• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Trying to understand the limitations of Helmholtz resonators in LF absorption

OP
D

Delrin

Member
Joined
Jan 4, 2023
Messages
52
Likes
46
Yes, that is exactly the same result I obtained - low to moderate effectiveness of a single HR box for LF absorption. For good results, it require big volume(s) devoted to HR box(es) - very limiting in a typical living room.

It's important to remember that those plots I posted do not show the impact of the resonator on the room acoustics as measured at the main listening position. Rather, they show the sound levels measured with a mic placed right in front of the port, with the port positioned fairly close to the speaker. This rather artificial configuration is intended to allow qualitative comparison of the responses seen with different versions of resonator.

This gives a very detailed depiction of the effect, but it also greatly exaggerates the effect compared with what would be seen at the listening position with the resonator installed as room treatment.

There may be pitfalls in measuring the way I have done, and I'm actually a bit surprised that nobody has come out to criticize this approach or flag some limitations (although I am trying to be clear that these don't provide a quantitative measure of absorption). If there's a way to do it better, or limitations to be aware of, I'd be happy to hear.

From the measurements I've taken, I don't think one can really say if my "best" resonator design (the flared pipe) would be effective or not as room treatment. If I arrayed five hundred of them over my back wall, they might well be effective (keeping in mind that this might not be the most economical or effective approach). It's very clear, however, that a single instance of the "flared pipe" design will have no significant impact at all on room acoustics.

One other point, related to my crude tests where I dropped cotton balls into the tube, is that the placement of an absorber could be important and simply letting them pile up at the bottom of the cavity may not be optimal. It may well be better to affix any absorber close to the resonator neck, where airflow velocities should be higher (as sarumbear has described).

Thanks Vladimir for keeping up the discussion, I just want to make sure it's clear I'm not measuring an actual room treatment.
 
Last edited:

Vladimir Filevski

Addicted to Fun and Learning
Joined
Mar 22, 2020
Messages
550
Likes
712
It's important to remember that those plots I posted do not show the impact of the resonator on the room acoustics as measured at the main listening position.
Yes, that is very clear. That is the method which I also used to determine the exact absorption frequency of the HR box. The impact on room acoustic is modest.

From the measurements I've taken, I don't think one can really say if my "best" resonator design (the flared pipe) would be effective or not as room treatment. If I arrayed five hundred of them over my back wall, they might well be effective (keeping in mind that this might not be the most economical or effective approach).
That is exactly the way how the 11th-19th century churches in my country (and over wider area) were acoustically treated - many clay pots-jugs built inside the church walls, with only openings protruding outside. Pots/jugs are partially filled with ash as a damping material.
 
OP
D

Delrin

Member
Joined
Jan 4, 2023
Messages
52
Likes
46
Yes, that is very clear. That is the method which I also used to determine the exact absorption frequency of the HR box. The impact on room acoustic is modest.


That is exactly the way how the 11th-19th century churches in my country (and over wider area) were acoustically treated - many clay pots-jugs built inside the church walls, with only openings protruding outside. Pots/jugs are partially filled with ash as a damping material.

Ok thanks for clarifying!

The use of these techniques in 11th century churches is really fascinating. I have heard a men's orthodox choir at a monastery in Russia (not making any presumptions about where you are from) and there is some serious sustained bass. It's easy to imagine they would have known about room modes even back then, and probably recognized that time smearing of modal frequencies is problematic for a choir where you need to be able to synchronize with the other singers.
 
Last edited:
OP
D

Delrin

Member
Joined
Jan 4, 2023
Messages
52
Likes
46
Anything that requires simulation of a physical phenomena I suggest using Wolfram Mathematica.

Here is what Wolfram gives as an example. Imagine what can be done by using a custom front end and using the actual architectural model of the room that the builders used to build it.

:mad: I installed Mathematica and was quite stoked to experiment with the RoomEigenFrequencies example you linked, only to find the notebook points to a non-existent support file. So far as I could tell, I selected all documentation options in the installation. It can find the notebook no problem, but the directory referenced for the 3D models does not exist. There's nothing obvious in the app settings like "Download Support Files".

