I've only made preliminary tests with that, because I only have one 18" speaker available (which normally is in one of my bassguitar cabinets).bushspeakers said:
The idea is simple: I used a simple frame (a ladder, in fact) from where I could suspend the driver (from the basket beams) with a few rubber ribbons so that it can float, especially forwards/backwards. The rubber ribbons (amount and tension) were chosen to give a very low resonance frequency, sub-Hz regions. Completily undamped (this is a issue, one would certainly need some damping, preferably aperiodic).
The mass ratio of driver to cone+airload then gives the amount of opposite movements of cone vs. whole driver. Therefore, heavy driver with light cone is best. Light cone is also all what's needed, since there is not much airload, especially on the cone edge -- this will surely help reduce distotion also. The advantage I see is that there is practially no energy storage other than in the driver itself (one could dampen the driver frame to go further) and there is absolutely no coupling to the floor or any other structure. The disadvantage is low output and the overall dipole problems with excursion and air noises etc. which makes driver choice critical. To get some resonable SPL you can't have much distance (I had ~ 2 feet, woofer set up behind listening pos.) and you need a lot of EQ, to make the response flat. I chose to drive the transducer with high source resistance to get a resonable Qts (~0.6 or so) and had to use the dipole shelving filter but not down to the lowest frequencies, only to the point when the proximity effect starts (as described on Mr.Kreskovsky site).
I tried this as a mono subwoofer up to 90Hz with my Tannoy minimonitors and the result was very pleasing. I've heard several OB systems before and am used to the raw driver sound so I knew what to expect, but it was really better than expected. The cleanest and "quickest" bass I've heard, and that with a purely experimental setup, nothing tweaked. The most remarkable quality was the effortless way the bass is projected into the room without overloading it, and almost no modal problems, thanks the near-field setup. And that sense of deepness.
No comparsion ie to my JBL floorstanders, which have "full" bass but of the diffuse "boom-boom" variety (and that after a lot of placement optimization and room treatment).
For a practial solution one would probably need four 18" or 15", two per side or set up as 4 distinct subs surrounding the listener. And a small amount of shaped baffle to get more SPL and also to provide a smother short circuit path for the air, to get less moving air noise from the driver edge.
To make something really useful with this idea, it will need quite a bit of experimentation, and it's a very inefficent approach. For the expense of the drivers and the efforts of EQ'ing them etc one can surely build a bunch of plain monopole subs and go the distributed placement road. OTOH, this effertless -- soundwise -- dipole bass sound has something unique to it... I definitely got bitten by the dipole bug from the day I heard, guess what .... Linkwitz' Orions...
- Klaus
This makes me think of a story told by a friend.
He said, once he obtained a lengendary EV 30W. He and people around couldn't wait to hear it. So they just hooked it up and play.
Without any box or baffle, just naked driver alone, face-up, magnet on floor. And no filtering and no EQ at all.
He said the sound was amazing, everyone there was stunned -- how can a naked driver produce bass like that?
I wasn't there, what a pity!
He said, once he obtained a lengendary EV 30W. He and people around couldn't wait to hear it. So they just hooked it up and play.
Without any box or baffle, just naked driver alone, face-up, magnet on floor. And no filtering and no EQ at all.
He said the sound was amazing, everyone there was stunned -- how can a naked driver produce bass like that?
I wasn't there, what a pity!
CLS said:Thanks for the info, Rybaudio.
However the measuring range seems too narrow for my applicationI'll need constant response over a wider range than a couch, say, maybe 10 seats, or simply walking around within a few steps.
There's no reason why you wouldn't be able to do that. In order to get the responses above I placed the subs so that they excite the modes in the room in a very "complimentary" way. I could slide the couch back a little, put another one in front of it and adjust things a little to get a similar response over 6 seats. When I first started messing around with this I didn't have any furniture in the room and measured a large grid of responses for a few different sub placements to try get a feel for how things worked... obtaining a consistent response over a large area is definately possible. The only way I've ever been able to do that is using multiple subs... not directional subs.
It's actually a very convenient layout. The room is an awkward shape... it's an apartment where the living room, dining area and kitchen are all one meandering area, but it does behave somewhat intuitively. A (crude) drawing is attached. The "guess" was that the main listening area behaves modally somewhat like a rectangular room, and thus I have the subs arranged to take care of all the modes in the two lateral dimensions except for the even order ones in the long direction... the only one that matters (2nd order) ends up being at about 45-50 Hz and that is the EQ I have in place. If I flip the phase on either the front or back they drive that mode with opposite polarity and it isn't a problem but then the odd order modes get driven in phase so I get a nasty peak down low and at 70 Hz. With the subs in the back, there is some sort of modal behavior with the dining/kitchen area but it doesn't effect the listening area too much... mainly there's a slight side to side variation in the 20-30 Hz region where the main listening couch is. If you move it forward a few feet the effect goes away pretty much entirely. The response is not all that good where the fouton and chair are, but I don't care... they are located where they are for (1) damping near a boundary and (2) providing a "social" layout. If I was going to do a theater I'd probably do two rows. The next step is to try some more LF absorbtion to further smooth things.
