Treatment of a small boxy room - REW file attached

#106

Greyhound wrote:Source of the postThis may or may not account for the total lack of acoustic effect from the MDF boxes, despite their ability to move a candle flame (could be related to turbulence?). It might be relevant that the bottle mouths, which do create an acoustic effect, are smooth plastic. Also the body of the bottle cavity flares smoothly into the neck, whereas holes in 3/4" MDF have a sharp corner (which may or may not matter, like everything else).

Would it be possible to compare the results with the same bottle, but with stoppers made from different material? One could be a "cork" made of mdf with a hole drilled through the middle. Another could be a plastic "cork", or perhaps one of those rubber stoppers with a hole through it I remember from school chemistry classes :geek:

#107

endorka wrote:Source of the post
Would it be possible to compare the results with the same bottle, but with stoppers made from different material? One could be a "cork" made of mdf with a hole drilled through the middle. Another could be a plastic "cork", or perhaps one of those rubber stoppers with a hole through it I remember from school chemistry classes

This is a really good idea - I was thinking about something like this, but the other way around. The idea was to make little plastic "nozzles" that could be pressed into a larger hole in the MDF face. The actual hole size would be determined by boring out the centre of the plastic plug. I've just made one, using an engineering plastic called Delrin on a small lathe. The holes are very clean and corners could be easily beveled or rounded if it would help with turbulence etc.

For initial tests, I also cut some 3" ABS pipes to have equivalent volumes to the MDF cells. These were capped, and I pressed the small nozzle into a hole in one of the caps.

Here are some pictures of the setup:

As noted above, the "nozzle" could also be press fit into a hole in the MDF to provide a cleaner and more controllable "neck".

I have not got round to thorough testing of this yet. A quick initial test was underwhelming, but there are a lot of parameters to control.

#108

I've been reviewing the physics on harmonic oscillators (driven, damped, etc) and one point came to mind that may or may not be relevant.

Helmholtz resonators are equivalent to a mass-spring system, where the mass is the air in the neck (with corrections) and the spring is the compressible air in the cavity. The mass scales with neck volume, and the spring stiffness decreases as the cavity size gets larger. The inverse of stiffness is compliance, and I find it a bit easier to think about a larger air cavity being more compliant.

In theory, a particular resonant frequency can be achieved using any combination of mass and spring compliance. A lower frequency requires the product [ mass x compliance ] to be larger, as the resonant frequency is equal to

f0 = 1/( 2 * pi * sqrt(m * C) ) where mass is in kg and C is the compliance in metres/Newton.

Helmholtz absorbers based on small-ish holes in MDF with a large backing cavity are like a very small mass bouncing on a rather compliant (floppy) spring. It's easy to imagine how such a construct would be prone to damping by even small resistive forces and indeed the damping ratio for this configuration will tend to amplify the influence of viscous effects (relative to a larger mass on a stiffer spring with the same resonant frequency).

It's possible that the bottles resonate so much better (than perforated MDF) because (i) they have a more suitable mass/spring distribution and (ii) the smooth, flared neck-cavity interface presents less resistance to airflow (the larger neck diameters should also reduce the relative influence of boundary layer effects).

#109

I added a tiny bottle, which has a neck volume similar to that of the MDF perforations. It resonates like crazy around 360 Hz (can get a nice pure tone blowing across mouth, and it interacts with a frequency sweep as shown in the red curve above).

#110

endorka wrote:Source of the post Would it be possible to compare the results with the same bottle, but with stoppers made from different material? One could be a "cork" made of mdf with a hole drilled through the middle. Another could be a plastic "cork", or perhaps one of those rubber stoppers with a hole through it I remember from school chemistry classes

Some digging required to actually extract any practical meaning, but apparently this is "a thing"! :ugeek:

(from The Journal of the Acoustical Society of America, December 2017)

#111

yes it is.

#112

Despite my ramblings about mass-spring systems and the failings of MDF boxes, they do appear to resonate as indicated by the "candle test". Apparently the candle effect is related to something called "vortex shedding" which seems to be more pronounced when sharp-edged holes are used (eg see this paper). They show simulation results (Figure 4) demonstrating a flow jet out the mouth of a resonator with sharp edges, which is eliminated when both (inner and outer) edges of the neck are chamfered. I am assuming it's this kind of flow jet that causes the candle flame to deflect.

The paper shows an example where the vortex shedding associated with sharp-edged corners contributes a non-linear resistive effect that can reduce the effectiveness of a Helmholtz absorber. This is separate from the "linear" effects that are modelled in classical Helmholtz calculators.

In their experiments and simulations, they used rather small resonators (a few cm) targeting frequencies of several hundred Hertz. In this regime, high SPL was required to demonstrate non-linear losses.

My bottles, despite being acoustically reactive, do not have any observable effect on the candle flame. Given the number of YouTube videos showing bottles used to drive rotation through Helmholtz resonance, they must also have some degree of vortext shedding (presumably from the non-flared plastic edges at the mouth).

The candle flame effect suggests that my MDF boxes exhibit stronger vortex shedding, and presumably higher non-linear losses, than the various bottles. This might be accounted for by the different neck edge profiles, and is another possible explanation for the lack of acoustic reactivity from the MDF boxes.

#113

gullfo wrote:corners are ideal for resonant traps which support multiple frequencies - top and bottom - LF, middle - midrange. divide the sections to improve efficiencies rather than a single compartment (the differing openings / spacings will result in less effective absorption for a given set)

May I ask you to clarify this? Do you mean it's better not to stack resonant traps that are tuned to different frequencies in the same location?

