Treatment of a small boxy room - REW file attached

#121

gullfo wrote:Source of the post well, since we're going to down the rabbit hole.. fluid dynamics is a good start: https://en.wikipedia.org/wiki/Fluid_dynamics

Yes it could be interesting to play around in OpenFOAM with some of these hole geometries (with all the other rabbits lol).

#122

So there's a bit more progress to report (I really should have started a new thread).

Based on what I've learned so far, I wanted to build something that actually had an acoustic effect at my room's nasty modal frequency around 44Hz. It didn't have to be the final practical solution, but I just wanted to see anything that would have a real acoustic impact with close micing (I'm nowhere near considering room effects). To this end, I picked up a 3' length of 4" ABS pipe (I was using 3" diameter before). In my initial experiments I used the whole three foot length of pipe, just to see if I could get it to do something at 44Hz before possibly cutting it.

As before, I capped the ends and drilled a hole in the centre of one cap. I did frequency sweeps in REW with a measurement mic pointed at the hole. The cap end being tested was positioned so it was facing the single active monitor, at a distance of around two feet. I used a 1/2" measurement mic, at a 45 degree angle from the cap face and within around 1/8" from the port.

Here's a picture of the rather large tube, standing beside the smaller 3" pipe I was using before:

Here's a close-up of one of the chamfered holes:

(the cap shown in close-up has the chamfered 3/8" diameter hole that ended up working the best).

Now we can look at some REW sweeps. First here's the "control" sweep which was acquired with the tube flipped around to present the back side cap (with no hole) to the measurement mic. The 44Hz modal peak is clearly evident (remember this is not like a normal "listening position" measurement).

Next, here's the first "Helmholtz Resonance" sweep with the measurement mic in the same position, but the tube flipped around to present a 1/4" hole with sharp edges. We can see a weak Helmholtz effect around 30Hz, as well as two "pipe resonances" at higher frequencies (which is to be expected from a pipe this size):

Chamfering the ends of the 1/4" hole increased the Helmholtz frequency slightly, and made the dip a little bigger. The pipe resonances are still there:

Next, I increased the hole size to 3/8" with the first measurement keeping the sharp edges on the hole. Here, the Helmholtz frequency is getting pretty close to the modal frequency. The "phase reversal" is still not as well defined as was seen for the bottles though. The pipe resonances seem stronger now:

Finally, I added a chamfer to both ends of the 3/8" hole. The resonant "phase reversal" effect seems fairly strong now and it coincides closely with the modal peak. The pipe resonances are even stronger now:

FWIW here is a figure with all REW response curves overlaid:

So this is still a long way from a practical room treatment, but at least it does something! It should be possible to continue adjusting the geometry so as to get the pipe length down from three feet. I now feel incredibly naive thinking I could just drill some holes in an MDF box, based on an online calculator, and have it work right off the bat...

One slight change from my tests with 3" pipe is that the cap walls are now a bit thicker. This translates into a longer "neck length" for a given hole size. This could be relevant in optimizing the final neck geometry.

#123

Another way to visualize the striking effect of HR neck geometry is to look at the spectrograms in REW.

The figures below are from the 3' length of 4" pipe.

Here is the spectrogram for the 3/8" hole with sharp edges. You can clearly see the long decay tails for the two pipe resonances, but there's not much happening around the Helmholtz resonance at 44Hz:

Here is the spectrogram for the same 3/8" hole with chamfered edges. There is now a long decay tail around 44Hz, indicating a narrow band resonance:

Of course the pipe resonances may be problematic, but at least we have something that's sort of controllable now. Also important to note that there is no damping of any kind in the pipe (so the long tails are to be expected).

As a control, here is the spectrogram with the pipe flipped around to present the back side cap (with no hole) to the mic. You can see a bit of a tail at 44Hz from the room mode:

I'll reiterate that these sweeps were not taken at the listening position, so some of the ringing from room modes is less pronounced than it would be in a normal room measurement (where the LP is further from the monitors and located at the left/right midpoint of the room).

#124

note that one thing we like to do is to force the mode distribution into higher frequencies since those tend to be easier to mitigate with plain old absorption. so having the pipe create some harmonics higher up is usually easier to deal with in exchange for more reactance in the target frequency range.

some nice testing here !

