deanm
New Member
Posts: 38
|
Post by deanm on Aug 19, 2018 22:20:29 GMT
FIRSTLY, thanks for the existence of this forum. New user here, but I've stalked Ethan for a few years now. I'm finally getting around to sound treating my composing/mixing room - it's pretty small (10' x 12' x 8'), so I'll be probably only use broadband velocity bass traps. Planning on: 1 - 12" of fluffy pink suspended from the entire ceiling 2 - super-chunking the front left and right corners with fluffy pink (about 25" wide by 15" deep (at the sides) with an outer face of prolly about 36" -- I'll consult Pythagoras for the real measurement, if it matters 3 - 4" 705 (or equiv.) at L and R early reflection points, mounted 2" off wall 4 - 4" 705 (or equiv.) traps between the speakers and front wall, for SBIR in the rear wall, there is a closet (28" deep) in the left 2/3rds and a doorway out in the right 1/3rd, so: 5 - 4" of 703 (or equiv.) in the opening of the closet, so I'd end up with 4" porous in front, with 24" air/storage behind 6 - chunk the rear right corner with 4" 705 (chunk would deploy outwards as the door closes) SECONDLY - So, I bring this plan to you, esteemed gurus, and ask - does it make sense to do this? Can you think of better plans? Any pitfalls that I cannot see in my lack of experience and wisdom? The closet vs. door in the rear wall throws the symmetry off a little. Is it enough to worry about? I think I might end up with a "dead" room. (BTW, the floor is carpet, I cannot change that.) So I may end up putting a membrane over some of the ceiling. And THIRDLY - according to the AC calculator at www.acousticmodelling.com/multi.php - the ceiling and closet traps will give me great absorption (near 1) down to about 200Hz. The closet trap also gives good (.6) absorption at 47Hz (my fundamental length axial mode), and the ceiling trap gives .6 at 70Hz (fundamental height axial mode). Anyway, I was left unsupervised for too long, and I got thinking to myself: what if I put a panel trap tuned to 47 Hz at the back of the closet? A "trap within a trap", to further diminish the axial fundamental -- Has this approach been tried? I imagine it has, but I haven't been able to find anybody's experience with it. Ineffective? Overkill? Maybe the panel trap would be better against the front wall, vertically down the middle, between the SBIR traps? Thank you for your thoughts and time, -Dean p.s. - FOURTHLY - my wife says "hi" to Ethan's cat.
|
|
|
Post by Hexspa on Aug 20, 2018 0:36:30 GMT
Hey, deanm and welcome. I'm glad you're here with the rest of us Ethan stalkers. Let's get to it.
Your room is small. Forget about making it dead, focus on making it better. In most of our cases, deader is better.
Symmetry matters most from your ears forward. A bit of irregularity behind your head isn't the end of the world. Still, we can position your planned treatment better.
I say put your rigid super chunks in the front corners. Firstly, they're heavier so, if you have to move them, it'll be easier to put the fluffies in the rear. Secondly, I believe the front of the room is where you want to maximize absorption.
It actually does matter how wide your super chunks' faces are. Actually, their total mass matters a lot. You can use 15" insulation, which it seems like you will, but 24" is better.
Don't be stingy with the 4" panels. I have 20 rigid panels in my relatively medium sized room and it's barely enough. In your tight space, throw out the cat and displace it with more panels. Then again, you do have 100% coverage of your ceiling coming so you will have to take measurements to verify the need for more.
We don't talk about pressure absorbers here, or at least I don't. I advise you to stick with the itchy stuff before moving on to harder options. Lastly, acoustic simulations are good but putting stuff in and taking real measurements is probably better. Do you have a linear response condenser microphone with which to take samples?
