John Sayers' Recording Studio Design Forum Archive

Hi, i want to build a helmholtz resonator, but instead of making it as an angled wall or mounting it onto a wall, i want to make them as panels so they can be movable in the room.

- First, about the air gap. regularly there will be an air gap (a sealed cavity in a shape of a right triangle) between the angled resonator and the original room wall behind it. but since i'm making panels, should i make a sealed cavity as part of the panel? or can i just position the panels spaced out from a wall (the cavity won't be sealed)? if there's a need for a sealed cavity, i was thinking of just making it like wooden boxes that one side of the box is the resonator, like this:



Does the slat length plays any role here? i mean, should i make the stand horizontal and not vertical for the slats to be longer, or it doesn't matter?
Should i do the cavity as a rectangle like in the sketch or should i do it in a shape of a right triangle?

- How seald should the sealed cavity be? the wooden box is sealed enough?

- Is there some guide or a thread of how to build these? when i fix the wooden slats to the frame, should they be touching the fabric that wraps the absorption material, or can be a couple of millimeters away from it?

Hi, i want to build a helmholtz resonator,

What you show is more correctly called a "slot wall", since it is a bunch of Helmholtz resonators all set to different frequencies.

should i make a sealed cavity as part of the panel

Yes. The cavity MUST be sealed or the device will not work.

Does the slat length plays any role here?

Not really. A longer slot just means you are treating more that frequency more heavily.

Should i do the cavity as a rectangle like in the sketch or should i do it in a shape of a right triangle?

Depends what you want to achieve! If you use the rectangle, the Q for each frequency will be quite sharp. If you angle the back (triangular cavity) then you flatten the Q, making it into a more broadband device, rather than sharply tuned. In other words, each slot will cover a range of frequencies instead of just a single frequency.

- How seald should the sealed cavity be? the wooden box is sealed enough?

Tightly sealed. The rear panel needs to be massive (heavy, thick) and stiff, and all the joints need to be carefully sealed with caulk. Any air leak at all reduces effectiveness, and might even wreck it entirely.

- Is there some guide or a thread of how to build these? when i fix the wooden slats to the frame,

You should probably buy the book "Master Handbook of Acoustics" by F. Alton Everest, to learn the theory behind them and how to make them. There are many examples on this forum, too.

should they be touching the fabric that wraps the absorption material, or can be a couple of millimeters away from it?

The cloth and insulation should NOT touch the slats. The way it works is that the "slug" of air in the slot vibrates back and forth at the resonant frequency, and therefore it moves a small distance beyond the surface of the slat on each vibration, before it goes back in again. If you put cloth right up against the slats, you interfere with this process, damping the vibration, smearing the Q and generally damaging the effectiveness. So the cloth must be well clear of the slots and slats.


Finally, please read the instructions on how to post and use this forum right here (click here). You forgot something!



- Stuart -

Soundman2020 wrote: Finally, please read the instructions on how to post and use this forum right here (click here). You forgot something!



- Stuart -

Thanks for the detailed answer, its been very helpful!

if by you meant that i didn't add my location, i added it, my bad.

i have one more question that i forgot to add, does the slats order matters? i mean, does it matter if i put the 8" slat on the top or in the middle? i understand that every slat responds to a different frequency, i could also see it from the excel Helmholtz calculator i downloaded from here, but i didn't see anything about the order of the slats widths.

i have one more question that i forgot to add, does the slats order matters? i mean, does it matter if i put the 8" slat on the top or in the middle? i understand that every slat responds to a different frequency, i could also see it from the excel Helmholtz calculator i downloaded from here, but i didn't see anything about the order of the slats widths.

Ideally, you should design the slot wall so that each slot is close to the position in the room where the pressure component of the sound wave peaks, for that mode. It ain't that easy to do, and the wall will still work anyway even if you don't get it right, but for maximum effect each slot should be more or less where the pressure peaks are that correspond to its specific frequency. And of course, you do want to keep each slot away from the pressure NULL for that frequency! Obviously, if an entire slot falls in a pressure null, then that slot won't do much at all!

- Stuart -

Soundman2020 wrote:

i have one more question that i forgot to add, does the slats order matters? i mean, does it matter if i put the 8" slat on the top or in the middle? i understand that every slat responds to a different frequency, i could also see it from the excel Helmholtz calculator i downloaded from here, but i didn't see anything about the order of the slats widths.

