Login / Register
 
My Big Soffit Trap Results
New Reply
Subscribe
pleventi
Thread Starter
#1
5th March 2012
Old 5th March 2012
  #1
Gear interested
 
Joined: Sep 2011
Posts: 22

Thread Starter
pleventi is offline
My Big Soffit Trap Results

I've (finally) started to add accoustic treatments to my 11.2 HT room. In this post, I'll share the results of adding two large soffit/bulkhead bass traps (16x18" Roxul Safe N Sound, 28 feet worth) to the room. Both quantitatively and qualitatively, the results were very noticeable -- but a little disappointing when it comes to <100 Hz.

Me
I'm a computer engineer, and tend to fall into the "if I can't measure it, it's probably in my head" camp, plus I only care about measureable differences that actually are noticable qualitatively. I'm a minimalist when it comes to speaker wires, power cords, etc. But I've seen enough evidence posted here and elsewhere that accousitc room treatment can make a major difference. I've read a lot of threads here and elsewhere, as well as the Master Acoustics textbook. I'm no expert, but I've learned a little bit.

My Room & System
The room is roughly 15x17x12' (irregular) room in basement, used primarily for home theatre use -- nearly 100% movies. You can see irregularity of room in photos below. There is a 4' wide hallway that opens into the room; I may eventually go all-out to make the perfect HT room by closing this off, but for now other functions and asthetics/light rule out that possibility. The room has laminate-over-concrete floors, drywall-over-studs-over concrete on three sides (with standard insualtion), and the only absorptive elements in the room (for now) are a leather sofa and a small area rug in front.

I'm running a 11.2 setup with Paradigm Studio 60 fronts, CC-590 centre, Studio 20 wides, SE1 for heights, ADP-590 surrounds, SA-25 in-wall rears, and 2 SVS PC12-PLus subs. Definitely overkill, but figured I'd have some fun; depending on how I like the Audyssey DSX effect, I may end up removing the hegihts/widths and using elsewhere -- but so far I like it. A/V is a Denon 4311ci, and amp is a used B&K 7250 I picked up used for a steal.

I use Audyssey XT32 correction with 8 positions, but that is disabled for these tests. I believe in room equalization, though it can only fix some problems, and doesn't help with reverb. It is the final step in prepping a room.

My Goals
I'm not looking for a perfect room. Given the constraints of my room, and some placement constraints relating to aesthetics, build a room that sounds good without breaking the bank on acoustic treatments. From a stats perspective, I'm looking for a RT60 in the 200-300 ms range flat to within +-20% across frequencies, and ideally no horrible nulls (> 10 dB) in the FR plots. Everything else shouldbe EQ treatable.

Treatment Plan
See pictures below for front and back picture & plan, and left and right sides of the room. I will be adding 18x21" soffit/bulkheads to front and back wall-ceiling boundaries, containing 16x18" Roxul Safe n' Sound (40 kg/m^3, pretty light for rockwool). Left and right front wall-wall boundaries will have 18x18" pillars containing 16x16" Safe n' Sound from 4' to the ceiling. In those corners are my sub-woofers, which will be hidden behind acoustic fabric. I will also be eventually adding a 9x5x1' riser to the back of the room, and configuring it as a broadband bass trap. Depending on how things go after Phase 1, I may add additional soffit/superchunk treatment to rear wall-wall corners or perhaps straddle panels on side wall-ceiling boundaries.

I will also be adding FRZ treatments to front, side and rear walls. Exact type and location TBD; this will be subject of a later posting.

Measurement Methodology
I used REW (fantastic software) with a calibrated ECM8000 from cross-spectrum labs, a USB Creative SoundBlaster X-Fi 5.1 sound card, and a ART Tube MP mic amp. I have marked the locaiton of my front speakers, sofa, and boom stand with tape in the event I need to reposition, and subs are in fixed locations. I use a 2 sweeps from 20-200 for LF and 20-20000 for HF, with mic pointed at ceiling. All equalization (Audyssey, dynamic EQ, etc) is disabled for measurements.

For my full tests before and after, I measured in three positions -- left seat, center seat, and right seat -- without moving the mic stand and relying only on rotation of the boom arm to improve reproduceability. I measure the center position twice (first and last) to see how much variability there is due to small differences in mic position.

During my soffit build, I did all the pre-work until filling the insulation. Then I performed measurements with no insulation, with the back soffit filled, and then with the back & front soffit filled. I did not move the mic at all in between (measuring only the center seat). This eliminates any positional variability from these measurements.