Sorry for the kvetching - I was ready to invest time on this but it doesn't reflect well on Wolfram that they ship examples that are broken like this...

(I'm sure I'll get it to work, and am still grateful for the tip)
 
Last edited:

DonH56

Master Contributor
Technical Expert
Forum Donor
Joined
Mar 15, 2016
Messages
7,835
Likes
16,497
Location
Monument, CO
:mad: I installed Mathematica and was quite stoked to experiment with the RoomEigenFrequencies example you linked, only to find the notebook points to a non-existent support file. So far as I could tell, I selected all documentation options in the installation. It can find the notebook no problem, but the directory referenced for the 3D models does not exist. There's nothing obvious in the app settings like "Download Support Files".

Sorry for the kvetching - I was ready to invest time on this but it doesn't reflect well on Wolfram that they ship examples that are broken like this...

(I'm sure I'll get it to work, and am still grateful for the tip)
Just ping Wolfram support, they are pretty responsive. May require an add-on or have changed files and you need to point elsewhere.
 
OP
D

Delrin

Member
Joined
Jan 4, 2023
Messages
52
Likes
46
Just ping Wolfram support, they are pretty responsive. May require an add-on or have changed files and you need to point elsewhere.

Thanks Don - I did post a query on their tech support forum. It seems one guy had the same issue a couple of years ago.

It could be that I am misunderstanding what is to be expected from the "notebook" paradigm. The Room Mode notebook opens with renderings of all the simulations, from start to finish, including graphics with interactive buttons allowing you to select which mode is visualized. Normally I'd want to tinker with the code and try changing things, but this isn't possible due to the un-findable "SupportFile" directory.

It's probably something silly - clearly the software is very deep, and I should probably learn more about its features before diving in.
 
Last edited:

sarumbear

Master Contributor
Forum Donor
Joined
Aug 15, 2020
Messages
7,604
Likes
7,314
Location
UK
Ok thanks for clarifying!

The use of these techniques in 11th century churches is really fascinating. I have heard a men's orthodox choir at a monastery in Russia (not making any presumptions about where you are from) and there is some serious sustained bass. It's easy to imagine they would have known about room modes even back then, and probably recognized that time smearing of modal frequencies is problematic for a choir where you need to be able to synchronize with the other singers.
Go back a bit more. Knowledge in acoustics had been around since the 5th century BC...

 
OP
D

Delrin

Member
Joined
Jan 4, 2023
Messages
52
Likes
46

sarumbear

Master Contributor
Forum Donor
Joined
Aug 15, 2020
Messages
7,604
Likes
7,314
Location
UK
I have often wondered what heuristics these ancients used to implement what seem to be quite rational acoustic designs. I gather this is an active topic of academic research, but have not read much on it. Thanks for sharing the paper!
I doubt very much it is heuristics. The survived bits of information tells us they did experiments that lasted for years. Each project was a lifetime job for someone.
 
OP
D

Delrin

Member
Joined
Jan 4, 2023
Messages
52
Likes
46
I doubt very much it is heuristics. The survived bits of information tells us they did experiments that lasted for years. Each project was a lifetime job for someone.

I suppose trial and error is itself an heuristic... they must have formed some mental model, and devised ways of assessing the performance.
 

sarumbear

Master Contributor
Forum Donor
Joined
Aug 15, 2020
Messages
7,604
Likes
7,314
Location
UK
I suppose trial and error is itself an heuristic...
Heuristic technique, is self-discovery that employs a practical method but thevresult is not expected to be optimal or perfect. It is a method for reaching an immediate, short-term goal. In the case of the ancient Greeks, nothing was short-term and they aimed, and achieved perfection.

they must have formed some mental model, and devised ways of assessing the performance.
Like every great scientist did throughout the history.
 