Thanks.
|
|
deanm
New Member
Posts: 38
|
Post by deanm on Aug 20, 2018 3:12:31 GMT
Hi Hexpa, Thanks for the quick reply and info. There are only 2 superchunks planned - the fluffies I'm thinking about putting in the front corners. That's where I have the most room available for them. The other chunk I mentioned will be a 4" panel of 703 or 705 that attaches to the door, like this: www.youtube.com/watch?v=MkZgBc_LLaMSo I'd do something similar to this on the rear right corner, floor-to-ceiling. The fourth corner (left rear) is in the closet, which is already planned to become a giant velocity trap. The superchunk faces (in the front) truly will be about 36" wide - that was a close approximation with a tape measure between the two imagined points on the corner walls. I haven't done the math to get the numbers exact. (I'll do that, of course, before I begin construction.) They'll extend inwards along the front wall to abut against the SBIR traps. I'll prolly keep them at 4"thick there (the inner "corner"), to match up with the SBIR trap. I'll do a diagram or sketch-up or something of how I envision the room layout, and upload it to clarify. Not doubting or challenging, but fishing for knowledge: how is 24" better than 15" for the superchunk depth? merely 12" of fluffy pink gives nearly 1.0 A.C. down to 200 hz according to: www.acousticmodelling.com/multi.php (300 mm pourous layer, resistivity = 5000) ....and even down to 125 hz according to: www.bobgolds.com/AbsorptionCoefficients.htm (although there's some confusion here because of A.C. > 1) Actually, I think I see your point: I just ran the numbers for 15" vs 24" on the AC calculator, and they're very close.... UNTIL you get down to 50Hz. Then 24" has a measurable advantage. It also just struck me that we're not just talking about the single edge increasing from 15" to 24". It would increase the depth of the entire chunk considerably, giving me more area where the distance from the face to the wall is at least 12". Okay. 24" thick superchunks makes sense. (how did I arrive at 15", anyway? I wonder...) I like hearing you say "deader is better" - that's reassuring to me. I'm for absorption. As I said, if I do manage to overdo it with the highs/mids, I can add a membrane to a few places on the ceiling. I have a *really* funky shower curtain that is begging to be recycled, after a long dip in boiling penicillin. The cat inclusion is probably not negotiable, but she's pretty absorbent. She actually transforms some of the low frequency energy into higher-frequency meows. These should, in theory, be easier to mitigate. I do have a decent omni condenser calibration mic (Dayton EMM6). I'll try to upload some REW measurements this week. The question about pressure traps really boils down to: has anyone here tried combining a pressure trap with a velocity trap (like, in the same real estate), and if so, how did it turn out? It was one of those Ideas I Had In The Shower, but I def wouldn't do this without measuring the room first, with the itchy stuff up. The more I think it through, there would have to be a compelling reason to do this, rather than just put the pressure trap on its own piece of wall. So it's prolly just an academic question. Unless, after fluffing the ceiling, the modal ring at 47Hz was unbearable, and I didn't want to build a pressure trap on the floor. Sorry if this is long-winded, dude, but it really helps me think it through. Thanks again, -d
|
|
|
Post by Hexspa on Aug 20, 2018 20:11:20 GMT
No problem, deanm. Have you seen my progress logs? I know all about thinking out loud. We've just been over modal ringing in the first audible octave in this thread here. Pretty much, no respected source cares or makes much of an issue of it. Everyone gets the same advice to use velocity absorption and you can read that thread as to why.
