The No.1 Website for Pro Audio
4 days ago
#1
Lives for gear

Let's discuss membranes touching the absorber

Hi there,

what happens (in terms of physics and acoustics) if we have a porous absorber that is covered with a limp membrane, or rather: wrapped into a limp membrane, the membrane touching the absorber?

Let us assume the membrane would have a weight of say 0.3 to 2 kg/m² (most probably between 0.7 and 1 kg/m²) and would consist of soft material (PVC and the like).
EDIT: I do not mean steel membranes. I mean membranes from very bendable material, like PVC, roof rail, linoleum, leatherette (as examples). /EDIT

Comparision with a porous absorber:

The porous absorber is based on air velocity. The basic effect happens when air molecules are slowed down while streaming through fibre or foam, dissipating kinetic energy into heat. The air velocity is lowest at the wall, so generally speaking the first few centimeters of the porous absorber (counting from the wall) don't have much effect.
If we now now simply wrap this absorber into a membrane then it will work similar to before (so something will happen), but with additional effects: the pressure difference between the inside and the outside of this package will cause the membrane to move, and the result will be a mass-spring system with the air and the fibres in the package functioning as a spring. However if the absorber does not have a rigid backwall then the stiffness of this spring will be quite low. The membrane is heavily damped though, due to internal friction in the membrane material plus the direct contact to the fibre absorber. One more additional effect is the direct reflection of higher frequencies at the membrane.

Comparision with a (conventional) membrane absorber:
The membrane absorber usually is constructed in such a way that the membrane can swing freely. This makes the system quite resonant, with the air in the cavity acting as a spring and the absorber inside acting as a damper. In contrast to this a membrane touching the absorber should be much more dampened, plus if it is just a "wrapped cushion" without a rigid backwall then the spring in the mass-spring system should be lower. The effect should be less resonant (lower Q) but whatever happens should happen at lower frequencies.

Comparision with a VPR:
At first sight a VPR has perhaps the most resemblance to a "wrapped absorber". It consists of porous material on which a metal plate (1 to 2.5 mm) is glued, and the back of the porous material must be tightly fastened to the wall, as I understand. The sound makes the steel plate move, and the air friction in the underlying absorber dissipates the energy. However the difference is that the VPR is quite resonant due to the internal stiffness of the steel which produces lots of Eigenmodes (resonance patterns) in the steel panel. Thus in order to work for low frequencies VPRs must be quite big and are also heavy.

What do you think?

Is there any systematic treatise or measurement series known in literature? Is there any formula or calculator available?

Please, I would prefer to discuss this from a scholarly point of view.

Discuss!

Last edited by Hannes_F; 4 days ago at 09:04 PM..
4 days ago
#2

Quote:
Originally Posted by Hannes_F
Hi there,

what happens (in terms of physics and acoustics) if we have a porous absorber that is covered with a limp membrane, or rather: wrapped into a limp membrane, the membrane touching the absorber?

Let us assume the membrane would have a weight of say 0.3 to 2 kg/m² (most probably between 0.7 and 1 kg/m²) and would consist of soft material (PVC and the like).

Comparision with a porous absorber:

The porous absorber is based on air velocity. The basic effect happens when air molecules are slowed down while streaming through fibre or foam, dissipating kinetic energy into heat. The air velocity is lowest at the wall, so generally speaking the first few centimeters of the porous absorber (counting from the wall) don't have much effect.
If we now now simply wrap this absorber into a membrane then it will work similar to before (so something will happen), but with additional effects: the pressure difference between the inside and the outside of this package will cause the membrane to move, and the result will be a mass-spring system with the air and the fibres in the package functioning as a spring. However if the absorber does not have a rigid backwall then the stiffness of this spring will be quite low. The membrane is heavily damped though, due to internal friction in the membrane material plus the direct contact to the fibre absorber. One more additional effect is the direct reflection of higher frequencies at the membrane.

Comparision with a (conventional) membrane absorber:
The membrane absorber usually is constructed in such a way that the membrane can swing freely. This makes the system quite resonant, with the air in the cavity acting as a spring and the absorber inside acting as a damper. In contrast to this a membrane touching the absorber should be much more dampened, plus if it is just a "wrapped cushion" without a rigid backwall then the spring in the mass-spring system should be lower. The effect should be less resonant (lower Q) but whatever happens should happen at lower frequencies.

