How to tune a Helmholtz resonator?

[Rasmus Faber]

Hi guys!

I have been helping out a friend re-structuring his studio. He's got a couple of Helmholtz resonators of the following type:
http://www.diffusor.com/Basabsorbenter.htm

I'm not sure how to 1. tune them and 2. what to adjust for. Are these meant to work only for peaks, or can they also adress nulls?

Is there a mathematical formula with which to determine the size of the gap? Or would this be done by ear somehow?

Some tips would be much appreciated! Thanks!

Rasmus

[Soundman2020]

Since that is a commercial product, I would suggest contacting the manufacturer to find out how to tune them.

What you would normally do in a room, is to measure the actual response using REW, to see which mode needs the most damping, then place some of those devices on the surfaces associated with that mode, and tuned to the center frequency of the mode.

- Stuart -

[DanDan]

Well done Stuart, I was kinda stuck at 'where do I begin'.
I would use a Spectrum Analyser rather than 1/3 Octave, but irrespective of indicator, for HH, tunable in situ is wise IMO.
https://www.youtube.com/watch?v=GjdACNPLzzg
DD

[Rasmus Faber]

Thanks guys! Sorry, I should have said, I've measured so I know where the nodes are, I'm just not quite sure how to tune the traps to corresponds to those frequencies, i.e how to measure the response of the traps.
I did some more googling, and seems you can knock on the traps and measure the resonant frequency. Don't know if thats the best way, but seems to make some sense..

You're right though, contacting the manufacturer seems like the obvious thing to do, I just want to try and have all options open in the case the manufacturer contact isn't stable etc.

[DanDan]

Please take a look at the linked video. That noise is Pink Noise. The little meter is showing the levels at all the different frequencies. The RTA or Spectrum Analyser in REW will do that even better, without the 1/3 Octave Bargraph. Just move the door on the V Trap for flattest response on the graph.
Be sure to place the traps where the Modes are strongest. Again, REW has an SPL Meter.
DD

[Soundman2020]

I've measured so I know where the nodes are,

Actually, you should be placing your devices at the pressure anti-nodes, not the nodes! Helmholtz devices are pressure based, so the pressure anti-node of the mode is were you need to put it.

As Dan said, you can use REW to find the peak pressure location of the anitnode for any given mode. Another way to do it is to use a tone generator (such as the one in REW) set to the exact center frequency of the mode, then walk around the room with an ordinary sound level meter. It will read highest at the anti-node for that mode.

I did some more googling, and seems you can knock on the traps and measure the resonant frequency. Don't know if thats the best way, but seems to make some sense..

I'm not too convinced about that! You'd likely be measuring the resonant frequency of the wood, not the the resonant frequency of the Helmholtz tuning.

You could estimate that by putting a measurement mic inside the mouth of the unit, then running slow sine sweeps with REW. It it is working, you should see the reaction quite clearly.

You're right though, contacting the manufacturer seems like the obvious thing to do, I just want to try and have all options open in the case the manufacturer contact isn't stable etc.

The device is likely sold with a manual or at least an instruction sheet that tells you how to tune it, and what the frequency rang is. If yours did not come with one of those, I'd google that as well. It's probably available somewhere, on-line to download.

- Stuart -

[Rasmus Faber]

This forum is so amazing!! Thank you all so much for your advice!!!

One thing related. I'll treat the back of the room as per your advice (and the manufacturer's hopefully)

But then there's that "classic" floor to ceiling null which occurs closer to listening position, usually around 100hz right?

The question, could such a null be treated with a helmholtz resonator under the desk? Or am I off here?

[DanDan]

An 8' ceiling, approx 74Hz. A cloud will shift the resonance so don't go by the physical height.
Yes a tuned trap under your seat or desk should be helpful. Or use two V4s as speaker stands if the height is right.
I believe they sell just the plans so one can manufacture whatever height you want on site.
A little trick with REW. Obviously you can see the modal peaks on screen. So, if you engage 'frequency follows cursor' in the Signal Generator, you can slide the cursor about and the generated frequency will follow exactly, and rather elegantly, no glitching. Also, REW has an inbuilt SPL meter, and all these functions will run simultaneously.
DD

[Soundman2020]

But then there's that "classic" floor to ceiling null which occurs closer to listening position, usually around 100hz right? The question, could such a null be treated with a helmholtz resonator under the desk? Or am I off here?

The "100Hz null" is usually floor bounce. In other words, a reflection from the floor to your ears, occurring at a point about half way between you and the speaker. It's a phase cancellation, not a resonance, so no, it can't be treated with a Helmholtz resonator. Those only work for acoustic resonant issues, such as room modes, mostly. Helmholtz resonators don't do anything for phase problems, or reflections, or general bass build-up near surfaces, or other things. All they do, is treat one specific resonance at one specific frequency. If you have a floor bounce problem, you might be able to treat it with absorption (thick!) or with a suitable desk design that gets in the way of the bounce. On the other hand, floor bounces generally don't need treating much anyway, because our ears and brains are quite well attuned to them: your brain tends to tune that out... and since it is a very steep, very narrow dip, it usually isn't even audible. You can see it on the graph, but you'll find it hard to actually HEAR it with your ears.

- Stuart -

[Rasmus Faber]

This is such helpful info guys, thanks so much!!