My Open Baffle-Less Speaker Journey

[Paul]

Hi Vic, I remember when you started this journey and how interesting your bass speakers idea was. My memory is that the two facing bass speakers in each unit were wired electrically out of phase to each other, and as they were facing each other the two speakers moved back and forth together and the space between them never changed, is that correct?

At the time you weren't particularly interested in exploring why you were getting such powerful bass response when accepted thought was that the bass notes produced by the front and back of the 'speaker unit' would cancel each other out and produce weak bass. I'm just wondering if you've had any further thoughts on why the bass response is so good in your interesting and original twin bass speaker concept?

I have my own thoughts but don't really want to crash your thread with conjecture that I can't prove without building a replica bass system and making measurements. 

[TransFi]

Hi Paul

Ahh....the phantom baffle concept 

Yes, the motion of the drivers you describe is correct. No idea why the front & rear waves dont cancel out.....although if you keep the drivers in phase so they oppose each other, there is no bass.

The guys over on DIYAudio kinda studied the concept, but it got a bit too mathematical. I'm not so interested in why it works....it just does for me., and I only need to convince myself!

https://www.diyaudio.com/community/threads/the-circles-of-doom-open-baffleless-full-range-speakers.374043/

The bass is powerful, afterall 4 UM18s come into play, altho nowhere near as powerful as they would be in a box.....but then no resonances & much easier to construct.

My take on it is the bass cancels out at the sides, so you only have waves coming from the front & back of the units. Makes it easier to place in the room and doesnt incite as many room modes.....

Your take would always be interesting!

Cheers

Vic

[Paul]

I’ve taken a few days to get my thoughts together before replying. The sentence below in your last post is at the heart of what originally got me thinking about how your bass speakers work.

My take on it is the bass cancels out at the sides, so you only have waves coming from the front & back of the units. Makes it easier to place in the room and doesnt incite as many room modes.....

I think there are two possible things that could be happening. One of them is that what you say above is correct. Normally, for instance, a full range driver not mounted in a baffle sounds pretty tinny, but if you put your ear as far into the cone as possible you get to hear some bass. So maybe the fact that your system is bi-amped and the bass amps are many, many times more powerful than the amps powering the rest of the frequency range means that your bass is powerful because of the sheer, irresistible horsepower driving it overcomes the effects of bass cancellation enough for the bass speakers to match the relatively feeble speakers above.

But still, it doesn’t seem right. I think cancellation of out-of-phase omnidirectional bass notes applies, no matter the power of the speakers, when there is nothing physically separating the front and back edges of the membrane generating the omnidirectional bass notes.

Which brings me to the ‘phantom baffle’ conjecture. My thought was simply that the only difference between a single speaker producing out-of-phase bass frequencies from the front and back surfaces of its cone and two facing speakers moving in parallel and producing out-of-phase bass frequencies from their outside cone surfaces is the space between the two speakers.

So what could be happening in the space between the two parallel moving speaker cones? It is easy to think that because there is no sound coming from the gap between the speakers that the air between the speakers would be inert/inactive/dead. But the air in that confined space is receiving the same amount of energy from the inner cone surfaces as the whole of the room is receiving from the outer cone surfaces. That air is being bounced around very energetically and that air has mass. It may not be making any sound but it must be having an effect on the air adjacent to it in the gap between the two drivers.

This is where the ‘phantom baffle’ thought came from. If that highly energised but acoustically neutral air effects the outside air around the gap between the two drivers, that effect could extend for a little way beyond the drivers, becoming ever more diffuse and eventually petering out. It would have a soft boundary which might contribute to making it easier to place in a room and not exciting as many room nodes.

This is all conjecture of course and irrelevant in that the only important thing is that your bass idea works and provides you enjoyment, regardless of how it works. I have enjoyed thinking about how it maybe works though and have even done a little drawing.  

[TransFi]

Interesting reply, and I agree with your speculation!

Also:

........your system is bi-amped and the bass amps are many, many times more powerful than the amps powering the rest of the frequency range means that your bass is powerful because of the sheer, irresistible horsepower driving it overcomes the effects of bass cancellation enough for the bass speakers to match the relatively feeble speakers above......

