Helmholtz resonator in second row riser

Jason Nugent · « **on:** 31 Jan 2006, 03:48 pm »

Greetings,

First post in this circle, so hello to all

I'm in the middle of constructing a dual listening room/HT in my basement in a house that my wife and I bought two years ago. It's a completely unfinished room at the moment, with nothing but studded walls (2x6 double stud walls, stuffed with standard R-12 fibreglass insulation). The walls will be drywalled with 2 layers of 1/2" drywall and a Green Glue sandwich. Not so much to keep noise from getting out, but to keep various sounds from the house from getting in.

Anyway, my question comes from obviously wanting to tame low frequency issues in the room. The room measures approximately 23 x 14 x 8 (ish - there will be a suspended acoustic tile ceiling). For the "HT" part of the room, the second row of seating will be on a riser measuring 8" high, by 9' wide, by 7 feet deep. The top of the riser will be covered with two layers of half inch plywood. I would really like to turn this riser into a resonator. I was wondering if anyone could provide me with advice or turn me onto a resource that could tell me how I should partition this with cross members internally, and how large to make the slots or holes in the ends. My idea is to stuff it with fibreglass, cut holes or slots in the ends and then carpet the whole thing, including the sides. I've been reading through "The Master's Handbook of Acoustics" by Everest (great book so far). He discusses resonators, but doesn't get into creating resonators that are filled with material. There is mention of a fudge factor for resonators that are not filled with air, but I didn't see any other mention of it.

I plan to also use bass traps in the corners of the room where I can, and a few wall panels to reduce first order reflections for two channel listening.

Thanks in advance.

Jason

ctviggen · « **Reply #1 on:** 31 Jan 2006, 04:47 pm »

Hey Jason,

That is a very good idea! I don't think I can help you much, though. Here are two websites that semi-discuss rectangular/box-type helmholtz radiators:

http://members.nuvox.net/~zt.robgrow/helmhltz/helmhltz.html
http://www.audioholics.com/techtips/roomacoustics/Helmholtzresonatorabsorber.php

As your radiator would be in the middle of the room, you won't get a benefit of corners, where all modes meet. I still think it's worthwhile, though, as you should get some bass/high frequency trapping and it should be relatively easy to do. The hardest part will be figuring out how to make it.

I also have the Acoustics Handbook. There are a lot of good ideas in there, but it's short on a lot of substance at times. For the cylindrical helmholtz radiators, for instance, he says you can use ETF to design one with a Q=1. Well, I tried for a while to design a radiotor using ETF and never got close to a Q of 1.

Jason Nugent · « **Reply #2 on:** 31 Jan 2006, 05:25 pm »

Bob,

Thanks for the links - I will check them out.

I'll get the benefit of one corner, at least, because the riser will be up against the back wall of the room. I plan to build it by attaching 2x4's to the floor in the square shape for the bottom of the riser, and then attaching 2x8's "on edge" to the outside of the frame, and then partitioning it off with more 2x8 cross members.

I'd like to do a halfway decent job with getting it right becuase it'll be hard to test until the room is done and everything is drywalled, and by then, changing the size of the riser or taking it apart will be difficult. I'll let you know how it turns out.

Jason

Jason Nugent · « **Reply #3 on:** 31 Jan 2006, 07:23 pm »

Bob,

I've been reading the sites you mentioned, and it seems to me that the height of the slot you have to make in the side of the resonator makes no difference - just how wide it is. It also seems like Im going to have to really partition up the interior of the riser because anything much over a cubic foot or two has a resonant frequency in the low teens.

Time to haul out the graph paper.

Jason

bpape · « **Reply #4 on:** 31 Jan 2006, 07:42 pm »

Actually, that's a common thing to do in higher end rooms and a very good use of space. To make it work, you'll need to make sure that the area below the top is sealed air tight or it won't work right. Also, the amount of insulation inside will change the tuning frequency.

Lastly, the thickness of the top plate absolutely changes the tuning frequency. Things that change the center frequency are:

Cavity depth
Face thickness
Slot Width
Slot Spacing
Amount of insulation inside

Bryan

Jason Nugent · « **Reply #5 on:** 31 Jan 2006, 07:53 pm »

Bryan,

I plan to put down a bead of acoustic caulk between the riser and the floor, and on all internal baffle edges, as well as along the top before the plywood goes down. Would this qualify for making it air tight? If need be, I can also line the whole thing with 6 mil vapour barrier.

I didn't realize that insulation changes the tuning frequency - I always assumed it just lowered the Q value and widened the absorption band. Hmmm. Is there any way to take the guesswork out of that?

I had a question about the orientation of the slots. Do they need to be cut on the side of the resonator that is parallel to the mode Im trying to reduce? For instance, my ceiling of 8 feet has a fundamental at 70Hz. Would I need to cut slats in the top of my resonator to tackle that one? Is that why the thickness of the top plate affects the frequency - because it determines the depth of the port? Or does it affect all resonators, regardless of whether or not it's used as the "port side".