Results in next post.
Attached Thumbnails
My Big Soffit Trap Results-front_plan_lowres.jpg   My Big Soffit Trap Results-rear_plan_lowres.jpg   My Big Soffit Trap Results-right_before_lowres.jpg   My Big Soffit Trap Results-left_during_lowres.jpg  
pleventi
Thread Starter
#2
5th March 2012
Old 5th March 2012
  #2
Gear interested
 
Joined: Sep 2011
Posts: 22

Thread Starter
pleventi is offline
Results (2 Soffits Only)
Prior to treatment, the room has nasty slap echo. If I clap my hands or sing, it's really noticeable. When watching movies, bass is muddy, dialogue is hard to make out when mumbly (syllabels run together). You can see this in the RT60 and waterfall plots. Also, note the really nasty 38 Hz null.

After treatment, the room sounds better to speak in, forget listen in. In movies, dialogue is much crisper; deep bass effects (fireworks, explosions) sound much better. My wife, who thinks I'm a little nuts, immediately noticed the difference unprovoked. I was frankly surprised that the difference would be so audible.

Quantitatively, you can see the before, during (rear soffit only) and after (2 soffit) results below. All are from centre seat. FR doesn't show much difference (some flattening in the 100-300 Hz range). Waterfalls and RT60 show improved decay, from 70 hz up.

Discussion
Huge difference to RT60 measurements, and some visible difference to decay in the waterfalls. It seems that most impact was limited to 100 Hz+, which is a little disappointing given the depth of these traps. But that said, the audible difference in the room is stellar, and for the $300 or so I has cost me, is the best audio upgrade I've made. Visually, I'm happy with how things look, with judgemnet reserved until I've actually covered the traps with acoustically transparent frames.

Help!
I can't help wondering if I would have had more success with a lighter insulation material given the trap thickness (16-18" deep traps, currently using Roxul Safe N Sound at 40 kg/m^3). If you have any informed opinions, shoot away.

I'd love to do something about my 38 Hz and 80 Hz nulls, since these will not be EQable. They are strongest in the center seat, but still ~13 dB in the left and right seats. Would a tuned helmholtz resonator help here (nulls still confuse me at times), or am I stuck with moving the subs?

Next Steps
(1) Add the front left and right columns (16x16" insulation). Re-measure.
(2) Play with sub-woofer locations. I'd really prefer to keep them in the corners for asthetics, but if I find a much better spot, may change my mind.
(3) Place thicker traps behind front sound stage, and 2-4" absorbers for FRZ off front speakers. Reevaluate. Likely going with GIK for these.
(4) If I've killed mids and highs too much, add a plastic membrane to the soffits/pillars or pick a less acoustically transparent fabric.
(5) Deal with back wall treatment as final step once I've seen where I stand. Lots of options there -- absorption, diffusion, have some fun with a DYI VPR, tuned absorbers,

PS I've attached my REW files for the No Soffit, 1 Soffit, and 2 Soffit test. If interested in any of the other cases, let me know.

PPS Thanks to all the knowledgable and experienced folks who share on these forums.
Attached Thumbnails
My Big Soffit Trap Results-rt60_t30_before.jpg   My Big Soffit Trap Results-rt60_t30_2traps.jpg   My Big Soffit Trap Results-spl_together.jpg   My Big Soffit Trap Results-spl_full_smoothed.jpg   My Big Soffit Trap Results-waterfall_before.jpg  

My Big Soffit Trap Results-waterfall_1trap.jpg   My Big Soffit Trap Results-waterfall_2traps.jpg   My Big Soffit Trap Results-decay_before.jpg   My Big Soffit Trap Results-decay_2traps.jpg   My Big Soffit Trap Results-spectro_before.jpg  

My Big Soffit Trap Results-spectro_2traps.jpg   My Big Soffit Trap Results-front_soffit_installed_lowres.jpg   My Big Soffit Trap Results-rear_during_lowres.jpg  

Last edited by pleventi; 5th March 2012 at 07:23 AM.. Reason: (Added after shot of traps)
#3
5th March 2012
Old 5th March 2012
  #3
Gear Guru
 
Glenn Kuras's Avatar
 
Joined: Jul 2005
Location: Atlanta, GA
Posts: 16,209

Glenn Kuras is online now
Quote:
(1) Add the front left and right columns (16x16" insulation). Re-measure.
That will help

Quote:
(3) Place thicker traps behind front sound stage, and 2-4" absorbers for FRZ off front speakers. Reevaluate. Likely going with GIK for these.
I would go with thick broadband on the back wall or something tuned.
GIK or Realtraps?