Comparision with a VPR:
At first sight a VPR has perhaps the most resemblance to a "wrapped absorber". It consists of porous material on which a metal plate (1 to 2.5 mm) is glued, and the back of the porous material must be tightly fastened to the wall, as I understand. The sound makes the steel plate move, and the air friction in the underlying absorber dissipates the energy. However the difference is that the VPR is quite resonant due to the internal stiffness of the steel which produces lots of Eigenmodes (resonance patterns) in the steel panel. Thus in order to work for low frequencies VPRs must be quite big and are also heavy.

What do you think?

Is there any systematic treatise or measurement series known in literature? Is there any formula or calculator available?

Please, I would prefer to discuss this from a scholarly point of view.

Discuss!
I read a bit on these things and get the impression that much is said and little is understood/investigated. But I will crawl in dust if I'm wrong.

If such a thing should do something predictabel you have to know about the weight of the steel and the parameters of the foam. And it will only be efficient in a narrow band.
I read something about modes in the plate doing someting and the placemant to waves is important.
I guess I don't understand but sound moves in steel something like 17.5 times faster than in air, so these modes will be fairly high freq, and a thin layer of foam, cloth or whaterver will suffice to absorp these anomalies.

Maybe they would be adequate as hangers in a Newell room? There absorbing stuff with a baffle inside is doing marvelous thingies, but not real.
4 days ago
#3
Lives for gear

Thank you Bert for your participation!

It helps me to be more precise in my first post: I do not mean steel membranes. I mean membranes from very bendable material, like PVC, roof rail, linoleum, leatherette (as examples).

The steel membrane may often be described as limp, but as you indicate it has quite an ability to transmit sound internally, and also it has a strong elasticity. The result is that it can be excited to resonance, while the first mentioned materials can not.

4 days ago
#4
Lives for gear

I've never read this whole thread, but i think you'll find it interesting.
Tim's Limp Mass Bass Absorbers
4 days ago
#5

It’s one of these endless threads where people constantly are reffering to other endless threads. If there is a simpele analytic solution sustained with measurements I have not yet found it.
3 days ago
#6
Moderator

With a 1kg/m² "light" membrane, if the insulation in the cavity or in front of it touches the membrane expect efficiency to collapse and Q to get very wide - depending on how much interference there is (where, how much is touching, in one or many areas and what pressure it puts on the membrane).

The heavier the membrane, the less it tends to be a problem.

A 1kg membrane using a material with reasonable behaviour (not under or over damped) with a 20cm sealed cavity would have a center frequency of around 134Hz. Not exactly a LF absorber. 6kg/m² would get you to ca. 54Hz, and be less sensitive to such interference, but certainly not immune.

Build it right!
3 days ago
#7
Lives for gear

Yes, this brings some light into the issue, much appreciated food for thought.

Quote:
Originally Posted by Northward
With a 1kg/m² "light" membrane, if the insulation in the cavity or in front of it touches the membrane expect efficiency to collapse and Q to get very wide
Why is this the case? Answer: This is both because the usual membrane absorber (with a free swinging membrane) is a highly resonant system, which makes it so effective at its tuning frequency.

Side thought, just thinking loud: If we want to tame resonant issues (modes, standing waves) then a highly resonant system works well because it has time to get into a swinging mode. But what about SBIR issues (first interactions of - mainly - bass waves with boundaries, causing nulls ((and peaks))) which are gone before a resonant system has time to form a quasi-static state? For taming SBIR issues we might need fast responding traps which handle the first wavefront, not necessarily highly effective traps that need time to get going.

It appears that for such issues it would be best to have the system critically dampened, which means: dampening without overswinging (but just enough that the system can return to zero within one wavecycle). The (resonant) overswinging may be what produces the high effectivity of tuned traps with high Q, especially when tested with sine waves, however for SBIR issues they might be too little dampened.

Quote:
Originally Posted by Northward
- depending on how much interference there is (where, how much is touching, in one or many areas and what pressure it puts on the membrane).
Yes, any force on the membrane will make its pitch higher, just like with a drum.

