Basstraps not effective?

[atledreier]

Hello.

I've been playing around with REW and my basstraps today, to measure the effect.
Well, set up REW, took a few readings of the room as it is now (1M, 8sweeps), and then removed all the acoustics panels I have.
That is one corner trap (60cm face, 85cm tall, one 2'x4'x4" sheet cut in 8), two 4" sheets straddling corners, four 1,5" panels in reflection points and one 1,5" panel straddlign the last corner.

Took a few more sweeps (1M, 8 sweeps again), and overlayed the plots. And they were virtually identical. There was minute differences, most likely due to inaccuracies in the measuring chain, but really minute. I was expecting a dB here and there, at least.

So I started moving things around, placing traps at odd locations, floor to wall corners, opposite corners, basically wherever I could get to. And on a few of the measurements there were slight differences, some in the 0.5dB region, but very little. Are 4" Rockwool THAT ineffective?

The thinner panels that I've used to kill reflections really help, my impulse response tell me that, but the basstraps are really just a waste. Do I just need more? I would think even the three thick panels I have would show up SOMEWHERE on my graphs?

I have another 12 4" sheets on the way, and will make proper tri-traps from them, but I'm having second thoughts now. Might just dress them up and use them as general absorbtion, if they are not gonna do anything to my messed up bassresponse.

[bpape]

Well, you're getting about 20% of the corner area. That will be helping in the decay times - have you looked at the differences in the waterfalls? Also, it may well be that a large part of your issues are either:

- Nulls/peaks off the rear wall where you have nothing.
- Height related
- SBIR related (no panels directly behind the speakers
- speaker and/or seating position related.

Bryan

[atledreier]

I've had panels on the rear wall, the rear corners, floor to ceiling in front or rear corners (as much as I have panels for now). I've also had panels directly behind the speakers and in the 1st reflection off the front wall. Frequencyresponse was the same, and impulse response was the same. I have a strong reflection coming off the ceiling (of course) and another pretty strong reflection off something else that I just can't figure out. After 7ms there's a nice clear peak off both speakers. I thought it was the rear wall reflection, but putting panels up didn't help at all.

What do you mean about height related and what is SBIR?

Even if some of my problems are seating related (70Hz comes to mind), I would have thought at least SOME part of my curve would be affected by treatments?

[Lobotomy^]

Some says that corners and near walls are worst place to put wool as acustic treatment. Wool will damp air movement, but sounds energy is mostly in form of pressure near the walls. Best place for wool to be effective is in the middle of two walls that are causing standing wave. However that is mostly effective against that single standing wave.
Basstrap that is airtight everywhere else exept for air movement through wool might be effective in place where sound energy is in form of changing pressure.

I did lowered roof of my concrete HT. 100mm wool with 100mm air gap in the middle of room and up to 350mm air gap at sides. Wool isn't always decreasing sound pressure, even on high frequencys.

Heres some measurements of my room. Mic was on listening point, both speakers and sub on. Red line is without any part of the roof, green is with 100mm airgap roof in place (75% of area of roof) and blue is full roof in place.


Effects are visible, but not easy to understand.

[atledreier]

Lobotomy^ wrote:

Effects are visible, but not easy to understand.

Exactly.

I'm gonna make my traps so I get 90% corner coverage and see what happens then. If I don't see anything then I'm sending the bill for the panels to Bryan.. I'm sure I'll se SOMETHING, then. If not, well.. It's a lesson learned.

I was under the impression seating-related problems could be remedied or at least reduced with traps and treatments, but Bryan's post seems to suggest otherwise...

[bpape]

Sorry - you can 'trap' till you're blue in the face but if you're sitting in a 20db null, many times you just can't treat it - you have to move. Things like sitting against a wall you're not going to 'fix'.

You've tried a lot of the things that need to be done but not enough in any combination at any single time. Things all work together. SBIR is a ragged frequency response in the bottom end based on the proximity of the speakers to the wall behind them and the wall beside them.

Quite honestly, my advice to you is to stop and remove all treatment from the room. Take a measurement (frequency, impules, and decay). Play with the seating position and learn where the issues are coming from and which are seating related. Fix what you can.

Then, start playing with speaker/sub locations (actually 2 different steps - speakers first) and get that the best you can. Again, learn what's happening when you move things. Then go back and tweak seating position, then back and forth until you get the best you can.

By that time, you should know what's speaker related, what's seating related, and what didn't change no matter what you did. THEN we can start looking at why they're happening and how to treat them.

Sometimes in some rooms, you just can't fix everything. Things like usability, walkways, WAF, etc. simply won't let you do what needs to be done - so we do the best we can.

Take your time and learn your room and setup before you do anything else. It'll pay off in the long run.

Bryan

[atledreier]

I was afraid of that...

My wife is gonna kill me..

That means moving the projector and screen... again... The ceilign is starting to look like swiss cheeze at this point...

I'll spend the weekend making corner traps anyway, they can't hurt, I presume? Then I'll ditch the wife and kids, lock the door and measure and move stuff around until I know what's what.

I'm sitting about 35% of the room length from the back wall (185cm in a 530cm room) right now, I would think that was about ideal? Have experimented very little with seating position, basically went for your recommendations on my room in another thread (dunno which one, it was one I started). I realize you can't know everything about my room, of course. I'm not blaming anyone for my difficult setup!

