Bass Traps... the results

[sploo]

Intro

From my previous thread "Yes, it's another Bass Traps thread...":

When you're done with your project please come back and report here, to tell me I was right.

If that gets the sound anywhere near to what it is in bigger rooms I'll be writing 'Ethan's right' in very large flashing characters

Well, a promise is a promise (see attached)
EDIT: You'll have to click on it for it's 'garish goodness'.

Results

OK, on to the serious stuff...

At some point I plan to put together a webpage detailing the construction of all the traps, but for the moment, here's some results (and questions for those in the know).

These first three images were taken without any form of room treatment present. They show the room's low frequency response, full frequency response, and the low frequency response/decay times respectively:

http://spikyfish.com/01 no absorbers - low range.gif
http://spikyfish.com/02 no absorbers - full range.gif
http://spikyfish.com/03 no absorbers - 3D.gif

You can see a 30dB peak-to-null in the low frequency response, lots of comb filtering in the full range, and a reasonable decay time.

This second set of graphs show the room's response after the addition of 6 thin high frequency traps (3 on each side wall), 4 slightly thicker high frequency traps (currently on the floor, but will be on the ceiling) and 4 room corners treated with solid fiberglass traps.

http://spikyfish.com/04 all traps - low range.gif
http://spikyfish.com/05 all traps - full range.gif
http://spikyfish.com/06 all traps - 3D.gif

The low range response has been smoothed, and the null around 50Hz has been reduced, however there's now a huge null around 175Hz. Anyone got any ideas why?

The full range response has obviously been improved, with much less comb filtering evident.

Finally, the 3D plot is slightly smoothed, and decay times are reduced.

I then moved the mic about 0.5m (1.5') forward, and took another set of readings.

http://spikyfish.com/07 all traps (0.5m forward) - low range.gif
http://spikyfish.com/08 all traps (0.5m forward) - full range.gif

This has a positive effect on the low frequency response, but now comb filtering is much worse in the audible range of the spectrum.


Conclusions and thoughts

1. Playing music in the room is now greatly improved - much less disorienting, and with much less 'boom'
2. The comb filtering recorded after moving the mic may be due to my high frequency traps no longer being at the first reflection points, so I need to check this and re-test.
3. Annoying null created at 175Hz.
4. Frequency response for listener is unfortunately still very position dependant, but greatly improved from the room with no treatment.

Hope that's of use to someone...

 

[bpape]

Sorry but FR will ALWAYS be position dependent. The idea with the absorbtion is to make it less so.

You still appear to have some seating position/speaker position issues but it's better. The waterfalls are somewhat better but still not tamed to where they should likely be.

Be patient. Move the sub around to help smooth the response curve (don't forget the third dimension and get that sub a foot or so off the floor). Make sure your seating position is in a decent place (stay away from 1/2, 1/3, 1/4 of the room dimensions) 7ths work decently. 38-40% also works well.

Unfortunately, tuning your room is not a simple exercise. You can calculate a good starting point but you still need to play with it over time, document what each thing does better and worse, and then make an informed decision as to what you're willing to compromise on for the best overall performance.

Hang in there. It's worth it.

 

[Ethan Winer]

sploo,

> there's now a huge null around 175Hz. Anyone got any ideas why? <

I've seen this before and I believe I know the cause.

The low frequency response at any given location in a room is the result of the original sound plus many different competing reflections. The original and reflected sounds all combine to give a single level (of course) at any given frequency. Some of the combinations create peaks, some create nulls, and sometimes a reflection that would have made a null combines with one that would have made a peak. So together the result is somewhere in the middle.

If one of your traps reduces a reflection that had negated a null from a different reflection coming from somewhere else, the null now appears. So the solution is to add more traps, to reduce the reflection that causes the null. Again, that null-causing reflection was always present, but it was partially cancelled by another reflection that has now been trapped.

This is a big reason to use waterfal plots to assess the improvement from adding bass traps. With this graph adding traps will always show some improvement, even if the raw response is made worse.