Quote:
Originally Posted by Northward
The heavier the membrane, the less it tends to be a problem.
Yes, however the heavier the membrane, the bigger the reflection of the first wavefront due to the inertia of the membrane.

Quote:
Originally Posted by Northward
A 1kg membrane using a material with reasonable behaviour (not under or over damped) with a 20cm sealed cavity would have a center frequency of around 134Hz. Not exactly a LF absorber. 6kg/m² would get you to ca. 54Hz, and be less sensitive to such interference, but certainly not immune.

Build it right!
Thanks for the hints, much appreciated. I knew that you would have some knowledge here!

Great replies so far, keep them coming!

Last edited by Hannes_F; 3 days ago at 03:04 PM..
3 days ago
#8
Moderator

Quote:
Originally Posted by Hannes_F
It appears that for such issues it would be best to have the system critically dampened, which means: dampening without overswinging (but just enough that the system can return to zero within one wavecycle). The (resonant) overswinging may be what produces the high effectivity of tuned traps with high Q, especially when tested with sine waves, however for SBIR issues they might be too little dampened.
Hence realistically the best membranes will have an efficiency of around 0.6 (which is plenty enough within a multilayered wall) as this is typically where the system will have the optimum damping / optimum Q. Left or right of this and you're giving something important up. There's no free lunch in Acoustics.

Quote:
Originally Posted by Hannes_F
Yes, however the heavier the membrane, the bigger the reflection of the first wavefront due to the inertia of the membrane.
This is where deciding to tune a system by increasing the cavity depth or the membrane weight can make a difference.

But in a multilayered system, it's seldom an issue.
3 days ago
#9

A membrane absorber for absorbing SBIR?
Tell me why that ain't silly.
3 days ago
#10
Lives for gear

Perhaps let us discuss this in a scholarly way, and with some grace if possible.
3 days ago
#11

Quote:
Originally Posted by Hannes_F
Perhaps let us discuss this in a scholarly way, and with some grace if possible.
I'm from Holland so I'll wait a little bit :-)
3 days ago
#12
Lives for gear

Quote:
Originally Posted by bert stoltenborg
A membrane absorber for absorbing SBIR?
Tell me why that ain't silly.
I brought up that same issue once before, and only got heated non-answers in reply... but no actual explanation as to why it wasn't silly... So, just like you, I'll be waiting too, with bated breath! Hopefully, there's a valid explanation, with some theory to back it up...

- Stuart -
2 days ago
#13
Moderator

Quote:
Originally Posted by bert stoltenborg
A membrane absorber for absorbing SBIR?
Tell me why that ain't silly.
Within a broadband multi-layered absorption system (full range system) it isn't silly.

On it's own, it is silly.
2 days ago
#14
Lives for gear

Maybe a silly question, but when referring to “multi-layered system” are we referring to broadband in front of a tuned trap? I had the idea of layering absorption like this on my back wall. Wish I had thought of it before I started my back wall, but is it a thing?
2 days ago
#15
Lives for gear

Yes it is a thing. Or rather, it can be.
Can also be two membranes as a sequence (but that needs fine-tuning).

But ... I also use it alone, or with a diffusor in front of it. Much is possible once you get into it. BTW I personally am not of the opinion that treating SBIR with membrane absorbers is silly because I see how it directly affects the frequency response. The condition is however, to treat larger areas. And whether this applies so anyone's demands depends on the circumstances, of course.

But I am not here to defend any opinion, rather I would stick to discuss it factually and physically.

Last edited by Hannes_F; 2 days ago at 07:56 PM..
2 days ago
#16

Quote:
Originally Posted by Brian M. Boykin
Maybe a silly question, but when referring to “multi-layered system” are we referring to broadband in front of a tuned trap? I had the idea of layering absorption like this on my back wall. Wish I had thought of it before I started my back wall, but is it a thing?
I hope I'm wrong, but i guess nobody will ever know, as Thomas is so professional that he will never give you any details on this.
2 days ago
#17

Quote:
Originally Posted by Hannes_F
Yes it is a thing. Or rather, it can be.
Can also be two membranes as a sequence (but that needs fine-tuning).

