|
Post by redrum on Jun 1, 2017 17:26:09 GMT
Hello! I have a question about the specs over time of your membrane traps. You’ve suggested numerous times that the thin membrane itself is glued to the fiberglass panel, but that you are not 100% sure (as far as I was able to find online) as to why the membrane actually extends the low end absorption. I have two theories (my actual questions are beneath): A. The non-air-transparent thin membrane “bonds” together the entire surface of the fiberglass panel, turning the fiberglass into (in part) a panel absorber. Here the weight of the whole device becomes an important factor. Also, the panel shouldn’t be squeezed by the supporting frame. Or, B. The glue and the fiberglass become the “spring” system, which applies opposing force to the movement of the membrane. This would make the operation more similar to a limp mass absorber. In this case the depth of the fiberglass panel, and how tight the frame is are not very critical factors. Question(s) - 1. A or B? 2. 1. If the system is closer in operation to the B theory, wouldn’t the stability over time of the glue be very critical? Have you tested it say, 5 years after production? 3. If it’s closer to my A theory, wouldn’t the surface size and weight of the panel (and mounting) be critical? I see you have different sizes on offer (e.g. MiniTrap), but the absorption measurements are the same AFAIK. 4. What’s most confusing is that (after watching your MiniTrap production video), it seems to me that the thin membrane is not glued to the fiberglass panel at all, but only to the cover fabric - which rests freely on top of the fiberglass panel. I hope I am not being too annoying, I’ve been exploring acoustics for the past 6 months and it’s completely taken over my life Thank you!
|
|
|
Post by Ethan Winer on Jun 1, 2017 18:30:56 GMT
1. I'm leaning more toward A because a mass-spring system would be resonant. Even with a membrane the absorption of MiniTraps and MondoTraps falls off at low frequencies, with a curve similar to plain rigid fiberglass. 2. I haven't taken five year old MiniTraps back to IBM's lab for testing, but I can't imagine the glue would weaken enough over time to harm performance. I have several dozen MiniTraps in my living room home theater, and all are more than ten years old. The sound in this room is still great so I assume they still work well. 3. It's tough to see trends of size versus the absorption "peak" frequency because acoustic labs measure in third octaves which isn't fine enough. Maybe if we had also measured 2x2 MiniTraps, half the size, we'd spot a trend. But thickness is also a factor, so comparing MiniTraps to MondoTraps won't work because there are two variables - the length and the thickness - and both dimensions are probably too close to separate given the limited lab resolution. 4. The membrane has glue on both sides, so it bonds to both the fabric and to the rigid fiberglass. No, this is not annoying at all!
|
|
|
Post by redrum on Jun 1, 2017 19:06:35 GMT
Nice to know that my line of thinking is OK, I was afraid that what I wrote is just gibberish 1. I'm leaning more toward A because a mass-spring system would be resonant. Even with a membrane the absorption of MiniTraps and MondoTraps falls off at low frequencies, with a curve similar to plain rigid fiberglass. Hm, I probably have my terms wrong, but with the "A" theory, which you say is probably more correct, I had a panel (diaphragmatic) absorber in mind - which is also resonant.... And you say that A is probably correct because there's no resonance. 1. Could you please pinpoint the source of my confusion and correct my course? Aren't both diaphragmatic and limp mass systems resonant (one more than the other)? 2.Are you saying that your traps are neither, because your membrane just "enhances" the absorption characteristics already "built-in" the fiberglass? 3. More importantly, what you said regarding weight and size being a factor is really relevant IMO... IF the membrane indeed makes the whole "system" turn into (in part) a diaphragmatic / panel absorber, then the size and weight of the system would drastically change the absorption characteristics... 4. On the other hand, if it's because of the glue/fiberglass system applying resistance to the limp membrane movement, then the size/weight of the ENTIRE system wouldn't matter - but it would be EXTREMELY sensitive to type of glue, membrane and fiberglass. OK, I understand that I'm maybe in over my head here, and overcomplicating things Any help would be appreciated Many thanks!
|
|
|
Post by Ethan Winer on Jun 1, 2017 19:32:07 GMT
1. I'll guess that the membrane "catches" more of the wave, and the rigid fiberglass damps the membrane. 2. Yes. We experimented with several materials and thicknesses, and what we ended up with worked the best. I honestly don't know what else to tell you other than my standard advice: Buy a bunch of materials, run REW sweeps in a small room with different materials and arrangements, with the microphone in the exact same place, then come back here and show us what you found.
|
|
|
Post by redrum on Jun 1, 2017 19:40:26 GMT
I honestly don't know what else to tell you other than my standard advice: Buy a bunch of materials, run REW sweeps in a small room with different materials and arrangements, with the microphone in the exact same place, Perfectly valid, I'll be doing a lot of that in the coming week, as I have access to a really good room for testing purposes. Nice thick walls, etc.. Was a butcher's shop refrigerated (huge) room originally then come back here and show us what you found. I'll show you mine if you show me yours Just joking, you've been very generous, thank you again!!!
|
|
|
Post by Ethan Winer on Jun 5, 2017 14:23:05 GMT
Remember that it's best to use a small room, so adding / moving only a few panels makes enough difference to measure the changes reliably.
|
|
|
Post by redrum on Jun 6, 2017 14:12:37 GMT
Remember that it's best to use a small room, so adding / moving only a few panels makes enough difference to measure the changes reliably. Yes, thank you. When I said the room is "huge", I meant it's huge for a refrigerator It's actually around 20 m2, so not that huge actually. I was thinking around 20 panels (1m x 0.6m) should be enough for testing. Cheers!
|
|