Quote:
(4) If I've killed mids and highs too much, add a plastic membrane to the soffits/pillars or pick a less acoustically transparent fabric.
That would help.

Quote:
(5) Deal with back wall treatment as final step once I've seen where I stand. Lots of options there -- absorption, diffusion, have some fun with a DYI VPR, tuned absorbers,
See number 3. You could put a mix of absorption and diffusion.

Over all you are on the right track. Just keep in mind that the low lows are going to take something tuned or a lot of broad band.
__________________
Glenn Kuras
GIK Acoustics USA
GIK Acoustics Europe
http://www.gikacoustics.de (German Translation)
404 492 8364 (USA)
+44 (0) 20 7558 8976 (Europe)

Built in Slat design (Scattering/Diffusion) on all Bass Traps click here
#4
5th March 2012
Old 5th March 2012
  #4
Gear interested
 
Joined: Oct 2011
Posts: 23

wgb113 is offline
Materials

What materials/construction did you use for the soffits? I'm getting ready to make some out of the Roxul Safe & Sound and keep changing my mind as to what type/size wood to use to construct the frame and how to hang them on my existing wall.

Bill
pleventi
Thread Starter
#5
5th March 2012
Old 5th March 2012
  #5
Gear interested
 
Joined: Sep 2011
Posts: 22

Thread Starter
pleventi is offline
Thanks for the advice Glenn. And Bryan, who has helped me offline as part of my fine-tuning leading to an order. Assuming I don't go DIY for the FRZ (I'm pretty bad at clean fabric work), I'll definitely be going GIK.

As for Bill's question on construction technique: I built the soffit frames using 2x2 finger-jointed pine, lined with thin plastic (painter's drop sheet from Home Depot). I screwed and glued the joints, and assembled the L-shaped frame including plastic lining on the bottom prior to hanging it. Used Roxul Safe N Sound 16" for Steel Studs (16.25" x 48" x 3") lying on its side to stuff the trap. And plan to finish the traps with 1x2" frames, attached with velcro.

Detailed instructions:
  1. Cut 3 2x2s to length; for a 8' section, no cuts necessary. These will be the back, top, and front-edge pieces. Be sure to pick pieces with relatively little warpage (look down the length with your eye). If it is warped, make sure the curved direction is up or down, not out, when placed in the soffit, otherwise it will hard to make your soffit square.
  2. Cut 8 2x2s for the bottom braces; in my case, these were 16 3/4" long (for a soffit that will be 20.5" deep fully assembled)
  3. Cut 5 2x2s for the front braces; 14 3/4" long (for soffit that will be 18 1/4" tall fully assembled)
  4. Screw and glue the bottom braces to the back and front-edge 2x2s at 3", 16", 32", 45", 51", 64", 80", and 91" along the length, using 3" wood screws. Easiest way to do this was mark and drill the 3/32" holes on the two lengths, screw in all the braces on one edge piece, then flip over and screw the other edge piece on, each time lining up the pre-drilled holes with the center of the brace. This way you know you're building an approximately rectangular frame.
  5. Screw and glue the front brace to the front-edge and top 2x2s at 1", 24", 48", 64", and 95". A tighter spacing is a little sturdier, but makes getting the insulation in a pain.
  6. Line the inner bottom with very thin plastic sheeting, stapling it into place reasonably tightly. This both contains the rockwool fibres, and makes it easier to slide in.
  7. To make installation easier, attach to small 2x2 pieces to the wall such that their top edge is 17.75" from the ceiling. This will serve as a temporary perch for the back-edge of your soffit. These should be screwed into studs, which you will want to mark out at this time to make screwing later easier.
  8. Measure and mark a line on the ceiling for where the top-edge will be; this should 19.75" from the wall. This way you can be sure your soffit is straight when you screw it in (the frame will flex a fair bit). At this time, you may want to determine stud locations in the ceiling, and measure where these are.
  9. Pre-drill and partially screw in some 3" screws at the pre-determined stud locations into the top-edge and back-edge of the soffit for attaching to your ceiling and wall. This way you can install it without a helper -- just rest on the perches, screw in a few screws, then do the rest.
  10. Lift soffit into place, rest on the perches, and screw top-edge into studs/ceiling, ensuring that the front aligns with your pre-marked line. Screw it in a bunch (I used about 8 screws top and back).
  11. Repeat for other soffit sections; make your soffits 0.5" shorter than you measure just so you have some play room for non-square walls, etc. When you place multiple soffit segments together, don't forget to staple plastic between them.
  12. Fill with insulation. I purchased Roxul Safe N Sound 16" for Steel Studs (16.25" x 48" x 3" sheets). One bale of insulation has 12 sheets, which at six sheets per 48" (lying on their side) fills the soffit. If you need to cut it, use a rockwool saw ($5 serrated blade) or a long box cutter knife with a metal ruler. It cuts quite easily when compressed.
  13. Staple plastic onto the front face of the soffit to hold in the fibres, and trim plastic.
  14. Build frames with 1x2s to cover the bottom and front of the soffit. Cover with acoustically transparent fabric (for broad band) or pretty much any fabric (for bass only). This adds another ~0.5" to the width and height of the soffit. I haven't tried this yet; hopefully it works out!
  15. Attach frames with velcro to soffit. Remember to add a few fabric loops you can pull on to detach the frame in future if needed.