OP
D

Delrin

Member
Joined
Jan 4, 2023
Messages
52
Likes
46
Heuristic technique, is self-discovery that employs a practical method but thevresult is not expected to be optimal or perfect. It is a method for reaching an immediate, short-term goal. In the case of the ancient Greeks, nothing was short-term and they aimed, and achieved perfection.


Like every great scientist did throughout the history.
Oh to have been a humble fly on the wall...
 

sarumbear

Master Contributor
Forum Donor
Joined
Aug 15, 2020
Messages
7,604
Likes
7,314
Location
UK
Oh to have been a humble fly on the wall...
You have to be an apprentice for life :)

We are more inclined to learn like this:

 
OP
D

Delrin

Member
Joined
Jan 4, 2023
Messages
52
Likes
46
You have to be an apprentice for life :)

We are more inclined to learn like this:


Indeed we are now YouTube apprentices for life ;).

I had watched this video before, but would have liked to have seen a demonstration at the end showing whether it had any effect on the targeted room modes. At least one poster in the comments asks whether the resonators worked as intended, but there is no response on this question (the video was posted around eight months ago). It may well have yielded spectacular improvements, but the "radio silence" on outcomes is a bit concerning.

I probably need to look harder, but I have yet to see anything on forums like this one or on YouTube where someone documents the details of their build, and then goes on to conduct controlled testing that demonstrates a significant beneficial (or any) effect on overall room acoustics (as depicted in a spectrogram or waterfall plot). In the academic literature there are papers in which the absorption coefficient is measured, but this is just a surrogate marker for the real outcome of interest which is the control of modal peaks at the listening position(s). Translating absorption coefficient into room performance seems like an inexact science, or at least this quote from Cox & D'Antonio suggests so:

"One problem with modal control using resonant absorption is knowing how much resonant absorption to use. Although the theories set out in Chapter 6 allow the absorption coefficient for Helmholtz absorbers to be estimated, the meaning of absorption coefficient at low frequencies is problematical. (Even more problematical is the lack of good prediction models for membrane absorbers, but that is another story.) At low frequency the sound field is not diffuse, and consequently the effect that the absorber has is not calculable through simple statistical laws."

So if anyone can post a link to a forum posting or YouTube video with controlled testing of a Helmholtz absorber that demonstrates a measurable benefit to room acoustics, I will buy them a virtual pint!

I understand that some large Helmholtz absorbers are built into the room, which would make it difficult to remove them for an A/B comparison. Nevertheless, I would think it should be possible to make a controlled test by blocking the mouth of the resonator(s), which should negate the mechanism of action.

Edit: There is a somewhat well-known example of a Helmholtz resonator in Chapter 12 of "Master Handbook of Acoustics" by Everest and Pohlmann. The example is purported to address a mode at 47Hz and they show waterfall plots that demonstrate greatly improved ringing. The resonator is based on a large concrete forming tube (commonly branded Sonotube in North America) with a single port at the bottom. I don't really consider this to meet the standard laid out above, as the implementaiton details are somewhat sketchy as is information about the room. Moreover, there is at least one online account from someone who did their best to replicate this design and found no effect on room acoustics.
 
Last edited:
  • Like
Reactions: MAB
OP
D

Delrin

Member
Joined
Jan 4, 2023
Messages
52
Likes
46
Ok so realizing I hadn't looked that hard on YouTube, I spent a painful hour going through the results from searching "Helmholtz Bass Trap".

Below is a sampling of the hits that came up, with some comments on each:

================================


Guy builds a "slat resonator" that is not airtight, and it seems like no thought was given to controlling the cavity volume. There are no measurements, and no discussion of whether it does anything acoustically. Several commenters point out that it's not really a Helmholtz resonator. No other followup videos about HR on channel.

================================


Gives a rather fluffy description of room modes, never actually builds the resonators, no results shown. Commenters love the video, as the guy has a great video presence. No other followup videos about HR on channel (so we never know if he finished building it or if it worked).

================================


Talks about resonant traps at around 19:00, but doesn't actually try it. He advocates convincingly for just using porous absorbers due to value, simplicity etc.

