Hey Duke, those are very good points.
I'd like to have your comments about the CABS system as presented in this paper: http://vbn.aau.dk/files/62729209/LF_sound_field_control.pdf
I find what they have demonstrated to be a compelling argument for the DBA/CABS method. I have attached a couple of their graphs (these are MEASURED, not just simulated)
Here's what I see as the main takeaways from these measured response curves, and the overall implications of this paper:
1) The main room nodes aren't just smoothed out, they are eliminated. The attached curves show a MASSIVE reduction in room modes - as in near zero.
2) The positional SPL variations are almost completely eliminated throughout the room - see the attached charts. We're looking at +/- 1 dB over the great majority of locations on the rom - and as you pointed out, it's very, very hard to get that kind of response with just damping and tube traps. Near impossible in fact.
3) There is no need to EQ or average out or smooth out the frequency response with additional woofer locations, it is already near perfect.
4) CAPS also reduces the broad-band bass transmission into other rooms by 1.5dB to 5dB
5) The "complex impedance" differences in the lab walls (previously offered as an objection) don't look like they affect the measured performance by a whole lot - the variation of the in-room response looks to be very, very low - and if we read the paper, at least these walls were described as being relatively flimsy.
Overall, it looks very much like the DBA/CAPS arrays present a very compelling improvement over conventional damping and bass-traps.
Like Duke said, starting at +/-9dB and then achieving +/-3dB throughout the room is very hard to do with passive absorption methods...
And then when we look at the +/-1dB that the CAPS method shows, throughout the room...
I will let you draw your own conclusions, but it sure looks clear to me that the DBA/CAPS has some MAJOR performance advantages.
I'd like to have your comments about the CABS system as presented in this paper: http://vbn.aau.dk/files/62729209/LF_sound_field_control.pdf
I find what they have demonstrated to be a compelling argument for the DBA/CABS method. I have attached a couple of their graphs (these are MEASURED, not just simulated)
Here's what I see as the main takeaways from these measured response curves, and the overall implications of this paper:
1) The main room nodes aren't just smoothed out, they are eliminated. The attached curves show a MASSIVE reduction in room modes - as in near zero.
2) The positional SPL variations are almost completely eliminated throughout the room - see the attached charts. We're looking at +/- 1 dB over the great majority of locations on the rom - and as you pointed out, it's very, very hard to get that kind of response with just damping and tube traps. Near impossible in fact.
3) There is no need to EQ or average out or smooth out the frequency response with additional woofer locations, it is already near perfect.
4) CAPS also reduces the broad-band bass transmission into other rooms by 1.5dB to 5dB
5) The "complex impedance" differences in the lab walls (previously offered as an objection) don't look like they affect the measured performance by a whole lot - the variation of the in-room response looks to be very, very low - and if we read the paper, at least these walls were described as being relatively flimsy.
Overall, it looks very much like the DBA/CAPS arrays present a very compelling improvement over conventional damping and bass-traps.
Like Duke said, starting at +/-9dB and then achieving +/-3dB throughout the room is very hard to do with passive absorption methods...
And then when we look at the +/-1dB that the CAPS method shows, throughout the room...
I will let you draw your own conclusions, but it sure looks clear to me that the DBA/CAPS has some MAJOR performance advantages.
Attachments
Interesting about the anechoic chamber!
In a LEDE design, the reflection are brought back to the room but have been significantly reduced. the reflections are then secondary reflections and wont create comb filtering and similar problem at the listening position. You however need the space required to implement LEDE but a well designed LEDE room would be as good as any proven treatment method. ITs actually what Id try to implement if I were to start again.
I was skeptical but that chart on the right is really convincing. (Not hard to make a single seat sound good, as on the left chart.)
But my intuition has trouble digesting the concept of that kind of active treatment for a room with waves and phases bouncing around in near-infinite variety. Or what does it sound like? Like EQ? Like a bass-reflex box with a residual one-note boom despite the sim FR?