--Ethan

 

[sploo]

bpape,

I suppose I knew the response would still be position dependant, I was just hoping I would've made it less position dependant.

Hmmm, yes, a sub. Well now, there's a thing. I have neighbours, so I decided against a sub. My main speakers don't sound like their too far down at 25Hz, and the setup will certainly produce some fairly crunching bass (and I mainly listen to music). Thus, I've never worried about a floor shaker.

Good point about the seating position. I had read an article (by Ethan I think) about the 38% rule. I'll check.


Ethan,

That sounds logical. The problem is that I don't have any more room (that is, WAF room) to place traps, so I might have to live with it.

What I was surprised about was the lack of effect on the room response between 200-6000Hz. It's a bit better, but not much.

I guess the most important thing is that it's a huge improvement. So I should be thankful for that.

I suppose I could also try taking some measurements with moving the speakers round a bit (not that there's much space in which to move them).

Thanks both for the continuing info.

 

[bpape]

A sub is not always necessarily needed for raw extension nor for sheer output. What it does provide is a way to allow movement of the bass generator without moving the mains. Unfortunately, where the speakers image the best and fit in the 'decor' the best is rarely the place that produces the smoothest response curve.

Try this. Move your speakers away from the front wall a foot or 2 - then remeasure. See if the 175Hz issue moves. Don't change anything else.

We're having fun now...

 

[sploo]

Try this. Move your speakers away from the front wall a foot or 2 - then remeasure. See if the 175Hz issue moves. Don't change anything else.

They're currently about a foot away from the side walls and 2 foot from the front wall. Unfortunately, I don't have much leeway as there's a centre speaker in the middle, and a rack with the amp and CDP between the center and front right speaker (and the room's about 9' wide). I will try moving then round a bit though.

This evening I checked up on my listening position and high frequency panels. The listening position could do with being just a touch further forward (I was at 44% not 38%!). I'm not sure how seriously to take myself there... somewhere between and

Somewhat ironically, I discovered that the 1.5' change in listening position had put the two reflection points on each side wall (R speaker on R wall, L speaker on R wall, and, obviously L on L and R on L) right in the two gaps between the three acoustic panels!

I've moved them to center on the new reflection points and my perception was that it sounded better, I'll take another frequency response measurement tomorrow night.

We're having fun now...

No... you're having fun. I'm standing in a small living room full of fiberglass dust, being deafened by speaker measurement tests, using one hand to stop an acoustic panel falling off the wall, another to move a speaker around, and um, another to start the analyser

 

[jaxvon]

Wouldn't that third hand be the woman of the house? But do you trust her to touch your analyzer....

 

[bpape]

That IS fun...

Not sure how tongue in cheek you were about position but 44% to 38% will definitely make a difference.

Also, I wasn't expecting you to make permanent changes in speaker location - just temporary ones to help identify what was causing the issue.

 

[sploo]

Wouldn't that third hand be the woman of the house? But do you trust her to touch your analyzer....

She can touch my equipment whenever she wants .

Not sure how tongue in cheek you were about position but 44% to 38% will definitely make a difference.

Sufficiently 'un' tongue in cheek that I'm going to move the listening position, but sufficiently tongue in cheek to understand that some people will think I'm nuts .

Also, I wasn't expecting you to make permanent changes in speaker location - just temporary ones to help identify what was causing the issue.

How would I interpret the results? I.e. if I moved the speakers 6" forward and it removed one null but created another, what would that tell me (other than I need more traps)?

 

[bpape]

Not necessarily - but it would tell you if the bigger issues are speaker position related or not.

If you can get to a point where you have eliminated the deep nulls, peaks can be tamed much easier - like with a parametric EQ. Depends on where they are (mains or sub) and how much you can do by modding seating position.

 

[sploo]

If you can get to a point where you have eliminated the deep nulls, peaks can be tamed much easier - like with a parametric EQ.

I was planning to get a parametric EQ at some point, as I fancy having a play. I also want to test a - possibly dodgy - hypothesis that the different sound produced by the analogue stages of expensive CD players is down to artificially induced frequency responses.

