sides of the wood. One good way is to place the wood over a doorway opening, with the microphone and speaker on opposite sides. Then run the software test sweep and you'll easily see at what "crossover" frequency the wood transitions from reflecting to passing.

--Ethan

[Puncher]

thanx for eplaining and also the nice method for testing it myself. i have all the gear - just need the wood

[Ethan Winer]

^^^ Excellent, and please post your result graphs here. I'm interested myself to see what you get.

--Ethan

________________
The Acoustic Treatment Experts

[SAC]

The limitation of this method is one that plagues many test facilities and shops – especially with regards to low end absorption! And that is precisely the limiting factor of the size of the test sample.

What one is more likely to discover rather than the actual characteristics of the material itself is that sound has size, and as soon as the stimulus frequency reaches the point where its wavelength is larger than the sample’s smallest dimension and begins to flow around the sample, this will be interpreted as ‘absorption’ as less will be reflected – when it is nothing of the kind.

For the prescribed method to have meaning, a sample would have to be provided that is larger in all dimensions than the wavelength is long. Thus, if one wants to evaluate a sample down to, say, 50 Hz, it will have to be greater than 22.51 feet in length and width.

Additionally, the space in which the test is being conducted, aside from being able to accommodate the sample, would also have to be sufficiently large to prevent reflections of the test stimulus from being reflected by other boundaries.

But on the bright side, one may discover, upon noting and converting the frequency corresponding to the change in energy being reflected, that they have found yet another method of roughly determining the approximate 2space dimensions of the sample.

[Puncher]

ok i tried something out just for fun


wood used is plywood poplar. 0.15748 inches thick, 19.68504x19.68504 inches size

some of the first reflexions from the sidewall should be catched with 3.93701 inches thick acoustic foam

best to discribe testmethod is a picture
on the right you see the mic, which is a behringer ecm8000



first a testrun without the wood


the with the wood

[SAC]

Note the change in behavior between 600 and 700 Hz.

And note that the wavelength of 19.69 inches = ~ 686 Hz.

Spooky, isn't it...

[Puncher]

Originally Posted by SAC

Note the change in behavior between 600 and 700 Hz.

And note that the wavelength of 19.69 inches = ~ 686 Hz.

Spooky, isn't it...

yes indeed - something evil is happening here

but i wonder if it can tell me something about the high end. there seams to be some kind of roll-off over 5 khz

made some additional absorbers last weekend out of mineral wool.

i though about covering them with some thin wood since i don't wanna kill to much of the high end and already have some acoustic foam in the studio...which is a home studio to point this out

aight

[SAC]

Simply put, absorption is more effective the higher the frequency of the incident signal.

But I fear you are looking at the reflection issue from a perspective that will not greatly benefit your efforts. Stop worrying about reflection frequency!

And start looking at the reflected energy in terms of arrival time and gain as provided in the ETC measurement.

The goal is not to 'EQ" the room, but to control the arrival time and intensity of reflected energy - and thus we only want to damp that energy that we cannot effectively redirect for later use in creating a diffuse soundfield to then be redirected (laterally as much as possible) back to the listening position in a well behaved exponentially decaying manner after the effective termination of the initial signal delay gap (a process often termed a 'Haas kicker').

[Puncher]

and ETC Stands for this kind of software based measurement like i used it?

[SAC]

You posted a frequency response.

The ETC - Envelope time curve is derived from the impulse response.

Not knowing the characteristics of your platform, you might also look at Room EQ Wizard (free) or FuzzMeasure for the Mac.

[macc]

Here's an ETC:




Top one = Bad

Bottom = Good

The bottom one would, as a result of having a better ETC, also happen to have a better freq response.

[Ethan Winer]

Originally Posted by SAC

The limitation of this method is one that plagues many test facilities and shops – especially with regards to low end absorption! And that is precisely the limiting factor of the size of the test sample.

Good point, and 20 by 20 inches probably isn't large enough for this test. That's why I suggested a large piece of wood in a doorway. If you have only a small piece, find a window that's even smaller and do it that way with the microphone outdoors or whatever.

--Ethan

________________
The Acoustic Treatment Experts

[Puncher]

Originally Posted by macc

Here's an ETC:




Top one = Bad

Bottom = Good

The bottom one would, as a result of having a better ETC, also happen to have a better freq response.

Thats a nice graph - sac's words in pictures

i like this forum

[DanDan]

A slightly different focus perhaps.
Many folk fear that their room may be 'too dead'.
IMHO, such an acoustic, coupled with a HF rolled off speaker response, makes mixing very easy. The mixes translate to the outside world. But each to his own.
The 'pro' version of this was the Non Environment concept for control rooms.
Again, great translating, but many simply don't like the 'deadness'.

Anyhooo. You can easily reflect some HF back into your room by bonding a membrane to the front of your traps. This only applies to angled traps which will not send an extra tweeter bounce to your ear. i.e. no membrane on the front of the RFZ or Cloud traps. Thick Paper should be fine. Glue it on with a spray fabric/insulation glue. SprayTac.
You could also consider using strips of wood with gaps across the fronts of the traps.
Take a look at John Sayers Productions
DD

[Puncher]

i wonder if i could take a 4 inch thick block of mineral wool, build a frame of wood which is also 4 inches deep just for the mineral wool to fit in, and then glue a thin wood panel to the front and the back of the frame to keep the mineral wool fibres in

the would panels should reflect high frequency on let mid and low trough. of course this construction would be air-proof somehow.

so i wonder will it absorb mid and low frequencies or can it even harm the sound since i created some vacuum inside without any calculation (i know resonant absorbers work this way)

these are not for RFZ - they would go a little bit behind on the side walls

[DanDan]

Puncher, with respect, to all involved. Could you try some searching and reading before posting questions?
Build a Better Bass Trap

DD

[Puncher]

yes sure, i'll try to search for this information. thanks for the link, i'll read trough it