Ben
But my intuition has trouble digesting the concept of that kind of active treatment for a room with waves and phases bouncing around in near-infinite variety. Or what does it sound like? Like EQ? Like a bass-reflex box with a residual one-note boom despite the sim FR?
Ben
Yeah, but... the measurements are showing the waves and phases are NOT doing that. The DBA/CABS acts like a unified planar wave moving through a tube, not like separate sources bouncing around.
Hey Ben, I also found it difficult to intuitively grasp the notion that the output from the 4 driver main bass array (and the back one also) would NOT be bouncing around. I would have thought something more along the lines of a 16 driver front wall would be necessary...
However, the actual measurements show that for all intents and purposes, below the frequencies defined by the driver spacing, their output sums to a flat or planar wave. And the measurements they posted aren't sims, they are actual measurements in a real room, and one with flimsy walls at that! (Yes, flimsy walls with complex impedances
)The original article I saw (still can't find it) DID have some subjective comments in it, the listeners reported it was the most natural bass they had ever heard, that it sounded like the room had disappeared.
Obviously I can't answer you from personal experience with the full DBA/CABS system... BUT.....I did build the mini-DBA/CABS (keep it simple approach) described as follows, and it was truly excellent. The room-boom completely disappeared. Also, some nasty suck-outs were gone as well....It sounded more like the room had expanded or disappeared!
With one subwoofer centered along the front wall, 1/2 way up the wall, and a second one centered along the back wall, also 1/2 way up, operating with proper delay and level adjustment, reversing the phase on the rear sub. Switching the cancellation sub in and out was a revelatory experience!
Switching it in created a very large "bubble" where the bass was subjectively very uniform and smooth, with tremendous speed and impact... no obvious coloration or suckouts whatsoever. And, while the center of the room was near perfect, what really surprised me was how wide and deep the "optimum listening zone" became... it was almost wall to wall, and ran nearly the length of the room. And it worked great in seated OR standing position.
Listen, if you have a fairly symmetrical rectangular listening room, you owe it to yourself to at least try this mini-DBA/CABS approach! All you need is two fairly identical subs and a Behringer DCX2496...
And measurements confirmed it was really good, no eq was needed. My living room was about 12ft (4m) wide, ceiling at 8ft (2.4m) and nearly 20 feet long, and this was within +0/ -2dB all the way from 28Hz to beyond 160Hz, in most areas of the living room. The ETC curves were also very good... not as nice as the CABS curves in the paper I referenced, but really good nonetheless.
This mini-DBA/CABS produced a large listening area that was very uniform all the way up to 160Hz which was the crossover imposed limit of the subs I was using. Utterly fantastic for home-theater with many seating locations.
Again, what I found especially notable was just how much of the room area was usable - nearly any practical seating location was in the sweet-spot.
So...how did THIS version sound? Utterly musical, warm and punchy with amazing pitch, rhythm and tonal purity, and very much as if the room was gone. I heard bass details I had never heard before, recording after recording. In later times I got into headphones because I had to travel a lot, and with the cans I was able to compare with the mini-DBA/CABS... and where the bass was concerned, the mini-DBA/CABS won every time!... it had the same level of detail but had a lot of body impact and slam, something the phones do not.
Sorry, I wish I could provide measurements, but a divorce occurred, lost the house and my sound system was the first to get liquidated... ouch
Actually, if I can find the money, I still want to build one of these big arrays with 16 drivers front and back... because I find the 80 or 100Hz Upper "limit" of the 4 driver arrays too restrictive....
Why? I think I would want to go with the 16 driver front and back walls to push the upper limit to more like 350-400Hz or higher.
The idea of being able to eliminate most of the passive absorbing devices really appeals to me. Then, with judicious use of of diffusers and some light damping, it might be possible to get something beyond merely excellent.
Of course, in stating this preference I am aware certain others here may object, but hey, to each his own.
Last edited:
Jack,
Did you build the subs into the walls or were they just centered on the wall?
If not built in, what was the size of the cabinets?
