Active room damping

Colakurt

Member
2017-12-28 6:37 pm
N/A
I have done some REW measurements and I can see that I have issues with standing waves at 47Hz and 65Hz among others. The waterfall and spectrogram plots reveal that I have issues with long reverb times for those frequencies.

I remember hearing about an active room damping system maybe made by Nelson Pass, which should consist of a relatively small active active speaker with a microphone and low pass filter.
The idea was to place the speaker in a corner and to have it play the inverse of whatever the microphone pick up. This way you should get the direct sound from your main system, but reverb and standing waves should be absorbed or cancelled out by the active speaker.

Have anyone head of such a system? And maybe have more info on how it works and how well it works?
 
Somewhere among the contrasting virtues of (1) Pass double-ended tube corner suppressor (also in Sweden), (2) the mega-price active absorbers, (3) motional feedback side-effects, and (4) anti-phase noise suppression (like in major powerhouses) there must be some actual data.

I can't read the Swedish. Maybe I can make google-translate help or if somebody could post the translation.

I'd be esp interested in the PSA Audio AVAA performance. What's the best way to measure how well it eats room resonances? Would you need FRs for a couple of locations in the room because not all locations suffer from eigentones??

B.
 
If you use the Panasonic WM61 capsule do the Linkwiz modification.
To reach low Fs of your woofer used for absorption add mass, preferably paint with epoxy close to the voice coil and stiff up the cone.
Do not use high order filters since they introduce phase shift.
Place the feedback microphone max 2" or 5 cm from the dust cap.
Turn up the volume (feedback) until it start to scream at high frequency., and then decrease until it is quiet.
Place the construction inside a tube trap or behind fiber glass to achieve best performance. And the trap works best in the corner!
The power consumption of the feedback amp is maximum 2 watt, on my absorbent when i play loud with my subwoofers.
 
But how do you know it is working? How is the sound of music different? How is the FR at your seat different?

It must be a simple matter to turn off the trap.

BTW, I find it helpful to use a mic stand esp if you need repeatable accurate locations, as would be the requirement here due to wave patterns. If you need readings in different locations, you can mark the location of the feet of the mic stand on the floor.

Footnote: I have a closet at one end of my music room and it is filled with my heavy winter motorcycling clothing. FR of the bass eigentones changes with the door to the closet slid open.

B.
 
Last edited:
The 1st step in dealin gwith room issues in the bass, as ASCTim has pointed out is multiple woofers as advocated by Toole & Geddes.

I fully expect that the active room damping you wish can be applied to those same subs to deal with any residual issues with appropriate DSP XOs so that there is no need for a separate box(es). Bascially described in at least simplistic form by Toole.

dave
 

Rallyfinnen

Member
2017-01-22 10:18 am
I have two room modes close to 40Hz I would like to improve, and I was inspired by this thread, and the Swedish discussion in the earlier link. I did some tests this weekend using a simple condenser mic kit, passive RC, volume pot, class D amp, and connected it to some speaker boxes I have. Mic FR was checked against an expensive measurement mic, and gave similar results to 20Hz after I replaced a cap in the mic amp.

However, I could not get good cancellation down to 40Hz. I got a boost in the low octave, and also in the upper bass/lower mid.
The upper frequency amplification (not cancellation) seemed to depend on the filter constant, and mic position, the lower part seemed to have some influence from the woofer resonance. It would absorb a little bit lower in frequency with a 10" heavy cone sealed sub than with a std 8" polyprop cone. I think the most cancellation I could get measured at feedback mic position was abt 15dB in the 100-200Hz area.

So, sadly it does not seem applicable for me, the room is pretty good in the higher frequencies, it's the 40Hz I want to get to.
I guess the limitations are caused by phase/delay in the feedback system, even if I'm no expert in this area.
I tried some extra weights on the woofer cone, but it only made marginal improvements. I thought the phase shift at woofer resonance was the cause, but it seems only to be a part of the equation.

