Midbass Decoupling vs Bipole- Newtonian reaction.

[joe carrow]

This post is to start discussion regarding the negation of newtonian reaction forces between a midbass woofer and its enclosure. As a speaker motor pushes on a woofer cone, it pushes back on the enclosure. The forces cause vibration in the cabinet, and this causes coloration in the sound.

As I understand them, here are a few ways to fight cabinet coloration from the newtonian reaction forces:

A common method to fight this effect is sort of brute-force; the cabinet is made to be very rigid and heavy so as to minimize vibrations. Ideally the cabinet is built from a material that is self-damped, such as a plastic.

Another method is decoupling, like the method used by B & W for some of their higher end speakers. The motor structure is allowed to move slightly in relation to the cabinet, expending the vibrational energy on a viscous damping material.

Thirdly, a bipole uses two speaker cones firing in opposition- one facing front, one facing back, both move in and out in unison. The reaction forces cancel, and if the motor structures are rigidly braced together inside the cabinet the forces don't disturb any cabinet walls.

Some things I wonder if anyone wants to tell us-

Can anybody comment on designs they have built/heard with similar drivers that used different methods? planet10, I think you did a few bipole/monopole fullrangers...

Does anybody know the relative magnitudes of the forces generated by internal sound pressure compared to the reaction forces?

Are there any rules of thumb regarding wavelength of the highest frequency played by a bipole?

I did search the forum for these, and it seems that I missed a few posts with good links buried deeply. If there are any good links out there, I'd love to see them

That's all for now- I hope someone else is as interested in this as I am!

Joe

[AndrewJ]

Joe,
the thread you missed is active right now!
Take a look at : Idea? Counterbalancing cones. :)

KEF was possibly the first to introduce de-coupled drivers, with the R105.
The idea has been used by many people many times.
De-coupling is only effective at frequencies above the resonant frequency of the suspension, but can still work extremely well.
I have used it both for my KEF designs and for Pioneer and TAD.
The best solution is the dual driver bipolar type, as the reaction force should be fully cancelled at all frequencies, as long as the drivers are well matched and identically loaded. But if you roll one of them off at a lower frequency than the other (as you might be tempted to do in a bipolar design), then the attendant phase shifts will negate some of the force cancellation

The reaction force is typically 10-30 times stronger than the force generated by internal air pressure at the frequencies of the cabinet resonances, and with appropriate choice of de-coupling you can readily reduce cabinet resonances by 20db or more.
I have many times demonstrated the effect of de-coupling upon the sound quality, taking a pair of speakers matched to better than 0.5dB and then de-coupling just one of the pair. The improvements are not subtle.!
I have written a white paper on the subject, and if I can find it I will post it in this thread.

Andrew

[planet10]

Originally posted by AndrewJ
KEF was possibly the first to introduce de-coupled drivers, with the R105.

I was working at a hifi store selling KEF new when those started showing up. Our feeling was that the speakers sounded better when the bolts were cranked down to defeat the lossy coupling. If one were to somehow ridgidly support the driver & still decouple the box that might work well (as used by the Japanese guys with the "egg" shaped full-range speakers -- can't remember the name off the top of my head.

I'm a big fan of bipoles. IMHO it makes quite a bit of difference. At 1st i was skeptical of full-range bipoles (LF bipoles where the wavelengths are omnidirectional is a no-brainer), but listening to many has made me a convert (even an evangelist ). I'm also a FR fan so i just let the drivers go up all the way. Granted -- the 2 rooms i do most of my listening in are great with bipoles. Since the lower the frequency, the more reactionary force you have to deal with, rolling off the back driver in an x.5 way fashion below the baffle-step won't lose you all that much in terms of force cancelation.

Big, massive cabinets, to my mind, often suffer from too much stored energy, cost, weight, etc.

dave

[sdclc126]

I have used a device on my loudspeakers called Micro-Scan - they look something like an amplifier heat sink and attach to the cabinet via a self-stick magnetic plate. Somehow they convert the mechanical resonances in the cabinet into heat (hence their apprearance).

They are (were) made by a company in Willits, CA (USA), and work extremely well - I auditioned them at a high-end store back in the 1980s - did an A/B test - one speaker with, one without. The rep just slid the balance from one side to the other - there was a very impressive difference - the speaker (a good brand as I recall) without the device had some resonance in the mid-bass region, and when we switched to the speaker with the device it was virtually gone.

Obviously individual results will vary - speakers with better cabinets will have less audible improvement, because there's less ROOM for improvement, while lesser-built speakers will have a more extreme improvement. But I was sold on these and have used them on my Yamaha NS-670s ever since, although they have very tight cabinets to begin with.

I don't know if these are still made - I haven't found any references on the web, nor seen any mentioned or for sale, but I would highly recommend them if they can be found, as they are "plug-and-play" - you just stick them on the back of your cabinets and you're good to go - no other mods needed.

Maybe others have experience with these or know of similar devices that might still be available?

[soongsc]

There are two aspects resulting from newtonian reaction. (1) Slight loss of resolution, and (2) excitation of box structural modes.

To minimize (1), the higher cabin mass to cone mass ratio, the better. This requires rigid attachment of driver to cabin. However, this kind of attachment also makes it easier for the driver to excite cabin structural modes.

To minimize (2), putting damping material between the driver and cabin to helps, but then the effective cabin mass to cone mass ratio goes down because the effective mass is now only the driver magnet and frame.

Ultimately, you would want the lightest cone you can get, the most heavy and rigid cabin, and find a way to damp the high requency structural modes.