Triple-leaf question for the experts

Eric_Desart wrote:

Rod, We not often disagree, but here I clearly do.

Eric, thanks for the education - I stand corrected and learned something new today - this has been a good day indeed. Sincerely, Rod

Eric, > I'm very sure that small leaks in Ethans design will not influence attenuation that much. I even believe it's not even worth it to seal them so accuratly < I don't doubt that. When I was selling a commercial version of those traps it occurred to me that the fiberglass inside could be replaced by a controlled amount of air leakage. In a tuned electrical circuit the Q is lowered by adding a resistor to the circuit. The wood panel is the same as an inductor, and the air trapped inside is equivalent to the capacitor, so adding a leak intentionally is like adding a resistor. Besides lower the Q and absorbing over a wider range, a leak will also help damp the panel's vibration to avoid "after" ringing. But it was easier and cheaper to keep building them as we knew worked, versus spending time and money to experiment with controlled leaks. --Ethan

I have some analagous data. What you see in the graph is the result of punching about 2" holes in one side of a steel stud wall to "vent" the cavity. Basically no change in the behavior at the MSM is observed. a steel stud wall is not the best choice of wall for this kind of experiment. In walls with steel studs, resilient channel, sound clips, or spring hangers used for decoupling, the studs or channel/clips/hangers provide some spring force which affects the location of MSM. This is most pronounced in channel and steel stud walls. The best choice of wall for this test would be a double stud wall. With respect to the seal quality on the panel traps, it should take only a few seconds to deliberately change the seal quality from very well sealed to poorly sealed and another minute to test noise reduction again. I'm going to post a proposed assembly at studiotips. I'll post it here as well.

rod gervais wrote:

Eric, thanks for the education - I stand corrected and learned something new today - this has been a good day indeed.

Rod many thanks for your kind words

Ethan Winer wrote:

But it was easier and cheaper to keep building them as we knew worked, versus spending time and money to experiment with controlled leaks. --Ethan

Yes Ethan, But you used panel traps as a part of a broadband absorption approach(set of 3 combined types), before the MiniTraps. Hence people didn't/don't bother about exact tuning. But for a huge amount of panel traps I see in groups, people try to target them to certain problem frequencies, In those cases you, and others refer to your related page and further to the standard formula integrated in the Master Handbook of Everest. As became more clear lately based themselves on an over-simplified formula, assuming that perfect sealing and respecting that formula should solve their problem. Well it doesn't. They have a fixed inflexible device, and only luck will have it tuned exactly. That formula assumes an:

1. infinite stiff or heavy backing;
2. an empty cavity;
3. a plain wave
4. at straight incidence;
5. the panel seen as a mass which freely moves as a piston.

Basically it's nothing else than a simple calculation of a mass-spring system where all those real life factors hardly come into play. For walls it's simple: Get MSM as low as practically possible. For panel traps to solve problem frequencies the aim is: Get it as accurate as possible. Important is that your design was just meant as a low frequent absorber (part of a whole room treatment), not exactly defined to which frequency it is tuned. Therefore the mini changes (if any to speak off) by caulking them or not can not be rated in better or worse. Therefore in most cases it hardly has sense to caulk them. On the net they interpreted/interpret it as some water tightness they need to obtain, simultaneously taking any 'after tuning' possibility away, (wrongly?) assuming that by following the rules and a simple formula they create exactly what they targeted. Based on some recent discussions at Studiotips I come to a shift of a factor of ca 0.85 to 1.00 for the resonance frequency ranging between filled and empty cavity. (still unlucky and uncertain with my own approach) John in another thread shows that the way the panels were fixed influenced the MSM. Hence your design is OK as part from your original goal (room treatment) but your related articles and plans, and the numerous references in groups are used as guidelines for a panel trap as a concept without reference to all these uncertainties, and the often occurring need for tuning. Hence in this very thread you said that if the outcome of Brian's tests (which I'm grateful for and very interested in) will show an influence on TL you'll make a notice of that in your articles. But then why don't/didn't you make such a notice already for standard double leave systems where you have a panel in front of a single wall? While the effect of a third leave is interesting in this context, for double leaf systems there's no doubt whatsoever. If your question is about influence on TL, then why only discussing triple leaf with a bit strange behavior, and leaving out double leaf with much clearer behavior? As I noted in this thread, I get the feel if the effect of a triple leaf principle is minor that that becomes another axiom: A panel trap hardly influences TL.