I tried 10Hz and 100Hz on the LP RC, and could see changes in gain setting, cancellation at different frequencies etc, but the mentioned limitations were still there. Cancelling low in frequency seems hard to do. As a side note, power and woofer stroke was never an issue, surprisingly little cone movement and normal listening levels.

Any suggestions what to try before I scrap the project?
 
Last edited:
I'm getting (more) confused.

A room has eigentones. An active absorber in a corner converts the corner into a big window thereby disrupting the room's eigentones, in so far as which ones relate to that corner (not all do, I suppose).

What I don't understand is the relationship between (1) active absorption and (2) adding out of phase sound to counter noise, as in an electric generating powerhouse. Logically (to me), (2) can't work because phases mix-and-match differently depending on location.

As with all room FR testing when close to an eigentone frequency, depends on where your measurement mic is.

BTW, the Geddes multi-sub approach (which everybody wants to do) does not affect eigentones. It works by providing heterogeneous FRs to your seat, extra woofer power, etc.

B.
 
The active absorber only reduces the reverberation time.
And it is true that there is not anti noise cancelling.
I think it is Nelson who says that: "it´s like truncating the corner of the room"
Another good qute he does is: "the sound pressure is pushing the membrane inwards and the feedback is helping the membrane on its way"

Thats why the arrangement only uses less than 1w in power most of the time...
But still, it really works!
 

Rallyfinnen

Member
2017-01-22 10:18 am
Do you have a cap in series with the input of the amplifier, make it bigger..
Try to maximize the phase margins.
Place the mic as close to the speaker as possible to minimize phase lag.
Do you have right polarity?
Place the active absorber in the corner of the room behind rockwool.

Yes, I guess there is a cap on the class D board too, only too small for me to see or change :) So I should try with another amp!

Actually, I'm just putting the mic amp in LTspice now, to try to simulate the behavior, maybe there is still LF phase lag in that too.

Yes, I played around with the mic position a bit. And got different results. It seemed that placing it 90deg angle to the cone close to the surround worked fairly well too. Seems to depend a little bit on filtering, woofer etc..

I played around in basta looking at woofer phase, and it seems that using a really low tuning on the LP RC can compensate for phase in the LF region too (a little bit of LT in there maybe), but that will likely create other problems higher up. I think I saw similar problems when swithching between 10&100Hz before too.

Yes, polarity was correct.

I have not tested it in the room, only in the garage, measuring how much it cancels in frequency sweeps from the garage stereo. I run one sweep without it, and one with, to see the difference. I see no use of it in the room until I see proper cancellation at 40Hz measured in the same position as the 'feeback mic'.

This is the mic I used:
Kop Forforstarkare med elektretmikrofon till ratt pris @ Electrokit

Waiting for this one, was out of stock:
Kop Adapterkort med electretmikrofon och forstarkare till ratt pris @ Electrokit
 
Last edited:
BTW, the Geddes multi-sub approach (which everybody wants to do) does not affect eigentones. It works by providing heterogeneous FRs to your seat, extra woofer power, etc.

Now think it through. What is needed to induce an eigentone? A point source radiating that specific frequency, right? Now what happens when you add a second point source, radiating that same frequency, at a different location in the enclosed space? The resonance is out of phase with one of the sources. Therefore, with spatial distribution of sources you have active cancellation of eigentones. Without feedback solutions that mess up easily. If and only if the placement and phase alignment of the sources is correct of course. But that's why everyone wants multisub systems.
 

AllenB

Moderator
Paid Member
2008-10-18 11:31 am
What I don't understand is the relationship between (1) active absorption and (2) adding out of phase sound
Labels mostly. An active absorber is a standalone device designed for a specific purpose.
BTW, the Geddes multi-sub approach (which everybody wants to do) does not affect eigentones. It works by providing heterogeneous FRs to your seat, extra woofer power, etc.
These are essentially the same thing.