|
|
deanm
New Member
Posts: 38
|
Post by deanm on Aug 22, 2018 12:36:49 GMT
Ok, so I've been doing a crash course on sketchup. Hopefully, this makes my ideas for the chunk and superchunks more clear. Let me know what you think. This is just a 2d top-down floorplan, so it doesn't show the fluffed ceiling, but it'll be there. I'll do more velocity traps on the side walls behind me, too, to tame flutter. I've been looking at the specs for OC 703 and 705 (from the OC product data sheet), and also using the AC calculator at www.acousticmodelling.com/multi.php to compare the two. The 705 seems, at best, to be only marginally better than 703, and it's sometimes worse. For example, the datasheet indicates that the AC for 4" of 703 (unfaced) at 125Hz is 0.51, while 705 is 0.60 - only .09 better, for about 1.5 times the price. Does this seem like it could be true? I've read many places about how "705 is vastly better!" and "you *need* more density to handle low frequencies", etc., but the numbers I'm digging up just don't support it. I realize that 125Hz is not a LOW frequency, musically speaking, but the calculator's range goes down to 20Hz, and its predictions don't support the argument for higher density either. Here are the resistivity numbers I've been using: Fluffy pink = OC Fluffy pink = about 5000 KPa.s/m2 OC 701 1.5 pcf = 24 kg/m3 = 8.000 MKS Rayls/m = 8.000 K Pa.s/m2 OC 703 3.0 pcf = 48 kg/m3 = 16.000 MKS Rayls/m = 16.000 K Pa.s/m2 OC 705 6.0 pcf = 96 kg/m3 = 30.000 MKS Rayls/m = 30.000 K Pa.s/m2 I don't remember for sure where I found these - they're not off the OC data sheet, so they could be wrong, and that could be what's throwing the calculator predictions off. Please shoot holes in this, and correct me if I'm off - I wish to learn, and to get this room tamed as best as I can - but correct me with numbers, or personal experience with both 703 and 705, or some sort of scientific reason why one's better than the other, rather than "I've always heard/read that..." I really wish I had enough money to try both, and measure each. -d
|
|
deanm
New Member
Posts: 38
|
Post by deanm on Aug 22, 2018 12:59:23 GMT
another question I wanted to ask about...
the HR824 has a rear-facing passive radiator for the low end. they should still be okay close up against the SBIR traps, right? my unsure understanding/reasoning goes like this:
low frequency waves propagate nearly omni-directionally, so it doesn't make much difference that the radiator faces the back instead of the front.
I only want, ideally, the sound that reaches me before it echoes off something else. so anything I can absorb behind the speaker before it bounces off the front wall is good.
So this is the "right" thing to do, as long as I'm not completely blocking the sound from getting around the cabinets...
...right?
-d
|
|
|
Post by Hexspa on Aug 22, 2018 15:56:30 GMT
I, and probably most of us, have also opened many tabs in search of the Acoustic Truth. Luckily for you, I might save you some time. Let's start with your closet. Can you leave it open and treat closer to the rear wall? You really aren't going to benefit much from a 4" panel spaced 7' from a boundary. In my experience, it will affect the respective mode but only to a small degree. I suspect you're better off treating the open closet just like a miniature small room. Either that or fill the whole thing with fluffy. Now let's talk about decoupling. Based on tests Ethan has done, decoupling is hype. Besides, all that crap on your desk will probably absorb any resonances if there are any. You'll figure something out with that big head of yours lol. So that swinging panel by the door isn't actually a chunk, right? It's just an angled panel. It has to be 'filled in' to be a chunk; hence 'chunk' and not 'slice'. Gas Flow Resistivity for absorption is always a range in real life. You can't find two sources with the same numbers. Even Cox/D'Antonio say that to find the actual absorptive values for any given specimen, you have to measure that specific one. They use tubes and stuff, don't even try it. You don't need to go this deep into it. Ethan demonstrated how 705 is better at low frequencies. 705 does outperform 703 in some applications. Even so, if you can double your coverage and/or thickness using the less dense insulator then you'll be better off than with the inverse. I've already been reamed here once about ports. Suffice it to say, ports are basically solid at or above their resonant frequency. Below that, air gets out but I don't think you're going to blow your whole project because your port location. SBIR is going to happen no matter what. The goal is to move your speakers so close that the frequency is higher and thus more treatable or to move your speakers so far that the mode is lower than your speaker puts out. More info on ports and their placement:
If your mains cross over at 120Hz then just over 8' gives you pretty much no fundamental interference. If I understand correctly, then your SBIR will affect the upper harmonics of the frequency and will be less strong. Just laugh if I'm wrong. Anyway, I suggest you read, or reread, Ethan's articles. Most, if not all, of your questions are handled there. I have a fraction of his experience and these are relatively technical questions. While I have studied and practiced a bit in this area, I prefer to keep it simple - make a bunch of panels, move them around till they sound good. If you get a problem then resort to math later. lol.