Ideally, you should design the slot wall so that each slot is close to the position in the room where the pressure component of the sound wave peaks, for that mode. It ain't that easy to do, and the wall will still work anyway even if you don't get it right, but for maximum effect each slot should be more or less where the pressure peaks are that correspond to its specific frequency. And of course, you do want to keep each slot away from the pressure NULL for that frequency! Obviously, if an entire slot falls in a pressure null, then that slot won't do much at all!

- Stuart -

Oh i understand. i asked that because someone gave me a design recommendation for a slat wall for my room, and he designed it as a full wall. and now that i want to design it as a panel i wanted to know what slots to take out of his design in order to make it fit in a smaller frame. according to what you said, i just need to "trim" the rectangle shaped panel i want from the middle of his wall design, and ignore the slots that left above and below the rectangle. Thanks!

Is there any general principle for positions of frequencies in a room (on the vertical axis)? for example something like: "Low frequencies are closer to the ceiling and the floor, and high frequencies are closer to the middle of the room" (just an example, not assuming its correct).

Is there any general principle for positions of frequencies in a room

Unfortunately not. It depends on each specific room, and which modes are problems. The dimensions of the room set the modal behavior, so each room is different. If the room is perfectly rectangular, then it is reasonably easy to predict the frequencies for each mode, and to determine the rough positions in the room where you can find peaks and nulls for that mode in the room. But if the room is not rectangular, then the best way to do it is with a pressure-based mic (not a velocity based mic, for obvious reasons) and some form of RTA. And even with a rectangular room, reality does not always match prediction very well, so there to you would be best off with a mic and an RTA. All of this only applies to axial modes anyway, of course. Tangentials and obliques are a bit harder to deal with.

But once again, all of this isn't totally necessary, since the slots are long and most likely some area along the slot will be close to a peak. You can measure or try to predict if you want to maximize, but the wall will still be effective, to some extent. Of course, this assumes that you put the device on the correct wall, if you are planning to treat specific problematic modes! It's no use putting it on the 0:0:1 wall if you are trying to treat a mode on the 0:1:0 axis!


- Stuart -

Soundman2020 wrote:

Is there any general principle for positions of frequencies in a room

Unfortunately not. It depends on each specific room, and which modes are problems. The dimensions of the room set the modal behavior, so each room is different. If the room is perfectly rectangular, then it is reasonably easy to predict the frequencies for each mode, and to determine the rough positions in the room where you can find peaks and nulls for that mode in the room. But if the room is not rectangular, then the best way to do it is with a pressure-based mic (not a velocity based mic, for obvious reasons) and some form of RTA. And even with a rectangular room, reality does not always match prediction very well, so there to you would be best off with a mic and an RTA. All of this only applies to axial modes anyway, of course. Tangentials and obliques are a bit harder to deal with.

But once again, all of this isn't totally necessary, since the slots are long and most likely some area along the slot will be close to a peak. You can measure or try to predict if you want to maximize, but the wall will still be effective, to some extent. Of course, this assumes that you put the device on the correct wall, if you are planning to treat specific problematic modes! It's no use putting it on the 0:0:1 wall if you are trying to treat a mode on the 0:1:0 axis!


- Stuart -

I understand, so can i get a name of a mode calculator that predicts and also gives me the location of the problematic frequency (just on the vertical axis - height) so i can decide where to put each slat?

Sorry to bring this old thread back up, but I'm exploring some H/holtz options as well and thought it's better to not start another thread on it.

Soundman2020 wrote:

should they be touching the fabric that wraps the absorption material, or can be a couple of millimeters away from it?

The cloth and insulation should NOT touch the slats. The way it works is that the "slug" of air in the slot vibrates back and forth at the resonant frequency, and therefore it moves a small distance beyond the surface of the slat on each vibration, before it goes back in again. If you put cloth right up against the slats, you interfere with this process, damping the vibration, smearing the Q and generally damaging the effectiveness. So the cloth must be well clear of the slots and slats.


- Stuart -

Stuart, how do you attach the slats to the studs in the "inside-out" wall construction method, if they are not to touch the fabric, which is stretched over the studs themselves? ~3mm thick, narrow packers along the length of the studs, before the slats come on.? Will this provide an extra edge between the slat and the stud, assuming the packers are not the same width as the studs, that might have an effect on the resonator?

I also have a couple of unclear things with the H/holtz resonators in general. I did some calcs in the spreadsheet provided on the forum here (thanks very much, BTW)

If I determine that i need:
-slot width - 1mm
-slat width - 200mm
-depth from wall - 90mm
-slat thickness - 25mm
to get the absorption Frequency of 74 Hz,
does this mean that the slot width of 1mm has to have 2 of the same 200mmx25mm slats on either side of it? What happens if I want to tune the next slot to a different frequency, using different width/thickness slats, how does the 200mm slat from the 74hz frequency affect this?...and so on.