Art
Hi Art, I was using two Velodyne Servo subs, similar to the current DD-10, centered on the walls. No built-ins. the cabinets were about 11"x11"x11" - or maybe a smidge smaller.
I tried them both ways, facing against the walls or facing out into the room, and didn't notice any great difference, so for aesthetics, used them facing out into the room.
Had them on heavy sand-filled cabinets, the back of the subs was about 3" from the walls.
Last edited:
With the center of the cone to wall distance of only about 20 inches, the wrap delay would still be well under 1/4 wavelength of 100 Hz, so would respond like a built in.
Interesting to hear that the DBA will work with only a pair of (good) subs.
This sounds like a bit of a breakthru. Major improvement in room bass by just adding rear woofers wired out of phase... I'm not sure I understand how they used delay though. If the distance from the front woofers to the rear woofers is 10ft., would I want to have a time delay that represents 10ft. (approx 10mS) AND a wired phase reversal? That would be my guess. Is anyone sure about this?
That is correct, except the rear speaker(s) are polarity reversed, not "wired out of phase"
.The DBA "pitcher/catcher" won't work well if the front and back speaker's phase response is not identical.
Last edited:
What's the difference between polarity reversed and wired out of phase?
Since the wavelengths are roughly 10 ft. and bigger below 100HZ, it seems that in a typical living room (that may only have 10 ft. from front woofers to rear woofers), you could get away with no actual electronic delay function, just the polarity reversal of the rear woofer (?).
Since the wavelengths are roughly 10 ft. and bigger below 100HZ, it seems that in a typical living room (that may only have 10 ft. from front woofers to rear woofers), you could get away with no actual electronic delay function, just the polarity reversal of the rear woofer (?).
Kreskovsky at Music and Design has study on something like that in the Tech Studies section of his website. I tried it and it sure beats the heck out using the subs in phase....
BUT....
For myself, I thought I noted a very definite improvement when the proper delay was built in,
Note: the level needed fairly careful adjustment.
In my case, it helped that the room was extremely symmetrical.
So from my limited experience, yes even if the living room is 11 or 12 feet long, sound propagating at 1100ft/sec, that 1mSec delay would be a definite difference maker. Mine was about 19.5 ft front to back and the delay was set to around 1.8mSec, it worked like a charm.
Again, what we are doing here is attempting to swallow up the wave, or at least as much of it as we can, looking at it primarily as a time function. As pressure arises at the back wall, the rear sub is creating an equal and corresponding anti-pressure, in perfect phase at all frequencies. Can't get that without the delay being right.
Granted with only double subs it's not the theoretical ideal, but in practice it worked so well as to make me wonder how much more bang we'd get for the money spent to get the true plane-wave.
Last edited:
Oh, also, for this to work properly you want your woofers as close to the front and back walls as you can get them. Not two or three feet out. And it seems to work best when you take the room length, not the woofer to woofer length as the distance for calculating the delay.
Delay Time in mSec = distance in feet/1100 ft/sec
Delay Time in mSec = distance in feet/1100 ft/sec
Last edited:
Bob,
I was being pedantic- "wired out of phase" is just an incorrect use of the word "phase", which is the relationship in time between the successive states or cycles of an oscillating or repeating system.
In the case of not using a delay on a reversed polarity sub at the back of the 10 foot room, it would cause a primary cancellation only at the 10 foot wavelength of 113 Hz, as that is the only frequency that would be completely "out of phase", while with a delay corresponding to the time of flight, all the frequencies in the delayed pass band (theoretically) cancel, if one assumes the sub output to be a plane wave, which can be approximated with multiple drivers arrayed on the small wall of a shoe box shaped room.
Art
Ding ding!! Very well explained.
Although I'm sure everyone already knows this since they watched the video I linked on page 2.
Last edited:
Very interesting. I'm not sure we'd actually get a plane wave below 100 Hz as the article claims (that looks to me like an optimistic assumption), but the measurements indicate a significant net improvement, so the fine points may not really matter.