But ... I also use it alone, or with a diffusor in front of it. Much is possible once you get into it. BTW I personally am not of the opinion that treating SBIR with membrane absorbers is silly because I see how it directly affects the frequency response. The condition is however, to treat larger areas. And whether this applies so anyone's demands depends on the circumstances, of course.

But I am not here to defend any opinion, rather I would stick to discuss it factually and physically.
Are you suggesting you understand what Northwards means?
2 days ago
#18
Lives for gear

Bert, yes by and large I think I know. But I will keep it decent here, and as I said, with grace, so I will not reveal more than the man has said himself.

BTW if somebody wants to play around with some values (with a free swinging membrane though, not touching the absorber), this is a great tool for a first guess that I have used much:
http://www.acousticmodelling.com/multi.php
2 days ago
#19

Quote:
Originally Posted by Hannes_F
Bert, yes by and large I think I know. But I will keep it decent here, and as I said, with grace, so I will not reveal more than the man has said himself.

BTW if somebody wants to play around with some values (with a free swinging membrane though, not touching the absorber), this is a great tool for a first guess that I have used much:
http://www.acousticmodelling.com/multi.php
???
2 days ago
#20
Lives for gear

Quote:
Originally Posted by Hannes_F
Yes it is a thing. Or rather, it can be.
Can also be two membranes as a sequence (but that needs fine-tuning).

But ... I also use it alone, or with a diffusor in front of it. Much is possible once you get into it. BTW I personally am not of the opinion that treating SBIR with membrane absorbers is silly because I see how it directly affects the frequency response. The condition is however, to treat larger areas. And whether this applies so anyone's demands depends on the circumstances, of course.

But I am not here to defend any opinion, rather I would stick to discuss it factually and physically.
Thanks. I’m not sure why I didn’t think of this earlier. I read everything I could get my hands on 6 months prior to starting actual construction on my room and either missed this when it was spoken of or it’s an acoustical black ops secret. I’ll be on my way.
2 days ago
#21
Lives for gear

Brian, hardly anybody officially uses it here, as far as I know. Those who do do not talk much about it, because nobody has joy in endless discussions, except those who enjoy opposing everything the do not understand.

I had used it in my old recording room, which had only ~ 1440 cubic feet. Still the result was quite decent given the size:
How's my ETC / ITG / TDS / RFZ?

Now I am preparing a new room, so I am reviving the old lines of thought, and perhaps with more understanding now it will become even a little better.
2 days ago
#22

Dzeezz, you creeping me out, dude.
2 days ago
#23
Lives for gear

Quote:
Originally Posted by Hannes_F
Brian, hardly anybody officially uses it here, as far as I know. Those who do do not talk much about it, because nobody has joy in endless discussions, except those who enjoy opposing everything the do not understand.

I had used it in my old recording room, which had only ~ 1440 cubic feet. Still the result was quite decent given the size:
How's my ETC / ITG / TDS / RFZ?

Now I am preparing a new room, so I am reviving the old lines of thought, and perhaps with more understanding now it will become even a little better.
You have to pick through the chaff to find the gems. Sometimes that means weeding through 10 pages of egos.
2 days ago
#24
Lives for gear

I can add that my interest in limp membrane absorbers began in 2008 when I read some remarks by the venerated Eric Desart in this very forum.

This is what he wrote then:
"By NOT sealing a membrane trap it goes deeper in frequency since you weaken the spring of the mass-spring system. You only get a lower Q and somewhat extended bandwidth.
A membrane damper does NOT assume a sealed cavity, it's only easier to calculate."

limp mass membrane

That is what I do, I do not seal them.
2 days ago
#25
Lives for gear
1 day ago
#26
Lives for gear

I’m gonna bite on this. My front wall has a 16 inch cavity in it. So naturally I tried to take advantage of that. With my room being approximately 1550 cubic feet I new bass absorption was going to be my Achilles heal. So what I did was make my front wall a huge absorber with the monitors inside the absorber. The first 6 inches is OC703 followed by 16 inches of pink fluffy. I’m sorry to report I don’t have before and after REW measurements. I’ve been wading through threads learning how to sharpen my REW skills. My initial measurements were all corrupt. I’m still not sure my most recent measurements are accurate either. Anyway, decay times are getting on the dry side and frequency response is flattening out. On the side first reflections I have 7 inch broadband ceiling to floor. I’m starting to get a better idea of what bandwidth in the lowend I need to focus on. I have the room to place tuned traps behind the 7 inch broadband.