If I'm using slats that span over a few stud centres (let's say studs are on 450mm centres), does this mean that I am creating, for instance, 4 h/holtz resonators over 1800mm (450 X 4) wide section of a wall?

I hope this makes sense...if you need any clarification, please let me know.
Thank you so much.

Stuart, how do you attach the slats to the studs in the "inside-out" wall construction method, if they are not to touch the fabric, which is stretched over the studs themselves?

You can put thin spacers between the studs and the slats. Of course, you so need to take the depth of that "shim" into account when calculating the frequencies for the slots.

However, I probably should not have been so vehement with what I said: you can have the cloth in direct contact with the slats, as long as you understand what it will do. If you want a very sharply tuned resonator that will just hit a specific frequency very well, then keep the cloth away from the slats. However, if you want a more broad-band effect, so the slot covers a wider range of frequencies with lower absorption, then leave no gap: put the cloth and the insulation up tight against the slats.

Here's a graph of both cases:
The cavity depth is identical here, at 90mm. The slats are also identical: 19mm, thick, 150mm wide, 5mm slots between them. The only difference is that for the blue curve, the cloth and insulation are placed 10mm behind the slats, while for the green curve, they are tight up against the slats. (In both cases, the insulation is 50mm of OC-703)- As you can see, the green curve shows much less absorption, but it also covers a much broader range than the blue curve (about two octave wider, in fact).

So it all depends on what you actually want to do. If you have a specific problematic frequency that you need to hit hard and tight, then keep the cloth and insulation away. But if you want to hit a broader range more gently, then put the cloth and insulation up close.

If I determine that i need:
-slot width - 1mm
-slat width - 200mm
-depth from wall - 90mm
-slat thickness - 25mm
to get the absorption Frequency of 74 Hz,

Right, but be careful there! Helmholtz resonators are very difficult to tune so precisely, especially at such low frequencies. If you make even a tiny mistake in building that, and the slot ends up at 1.5mm instead of 1mm, then you are about 10 Hz off! Way too far to be useful. If the depth is off by 10mm (say 80 instead of 90) then you are off by 5 Hz. Still a large difference.

Then there's the issue of effectiveness: in order to be effective, the resonant volume of the resonator needs to be about 1% of the room volume, at least, if you are targeting specific frequencies. Can you do that?

And finally, why do you need to hit 74 Hz? That's pretty low for a Helmholtz resonator. They don't work too well down that low.


Your other two questions are more about percent coverage: It works like this... If the open percentage (are of slots as compared to area of slats) is greater than about 10%, then the entire wall acts more like a broadband absorber, and should be calculated as though it were a perforated panel resonator. If the open area is less than about 5%, then it acts more like a series of individual Helmholtz resonators, each tuned to its own frequency. So once again, you need to take that into account when figuring out your slot wall plan.

- Stuart -

Thanks Stuart!
And I didn't really need to hit the 74hz, they were just some numbers I was punching into the calculator to see how low it would go, given a certain depth, so they were just there as an example.
My room isn't built yet, just researching the H'holtz a little more and realizing what you just said, that they are quite tricky to make work precisely, so some other methods might be in order (but that might be for another thread)... I just thought H'holtz would be a handy option for an "inside-out" wall design. We'll see...
cheers

I guess you are wanting to get the room as close to a professional control room as possible? Based on your room dimensions, theoretically you need a total of about 352 sabins of absorption to get the decay times within ITU specs for such a room. In real-world terms, that equates to about 32 square meters of perfect absorber. There is no such thing as a perfect acoustic absorber (except for an open window!), but you can get close with good fibrous insulation... if you know what the characteristics are! so figure on roughly 30 to 35 m2 of fiberglass or mineral wool panels. The total surface area of your entire room is about 64 m2, so you need to cover roughly 50% of the surface. However! 13m2 of that is floor, and it's pretty impractical to have mineral wool absorbers all over your floor (as well as being a really bad idea, psycho-acoustically), so that leaves you with the walls and ceiling. You'll need to cover about 60% of your walls and ceiling.

[NOTE FROM MODERATOR: THIS IS YET ANOTHER PLAGIARISM SPAMMER: TEXT COPIED FROM HERE: &t=19695&p=134777#p134777 SPAMMER BANNED, DELETED AND BLOCKED PERMANENTLY]