#125

gullfo wrote:Source of the post note that one thing we like to do is to force the mode distribution into higher frequencies since those tend to be easier to mitigate with plain old absorption. so having the pipe create some harmonics higher up is usually easier to deal with in exchange for more reactance in the target frequency range.

some nice testing here !

Thanks! That's reassuring about the higher harmonics.

#126

Ok now we are getting serious...

This gives the orange frequency response curve (cyan is the previous test with a chamfered 3/8" hole in a pipe cap):

Recalling the strong resonances from the bottles, I thought I would try and further augment the "flaring" en route to the resonator neck (using a few flared coupling sections). It turns out this helps a lot. Note that, rather than drilling a hole in an ABS pipe cap, we are now using an open section of 1.5" pipe as the "neck". This is much easier to tune, and seems to work very well.

This improvement might be related to my earlier musings about how having a larger mass of air in the neck could be advantageous. The very large cavity volume in the 4” pipe tends to decrease the "spring stiffness" but the much larger surface area of the 1.5” neck increases the force exerted by the air cavity at a given pressure (more than compensating for the increased cavity volume). The net result is that we have a much larger air mass, and a much stiffer "spring", than was the case for the small holes drilled in the MDF panel (even though the resonant frequencies for the two geometries are about the same). Of course there is also much less flow restriction and potential for turbulence in the larger pipe assembly. It's worth noting that the surface area of the open 1.5" pipe is equal to that of 36 of the small 1/4" holes that were drilled in the MDF panel (which, because of "port chuffing", weren't doing anything acoustically irrespective of their total surface area).

Of course there are some issues with the practicality of this now huge resonator cell, but it's a much better starting point than a useless MDF box

.

#127

Here are the frequency response curves obtained by varying the neck length of the "Hubble Space Telescope" (as I am calling this now huge assemblage of tubes) in 1/2" increments:

Resonance at the lowest frequency of 44Hz was achieved with the longest neck length (4.5").

#128

Hi,

I am not an expert in the physical aspects of acoustics. My complements for the details you try to bring in here.

My thoughts on this test are:

- Maybe it is not comparing apples to apples.
- You are using 2 kinds of resonating devices
- If I understand it correctly you are trying to figure out the impact on sound in the room using an empty bottle (thus having NO damping capability other than resonating as a one tone wonder) as one device and a single (small) version of a Helmholtz resonator, designed to dampen the volume of sound at a certain frequency (not a one tone wonder but having most impact on a small frequency range).
- Your conclusion " the small low-frequency resonator cells did not appear to do anything to the sound field picked up by the mic:" is may be not the real conclusion but it is "guided" by an expected outcome of the test.

- I would put the following to be considered:

1 - as your are working with "low" frequencies (200 Hz or somewhere around that), your damping material is not very good at damping at those frequencies. That's why we need air behind an absorber to start making the low frequencies getting impacted and lowered in dB level
2 - The volume damped with this HH resonator (VERY small one) is most likely VERY minimal to be measurable as the device is very small. I would love to have one HH resonator of this size giving me damping at a noticeble level. You would need a considerable area of HH resonator to create a good and measurable impact
3 - The Empty bottle resonates at the given frequency because of its physical properties when it is empty, fill the bottle with water, you get a higher pitch and so on. The HH resonator resonates by it's pipe (the hole I the wood) and impacted by the area behind it which then is impacted by the rock wool. Not the same device physically, different outcome.
4 - If the resonator (bottle) in you case was placed 180 deg out of phase at it's position with the wave we are handling than you would have a null left on that position in the room. If it is exactly in phase, you would amplify the wave. So the solution would be to have n bottles at one location where they all are 180 deg. out of phase with the wave, where n is the amount off bottles that have tamed it enough to get an even frequency response. As we are talking low frequencies (in my case I need to tame 34 Hz) we have a large area to use to tame (or am I wrong here?). So put 10 bottles in the corner and measure, if it works like we think, than 20, 30 40 and so on until we have reached the right level of cancelation.
5 - There already is a solution called anti-sound (if that's the right word for it), where an electric device collects a frequency and send out the 180 deg. out of phase frequency into the room. It cancels out and you hear nothing.

keep up the good work,

regards, Frans

I love what you are trying to find out here, being also very creative and think it over as detailed as possible. I am very interested in the HH resonator creation to tame low frequencies in my studio (currently in design phase), where I need to tame a few low frequencies.