When building segments shorter than 8' and longer than 4', use 3", 16", 32", and 45" brace placement for the long "half" of the soffit, enabling use of insulation without cutting. For the odd "half", be sure to put a brace at 3", N-3", and somewhere in between for adequate support.

If you want a shallower trap, I suggest adjusting depth in multiples of 3" since that's the size of the rockwoll sheets. Safe N Sound is also available in 15 1/4" wide x 47" sheets (for "wood studs"), and 23 1/4" and 24 1/4", so those are convenient dimensions to play with.

For maximum insulation, you could fill the 1.5" deep gaps between the stringers with more fluffy / wool / cotton batting.

BTW, you may be able to get away with fewer braces on the bottom -- the rockwool panels are stiffer (standing on their sides) than I expected.
#6
5th March 2012
Old 5th March 2012
  #6
Lives for gear
 
kasmira's Avatar
 
Joined: May 2011
Location: Phoenix, AZ
Posts: 1,269

Send a message via AIM to kasmira
kasmira is offline
I imagine a huge factor to why you aren't getting results as expected is because three of your walls are concrete. Bass frequencies will reflect off concrete much more than they would through standard drywall rooms, which explains your extreme nulls.

Did you use all of your speakers or just the subs on the waterfall test? You may be able to help those nulls dramatically by placing your subs in different places in the room.

Edit: Also, on top of that, it seems that you placed your soffits against the ceiling, which isn't concrete, and on the walls in the back by the hallway, which I'm assuming isn't concrete as well. The most beneficial spot for them may have been the corners that are concrete-concrete corners. I'm not very certain on this, but it may be possible.

They look great. Hopefully you can experiment a bit to get some better results. I have my insulation to build my soffits but don't have the wood yet. I will be making a similar thread with my experience as well.
pleventi
Thread Starter
#7
5th March 2012
Old 5th March 2012
  #7
Gear interested
 
Joined: Sep 2011
Posts: 22

Thread Starter
pleventi is offline
Not much I can do about the walls. They are concrete + studs + drywall with 4" insulation. The "front" wall and "back" wall (where the soffits are) is concrete most of the way up, but probably by the time you get near the top where the soffits are, switches to wood frame. Technically, my soffits are against a soffit (!) not a ceiling, since there was already a soffit there for some HVAC and plumbing; I briefly contemplated opening up the existing soffits and stuffing them with insulation (instead or in addition), but only possible for the back wall, and was going to be too messy.

I was using front L speaker + 2 subs for the waterfall, with crossover at 60 Hz. I've also done subs only and had (as you would expect) similar results. I previously had the subs both at roughly 1/4 along front wall, with similar results (though I'd have to dig it up). Now that the room is treated and my listening position is pretty much finalized, I will re-test sub locations. I may be able to hide one sub in the back left corner instead of front left.

Lets see how things go as I add more treatments.
#8
6th March 2012
Old 6th March 2012
  #8
Lives for gear
 
kasmira's Avatar
 
Joined: May 2011
Location: Phoenix, AZ
Posts: 1,269

Send a message via AIM to kasmira
kasmira is offline
In response to the wall configuration, I was just suggesting a better corner to treat the sub freqs may have been the two front corners, as both of those corners would be concrete to concrete (on the back side of the wall at least)
#9
6th March 2012
Old 6th March 2012
  #9
Gear interested
 
Joined: Oct 2010
Posts: 12

Paul Spencer is offline
Really good to see all the info there, pics and measurements. Very helpful for others to see and learn from.