Depends on where they are (mains or sub) and how much you can do by modding seating position.

Which brings me nicely to...

Last night I moved the seating position to 38% from the front wall. It certainly sounds good, very 'enveloping'.

I also recalculated the first reflection points for all the front speakers, using a formula on one of Ethan's pages.

The results are here:

http://spikyfish.com/09 all traps, at 38percent - low range.gif
http://spikyfish.com/10 all traps, at 38percent - full range.gif
http://spikyfish.com/11 all traps, at 38percent - 3D.gif
http://spikyfish.com/12 all traps, at 38percent - impulse response.gif

The low frequency response is as good as anything I've managed previously, and the full range is pretty good.

I being to understand why the comb filtering is still pretty poor below 8kHz - the wavelength is just over 4cm, and my side wall traps aren't that much deeper, so of course they wouldn't touch the lower stuff. To get down to 1000Hz I'd need 34cm (13.5") .

As a quick experiment, I raised my 'ceiling' traps off the floor (2" of board + 5" of gap instead of 2" of board + 1" gap). Didn't seem to make any difference to the response.

Jeez. Now I understand why Ethan's traps are rated so highly.

Meanwhile, back at the ranch... The waterfall graph shows a much longer decay at the bottom end (around 70Hz) which is a bit puzzling. But overall I'd claim it's a little better - less deep nulls.

One of the things I did notice, is a few fairly large reflections at 4.14, 4.83 and 8.16ms respectively. Given a speed of sound of 343m/s, I can understand the one at 8.16ms (approx 2.8m) as it's the distance from speaker to ceiling to listener (currently the ceiling traps are on the floor as I haven't worked out how to mount them yet).

I'm a little confused by the other two though. Unless I'm misinterpreting what the impulse response represents, I don't understand how I can get reflections at 1.42m-1.66m, as this is less than the distance from speaker to listener. Any thoughts?

 

[bpape]

The impulse response shows you the difference between the original direct signal and the reflected one. If you show a spike at 1.4m, then this is a path that is 1.4m longer than the direct path from speaker to mic.

Looking at your waterfalls, it's very clear that you still need significantly more broadband bass absorbtion.

Also, when you plot the sub + mains, try to run it just to 400Hz or so to see better what's happening. The sub response isn't that bad. You have a null around 40ish. Other than that, a couple bands of parametric EQ to tame the broader peaks will level it out nicely.

 

[sploo]

The impulse response shows you the difference between the original direct signal and the reflected one. If you show a spike at 1.4m, then this is a path that is 1.4m longer than the direct path from speaker to mic.

Understood, thanks.

I'm pretty certain the first pair of impulses are speaker->ceiling->listener, so that should be sorted when I move the absorbers from floor to ceiling (and I'll put a rug on the floor to help that side).

The other one is, I think, a reflection from behind the listening position (the corner of my fireplace). I'll check the theory with a panel.

Looking at your waterfalls, it's very clear that you still need significantly more broadband bass absorbtion.

Unfortunately, whilst I've not yet hit audio nirvana, I've hit the room's aesthetic limit (or at least the WAF limit), so I'm probably going to have to live with what I've got. What I've got however, is a whole world better than it was before, so I'm not too worried.

One thing I'm going to try is to use some of the leftover dense fiberglass to bolster the back of a couple of the high frequency traps. I'm hoping I might be able to extend their ability to remove first reflections to a lower frequency.

Also, when you plot the sub + mains, try to run it just to 400Hz or so to see better what's happening. The sub response isn't that bad. You have a null around 40ish. Other than that, a couple bands of parametric EQ to tame the broader peaks will level it out nicely.

Bass boom really isn't too bad now - I've noticed a couple of louder 'notes' when music plays, which are probably the peaks.

I've been considering an EQ. Something that would handle several channels would be very useful (i.e. analog stereo and/or discrete analog surround in/out and digital in/out). This would allow me to trim both music/DVD-A from the CD/DVD player, and compressed Dolby/DTS movie soundtracks (via coax) from my PC.