I've been playing with the numbers on a spreadsheet on one of the links mentioned by Bob. It seems that to tune a 70 Hz fundamental, I'd need a slot 0.1 inches wide, cut in a slat 1.5 inches deep, which has an effective depth of 1.8 inches. The depth behind the port would need to be a whopping 51 inches, which is doable, provided I can use one of the long sides of the resonator for a mode that's perpendicular to the slot opening.

I guess Im looking for a better formula, or something specifically better for my application. Big square box sitting on the ground, with slots or holes (doesn't matter which) drilled in various ends. The inside would be partitioned off accordingly. Most of the things I've seen account for slot width, but not slot height, or more than one slot. Some reference depth of the resonator, others reference volume. If depth was the only concern, volume could change by changing the other two dimensions. I'm guessing that the "depth" ones refer to cylindrical resonators, not cubic ones.

Jason Nugent · « **Reply #6 on:** 31 Jan 2006, 08:00 pm »

Hmmm, according to another source (Terry Montlick on AVS), completely filling the baffle with insulation increases the effective volume by 1.4. Good to know.

Jason Nugent · « **Reply #7 on:** 1 Feb 2006, 01:10 pm »

Hehe.

After a few hours of playing around and some more reading, I think i have this figured out. Well. I know enough to be really dangerous.

I asked if the slots need to be cut on the side of the resonator that interacts with the room mode. I believe the answer is yes.

It turns out that in many calculators and spreadsheets on the net, there's an error that has been propagated for a while. The first link that Bob mentioned has this error.

Instead of this:

f = 2160 * sqrt ( r / (( d * D ) + ( r + w )))

The forumula is supposed to be:

f = 2160 * sqrt ( r / (( d * D ) * ( r + w )))

* instead of + there.

Subtle, but it can move the tuned frequency by as much as an octave and a half. This is what was messing with my first set of numbers.

Anyway, it looks like I can cut 5 mm slots in the edges of my resonator, spaced 150 mm apart, and build a cavity behind them with an effective depth of 330 mm for 80Hz. For the vertical mode around 80 Hz, the depth of the cavity is less (which makes sense because the riser can only be physically 7.5 inches deep inside), but the slots are further apart at 285 mm.

Those numbers seem to agree with me.

Sorry for talking to myself.

Jason

bpape · « **Reply #8 on:** 1 Feb 2006, 03:14 pm »

Just to make sure, I'd put down a weatherstrip on the bottom of the perimeter and the joists and then caulk both sides of them. I don't think you'll need the vapor barrier. Also, don't forget to caulk the ends of the joists where they attach to the perimeter frame.

Yes, the + instead of * error is on a LOT of sites and in a few books even.

Normal construction for this type of absorber puts 1" of 703 or equivalent directly on the back face of the slotted board (this would be down inside your riser). That is so little that it doesn't change the design frequency much but still provides the damping. I trust Terry's numbers and in the case of a riser, more insulation is a good thing to stop random ringing.

Jason Nugent · « **Reply #9 on:** 1 Feb 2006, 06:04 pm »

Thanks, Bryan. I appreciate your input. I will put weatherstripping under the riser along the perimeter.

By the way, when doing my searching on the net for a decent web-based room mode calculator with graphing capability, I couldn't find one, so I created my own. I know that there are excel spreadsheets that do grouping in a chart, and there are windows and dos apps that do it as well, but not everyone has Excel or wants to download a spreadsheet or an app, and not everyone (like me) runs Windows.

Right now, it just does axial modes, but I plan to add the other ones soon. I hope people find it useful, and I'm open to suggestions.

http://malhavoc.homeunix.com/modes/

Jason Nugent · « **Reply #10 on:** 6 Feb 2006, 02:35 pm »

Well, I'm getting ready to build my riser/resonator this coming weekend. Here's the plan, as it stands, currently:

1. I'm going to frame out the base of the riser with 2x4s directly screwed to the concrete floor in the room. There will be weather stripping underneath it, and acoustic caulk around the inside edge on the bottom.

2. a 2x8 frame will be screwed on edge to the outside of the frame. Again, inside edges will be sealed with caulk.

3. 2x8 joists will be run across the inside, since it has to support the weight of the seating for the second row.

4. internal baffles will be filled with R-12 fibreglass.

5. 2 layers of half inch plywood will be attached to the top.

6. Slots measuring 1/16" of an inch will be cut 9 and 3/4 inches apart perpendicular to how the joists ran inside. With an internal effective depth of 10.5 inches, this works out to a resonance frequency of about 50 Hz.

The whole thing will measure 9 feet by 7.5 feet.

Does this seem at all logical? There are a few areas I'm still fuzzy about. I'm not sure how tightly to pack the insulation. And I'm worried that if I screw this up, I can do way more harm than good in the room. Can I?

Thanks all. I appreciate your input.

Jason

Jason Nugent · « **Reply #11 on:** 6 Feb 2006, 02:38 pm »

By the way, I plan to supplement the room's acoustics with eighth nerve and real traps products, once everything is drywalled.