..., and only luck will have it tuned exactly.

What about creating the panels with a higher theorical center frequency and then installing a tuning/damping device like on a snare drum ? Alternativly, a sharp panel could maybe be tuned by affixing a certain amount of roofing felt on the membrane, down to pitch. Using a contact microphone a blast of pink noise and a spectrum analyser should help to fine tune for a specific frequency. My feeling is some commercial membrane traps (RPG Modex) works on this principle, maybe using an ajustable spring ? Any other though on fine tuning those beasts ?

Eric, > Hence people didn't/don't bother about exact tuning. < Those traps are pretty "broadband" when fitted with insulation as shown in my plans. Each style (low, high) covers about an octave, so there's a fair amount of tolerance built in. More to the point - and I've built three different studios using those traps - they work very well for their intended purpose as bass traps. Could a different design work even better? Maybe. There are dozens of bass trap designs including all the variants, and I have direct experience measuring only a few of them. > if the effect of a triple leaf principle is minor that that becomes another axiom: A panel trap hardly influences TL. < Well, if it turns out that adding a series of high-bass and low-bass panel traps on a standard sheet rock wall does not reduce transmission loss by much, then "hardly influences" is the correct conclusion, no? --Ethan

Ethan Winer wrote:

Eric, > Hence people didn't/don't bother about exact tuning. < Those traps are pretty "broadband" when fitted with insulation as shown in my plans. Each style (low, high) covers about an octave, so there's a fair amount of tolerance built in. More to the point - and I've built three different studios using those traps - they work very well for their intended purpose as bass traps. Could a different design work even better? Maybe. There are dozens of bass trap designs including all the variants, and I have direct experience measuring only a few of them. > if the effect of a triple leaf principle is minor that that becomes another axiom: A panel trap hardly influences TL. < Well, if it turns out that adding a series of high-bass and low-bass panel traps on a standard sheet rock wall does not reduce transmission loss by much, then "hardly influences" is the correct conclusion, no? --Ethan

You only read what you want to read don't you .....?????? In fact that low bass trap showed a very high 1 1/3 octave behavior, and how high that peak exactly was is hidden by the way a lab measures as I mathematically simulated at Studiotips. My comments were NOT that these traps didn't function in relation to the purpose of alternative broadband absorption. Hence you don't need to repeat what I said myself. But I contribute for others. I DO know your way of reasoning. I know you're looking for a new slogan, which I why I reactivated this thread after your first reply to Brian, which was more than enough answer for you already. How people interpreted your pages in function of solving problem frequencies and what a panel trap is all about, is easily to check all around the net, including here at this forum, even in this very thread.

Eric, > You only read what you want to read don't you .....?????? < Are you not able to discuss anything without insulting me? > In fact that low bass trap showed a very high 1 1/3 octave behavior < Okay, we'll call it 1 1/3 octaves. And your point is? > how high that peak exactly was is hidden by the way a lab measures < That has been my point for many years now. Unless you measure an absorber at a resolution much greater than 1/3 octave, it's impossible to identify what it's really doing.

My comments were NOT that these traps didn't function in relation to the purpose of alternative broadband absorption. Hence you don't need to repeat what I said myself. But I contribute for others. I DO know your way of reasoning. I know you're looking for a new slogan, which I why I reactivated this thread after your first reply to Brian, which was more than enough answer for you already. How people interpreted your pages in function of solving problem frequencies and what a panel trap is all about, is easily to check all around the net, including here at this forum, even in this very thread.