================================


Recording engineer shows slat resonator boxes that he built, and reports a subjective improvement but gives no measurements. At the end he says "the acoustician will come back later to check it" but there is no followup in comments or other videos on the channel.

================================


The video shows fabrication of boxes (very similar to my useless MDF boxes) but there is no testing or any indication of the result. A commenter asks if it works, and the video poster replies (with commendable honesty) that he doesn't know.

================================


This is the last in a nice series of three videos talking about the theory of HR and fabrication of a flush mount installation that incorporates slat resonance absorption. REW measurements, including waterfall and spectrogram, are shown at the end and demonstrate considerable improvement in frequency and time domains. Because the treatment incorporates several different acoustic techniques, it's hard to attribute improvements specifically to Helmholtz resonance. The presenter seems to have consulted with academic/professional acousticians, and (in an earlier video of the series) goes into some detail about the judicious use of the online calculator at http://www.acousticmodelling.com. He relates some guidance he has received regarding which models to select, in a clear and credible way.

The slat assembly covers an entire wall, so area coverage would be good. If I had to copy something I saw on YouTube, this would be a contender.

Although this last video is promising, I'm not sure it entirely meets my standard of "a controlled demonstration in which Helmholtz resonance is shown to address modal peaks". It feels a little "fast and loose" in spots and I need to watch the three videos in his build series more carefully. There is no real description of the room, or what other treatment is included in the "before" and "after" measurements. Because his design is modular, he could have done some "close micing" tests to assess the acoustic reactivity around the resonator slots (nothing like this is shown or discussed).

I would recommend this video series as one of the better YouTube accounts of a Helmholtz absorber project (keeping in mind my search was far from exhaustive). It's in French, but I think you might be able to get subtitles in your preferred language. I will try and do a more detailed review of this video series, summarizing highlights and pointing out concerns, in another post.

I will not comment on the presenter's choice of headgear :).
 
Last edited:
OP
D

Delrin

Member
Joined
Jan 4, 2023
Messages
52
Likes
46
Taking a break from YouTube, it occurred to me that I've been throwing around the terms "linear" vs "non-linear" in relation to physical phenomena associated with Helmholtz resonators. A simple distinction would be that linear effects do not undergo any qualitative changes as the sound amplitude is varied, whereas non-linear effects should exhibit qualitative changes in the frequency response seen at different sound levels.

It occurred to me that this could (and should) be tested, so I did some frequency sweeps on one of my small MDF boxes at different sound output levels. I used a box that resonates at 48Hz (as predicted by theory and confirmed by the candle test) with a neck diameter of 1/4" and length of 3/4". This particular neck is a plastic insert press fit into a larger hole in the MDF.

Interestingly, as the SPL of the sweep is decreased, we can see the emergence of some small acoustic reactivity at the resonant frequency. This fits with the notion of non-linear effects, and further suggests that some neck geometries are just "bad" even if they do support resonance.

Box HR Non-Linearity (1).jpeg


As per my loose definition of linearity given above, you will note that all other qualitative features of the frequency response remain constant despite the shifts in level (although the lowest curve, in yellow, is getting a bit noisy).
 
Last edited:
OP
D

Delrin

Member
Joined
Jan 4, 2023
Messages
52
Likes
46
Here's a bit more info from Andy Mac Door's video series that I referenced above. He has numerous videos on acoustic treatment (all in French). I do not know his credentials, and can't vouch for the accuracy of anything, but in a number of respects his videos seem above the YouTube average in terms of research and execution (while also speaking on a level that is good for an informed DIY crowd). He raises a few interesting points, some of which might be debatable, and I thought it might be helpful to post some notes here in this English-language forum. It's a bit vague, but the videos imply that he's either worked with a credentialed acoustician or at the very least has done a fair bit of research.