Oh, and they are mono; I don't see much point in stereo bass personally... I have a hard enough time getting one channel smooth and consistent over a listening area.
Oh, and they are mono; I don't see much point in stereo bass personally... I have a hard enough time getting one channel smooth and consistent over a listening area.
Attachments
gedlee said:
Hi gedlee, i would really like to know how you damp the room, especially in the lower frequencies. This topic is much too often ignored and after an internet search i only found rule of thump solutions.
As the topic is cardioid bass, which in itself is an attempt to controll the bass in the room, i think room treatment is directly related.
Thanks for sharing. I believe this is a pretty good reference for a normal household usage.
However, in this case, I need the bass to do upto maybe 150~200Hz, so it has to maintain stereo.
If I have to fill the bottom octave holes left by dipole or cardioid bass, I will consider more on this.
However, in this case, I need the bass to do upto maybe 150~200Hz, so it has to maintain stereo.
If I have to fill the bottom octave holes left by dipole or cardioid bass, I will consider more on this.
I though I post this for anyone interested. The first link is to an old cadrioid, dipole, monopole woofer design I used to have avaliable. No longer available, but the concept may be of interest.
http://www.musicanddesign.com/craw.html
The second link is to a discussion of power matching between dipole midrange and a woofer system.
http://www.musicanddesign.com/PowerMatching.html
The third link addresses power radiated and polar pattern in terms of directivity index. This last study may be of interest because it breaks down the variation from monopole through cardioid to dipole interms of the relatibe strength of the dipole and monopole sources.
http://www.musicanddesign.com/VariabDF.html
http://www.musicanddesign.com/craw.html
The second link is to a discussion of power matching between dipole midrange and a woofer system.
http://www.musicanddesign.com/PowerMatching.html
The third link addresses power radiated and polar pattern in terms of directivity index. This last study may be of interest because it breaks down the variation from monopole through cardioid to dipole interms of the relatibe strength of the dipole and monopole sources.
http://www.musicanddesign.com/VariabDF.html
MaVo said:
To damp a room at LF you have to do it structurally. There is no effective acoustic solution ala tube traps etc. I make the walls highly damped to LF vibration by using Constrained Layer Damping in the walls them selves. The details of construction are given in my book on Home Theater. I have found this technique to be very effective at the LF mode problem. This technique and multiple subs makes for the best bass that I have heard.
john k... said:
John - interesting. All of your studies seem to be in a free field condition. For LF isn't this unrealistic? The room is the dominate factor at these frequencies and to exclude it seems like a problem to me. When I studied the room and how it couples to the LF sources I continually find that monopoles work the best - but admitedly the bigger factor is the room and source number and location.
CLS said:Sorry I don't understand how to use several dipoles to do cardioid.
Your pic shows one dipole woofer and two boxed woofers.
Since a boxed woofer can reach much higher outputs (down low) as compared to a dipole woofer you need something like 2-3 dipole woofers for each boxed woofer if your goal is cardioid response down in the deep bass. They should also be equalised for the same on axis response (more or less).
/Peter
Regarding room resonances and their damping, IME there are several effective worthwhile methods like membrane absorbers and 1/4 wave resonance traps. Helmholtzabsorbents works fine to but they must be bigger to give the same result as a membrane absorber.
Best is always to integrate the losses in the room when building it but that is something that is not an option for many people.
Multiple subs with some traps and a bit of EQ goes a long way.
/Peter
Best is always to integrate the losses in the room when building it but that is something that is not an option for many people.
Multiple subs with some traps and a bit of EQ goes a long way.
/Peter
MaVo said:
Thanks alot, i will study this topic
To be fair SOME LF absorption can be achived with tuned absorbers, but they take up a lot of room and don't work all that well unless they are sharply tuned.
But contrained layer damping could be added to any existing room with only an inch or so off of 1/2 the walls. Its a bit of work, but doable. And it makes the wall far more sound proof which is a very good thing in an attached home. You only need to damp one each of two opposing walls. Note that the wall itself should be fairly reflective so as to not overdamp the HF. Its the supporting structure that is damped and this only gets excited by the very LFs.
My plans up till now were to use several plate resonators for damping of the modal range and then to tame the remaining modes with tuned absorbers.
The resonator i intended to use is described here: http://www.google.com/patents?id=JX4XAAAAEBAJ&dq=5975238
Building CLS walls is alot more complex, but the product will be very friendly for the eye, so i like the idea. Additionally, i think its good to alter the room to improve the quality of the reproduction instead of compromising the subwoofer(s) for this.
The resonator i intended to use is described here: http://www.google.com/patents?id=JX4XAAAAEBAJ&dq=5975238
Building CLS walls is alot more complex, but the product will be very friendly for the eye, so i like the idea. Additionally, i think its good to alter the room to improve the quality of the reproduction instead of compromising the subwoofer(s) for this.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.