You've certainly piqued my interest. I have a lot of projects on my plate right now, but unless some fatal flaw becomes obvious to me, I'll give it a try later this year.
in my experience with playing with stereo subs, I never achieved satisfactory bass with subwoofers only or fullrange speakers. Its only when adding deep bass traps that I got what I wanted in my bass.
I have not tried the multiple subwoofer method from Geddes, but did have stereo subs, mono subs, in my system.
simply treating my room with many good bass traps changed the quality of my bass dramatically, something subwoofer never did for me. In my room, I actually wasnt able to create a truly satisfying bass sound either without subwoofer, with stereo subwoofer or single subwoofer. I found my bass always innacurate and always a bit boomy, lacking punch and exactitude. Bass traps make the bass sounds very tight, punchy, controlled in a way hard to explain. its a very dry type of bass. nothing like adding subwoofers.
To put it another way, you are not likely to transform +/- 9 dB bass region frequency response to +/- 3 dB simply by adding damping material
In my room, bass traps helped to bring the FR of my bass at my listening position something 3-4db more flat then before from between 60hz to 200hz.
I also found that subwoofer really helped to make the bass flatter, but only within the range of the subwoofer and it still didnt brought the performance and the tightness and focus of the bass traps across the entire range of the bass.
I have to say that this comment make me perplex: how high do you use your subwoofer? Yes, subwoofer really do flatten the FR, but what about the bass frequencies not covered by subwoofers?
We cannot even detect the presence of bass energy from less than a single wavelength, and we must hear several wavelengths before we can detect pitch. So by the time we can even hear the bass, the energy has already bounced off of several room boundaries and the room's effects are already all over the frequency response.
maybe its a stupid question, but how we can then perceive bass in headphones?
If we accept the in-room steady-state frequency response curve as correlating well with perception at low frequencies, and I think this is reasonable, then the amount of improvement we can expect from adding subwoofers (widely distributed) is greater than the amount of improvement we can expect from adding damping.
Id like to know anyone who has treated their room adequately with bass traps and added subwoofer who could comment on that.
This is at least is far from my experience but id have to admit that people experience that I read is more about absolute: only bass traps vs only subwoofers. I really wonder who did both: truly treat adequately with bass traps his room AND added multiple subwoofer.
Damping is pretty much always desirable at low frequencies, but too much damping quickly becomes undesirable at high frequencies. So it has to be done in a way that adequately preserves the shorter-wavelength reverberant energy.this is over simplification.
At the listening position, you dont want any early reflection at any frequencies. you dont want mids nor highs early reflection but you do want and need secondary reflections. You dont want to allow any early reflections at the listening position that are at least not attenuated around -10db at 15ms.
this is a very interesting thread:https://www.gearslutz.com/board/stu...erstanding-precedence-effect-etc-shift-3.html
Last edited:
Youknowyou,
As far as hearing pitch, several wavelengths at any frequency are required, but you would hear 10 cycles at 400 Hz in the same time as one cycle at 40 Hz. Headphones, like any room or chamber also have modal response, but far above anything perceived as "bass". You hear the bass as recorded using (good) headphones, you hear bass plus room modal response in a room. The tricky bit is how to reduce the modal peaks and dips without making the room too "dead" in the high frequency.
Art
I personally have not found that tricky to make my room not too dead with making sure to not over use absorption: use absorption only for early reflection points. anywhere else use diffusion or bare walls. All reflections are not bad, simply those that are early reflections at the listening position are.
The tricky bit is how to reduce the modal peaks and dips without making the room too "dead" in the high frequency.
As I said, I did the mistake my self by over treating my back wall resulting in a very uncomfortable sound.
Its very important to make sure that bass traps are reflective for mid and HF. For bass traps you can either use a method of pattern slotted wood on your bass traps. Using FRK or kraft paper over my bass traps did the trick for me. You can also use pressure based bass traps that are sealed so the foam is entirely covered hence not absorbing HF.