In other words, if you have say, an axial mode of 34hz and a tangential mode of 78hz, is it better to keep all the 34hz resonant traps in the corners and all the 78hz resonant traps at 50% of the room's length/width/height? Or is it OK to leave some space in the corners for 78hz traps?

Just want to say a massive thank you to all who have contributed to this thread, I've learned a ton! Love this forum.

Greyhound, good luck with your build!

#114

you can definitely stack them. the idea was that having the separate chambers for specific target frequencies is generally more effective than a single chamber as that approach "averages" the overall opening whereas the individual chambers are focused.

as far as positioning - yes, targeting the modal points is more effective than (example) simple stacking into a corner -- where modes terminate is not as effective as a mid-point in the room to attenuate a length mode. however, depending on the room space, a 2ft deep box in the middle of the room might not be good either :-)

#115

gullfo wrote:Source of the post as far as positioning - yes, targeting the modal points is more effective than (example) simple stacking into a corner -- where modes terminate is not as effective as a mid-point in the room to attenuate a length mode. however, depending on the room space, a 2ft deep box in the middle of the room might not be good either

I'm still trying to get my head around this as well. You are saying it's preferable to put the HR in the middle of the wall, to address the length mode. I understand this could be true in the general case where the length and width are different, and the vertical (wall/wall) corners will see pressure maxima from two different length modes.

In my case, the length and width of the room are unfortunately the same (12'), so the corners would seem like a reasonable place to put an HR absorber targeting these two (nominally identical) room modes.

I thought that a pure length mode would have a pressure maximum all along the facing wall. I'm trying to make sure I understand what you mean by referring to a corner as the place "were modes terminate".

Thanks again for all the helpful information - it's a lot to digest but the discussion is really helpful in this regard!

#116

If you've been following this thread, you'll have noted a lot of head scratching in regards to different Helmholtz resonator neck profiles (ie sharp corners vs flared vs chamfered etc). I found time to do a few more tests, and think there might be something to this.

To recap, resonators with smoothly flared necks (eg bottles) seem to be acoustically reactive while producing minimal candle deflection (caused by the phenomenon of "vortex shedding"). Resonators with sharper hole edges (eg MDF boxes) appear to show little acoustic reactivity, but produce large candle deflections at resonance.

The holes in the MDF boxes, in addition to having sharp edges, can also be characterized as "long and narrow" in the sense that the neck length is three times the neck diameter for the configuration required to target my particular room mode.

In place of the MDF boxes, I started using capped ABS pipes with similar volumes to the MDF constructions and a machined plastic insert for the neck. This is more controllable and easier to ensure air tightness.

First, I mated an ABS pipe with a neck insert that matches the MDF constructions: 1/4" diameter x 3/4" long, with sharp hole edges. This behaves very much like the MDF boxes, in that there is very little acoustic reactivity (compared with the backside of the tube as control) but a very large candle flame deflection (ie vortex shedding).

Chamfering the edges of the 3/4" deep hole insert led to a slightly larger departure from the "control" frequency response, but it was still nothing like the bottles.

I then removed the plastic insert, and considered the hole through the ABS pipe cap as the resonator neck. This neck, with a diameter of 3/8" and depth of around 3/16", could be considered "shallow and wide" compared with the "long and thin" insert (which mimics the MDF hole). While this inevitably changed the resonant frequency, it also had the effect of greatly increasing the acoustic reactivity and also of greatly diminishing the candle flame deflection at resonance.

I then added a modest chamfer to the shallow hole in the MDF cap, and observed an even larger increase in the acoustic reactivity (keeping in mind that this changed the resonant frequency again, but only a bit). In this configuration, the candle flame deflection was completely eliminated.

Here are some photos of the neck insert (after chamfering) and the ABS hole with and without a chamfer. In all cases, a chamfer was applied to both the inside and outside ends of the hole.

And here are the frequency response curves for the various HR apertures:

What I've been calling "acoustic reactivity" is manifested as the bipolar excursions in the orange and cyan curves around 130-150Hz. The increase in reactivity when going from a long, narrow hole with sharp edges (red curve) to a shallow, wide hole with chamfered edges (cyan curve) is very evident. Another observation is that, with the chamfered 3/8" hole, it is possible to excite a pure tone by blowing across the opening (something I've been unable to do with long/narrow/sharp-edged necks).

So at this point, a reasonable working theory seems to be that long narrow holes with sharp edges are prone to vortex shedding, which impairs the acoustic reactivity of a Helmholtz resonator. MDF panels in 3/4" thickness have been a common material for DIY attempts that I've read about, and when targeting lower frequencies you will inevitably end up with the (apparently) undesirable long/narrow/sharp-edged hole. The issue is related to non-linear turbulence effects that are not modelled in popular HR calculators, which will happily predict excellent absorption properties in what I am christening the "candle snuffer" regime.

My next steps will be to adapt the ABS tube geometry to get into the frequency range of interest, and then get back into damping.

Thanks to Endorka for the suggestion of experimenting with hole inserts!

#117

Thank you for sharing all this knowledge and experiments. I'll stay tuned!

#118

nicely done. i was going to mention "ported speakers" a while back ;-)

#119

gullfo wrote:Source of the post nicely done. i was going to mention "ported speakers" a while back

Yes it seems like I've rediscovered "port chuffing" :geek:

. This is said to happen at high SPLs in ported subs and is audible. Interesting that it crops up silently (unless you're a candle) in these smaller apertures. I guess there's a higher degree of flow restriction and turbulence sets in at lower SPLs.

#120

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

Soundman2020 - Studio Design Forum