Chris (WmAx) has mentioned some Behringer products. Any suggestions?

Cheers!

 

[Tom Andry]

Unfortunately, whilst I've not yet hit audio nirvana, I've hit the room's aesthetic limit (or at least the WAF limit), so I'm probably going to have to live with what I've got. What I've got however, is a whole world better than it was before, so I'm not too worried.

I have to comment on this. So often it is easy to get wrapped up in trying to get the "perfect" response or the "optimal" sound. Far too often I see little tweaky comments in threads that are impractical for just about anyone except independently wealthy bachelors with multi-room mansions and the nearest neighbor at least a mile away (I'm not saying that is what is happening in this thread, I'm speaking generally here).

We can do what we can do. Is my room perfect? - not by a long shot. Has the room treatments and advice I've gathered on Audioholics helped? You bet your rear end it has. At some point all (or most) of us are forced to say, "That's as good as it is going to get," and just live with it. I know that it took me about 6 months to get to that point, it sounds like you are almost there as well. Now it gets fun, you can actually sit back and ENJOY your system!

Congrats!

OPRA (how the heck did you get your SigOth to agree to panels on the ceiling?! I'm so jealous!)

 

[sploo]

I have to comment on this. So often it is easy to get wrapped up in trying to get the "perfect" response or the "optimal" sound. Far too often I see little tweaky comments in threads that are impractical for just about anyone except independently wealthy bachelors with multi-room mansions and the nearest neighbor at least a mile away (I'm not saying that is what is happening in this thread, I'm speaking generally here).

Absolutely. After I'd first tested the high frequency panels, and found they changed the sound from terrible* to pretty good, I was tempted to leave it there. However, I'd already ordered the panels for the bass traps.

The effect of the bass traps is much more subtle. I wasn't convinced I could hear any difference (despite the graphs showing the changes) until I played a film. Now, I can hear dialogue, and the loud bits don't have you diving for the volume control. It's not perfect, but it's much better than it was.

* And I do mean terrible. I mean really, utterly downright unpleasant. A friend commented that the sound (for complex music) made him feel ill.

We can do what we can do. Is my room perfect? - not by a long shot. Has the room treatments and advice I've gathered on Audioholics helped? You bet your rear end it has.

Yup. After setting up the audio gear, and being somewhat dissapointed, I searched for ages to work out why it was bad. This forum (mainly Ethan, bpape and yourself) opened my eyes to acoustic room treatement.

I read a quote - probably from another site related to Ethans work, that the pros shoot for a 6dB peak/null range, and if you manage 10dB in your 'normal' room, you're doing well.

At some point all (or most) of us are forced to say, "That's as good as it is going to get," and just live with it...

I'm pretty certain I can get a few more improvements with relatively unobtrusive tweaks, but I am (hopefully) nearing the 'just sit down and listen to it' point .

(how the heck did you get your SigOth to agree to panels on the ceiling?! I'm so jealous!)

"I have these panels"
"Yes."
"Those go on the side walls..."
"Yes."
"...and those go on the ceiling. Or I could just leave them on the flo..."

So, after I regained consciousness...

Actually, I've not purchased a ball'n'chain yet, so I have a little more freedom . Also, the room has white walls and a white ceiling, so I've made the panels white so they don't intrude too much. Not that I had much choice - they're in the main living room (our UK houses tend to be rather small by US standards).

 

[bpape]

Aesthetics and functionality must play a role in any non-dedicated space moreso than a dedicated one. If you've done as much as you can with treatments, then your options for improvement will com via placement and EQ. We all have limits to work within - we do what we can with the space we have to work with. I'm hardly a millionaire nor do I have a mansion. What I DO have is:

1. A dedicated space that is mine to do with as I please with no flack from the significant other (other than redoing the house structurally to remove a lolly pole)

2. A very understanding spouse - as long as it's in my room

I'm glad that it's improved for you. The things you describe are exactly the kinds of things that should improve. Sounds like you're well on your way.