I don’t see why using this approach would not combat SBIR. I could go as far as to place tuned traps in the front wall, rear wall, and cloud in first reflections points behind the broadband covering the bandwidth of the rooms peaks and nulls. It’s a multilayered system as Northward seems to be hinting to.

Am I being naive/smoking crack, or on to something here????
1 day ago
#27
Lives for gear

Brian, yes maybe you are a bit naive but we all are more or less Without some naiveté one would not start such a business at all, so it is good.

Treating walls completely generally is a good idea. As long as there are untreated parallel parts of walls they will cause nulls in the frequency response. Also baffling generally is good. You might then want to cover those parts of your absorbers that are not in first reflection points with something (foil, wood, slats, gypsum panel) in order not to suck out all the energy from the room. Try to stay around 0.2 sec for T20 in all the spectrum if possible.

If you say, the first 6 inches are OC703, do you mean the first counted from the wall, or the first counted from the room? Anyways here is the calculation; first fluffy and then OC 703 (counted from the room) works considerably better in the bass:
http://www.acousticmodelling.com/mli...=150&v24=16500

Putting bass membrane absorbers behind broadband coverings generally is a good idea. If we do not care about the bass then our treatment will be lowpass treatment, plain and simple.

Quote:
Originally Posted by Brian M. Boykin
I’m gonna bite on this. My front wall has a 16 inch cavity in it. So naturally I tried to take advantage of that. With my room being approximately 1550 cubic feet I new bass absorption was going to be my Achilles heal. So what I did was make my front wall a huge absorber with the monitors inside the absorber. The first 6 inches is OC703 followed by 16 inches of pink fluffy. I’m sorry to report I don’t have before and after REW measurements. I’ve been wading through threads learning how to sharpen my REW skills. My initial measurements were all corrupt. I’m still not sure my most recent measurements are accurate either. Anyway, decay times are getting on the dry side and frequency response is flattening out. On the side first reflections I have 7 inch broadband ceiling to floor. I’m starting to get a better idea of what bandwidth in the lowend I need to focus on. I have the room to place tuned traps behind the 7 inch broadband.

I don’t see why using this approach would not combat SBIR. I could go as far as to place tuned traps in the front wall, rear wall, and cloud in first reflections points behind the broadband covering the bandwidth of the rooms peaks and nulls. It’s a multilayered system as Northward seems to be hinting to.

Am I being naive/smoking crack, or on to something here????

Last edited by Hannes_F; 1 day ago at 05:40 PM..
1 day ago
#28
Lives for gear

Quote:
Originally Posted by Hannes_F
Brian, yes maybe you are a bit naive but we all are more or less Without some naiveté one would not start such a business at all, so it is good.

Treating walls completely generally is a good idea. As long as there are untreated parallel parts of walls they will cause nulls in the frequency response. Also baffling generally is good. You might then want to cover those parts of your absorbers that are not in first reflection points with something (foil, wood, slats, gypsum panel) in order not to suck out all the energy from the room. Try to stay around 0.2 sec for T20 in all the spectrum if possible.

If you say, the first 6 inches are OC703, do you mean the first counted from the wall, or the first counted from the room? Anyways here is the calculation; first fluffy and then OC 703 (counted from the room) works considerably better in the bass:
http://www.acousticmodelling.com/mli...=150&v24=16500

Putting bass membrane absorbers behind broadband coverings generally is a good idea. If we do not care about the bass then our treatment will be lowpass treatment, plain and simple.
This is an abbreviated post to try and cover my knowledge base.

I’m aware of parallel walls.

I read quite a good deal of Floyd Toole. I decided on a Davis and Davis approach of heavily treating the first half of the room and first reflections. Seemed the most practical for my level of understanding and budget. However, I believe my room is more inline with a no environment though.

I’m also aware of fluffy before OC703 but for aesthetic reasons and construction reasons ignored it. It was an educated compromise.

