Looking for help with a Helmholtz Slat/Slot Resonator

[Matt C.]

so i am thinking about building a big (maybe 6'w by 5'h by 8"d) helmholtz slat resonator to put in my live room directly behind where drums are usually set up. i am intending to build two or three seperate standalone boxes instead of one big one, with angled fronts so the depth of the cavity varies from about 3" to about 8", making them broadband absorbers. i understand the general concept and design of these things, but i have a few questions about specifics, so if anyone has some insight into these absorbers, let me know. i will try to post some simple diagrams when i have more time so that this makes more sense.

my questions are:

1. when putting the insulation inside the box, should i place it flat against the back wall of the resonator (far away from the slats/slots on the front), or should i put it right up against the back side of the slats, at the front of the air cavity? or kind of freestanding in the middle of the cavity? i have seen diagrams showing it both ways. my limited knowledge would say to put it more toward the front of the cavity, for the same reason that it's good to have absorber panels mounted with a gap between it and the wall - it can lower the absorption frequency. any thoughts on insulation placement here?

2. i am also expecting to use all different sizes of wood slats on the front, with different slot widths, again to make it a broadband absorber instead of specifically tuned (and so the irregular wood slats will provide some HF diffusion). my question is, when placing the slats on the front, should i put them on in pairs of identical sizes (two 1x2s placed together, then two 2x4s, then two 1x1s, etc). this way those slots will have the same depth of wood on each side, instead of irregular slot depths (sorry if this is an unclear explanation, i'll try to post a diagram soon).

i think those are my main two issues. i will try to post some diagrams to make it all more clear...

[snoopy23]

I built one of these in my studio.I't about 12' wide and 8'tall. On the advice of an experienced audio designer, I made each slot a different length. It was his opinion that spreading similar sizes apart offered a more even diffraction to the whole room, rather than focusing similar length waves in the same area. His reasoning is that this will decrease the chance of uneven sounding spots in the room where some frequencies may be bouncing from the same location in the same direction. Also, I placed my insulation on the back of the enclosures, it seems to work just fine.
Hope it helps!

[Matt C.]

thanks for the information. when you say "slot length", do you mean the vertical height of the gap between boards, or the front to back depth, or the actual horizontal length across the whole box?

here are some quick diagrams illustrating my questions (sorry, but my graphic design skills do not extend beyond MS Paint).

the first picture shows a side view of attaching the wooden slats in pairs of similar depth, so that the open slot between them has "walls" of the same size on each side (slots 1,3, and 5 in the picture).

the second picture shows the depth of the slats changing every time. realistically it would probably not be so random (it would probably just go back and forth between .75" and 1.75"), but you get the idea. this way every open slot has two different sized "walls".

not sure from a resonating/absorbing standpoint which is preferable (or whether it is worth changing the slat depth at all). mainly i am trying to get away with making the front of this thing as random as possible to create diffusion in higher frequencies.

[snoopy23]

That design looks like it will work fine. Mine looks more like a side view of a saw blade, with the depth at the end an even four inches and the width of each "blade" a different length. I went from 2" to 10" in varying intervals and did not repeat the same length twice. The width is measured at the 'base' of each blade. I then stuffed each compartment with insulation and pulled it to the front as best I could to create a row of absorbtion chambers. This is crude, but from above it looks like this:


|\|\|\|\|\|\|\|\|\|\|\|\|\|\|\|\|\
------------------------------------
This design works pretty well for me, my room (especially drums) sound live but don't have that annoying wall reflection that makes it sound "demo" quality. Sometimes it is a bit too dry, but I am experimenting with putting up pieces of plywood n certain areas to increase the sonic depth of the room. The dryness also works well with a little reverb, but I try to "keep it real" when possible. Of course I am fortunate enough to have a high, peaked roof in my tracking room, but this should noticeably improve the sound in most any room. Hope it helps!

[JWL]

1. when putting the insulation inside the box, should i place it flat against the back wall of the resonator (far away from the slats/slots on the front), or should i put it right up against the back side of the slats, at the front of the air cavity? or kind of freestanding in the middle of the cavity? i have seen diagrams showing it both ways. my limited knowledge would say to put it more toward the front of the cavity, for the same reason that it's good to have absorber panels mounted with a gap between it and the wall - it can lower the absorption frequency. any thoughts on insulation placement here?

Insulation should go against the FRONT of the resonator, just behind the slats.

2. i am also expecting to use all different sizes of wood slats on the front, with different slot widths, again to make it a broadband absorber instead of specifically tuned (and so the irregular wood slats will provide some HF diffusion). my question is, when placing the slats on the front, should i put them on in pairs of identical sizes (two 1x2s placed together, then two 2x4s, then two 1x1s, etc). this way those slots will have the same depth of wood on each side, instead of irregular slot depths (sorry if this is an unclear explanation, i'll try to post a diagram soon).

In his book, Rod Gervais argues that there is no real benefit to doing this. If you use different slat spacings/depths, then in order to calculate it you simply average them. He recommends using boards the same width, depth, and spacing. Certainly it is simpler to construct this way, and there is no real benefit to varying it (according to him).

That said, I've never tried both ways side-by-side to compare them.

[JWL]

Also, varying the slat widths/depths at random will NOT produce diffusion. It might scatter the sound somewhat, but diffusion is a mathematical sequence that must be built accurately or it won't work as a diffusor.

The frequencies at play here are also pretty high. Any sound with a wavelength longer than the width of the slat will diffract around the slat and be absorbed by the resonator. For reference, if you use a 1x4 (actually 3.5" wide) this wavelength is around 4k. Diffusion can be effective in this range but it's more useful down to under 1k.

[Matt C.]

i guess a little bit of sound scattering is what i was hoping for, i know this sort of thing won't be as effective as a real QRD diffusor (in fact i'm wondering whether i should just build one of those instead). but it seems like making the front slats completely uniform would be just as bad (at higher, non-absorbed frequencies) as having a plain flat wall there, maybe causing some ugly reflections. but maybe not, i might be over thinking it.

also, regarding the issue of varying the slat size/depth and slot size, wouldn't this potentially create a larger range of absorption? or am i missing some details here? here's an example, based on the helmholtz calculator spreadsheet i have:

if i use all uniform 2"x4" slats (actually 1.5x3.5), with a uniform .25" gap between all of them, then (because of the varying depth of the air cavity from 2" to 8" deep) the absorption range will be from 147hz-294hz. but let's say i use the same uniform 2x4 slats, but half the slot gaps are .25" and half are .15". wouldn't the addition of these smaller slots expand the absorption range to 115hz-294hz, since smaller gaps provide a lower resonant/absorbed frequency?

thanks for the help!
i should really read Rod's book...

[Matt C.]

(previous post edited to contain actual calculations.)