Firstly I see that you've chosen a window of 600ms. One can dramatically change the impression with the settings for a waterfall. Set both time and window to 300, then 600 then 1200 and you will see the same data result in 3 quite different impressions!

I also see that you have a bit of a problem at 40 Hz. This is where I'd start measuring each sub individually in different spots and looking for a location that avoids that big dip you have. When you find a couple of spots that look like they could work together, then you measure them together. It's also possible that you can create dips caused by phase issues where mains and subs interact.

Is it possible that you could try shifting the listening position? It looks quite close to the front wall. Then again, maybe I'm being misled by the wide angle lens because it looks different in different photos!

Regarding your front wall corner traps, here is an idea that you might not have considered. I'm doing a similar thing - building very large traps which will look like a studio soffit speaker mount. In my case I'm building horns that will sit flush in a bass trap at 30 degrees and about 1.2m wide. Looks like you could do a similar thing. You build them around your speakers so they sit flush, and then you get a very big trap that works down low.
__________________
AUDIO BLOG .:::. Writer for Hifi Zine

Red Spade : www.redspade.com.au
creative web design studio
pleventi
Thread Starter
#10
6th March 2012
Old 6th March 2012
  #10
Gear interested
 
Joined: Sep 2011
Posts: 22

Thread Starter
pleventi is offline
Wide angle lens mess with your mind :-) I've attached an overhead layout of the HT room. (EDIT: I've changed it to reflect current layout)

The seating position really can't change much -- maybe forward 1' at most -- especially since I plan to add a projector eventually.

On the subs, I'll do some measurements separate from the main (which shouldn't change the 40 Hz null), with them ~1/4 along front wall, and ~1/4 down the side wall, and finally with one in the back-left corner (though I'm worried about this giving a bass massage to the occupants there!). I'd really prefer to have 'em in the corners for asthetic reasons, but that 40 Hz null is crazy deep and wide.

BTW, it is the Front, Right, and Back walls that have concrete under the drywall.

Quote:
Really good to see all the info there, pics and measurements. Very helpful for others to see and learn from.
Thanks! I don't see many true before-and-after comparisons; usually people have moved equipment, mics, furniture, etc. Even when I measure my room today, it doesn't read the same way as my last measurement -- I've moved one box that was sitting behind the couch, and someone bumped the mic so I had to carefully reposition it based on my tape markings, and the temperature & humidity are different. When I post more results tonight, you'll see how much things moved around.

Quote:
Firstly I see that you've chosen a window of 600ms. One can dramatically change the impression with the settings for a waterfall. Set both time and window to 300, then 600 then 1200 and you will see the same data result in 3 quite different impressions!
The limited resolution of REW means causes me to loose too much resolution beyond 600 ms window, and 300 ms window just doesn't capture the resonance length in my room. Haven't played with time much. Besides showing me a different absolute view, is there are reason to play with this?

I was mostly concerned with showing an apples-to-apples relative comparison before and after, so I figured as long as I used the same settings I was fine.

Quote:
Regarding your front wall corner traps, here is an idea that you might not have considered. I'm doing a similar thing - building very large traps which will look like a studio soffit speaker mount. In my case I'm building horns that will sit flush in a bass trap at 30 degrees and about 1.2m wide. Looks like you could do a similar thing. You build them around your speakers so they sit flush, and then you get a very big trap that works down low.
I have a window on the right front wall -- if I want a symmetric setup (for visual reasons), then I'm limited to about 20" deep on that right corner. Otherwise I would have gone with 34" faced SuperChunk style angled traps. But as it is, I would have had to start it 34" then drop to 24" face at the window. Also, the wall-ceiling boundary is complex in that corner, another reason I'd be limited on depth on the upper portion.

If I end up with my subs not in the corners (e.g. roughly 1/4 along front wall), I may revisit my corner traps. First I'd just stuff 18"x18" into the corners where the subs were going to go, and if I'm still not happy, I may got 34" SC for the first six feet and then switch to 18x18" pillars from there (uglier, but more effective).