I think the choice of woofer is important. The UM18's are not very efficient, but the have a very low Fs (19hz), and only 3db down:



Wasn't until I started using these inefficient, heavy coned, rubber surround long throw drivers that I got any DEEP bass.

Love the pic! This is mine:

[Paul]

Wasn't until I started using these inefficient, heavy coned, rubber surround long throw drivers that I got any DEEP bass.

That is interesting, I think I know what you mean as I've had bass speakers like that in conventional cabinets and the bass was very good. The best was a pair of speakerlab 7 speakers with a 12" and 10" driver running in the same cabinet space. Heavy paper cones and thick rubber surrounds giving deep, controlled bass if the amplifier had enough guts to drive them.

If I get the chance I'll run an experiment. I have 8 and 16 Ohm pairs of the same 12" drivers. It would be interesting to mount a single 8 Ohm driver on to a couple of poles, then mount two 16 Ohm drivers on another pair of poles, wired the same way as your speakers are, and connect the 8 Ohm to one channel and the pair of 15 Ohm drivers to the other channel of an amplifier. I could then run bass tones and use the balance knob to get an idea of differences in bass output between the single and facing pair configurations. Just for my own interest of course.

[TransFi]

Paul, did you get a chance to  read the DIYAudio thread I posted a link to?

They have gone through various permutations using single naked drivers, doubled up naked drivers, drivers positioned back to back & face to face. They have taken measurements and discussed the results, many reasoning along your lines. Very interesting analysis. The OP also does a similar experiment you are proposing:



You can see the volume is doubled up using the clamshell arrangement.

More discussion here:

https://www.diyaudio.com/community/threads/celestion-system-6000-vs-gr-research-dipole-bass.352298/


Still, some of them are only interested in increasing bass output, and one guy says may as well use an H-frame. The whole point of doing it this way is to avoid the pressurisation & ensuing room modes the frames produce, i.e. minimise room effects.....in my experience anyway.

[Paul]

Thanks Vic, I'll have a decent read of the thread. I had only looked at the beginning and the end of it.

[Toaster]

I don't really get this 'phantom baffle' thing at all. I'd just expect more output from two 'clamshelled' drivers, exactly as the DIY Audio graph above shows. Two motors and cones, so more sensitivity and power handling. The air between the drivers is energised by the drivers, in that it's moving back and forth, but it's not being rarefied or compressed to any significant degree. The output of the drivers will be progressively attenuated, as frequency decreases, by cancellation, just like any dipole. There will be the usual dipole null at 90 degrees.

The reason Vic's subs have sufficiently high output and bandwidth is due to a combination of factors, mostly plenty of cone area and excursion capability (swept area), and low Fs, along with the related ability to handle a lot of power. EQ too, of course. I expect the drivers also have a higher Qts than most subs, if they're designed for IB applications, which would mean they need less EQ to compensate for dipole cancellation (effectively greater sensitivity at low frequencies compared to a lower Q driver), which doesn't hurt in terms of how much power is needed at low frequencies, but it's the usable swept area that's doing the job. Amplifier power as such isn't the issue, except in the usual way that you need more of it to go louder.

The late Siegfried Linkwitz's site www.linkwitzlab.com covers the theory and implementation of dipoles really well.

[TransFi]

Hey John.....good to hear from you!

You know I dont pay much attention to the science or what the 'experts' say....I just suck it & see. After many trials & tribulations I finally found something that works to my satisfaction in my room, so I am happy 

[Toaster]

Cheers Vic!

Although you've got where you are through experimentation, your speaker system has turned out quite 'sciencey' by most people's standards. Then again, experimentation is sciencey!

[Punchrulle]

How do you determin the distance between the opposing um18s(or other woofers) in this application?

Have you tried to apply/push less power in to the um18 att the back then the one in front/facing it? Would that achieve some kind of cardiod ~ dipole mix?

How about putting som delay on the rear driver to simulating a bigger ”phantom baffle”?

[TransFi]

I dont think the gap between the 2 woofers makes much difference (at least below 10cm) as it all cancels out.

Not tried what you suggest. They are operating below 60hz now. Most of the detail, texture, slam etc is coming from the Focals.... the UMs just give that low-end rumble!