I’m also aware that large amounts of broadband will suck the highs out. Front wall corner traps have the foil facing the room. Subsequent traps on side walls will be constructed in the same manner. Once all the broadband is up I plan to add slats for scattering the highs that have rolled off. I plan to use a 5344..... design starting at the roll off frequency. I’ve found info on this forum for slat width to frequency.

What attracted me to this thread is understanding tuned traps better. I was not sure if I had the knowledge base to build my own and should maybe leave them to those who understand the math and just buy from a manufacturer.

Obviously with a small room where you don’t have time and distance to help with the reflection times, absorption is your friend.

So question: because the lower the frequency the more spherical the sound moves away from the source, will treating first reflections with tuned traps really help in controlling the low end? It seems it could, but also seems that it could not.

And I think this is the question at hand with using tuned traps to control SBIR.

500hz and below were not dealing with laser beam frequencies so the waves spread out like throwing a rock in a pond. All those lows will find there way back to the listening position unless the entire surface area of the room is treated with traps.

1 day ago
#29
Lives for gear

Yes you are hijacking the thread but it is down the drain anyways, so if it serves you then at least it is of benefit to someone.

Quote:
Originally Posted by Brian M. Boykin
This is an abbreviated post to try and cover my knowledge base.

I’m aware of parallel walls.

I read quite a good deal of Floyd Toole. I decided on a Davis and Davis approach of heavily treating the first half of the room and first reflections. Seemed the most practical for my level of understanding and budget. However, I believe my room is more inline with a no environment though.

I’m also aware of fluffy before OC703 but for aesthetic reasons and construction reasons ignored it. It was an educated compromise.
But trust in physics is trust in life.

Quote:
Originally Posted by Brian M. Boykin
What attracted me to this thread is understanding tuned traps better. I was not sure if I had the knowledge base to build my own and should maybe leave them to those who understand the math and just buy from a manufacturer.
Which might be a great idea.

Quote:
Originally Posted by Brian M. Boykin
Obviously with a small room where you don’t have time and distance to help with the reflection times, absorption is your friend.

So question: because the lower the frequency the more spherical the sound moves away from the source, will treating first reflections with tuned traps really help in controlling the low end?
No, at least not completely, and that is not at all what I am saying. All that I am saying is, if you have to absorb, then better do it broadband.

You can control the low end only with these two options (in combination):
1. Absorb standing waves where the modes are the most intensive
2. Treat the whole wall (edge to edge, top to bottom) for the low end if you want an even frequency response.

Quote:
Originally Posted by Brian M. Boykin
And I think this is the question at hand with using tuned traps to control SBIR.
500hz and below were not dealing with laser beam frequencies so the waves spread out like throwing a rock in a pond. All those lows will find there way back to the listening position unless the entire surface area of the room is treated with traps.
Exactly. So if the wave finds an untreated spot (read: untreated for bass) of wall or ceiling then BAM you will get a null, even if would not expect it in a laser beam picture.
1 day ago
#30
Lives for gear

Quote:
Originally Posted by Hannes_F
Yes you are hijacking the thread but it is down the drain anyways, so if it serves you then at least it is of benefit to someone.

But trust in physics is trust in life.

Which might be a great idea.

No, at least not completely, and that is not at all what I am saying. All that I am saying is, if you have to absorb, then better do it broadband.

You can control the low end only with these two options (in combination):
1. Absorb standing waves where the modes are the most intensive
2. Treat the whole wall (edge to edge, top to bottom) for the low end if you want an even frequency response.

Exactly. So if the wave finds an untreated spot (read: untreated for bass) of wall or ceiling then BAM you will get a null, even if would not expect it in a laser beam picture.
Got it. Your work here is done with me. I know what I need to do. Back on topic. Thanks for the small hijack.

## Welcome to the Gearslutz Pro Audio Community!

###### Registration benefits include:
• The ability to reply to and create new discussions
• Interact with VIP industry experts in our guest Q&As
• Get INSTANT ACCESS to the world's best private pro audio Classifieds for only USD \$20/year
• Remove this message!
You need an account to post a reply. Create a username and password below and an account will be created and your post entered.