Blah blah blah. Please let me know when you're ready to carry on a civil discussion about the science of audio. Thanks. --Ethan

Ethan Winer wrote:

That has been my point for many years now. Unless you measure an absorber at a resolution much greater than 1/3 octave, it's impossible to identify what it's really doing.

Please explain what I exactly was referring to (I not you), before telling it was your point for years.

Eric wrote: "I'm very sure that small leaks in Ethans design will not influence attenuation that much. I even believe it's not even worth it to seal them so accuratly with caulck or silicone as often described, assuming they are made relative good. And the shift in frequency caused by such minor leaks is probably smaller than the uncertainties involved in the whole construction of the trap versus his background, stiffness, wool filling etc. I've build traps, wittingly spaced from the walls (without backing) to get my frequency low enough. I DO use this shortcut of the air enclosed by the trap and its frame with the environment to tune them. " Ok dumb question, but would then that be a Hybrid device at that point? Sorry just was reading Rods book and read your post almost at the sametime.. Very cool idea, BTW. Glenn

Glenn, For me a panel trap in the context discussed here refers to a mass-spring system where the mass is a panel resonating on a spring which is the air layer. For me there is nothing hybrid about what I describe. I cause a panel to absorb at the resonance frequency of that panel. That for me is a panel trap. Ethan's design is a design of a panel trap, not THE design of a panel trap making other things a hybrid of the real thing and some diversions. I don't want to go into the use of the word trap here (disputed by John before), for me one can call it absorber, trap, whatever.

myfipie wrote:

Very cool idea, BTW.

Glenn in fact I tell the same thing for years already, only when pushing it seems to have some effect. You interpret the idea as cool, because since you came on the net the panel trap idea/concept was mainly defined in groups by pages and articles from 1 author with enormous penetration on the net. And for me the sealing obsession, when I read all these discussions and absolutes about it, were just one of these typical axioms from the first moment I saw them. There's nothing cool about my idea, it's just applying basic principles as I did from long before these articles existed. I know on what these formulas are based. I know the uncertainties. I am confronted with these uncertainties from the beginning, knowing it's difficult to grasp them all theoretically. Hence for me the logical thing ALWAYS was having some possibility to tune/adjust these things, when it mattered. And in basics there are only 2 things to work with, that's the mass or the spring or a combination of both. I don't make a hybrid. I make a panel trap, that's a physical principle/concept, not a predefined object or construction. Ethan's plans are an approach, not THE approach.

Just so I understand what you are saying about this kind of panel. Basically you would build it the same way as EW wood panel trap but leave the back open to space it off the wall to what ever frequency you want? If that is right how you do know how to space it to get a curtain target frequency? Sorry I did not mean to label that panel, but found it funny that I was reading Rods book about Hybrid panels just about at the same time I read your post. :lol: Glenn

myfipie wrote:

Sorry I did not mean to label that panel, but found it funny that I was reading Rods book about Hybrid panels just about at the same time I read your post. :lol:

Glen, remember - my book discussed velocity and pressure devices - the hybrid reference was in regards to treatments that work on both principles.......... Plywood faced panels are pressure devices regardless of whether they are 100% sealed or not........ any of the Helmholz " traps" are also pressure devices - and they couldn't be sealed and work properly........ Just something to keep in mind......... Sincerely, Rod

Rod wrote: "remember - my book discussed velocity and pressure devices - the hybrid reference was in regards to treatments that work on both principles.......... " It seemed like the panel Eric was talking about was on both principles, is it not? Like I said I was sorry for labeling them, and I really do not care what label you put them under, just thought it was a great way to build a panel. BTW Rod, really a great book. I bought a copy before, but really the first time I have had the time to read it all the way through. Very easy to understand. Can not wait for the next one!!!! Oh and thank you for page 215 thru 219 :D :D :D Glenn

myfipie wrote:

Just so I understand what you are saying about this kind of panel. Basically you would build it the same way as EW wood panel trap but leave the back open to space it off the wall to what ever frequency you want? If that is right how you do know how to space it to get a curtain target frequency? Sorry I did not mean to label that panel, but found it funny that I was reading Rods book about Hybrid panels just about at the same time I read your post. :lol: Glenn