|
|
deanm
New Member
Posts: 38
|
Post by deanm on Aug 22, 2018 22:12:13 GMT
Lol - yeah, it's in my nature to overthink and invent problems. Regarding the closet - it's only 1'8" deep, not 7'. Aside from my head, the drawing is to scale. The 7' refers to the ceiling height. (oh, man, i WISH i had a 7' deep closet... I've got 4 decades' worth of crap that I can't seem to throw away...) So that half of the rear wall will look like 4" of rigid, with 20" of air behind it, which gives good numbers on the simulator. Regarding decoupling - ok, got it. Probably not much to be gained by stressing over this. I have some old slabs of packing foam that I currently use - good enough. Also - on the subject of the desk - and this is probably just an academic question, since I probably can't do anything about it anyway, but... Are early reflections off the top of the desk an issue? If yes, any solutions? just pile on more random crap? I wonder if musical keyboards make good diffusers. hmmm... if the depth of the grunge between the black keys was piled up in a prime number relationship... Regarding the back corner - ok. New term learned: That'll be a slice back there, not a solid chunk. Anyway - any better ideas on how to treat the back corner? I'm going to extend the ceiling fluff back to the doorway, but there's a smoke alarm in the way that I cannot move or obscure. (HOA rules - I live in a condo.) So I'll work around the alarm, but I won't get full coverage on the ceiling back there. Sounds like you and I are in agreement w/ the density question. I'll most likely go with 703, since I'm closer to being able to afford that. Additional points of view are welcome. The HR824's passive radiator (it's a sealed diaphragm, not a port, if that makes any difference) takes up nearly the entire back side of the box, largely obscured - but not blocked acoustically - by the amplifier. While the amp doesn't block the bass, it probably deflects and redirects a lot it out the rear of the sides. This was my takeaway from SBIR too - glad to hear that we agree. So the planned placement should work pretty well. I'll leave at least as much clearance between amp and wall as Mackie left between the radiator and the amp. I have / do read Ethan's articles. As far as I can tell, I'm following them, and following his advice, but I lack confidence. I'd be interested to hear what he has to say about my plan - whether that's corrective, or reassurance that this looks like a good way to start. Same with advice from any additional gurus out there with enlightenment, or encouragement. Thx again Hexspa. Your input is encouraging. -d
|
|
|
Post by Hexspa on Aug 23, 2018 0:14:34 GMT
Cool. I hope someone else can chime in. As far as I'm concerned you're good to go.
Cheers.
|
|
deanm
New Member
Posts: 38
|
Post by deanm on Oct 11, 2018 5:14:16 GMT
This project oozes forward... Attaching the untreated SPL and waterfall images. It sucks about like I'd expected. This is both L and R channels; mic is more or less at mix position, and monitors are close to what's shown shown in the floorplan I uploaded earlier, although probably farther from front wall than shown. Here's what I'm guessing: The peak at 46Hz matches up with the 12' length fundamental mode. (mix position is about 38%, which is pretty close to quarter-wavelength.) The null at 57 is from being in the middle of the the 10' width. The Z axis (8' ceiling) fundamental mode is predicted to be about 70Hz. I can't tell if this is a peak or a null here... ..Probably because it's being overshadowed by the mother of all nulls at 78Hz. This one doesn't correspond to a predicted mode. Wavelength is about 14' half wavelength is 7', which is about the distance from mix position to rear wall... but if it's a reflection (traveling 14' back to mix position), that would make this a peak, not a null. Highest peak is at 127Hz, Wavelength = 8.9', half wavelength = 4.45'. This also is not a predicted mode. Maybe SBIR off front wall, maybe reflections off side walls, maybe reflection off ceiling? Combination of all 3? I've got a 35 dB valley-to-peak difference within an octave. Whee! Oooh hey, so maybe that null at 78Hz is also being caused by whatever's creating the peak at 127Hz. (78 * 1.5 is close to 127). So maybe my highest peak and lowest nulls are caused by the same reflections off side walls and ceiling (and possibly front wall, too) all ganging up on me. One thing I did notice that I think is good: the SPL graphs between solo Left and solo Right are similar. So I think this indicates pretty good symmetry. Anyway, this is all academic. Broad band bass traps are needed - we knew this already. But, the academia does interest me. Can anyone help me with my guesswork? Thanks, -d Attachments:
|
|
|
Post by rock on Oct 11, 2018 13:31:17 GMT
As far as guessing goes, if you haven't already seen this site; amcoustics.com/tools/amroc?l=12.13&w=9.94&h=8&ft=true&r60=0.6I have your dims entered. You'll see a little 3d room graphic below the room modes to help you visualize the pressure. Looks like you still have a lot of ringing too. Maybe extend your graph display out to get more time but either way, you're right, you need more LF absorption.