Lets see where I land after further bass trapping, and some sub placement!
Attached Thumbnails
My Big Soffit Trap Results-hometheater.jpg  
pleventi
Thread Starter
#11
6th March 2012
Old 6th March 2012
  #11
Gear interested
 
Joined: Sep 2011
Posts: 22

Thread Starter
pleventi is offline
Added left & right soffit/pillars

The Traps
Two 6' tall 16x16" Roxul Safe N Sound (40 Kg/m3), straddling the front right and front left corners, below my existing soffit traps. 4' left below to (potentially) house my two SVS subs. Construction is simple: Two 16x16" pine shelves in each corner, and 16.25" x 48" batts cut into 16.25"x16" squares and stacked. Outside clad in painter's drop sheet (thin plastic), stapled to shelves and wall. I'll eventually cover it in L-shaped frames made of 1x2" pine with accoustic fabric covering.

Measurments
Unfortunately, I took the 3- and 4-soffit measurements on a different day (humidity & temp) than the 0- 1- and 2-soffit measurements, and after the mic had been bumped and repositioned. So it is not perfect apples to apples.

New decays, waterfalls, and SPLs below. Also a comparison of 0- vs 2- vs 4-soffits. Let me know if you want more data.

Observations
4 dB improvement in 80 Hz null, and 7 dB improvement at 106 Hz, and 7 dB at 190 Hz. FR generally much smoother from 75 to 125 Hz.

The waterfalls look worse, but probably due to slight change in preamp settings for mic (note that overall SPL is higher). I've also included waterfalls with a 600 ms window in order to capture more LF detail.

Shifted my low null by 2 Hz; may be by reducing magnitude of one of the two nulls that are combining to form that deep null, or by change in room dimensions. I wish I could easily compute the "area" of the null since that would tell me if it just shifted or if it also got narrower (which it looks like)
Attached Thumbnails
My Big Soffit Trap Results-front_3soffits_installed_lowres.jpg   My Big Soffit Trap Results-rt60_t30_4traps.jpg   My Big Soffit Trap Results-spl_together_4_soffit.jpg   My Big Soffit Trap Results-waterfall_4traps.jpg   My Big Soffit Trap Results-waterfall_2traps_600ms.jpg  

My Big Soffit Trap Results-waterfall_4traps_600ms.jpg  
pleventi
Thread Starter
#12
6th March 2012
Old 6th March 2012
  #12
Gear interested
 
Joined: Sep 2011
Posts: 22

Thread Starter
pleventi is offline
About that 38 Hz Null

I investigated that 38 Hz null a little. First, my measurements to date have been 2 Subs + Front Left speaker and center (prime) HT seating position. Crossover set to 60 Hz in my A/V receiver, speaker set to "small".

Measurements
  1. Measure subs and main independently. You can see that in the first SPL chart. Solid black is combined measurement, other curves are for each speaker individually. One of the subs, in the front left corner (LC), is getting some -39 dB @ 36.7 Hz null action. The front right sub (RC) is getting -25 dB @ 41.7 Hz, plus some null at 37 Hz. The front left speaker is also affected by -17 dB @ 36.7 Hz null.
  2. Move sub around in the left corner a little to see if that helps. Second graph shows a few positions 14-42" away from corner. Pretty much anything other than right in the corner has similar performance, with a 10 dB reduction of the null and pretty similar response elsewhere. To properly investigate, I need to measure multiple listening positions, or do a "sub crawl".
  3. Try offending sub in back-left corner. Third graph shows this. No stimulation of the 37 and 42 hz nulls, but now 28 Hz null shows up and severe nulls at 100 and 123 Hz. Those upper frequency nulls are probably ok since speakers will dominate that far from the crossover.
  4. Measure front corners + left speaker compared to front right corner + back left corner + left speaker. Sure enough, the lower nulls are solved and there's some pretty flat FR to 100 Hz. But crazy nulls now at 100 and 190 Hz. EDIT: This doesn't make much sense, since the left speaker is unchanged; debugging in progress (maybe I've changed some setting???)

Discussion
Width of the front of my room: 185" = 15' 5" = 36.6 Hz. This explains why (a) I am sitting in a null in my central seating position, and (b) why the left and right seats see part but not all of the null (not shown here). It also explains why corner-loading experiences this null heavily. It doesn't explain why the right sub doesn't see the null as strongly as the left sub, but then again the room has irregular dimensions.