Glenn, A panel trap for me can be a lot of things. It can be mounted in between studs which are also used to mount standard absorption, which then are covered by continues fabric, hence you can't recognize or see them. It can be panels laid/mounted between the bottom flanges of I or H beams, it can look as the Ethan plans. It all depends. John also referred to an application he described here. How to know? First calculate it as good as possible theoretically, where by definition for closed panels I start from the wall MSM formula as entered in that Bob's thread, I adjust it listening to my belly knowing much more than I do, and depending on the situation. I ONLY should use the traditional formula (Everest and others) on very heavy walls, NEVER on drywall. I take care that I easily can adjust things. Hence if I don't integrate some standard posibility, I mount it temporary in a way I can easily adjust things until happy with the results. I once tried to aim for about 32 Hz (if recalling correctly) and then gave up, because once it was tuned correctly and damping high enough, it hardly absorbed anymore (was a costly, time consuming experiment). And I have the advantage of doing narrow band sound AND vibration measurements, but with single walls, which I have more experience with here than double walls you can also see it with narrow band sound measurements on the other side of the wall too (DIP in isolation)**. Hence you measure it with whatever and however it's practical. I easily can imagine that with pure sinus tones scanning the relevant frequency range you can find it back as well. It's like playing a sweep in a living room. Once you hit the resonant frequencies of different objects they start rattling/vibrating. Only with a sweep you don't know exactly the frequency (passing through). In DSSF3 sits a very extensive signal generator with some fun possibilities. In the freeware version a lot of function stop working after 30 days but the signal generator remains functional (is meant as freeware). http://www.soft32.com/download_13520.html ++++++++++++++++++++++++++++++++++++++++++++++++ ** For triple leaf systems the TL method is too uncertain too my taste (not enough experience with. The first time I was confronted with the TL behavior of triple leaf systems by adding an additional layer on the outside of a wall was between '80 to '84 somewhere (can't remember exactly). I helped people from a company called Agglorex which produces this bonded foam. The idea was to start using it as resilient skin walls to be glued to existing walls (hence the only connection is the foam itself). There are also mineral wool versions of this principle. The logical test back then was to check also (note that brick walls are standard here in private housing) how this should behave on drywall to improve TL (offices and utility buildings are made a lot here with non-load bearing drywall as separation). There I was first confronted with this strange triple wall behavior. In general the curve remained comparable with some dips at certain frequencies and others a bit better. Overall it became worse. But the traditional behavior one could expect from a theoretical stylized mechanical MSMSM system was far from recognizable. But the conclusion was: Don't do it. It doesn't work in function of TL. But that behavior isn't that clear for walls. This triple leaf system however does work nicely as the theory predicts on decouplers. I once worked for a factory which sold such decouplers mainly/often for elevator constructions (a lead block in between springs). It did what it was expected to do: Increasing high frequent insulation significantly, but it simultaneously increased the MSM frequency, which wasn't a problem since located far enough below the frequency of interest.

Thanks Eric for your reply Glenn

Glenn, You're welcome. I think it's worth to play with this signal generator. It's the most extensive I know (and free). You can e.g. scroll a pure tone back and forth, really scanning a limited area and searching. You can do a lot with that thing, and it's really fun to play with. You can phase and time shift two speakers, have a piano keyboard in that generator. to play correct musical frequencies and so on. Eric .

So, discussion closed? ;) After reading I couldn't help myself thinking; "There should be a new and improved article/book about trapping/absorption". (or something, u get the idea) The accoustics faq on EW's site, also discussing paneltraps etc, is a very nice article. Helped me have some more inside. But as Eric mentioned, it's one side of the story. And who are we to pick a side. It's good stuff to discuss and debate about (without calling eachother an ass) So with input from more than just one pers, a dedicated article/thread/faq whatever should be written. idea? 2 cents