I have been struggling with using theory applied in practice, while designing loudspeaker enclosures. I design an enclosure using the principles of the enclosure I want to produce. The loudspeaker itself does not know in what type of enclosure its is placed nor does it have any knowledge of our human science and theory on loudspeaker enclosure properties, but simply behaves by it's own physical properties combined with the properties of the environment it is in. I think I am building a transmission line, but actually my build has all the properties of a bass reflex enclosure. Then I am measuring and I do not get the outcome I was expecting by the theory, because my mind says "this is a Transmission Line", so I should measure TL behavior.

#129

Frans Wessels wrote:Hi,

I am not an expert in the physical aspects of acoustics. My complements for the details you try to bring in here.

Hi Frans - thanks for posting your comments! I've been dormant for a while with deadlines and other projects picking up more momentum than I expected.

I'm still interested in the physics of Helmholtz resonators and how this relates to their practical application for sound absorption. I'll try to follow up on your discussion points below!

My thoughts on this test are:

- Maybe it is not comparing apples to apples.
- You are using 2 kinds of resonating devices
- If I understand it correctly you are trying to figure out the impact on sound in the room using an empty bottle (thus having NO damping capability other than resonating as a one tone wonder) as one device and a single (small) version of a Helmholtz resonator, designed to dampen the volume of sound at a certain frequency (not a one tone wonder but having most impact on a small frequency range).
- Your conclusion " the small low-frequency resonator cells did not appear to do anything to the sound field picked up by the mic:" is may be not the real conclusion but it is "guided" by an expected outcome of the test.

These are all really valid points that I should be able to address. Because of the meandering learning path I've taken on this, my purpose has not been particularly clear.

I would say that my objective coalesced into something like "understand the basic physical mechanisms of Helmholtz absorption and how these mechanisms interact with room acoustics". This is why I devoted so much effort to being able to manipulate small resonant cavities even though these are evidently of no practical value for controlling room acoustics. The converse is that all practical absorbers (based on Helmholtz resonance) are based on the same physical concepts and so it should be possible to scale from single cavities up to arrays of cavities.

If a single resonator cell does not have any demonstrable acoustic reactivity under close micing, I think it's fairly safe to conclude that a whole wall full of them is not going to do anything to the room acoustics. I am fairly sure this is the case with my initial 1/4" hole array, and I am also confident that this is due to flow turbulence in the neck that is not captured by the standard "linear" Helmholtz calculators.

I think it's also safe to say that if an undamped cavity does not exhibit any of the hallmark signs of acoustic reactivity (primarily phase reversal in the frequency response), under close micing, then this is probably due to flow turbulence effects and no amount of added damping is going to mitigate an unsuitable neck geometry. On one hand this is not a huge deal, as many practical efforts start out with larger holes than I did and turbulent flow is probably not as much of an issue. On the other hand, there is probably some value to this insight as it may be a limiting factor in some absorber designs that could perhaps be improved by a more careful consideration of neck geometry.

I have learned a few things since my last post here, and some of these things relate to the excellent points you raise below and which I'll try to address:

- I would put the following to be considered:

1 - as your are working with "low" frequencies (200 Hz or somewhere around that), your damping material is not very good at damping at those frequencies. That's why we need air behind an absorber to start making the low frequencies getting impacted and lowered in dB level

In reviewing academic papers, forum posts, and YouTube videos, I've noticed that many of the "best" results appear to be from Helmholtz resonators in which the only source of damping is acoustically transparent cloth placed across the mouth of the resonators. My criterion for "best results" is that the the poster/author demonstrates a localized acoustic effect in close micing and a general effect on overall room acoustics.

2 - The volume damped with this HH resonator (VERY small one) is most likely VERY minimal to be measurable as the device is very small. I would love to have one HH resonator of this size giving me damping at a noticeable level. You would need a considerable area of HH resonator to create a good and measurable impact

If you scroll back though this thread (which is now rather long and somewhat disjointed), you will see one test I did with a tiny glass bottle. The little bottle showed a very strong local resonance effect (under close micing) even though nobody would expect this single small bottle to impact room acoustics. To paraphrase what I said earlier, the demonstration of localized resonance under close micing is "necessary but not sufficient" for a scaled up resonator array to impact room acoustics.

3 - The Empty bottle resonates at the given frequency because of its physical properties when it is empty, fill the bottle with water, you get a higher pitch and so on. The HH resonator resonates by it's pipe (the hole I the wood) and impacted by the area behind it which then is impacted by the rock wool. Not the same device physically, different outcome.