The Behringer Feedback Destroyer is a very inexpensive product that can produce good results for subwoofer tweaking. Personally, I wouldn't put it in the loop for my mains or surrounds. It's another one of those things that is a tremendous bang for the buck in terms of performance gained for the money spent.

If you want a parametric EQ suitable for your mains, I'd look elsewhere. I personally like Symmetrix products. However, almost all of them are pro stuff and only offer balanced connections and are not inexpensive. Probably one of the best out there is the TacT system but it's big-time money.

My advice would be to get a basic parametric EQ just for the sub. That's where a lot of your issues will be that can't be tamed without more treatments (and maybe even with to a certain extent). Just make sure that whatever you get will allow you to have the ability to put ALL of the bands centers between say 20 and 100Hz. Many parametrics have limited ranges for different bands so maybe only 1 or 2 out of 5 or 6 would be in the subwoofer range.

Plan on using 1 of the bands to take the top out of the sub at a steeper slope than the processor will allow. For instance, if your xover is at 80Hz and 12db/octave, you might have a band set at 1 octave width and cut almost all the way - centered at 160Hz. This will leave the sub untouched (other than introducting a little phase shift) but eliminate the lower level interaction between the sub above the xover and the mains.

Enjoy!

Bryan

 

[sploo]

What I DO have is:

1. A dedicated space that is mine to do with as I please with no flack from the significant other (other than redoing the house structurally to remove a lolly pole)

*sigh* If I could afford a UK house with a large enough room to dedicate to HT, I'd fill the thing with as many panels as it needed.

I'm glad that it's improved for you. The things you describe are exactly the kinds of things that should improve. Sounds like you're well on your way.

Yea. It's nice that things have matched the science. Even when I didn't understand a result, a little research would dig up the reason, and the room would always have the artifact/feature that caused the problem. I've certainly learnt a lot here, but also learnt a little of what I don't know!

My advice would be to get a basic parametric EQ just for the sub.

Hmmm, yes, a sub. Well now, there's a thing. I have neighbours, so I decided against a sub.

I'm still interested in an EQ though . I'll have a look through some of the music (as in musician) magazines we've got in the office.

 

[WmAx]

A

The Behringer Feedback Destroyer is a very inexpensive product that can produce good results for subwoofer tweaking. Personally, I wouldn't put it in the loop for my mains or surrounds. It's another one of those things that is a tremendous bang for the buck in terms of performance gained for the money spent.

If you want a parametric EQ suitable for your mains, I'd look elsewhere. I personally like Symmetrix products. However, almost all of them are pro stuff and only offer balanced connections and are not inexpensive. Probably one of the best out there is the TacT system but it's big-time money.

I can not comment on the current FBD[the previous model that is now discontinued measured superbly], but the Behringer DCX2496 and DEQ2496 work flawlessly, producing no noise or distortion whatsoever, when interfaced properly. The DCX2496 DSP device has the power to control/manipulate just about any complexity of set-up provided you can feed it's outputs directly to the amplifer inputs; it can be used simultaneously as a 3 way active crossover, parametric equalizer, shelving contour filter, delay/phase adjustment, dynamic vector e.q.[to linearize non-linear motor performance dependant on input power, or to use as a safety device to prevent over-excursion or overdrive of drivers] and can be daisy-chained to mulitple DCXs while using one as the master control unit. The DCX sells for approx. $250 USD at the time of this posting.

-Chris

 

[Tom Andry]

Chris,

It's like you can smell a Behringer question!

 

[sploo]

...but the Behringer DCX2496 and DEQ2496 work flawlessly, producing no noise or distortion whatsoever

Cheers Chris. I was looking at the Behringer site a few days ago, and it seems there are a few models available in the US that they don't sell over here, but I'll check up on those numbers.

Chris,

It's like you can smell a Behringer question!

*LOL* Actually, I remembered that Chris had posted a message about Behringer EQs a while back, but I couldn't find it, so I was hoping he'd chip in with the details .