Based on this hypothesis, treatment on the front left and right wall-ceiling/floor boundaries could help. Except it is 36 Hz, so no amount of fluffy is going to touch that. Could try a tuned solution, such as a Helmholtz panel resonators on the left and right upper walls near the corner -- I have nothing else going there. Any opinions on if this would help?

Moving my sub seems like a good idea, but it doesn't solve the problem that I'm seated nearly perfectly in a room null. And other locations seem to create new problems. Suggestions welcome!
Attached Thumbnails
My Big Soffit Trap Results-together_vs_seperate.jpg   My Big Soffit Trap Results-three_corners.jpg   My Big Soffit Trap Results-move_left_sub.jpg   My Big Soffit Trap Results-two_cases.jpg  

Last edited by pleventi; 7th March 2012 at 12:05 AM.. Reason: Note discrepancy
pleventi
Thread Starter
#13
7th March 2012
Old 7th March 2012
  #13
Gear interested
 
Joined: Sep 2011
Posts: 22

Thread Starter
pleventi is offline
After playing around Whealy's Porous Absorber calculator, I may pull out the Safe N Sound from one bulkhead and replace with fluffy. 40 kg/m^3 rockwool has a flow resistivity in the realm of 15k rayls/m2 -- based on results for other rockwools of similar density. For an absorber of the depth I'm using, probably something in the 5k range would be more effective. That will be my next experiment; I'll be sure to measure carefully before and after.
#14
7th March 2012
Old 7th March 2012
  #14
Gear interested
 
Joined: Oct 2011
Posts: 23

wgb113 is offline
What about just facing the bulkhead with S&S and leaving an airgap behind? If it's still not sufficient fill the gap with pink fluffy. If it still isn't improving things go all pink & fluffy.

What about measuring +/- a foot or so forward/back from that center seat see if that works?

Bill
pleventi
Thread Starter
#15
7th March 2012
Old 7th March 2012
  #15
Gear interested
 
Joined: Sep 2011
Posts: 22

Thread Starter
pleventi is offline
Creating an airgap in 2D is tricky -- Ideally I'd go 12" SNS with a 6x6" air gap in the corner. But to do that would require cutting some SNS sheets, which I'd rather not do for the experiment. Maybe I'll start with just pulling out two batts of SNS and leave a 6" air gap to one wall (but still soffit-to-celiing in the other direction) and measure what difference this makes (if any).

I have measured different listening positions around the center seat -- even 3' to the sides there's still a prominent null. I'll try some more measurements later. Looking at my results with sub in back-left corner, something was funny with that measurement -- the L speaker is introducing deep nulls that should be there in the other sub location too since the L speaker is in an identical position. Could be a measurement anomalie (bad settings on computer/sound card sometimes causes this). I'm guessing if I remeasure the front-right / back-left combo of sub, I'll find it much better overall.
#16
7th March 2012
Old 7th March 2012
  #16
Gear interested
 
Joined: Oct 2011
Posts: 23

wgb113 is offline
Your null at the center seat could be a result of the room's length related axial mode combined with the axial mode due to the width of the room if that seat's dead center.

Bill
pleventi
Thread Starter
#17
7th March 2012
Old 7th March 2012
  #17
Gear interested
 
Joined: Sep 2011
Posts: 22

Thread Starter
pleventi is offline
Definitely could be -- room is 16' 7" long and 15' 5" wide (at the front). 34.1 and 36.6 Hz respectively. Center seat is dead center width wise, but 63% in the room on the length mode; definitely going to be some length axial contribution (assuming room modes actually contribute to FR -- I gather there is some controversy in the forums on whether they do vs. just impacting resonance). There's no room to push the seat back in the room without eliminating (or scaling back) second row seating. And not much room to move side-to-side. Best I can do is try out sub locations and treat the hell out of the room.
pleventi
Thread Starter
#18
7th March 2012
Old 7th March 2012
  #18
Gear interested
 
Joined: Sep 2011
Posts: 22

Thread Starter
pleventi is offline
Anomaly = SBIR

I re-measured my sub-in-back-corner + front left, and still see the new dips at 100, 110, 190 Hz. So what's going on? How could moving my sub affect the FR above 100 Hz?

Answer: SBIR (I think).

