Bass array experiments with extra tower speakers

[neRok]

My pair of Fosi V3 Mono amps have finally arrived (well, they're at the post office waiting for me to collect them). I bought them to get my tower speakers back in to action, because they've been sitting dormant for a few years now. They were good value 3 way speakers from a now retired Aussie company (Aaron ATS-5's), so they've got twin 8" woofers and rear ports. Here's some near field tests I did previously;

Not too shabby all things considered. My PC rooms first length mode is at ~40Hz, so there's potential for cancellation there (although the port location might not be effective in this case). But really I'm more interested in helping the 60-120Hz range (my room treatment is doing good work about 150Hz), and the woofers are almost covering that range on their own.

What's driving this experiment is the recent thought I had of using an array of 8" drivers on the back wall instead of an array of full power 12" or 15" or whatever, because the 8" are presumably cheaper, shallower, easier to drive, and even cheapies can probably cover the 80-200Hz range where room modes pile up. I'm also thinking ahead for Dirac ART, where they won't manage to cancel the SPL and low-bass range of big fronts and subs, but they can probably do a decent job of handling ~70% of the waterfall. So this test should give some evidence towards that idea.

Also I think I will try a couple of different physical arrangements like below and see what that can actually do in room. I've posted these ideas before, and whilst I haven't followed the forum closely, I still haven't seen anyone actually trying multi subs in these sorts of arrangements;

Anyway, I better get cracking. Hopefully I can get some stuff done today.

 

[Supra2k0]

Do you made some tests yet?

I am on a similar approach as I wrote on another thread :

I am wondering how to optimize my bass response with two existing speakers that are not in use at the moment.
I would like to mainly compensate for the nulls/low dips below 30 Hz, 38 Hz, and 50 Hz, maybe also at 89hz at my main listening position.
I have already tried repositioning (as far as possible), PEQ, DIRAC, FIR, etc. Space for absorbers is limited, and it is still a living room. The room dimensions are 4.95 m x 3.80 m x 2.85 m, with the main listening position (MLP) being 2.40-2.50 m away from the front of the speakers. The front loudspeakers each have three 18 cm bass drivers, while the rear speakers under consideration each have two 18 cm bass drivers. The positioning could be nearly symmetrical.

What do you think about it?

 

[Purité Audio]

I would enter your room’s dimensions into REW’s ‘room simulation’ then add your speaker and listening position.
Keith

 

[neRok]

Do you made some tests yet?

I am on a similar approach as I wrote on another thread :

I am wondering how to optimize my bass response with two existing speakers that are not in use at the moment.

I haven't got around to it, but I think it's worth a go.

I was just having a play with REW room-sim, and I think good results are possible. In a "normal" rectangular room arrangement with speakers firing down the room length, the length modes are definitely the worst, and the best problem to target IMO. And so having any bass drivers on the rear wall seems like it can do a decent job, even if it isn't a "plane wave", and even if it doesn't have good symmetry. It seems just having the right amount of power at the rear can help, even from 1 source. There could be bass localisation issues though?



It was interesting moving the speakers around in room-sim though, because when you get the front-to-rear modes relatively under control, all of a sudden the width and height modes start playing a significant role in propping up the peaks that are <150Hz. An actuall "bass array" deals with all of that via the plane wave, which this method probably won't create. But that doesn't mean the results will be bad! "Don't let perfect be the enemy of good" and all that.

But also, room-sim doesn't have allowance for putting random low-pass on the rear drivers, so they are probably propping up the >200Hz region more than they actually would. Additionally, it doesn't allow EQ in its calcs. A bit of EQ on either driver could work wonders for those stubborn peaks/nulls. And if getting in to full DSP/IIR mode, you could be merging cancellation impulses with different delays/volumes/etc to target different sections. So 1 driver on the rear wall might be able to do a fair bit (if it's powerful enough). And if you've got 2, like you do, then it might even be better to have the 2nd at the front working in a sort of cardioid arrangement. That way it might reduce the "back end" of the direct wave, and then the rear driver can clean up the reflection!?!

Lots to ponder.

Edit2: I just remembered something else - the strength of peaks and nulls you are trying to cancel is *probably* important. If you just have basic EQ, you might try position the speakers and yourself to minimise the nulls, even if the peaks get worse, because you can then EQ down the peaks. But for DBA style cancellation, it's probably best to get them relatively balanced, so neither needs particularly more power than the other. So for example, +6dB peaks with -6dB nulls will be easier to cancel from a power perspective than +9dB peaks with -3dB nulls.
*I say probably important because there's a lot to think about beyond peak numbers. You gotta think about time! (ie, the impulse.) Because the rear drivers start cancelling "straight away", whilst the peak normal max (peak or null) is after some amount of reflections. So what does that mean? Dunno.

Edit: I saw you posted this in the other DBA thread;

fed by a miniDSP PWR-ICE250 with DSP

You don't have to get all that pricey gear to test this out. Any old tower computer with onboard 5.1 soundcard would do. Possibly even a regular old HT receiver will be ok if it can send full range to the surrounds (fed a 3+channel digital signal via PC/laptop/etc). Things like DAC "quality" and potential hums from RCA cables etc isn't a big deal if you're just at the proof of concept stage.

 

[Flaesh]

What do you think about it?

You don't need the rear array to radiate above the upper operating frequency of the DBA.

I would have two speakers on the rear wall and would combine them with a separate DSP to modify the frequency response of the rear speakers using PEQ. The DSP will also cut off the tweeter in the floorstanders. That's the plan

It is (is?) possible to make FIR low-pass filter for the rear speakers that does not change the phase response.