All of these systems reduce to what a physicist would call a "driven harmonic oscillator with damping". The trick is to accurately model how the different physical elements (mass, spring, damping) interact with each other and with the sound field in the room.

4 - If the resonator (bottle) in you case was placed 180 deg out of phase at it's position with the wave we are handling than you would have a null left on that position in the room. If it is exactly in phase, you would amplify the wave. So the solution would be to have n bottles at one location where they all are 180 deg. out of phase with the wave, where n is the amount off bottles that have tamed it enough to get an even frequency response. As we are talking low frequencies (in my case I need to tame 34 Hz) we have a large area to use to tame (or am I wrong here?). So put 10 bottles in the corner and measure, if it works like we think, than 20, 30 40 and so on until we have reached the right level of cancelation.

This is where the physics description of Helmholtz resonance gets complicated. The resonator presents both a "reactive" and a "resistive" impedance effect on an incoming wave. The reactive component will cause phase cancellation (reducing sound) above the resonance frequency and phase reinforcement (increasing sound) below resonance. At the exact resonance frequency, there is zero phase shift and the air in the resonator "neck" is happily bouncing in and out in perfect synchrony with the incoming wavefront. Most of the close micing tests I've done suggest that this neither adds nor subtracts from the local sound field but it's important to remember that those tests were done with the resonator in "free space" rather than "baffled" by a surrounding wall.

My understanding of practical Helmholtz absorption is that it exploits the fact that air flow through the neck is at a maximum when the resonant frequency is reached. Normally, a porous absorber is impractical for low frequency sound because it requires a high sound particle velocity that is only seen far from a wall/boundary. A Helmholtz absorber can be thought of as a transducer for converting the high sound pressure at a room boundary into high flow velocity within the neck of the resonator. Porous absorbent material within the neck will, in theory, become much more effective because it requires a high air flow velocity to remove energy. This is the resistive aspect and what makes a Helmholtz resonator act as a true "absorber".

5 - There already is a solution called anti-sound (if that's the right word for it), where an electric device collects a frequency and send out the 180 deg. out of phase frequency into the room. It cancels out and you hear nothing.

The classic commercial example of this, for control of acoustic modes, seems to be the PSI Audio AVAA. The concept is very elegant and the designers seem to have solid technical credentials. However, they are extremely expensive and online reviews seem a bit mixed (mainly in relation to the cost/benefit aspects - since conventional porous/resonant trapping can be attempted for a tiny fraction of the cost).

keep up the good work,

regards, Frans

Will do thank you! I'm working on this problem in parallel with a bunch of other science stuff I do in my day job. I will disclose that I've floated questions on this topic in a couple of other online forums, but the community spirit and ambiance of this forum is by far the most congenial and constructive!

I love what you are trying to find out here, being also very creative and think it over as detailed as possible. I am very interested in the HH resonator creation to tame low frequencies in my studio (currently in design phase), where I need to tame a few low frequencies.

I suppose I could call myself a "scientist" by trade (evidently not in acoustics), and I love it when elegant concepts like Helmholtz resonance find practical application that is embraced by different communities like people involved in music production and creative arts. I'm also fascinated by the social aspects of how we can be dazzled by "miracle solutions" when the underlying practicalities are often rather murky and easy to gloss over.

I have been struggling with using theory applied in practice, while designing loudspeaker enclosures. I design an enclosure using the principles of the enclosure I want to produce. The loudspeaker itself does not know in what type of enclosure its is placed nor does it have any knowledge of our human science and theory on loudspeaker enclosure properties, but simply behaves by it's own physical properties combined with the properties of the environment it is in. I think I am building a transmission line, but actually my build has all the properties of a bass reflex enclosure. Then I am measuring and I do not get the outcome I was expecting by the theory, because my mind says "this is a Transmission Line", so I should measure TL behaviour.

It has occurred to me that learning about loudspeaker design is probably very helpful (essential?) for learning about room acoustics and treatment. Perhaps the ultimate way to go is to abolish the distinction between room and speaker and come up with an integrated design with drivers built into the room structure (beyond flush mounting of commercially sourced speakers). This may not be cost effective and at any rate is way out of my league at this point.

I will try to post a better summary of my findings and conclusions at some point. Thanks for posting and the thought provoking comments!

Soundman2020 - Studio Design Forum