I'll do measurements of the room, both dimensions and acoustic without any treatments, FR, IR and decay at different positions for speakers and listening positions first. I'll post them here for analysis, so I can learn what is the result of what. I have a bunch of 4" panels on the way, maybe I should NOT make traps until I know they are needed? Maybe I need more panels other than the corners instead?

Then we can get to work on the subwoofer....

[bpape]

Seating doesn't sound like it's in a bad place. Try going forward just a little bit. Also, maybe consider moving off center maybe 6" or so side to side.

Why do you have to move the screen and PJ. We're talking small movements here. Moving forward should not cause you to move anything but the seating. Speaker movement is not a big deal unless they're in-walls and then you're kind of stuck.

Again, decay time is IMO as or more important than frequency response. Overly long decay times in the bottom end can cause dialog intelligibility issues, masked high end, tubby/muddy perception of bottom end, masking of extreme bottom end, etc.

Lastly, be realistic. If you can get a treated, un-EQ'd room to +/-5db, you're doing well.

Bryan

[atledreier]

I am off center in the room. My sweetspot is 160cm from my closest side wall, in my 390cm wide room.

Going much closer to the speakers/screen will bring the screen in too close, really. I have a 40" tv behind the screen, so I can't readily move it any further against the front wall either

I have measured the room with all treatments removed, including the 1st reflection panels. It sounds really hollow now, so at least it's doing some good up high.

I will post measurements shortly. Not sure what you need, exactly. I have saved the measurements, so I can basically give you whatever REW can generate.

[atledreier]

Ok, these are fronts full range, no sub, from my current sweetspot

185cm from rear wall (room 530cm)
160cm from closest side wall (room 390cm)
The speakers are placed symmetrical on the 160cm listening axis, if that made sense.

The impulse response for the left speaker (and also seen in the plot for both speakers) look strange to me. Is this a measuring anomaly or some explainable phenomena?

Fullrange frequency response, both speakers, 1/3octave smoothing


15-200Hz plot of both, left and right speaker


Decay left


Decay right


Waterfall left


Waterfall right


Impulse left


Impulse right

[atledreier]

Ok, I've been playing around wirh positioning again, and think I've found a better place for speakers and listening position. Moved the speakers forward, narrower and symmetric, and moved the seat back a little. Haven't really listened to it much, and haven't touched the subwoofer yet, but the fronts are looking better now.

NB! This goes to 500Hz, and the target is for bass limited, not full range as measured.

[atledreier]

Played some more with my setup. Moved the sub around and found a good spot around midway on the side wall, that's about level with my listening position. EQ'd it, integrated it (pretty easy, it was pretty much flat FR, as were the fronts), and sat down and had a good listen.

It sound like crap.

Well, not as bad, but before I have had a sound that just keep you coming back for more. I often spend far more time than I really have on music. Last night I just didn't feel like it. The subwoofer is easy to locate, being out on the side. I tried lowering cutoff to 60Hz. That helped a little. Then I activated the sub's internal crossover on top of the pre/pro crossover and realigned phase and distance. Again, better, but I don't know how much distortion I introduced with the dual crossovers. Probably a lot. My system measure within +-5dB pretty much across the board, but it just sounds a little nagging. Maybe I need a house curve. I'm afraid more subwoofer level will just bring out the localization issues even more, though.. I need to experiment more, I think...

[jerome]

atledreier wrote:

I have measured the room with all treatments removed, including the 1st reflection panels. It sounds really hollow now, so at least it's doing some good up high.

Did you try to put some of the treatments back before listening to music? That may be one reason why you do not get the sound you really liked.

Even if you now have a frequency curve that you like, the decay measurements may still be incorrect? Remember what Bryan wrote:
Again, decay time is IMO as or more important than frequency response. Overly long decay times in the bottom end can cause dialog intelligibility issues, masked high end, tubby/muddy perception of bottom end, masking of extreme bottom end, etc.

[atledreier]

Yes, I did, and my decay looks better than it did before I moved stuff around. The treatments seemed to work now too, with clearly visible effects bot in FR and decay. I have not applied a house curve this time around, but aimed for a linear frequency response. I have a slight increase towards the bottom end (+5dB at 15Hz) but nowhere near as aggressive as I did before. That could be it. I need to experiment more.

Any tips on that subwoofer localization issue?

[bpape]

Where is your xover frequency? From your original plots, it appears that you're running the mains full range - or was that just for measurement purposes? Your mains should be cut off at 80Hz and the sub brought in at 80Hz.

If you have mains and sub trying to all 3 do 80Hz down, they're fighting each other. Mains go where they need to for best locking to screen, imaging, FR from 80Hz up - which is rarely where they produce the smoothest bottom end response.

Bryan

[atledreier]

I used fullrange for measurement purposes only. I usually cut at 80Hz, and have completely disregarded frequencies below that for the current placement. The sub initially was crossed over at 80Hz. With the sub off to one side now, crossing it over at 80Hz makes it very easy to locate, and is very distracting. Crossover at 60Hz help a little, but it is still localizable. I'll post plots when I get home, I'm currently at work.

[jerome]

atledreier wrote:

Yes, I did, and my decay looks better than it did before I moved stuff around. The treatments seemed to work now too, with clearly visible effects both in FR and decay.

Could you post some measurements when you get home tonight? And a quick description of where you placed the treatments. That's kind of interesting to see your results

[atledreier]

Sure, I will.

I'm also interested in anything that will make my subwoofer less 'localizable' (is that even a word?).