|
|
deanm
New Member
Posts: 38
|
Post by deanm on Oct 11, 2018 16:53:01 GMT
Thanks rock - I agree with everything you said.
I've been using amroc to see which peaks/nulls are modal. very useful tool. it also helps me understand why every song ends up playing back in F#: my fundamental resonances are F#, sharp A or flat A#, and C#.
I have lots of ringing and flutter. I've looked at the graph extended out to 600ms. it didn't get any prettier.
|
|
|
Post by Hexspa on Oct 12, 2018 2:19:45 GMT
It's not that bad. My room looked like that too. Put in some panels, play with their positioning, then polish with some low frequency EQ.
|
|
deanm
New Member
Posts: 38
|
Post by deanm on Oct 12, 2018 5:55:23 GMT
Hexpa - yup. That's the plan. It can (probably) only get better.
To all Gurus:
I need some advice having to do with pressure absorption. I realize the "best bang for your buck" approach is focusing on broad-band velocity absorption, which is what I'm doing. I think I may also be able to get some pressure absorption for free (or just a few dollars more) out of the construction of the ceiling velocity trap, if I'm clever.
It seems to me (PLEASE correct me if I'm off here) that the simplest theory of a passive pressure trap is: 1 - get something that resonates sympathetically with a problematic frequency 2 - damp it
And when you do this, you take a portion of the sound energy at that problematic frequency, and convert it to heat.
Then, to improve the trap's efficacy: 3 - put it at a node for that frequency
Right?
Here's what I'm thinking. I'm building the framework for the ceiling trap, and I have these 2x4s suspended such that the bottom is 11.5" under the ceiling. They span from the left wall to the right, and are attached only at the ends. I haven't put any crosspieces in them yet. So, they resonate at 10 Hz when I thump them, moving frontwards/rearwards, for several seconds.
So I notice this and think, okay, I'll put in crosspieces to make it non-resonant.
"Hey, Dean," my brain says. "What if you don't? What if, instead, you: (step 1) find a way tune the 2x4s to 46 Hz (where your room rings the longest), and then (step 2) damp them by gluing them to the fiberglass? Even though you can't put them at the nodes (step 3), won't this give you *some* extra benefit, at little to no extra cost?"
So I ask you, Gurus, on behalf of my brain: 1. Could this steal some of the energy from my 46 Hz mode, converting it to heat, thus making the modal ringing problem better? 2. If I don't damp it enough, would it become another resonant mechanism to store the energy and release it over time, making the ringing worse? 3. Does gluing it to a lot of fluffy fiberglass damp it enough?
And/Or, maybe I'm missing some crucial understanding of pressure traps? I'm learning it as I can, but I'm unsure.
Thanks in advance, -Dean
|
|
|
Post by Hexspa on Oct 12, 2018 18:14:08 GMT
46Hz ringing don't really matter that much, bud. Ethan draws the line at 50Hz and the EBU standard for studios is like 65Hz.
|
|