Measurements
  1. FrontRightSub + BackLeftSub + FrontL. Gives ugly FR at 100+ Hz, but mostly fills my 36 Hz null.
  2. FrontRightSub + FrontLeftSub + FrontL. Back to smooth 100+ Hz, but ugly 36 Hz null. See comparison of (1) and (2) in first graph.
  3. Bump the subs out of the corners (3' off both walls). No improvement of null, and 100+ Hz badness is back. Comparison of (2) and (3) in second graph.
  4. Stuff front-left corner with 6" SNS + ~3-6" gap. Theory: Its SBIR. Result in third graph, showing (3) vs (4): Its SBIR -- much smoother response, though that could just be due to more absorption in general.
  5. FrontRightSub + BackLeftSub + FrontL + Fluffy in front-left corner. Graph #4 shows (2) vs (5). Which is preferrable? Trade-off +10 dB in untreatable 36 hz for a potentially treatable and narrower but more musical 110Hz null.

Given this, not sure if I prefer front or back sub placement. I'll play with FL speaker placement and my makeshift SBIR treatment to see if I can make that null go away.

Empirical acoustics is FUN!
Attached Thumbnails
My Big Soffit Trap Results-back_vs_front.jpg   My Big Soffit Trap Results-off_corners.jpg   My Big Soffit Trap Results-fluff_impact.jpg   My Big Soffit Trap Results-front_with_fluffy.jpg   My Big Soffit Trap Results-two_sub_options.jpg  

pleventi
Thread Starter
#19
8th March 2012
Old 8th March 2012
  #19
Gear interested
 
Joined: Sep 2011
Posts: 22

Thread Starter
pleventi is offline
Sub Null Solved

Moved both subs into the back corners of the room. Not perfect from a utility perspective, since this restricts traffic to infrequently used back door. However, this completely smoothed out the bass response.

FR Graph #1 shows treated room with both subs in front corners vs rear corners. Treatment is four soffit traps (front and back ceiling-wall boundaries + front wall-wall boundaries). I've stuffed the former sub pockets in the front soffits with 15" deep of SNS (I wisely left a 48" tall gap so I just stood 5 sheets upright). I also measured with 9" SNS + 6" air gap with slightly better LF results as calculator predicts.

FR Graph #2 shows front subs, untreated room vs rear subs, treated room. This is the "before" and "after" from beginning until now. This is fast approaching EQ territory! I've also included before and after waterfall plots. Getting smoother, but could be better! And before and after RT60s. Still room for more LF absorption.

Remaining treatment plan:
  1. Add 16x16" soffit traps in back wall-wall corners
  2. Add 6" + 2" air broadband absorbers to front walls to help with boundary effects; I have room for deep traps here to match TV + mount 8" depth.
  3. Add 4" side-wall broadband absorbers to treat early reflections (verified with ETC).
  4. Add cloud. Early ceiling reflections verified via ETC.
  5. Build riser w/integrated bass trapping.
  6. Reduce HF absorption of existing traps if necessary.
  7. Add additional wall-ceiling traps if necessary.
  8. Maybe watch a movie :-)
Attached Thumbnails
My Big Soffit Trap Results-front_vs_back.jpg   My Big Soffit Trap Results-before_and_after.jpg   My Big Soffit Trap Results-waterfall_before.jpg   My Big Soffit Trap Results-waterfall_after.jpg   My Big Soffit Trap Results-rt60_before.jpg  

My Big Soffit Trap Results-rt60_after.jpg  
#20
25th April 2012
Old 25th April 2012
  #20
Gear interested
 
Joined: Oct 2011
Posts: 23

wgb113 is offline
Any updates?
#21
25th April 2012
Old 25th April 2012
  #21
Gear Guru
 
Glenn Kuras's Avatar
 
Joined: Jul 2005
Location: Atlanta, GA
Posts: 16,209

Glenn Kuras is online now
I honestly think it looks pretty damn good. Yes you could make it better but it will going to take a treatment to gain a few db here and there. Pretty much if you can get it around 5-7 +/- db you are doing well.
New Reply Submit Thread to Facebook Facebook  Submit Thread to Twitter Twitter  Submit Thread to LinkedIn LinkedIn  Submit Thread to Google+ Google+ 
 
Thread Tools
Search this Thread
Search this Thread:

Advanced Search
Similar Threads
Thread
Thread Starter / Forum
Replies
Cojo / Bass traps, acoustic panels, foam etc
1258
rackdude / So much gear, so little time!
11
Psyentific / Low End Theory
14
rob99 / Remote Possibilities in Acoustic Music & Location Recording
5
Bruce Swedien / Remote Possibilities in Acoustic Music & Location Recording
10

Forum Jump

SEO by vBSEO ©2011, Crawlability, Inc.