I'll list the relevant claims made in Video 1/3 here, along with my editorial comments in italics:
  • A Helmholtz absorber should ideally cover the entire wall of the treated room [this jives with what I've read elsewhere, and seems intuitively correct since we are trying to neutralize a standing wave that also spans the entire wall; it seems at odds with reports of large tube resonators having a single localized port though]
  • In Andy's videos, he is treating a room in which the main modes of concern are at 55 and 110Hz. The Helmholtz trap series is devoted entirely to addressing the 55Hz mode. [This is relevant for many small listening/practice rooms, which will have modes around 50Hz]
  • The videos seem to be based on work by French acoustician Frédéric Poirier, who has a web site referenced in the video. Andy goes very quickly over a document which seems mostly focused on membrane resonators, although it does include material relevant to Helmholtz absorbers. Andy also has a web site and mentions that he has acoustic calculation spreadsheets available for download. [Note that Andy does appear to sell stuff on his web page, so I hope my linking here does not violate forum advertising rules (there is also free educational material that you can consume at your risk).] Andy also mentions that there is some good information on a French forum by a poster calling themselves "Alec Eiffel", who summarizes some of Poirrier's work. I haven't been able to find this through Google, but could always ask.
  • There are warnings about the perils of on-line Helmholtz calculators, many of which are purported to give incorrect results. He endorses www.acousticmodelling.com [which is well known and seems widely viewed as giving valid results].
  • A demo is given for www.acousticmodelling.com, which contains a few interesting points:
    • The "Multi-Layer Absorber Calculator" is used, rather than the "Helmholtz Calculator"
    • It is stated that the model "Komatsu (2008)" should be used in place of the default "Allard and Champoux (1992)". [Komatsu is certainly far more recent than any of the other options]. It is claimed that the "Allard and Champoux" model works well when porous material is used as a stand-alone absorber (which is not relevant here).
    • For "Helmholtz Model", it is recommended to use "Ingard/Allard" and not "Transfer Matrix" [I had been using Transfer Matrices, from Matlab code published by Trevor Cox].
    • Several iterations of modelling are done to arrive at a frequency of 50Hz. The settings are:
      • Layer 1 = slotted panel (thickness=50mm, slot width=4mm, slot spacing=64mm centre-to-centre)
      • Layer 2 = porous absorbent (thickness=400mm, GFR=Pa.s/m^2)
      • Layer 3 = air (thickness=150mm)
    • [It's not entirely clear if these are the settings used in the unit he actually builds, but it gives the idea]
  • The remainder of the video is devoted largely to the vagaries of lumber dimensioning (seems like Europe also has to deal with nominal vs real dimensioning).
  • He also discusses a spreadsheet he implemented to translate geometric parameters from non-parallelogram boxes to parameters suitable for inputting to the on-line calculator he uses.
  • In closing he states that the total absorption of the Helmholtz trap will be directly proportional to its volume (assuming that other parameters have been scaled correctly to achieve the desired frequency response). [This seems intuitively plausible, but it would be nice to see a more rigorous physical basis for such statements].
Hopefully this is of interest and can maybe stimulate some further discussion!

Here is the link to the video summarized above:

 

sarumbear

Master Contributor
Forum Donor
Joined
Aug 15, 2020
Messages
7,604
Likes
7,314
Location
UK
A Helmholtz absorber should ideally cover the entire wall of the treated room [this jives with what I've read elsewhere, and seems intuitively correct since we are trying to neutralize a standing wave that also spans the entire wall; it seems at odds with reports of large tube resonators having a single localized port though]
This not only seems odd with the tube resonators I use but also with practically every ‘bass-trap’ designed for and placed in a corner.

Don’t you think?
 

Geert

Major Contributor
Joined
Mar 20, 2020
Messages
1,938
Likes
3,526
This not only seems odd with the tube resonators I use but also with practically every ‘bass-trap’ designed for and placed in a corner.

Don’t you think?
Maybe we need to distinguish between what's 'ideal' (as mentioned by Andy Mac Door) and what's often good enough. All types of room modes have high energy in corners and intersections between walls/floor/ceiling. So tackling them in these areas can be pretty effective, taking into account the large wave length of the lowest room mode frequencies.
 
Top Bottom