Push Push Bass design

I wondered if someone can explain a bit benefits of the push-push design, each woofer placed on one side of the speaker.
For example MBL uses them in their 116. 120 and 126 models.
I understand it helps radiate sound as an omni source and as they shot in opposite directions the forces get cancelled out.
Sound wise, is this approach similar to Open baffle design?
Does it help to reduce room impact to the overall bass sound?
It is definitley not like an open baffle which would be a diplole. At low frequencies the push-push and the "normal" driver mounting behave alike.
The cancelling of the inertial forces is the main advantage of the design. I.e. less vibration that is fed into the floor (theoretically the vibration is even zero).
And if they are side-mounted narrow baffles with quite large drivers are possible.


What i call a push-pull is a symmetrically working loudspeaker. Most electrostatic loudspeakers are of this type, while most electromagnetic ones are not unfortunately. There is a partial fix tho: Two identical drivers with open baskets, in explanation which radiate out both front and back of the diaphragm, can be mounted with fronts close together, so that only a small air volume is trapped between the diaphragms. While a single driver would distort by pushing differently than pulling, two drivers in a push-pull arrangement do not. This is also related to compound or isobaric arrangements, which have some things in common with the push-pull arrangement. The coupling of both diaphragms is complexed at higher frequencies, tho, causing resonances, so you see it mostly in subwoofers.
Seems to me its just what they call it when its woofer/sub-woofers. Othervise, its bipolars.

The drivers only job, is to create over/underpressure. Much of which takes the easy way around and over the speaker. Especially in ported enclosures, where your much needed pressure gets cancelled. Driver pushes, ports sucks it right back in .. etc.

The push-push/bipolar is the opposite. The soundwaves from the driver reinforce each other instead. Allowing rather small speakers to generate low frequencies, and high output.

I have a couple of Paradigm Reference Eclipse BP. 18-22kHz, from nothing more than a pair of 8" woofer/mids. The woofers are mounted in front and back in a baffle barely wide enough to fit the drivers.

In my experience.... there is still a sweetspot, but the sound is more uniform with a flatter response in the whole room.

I do not agree that the main advantage is less vibration, but the theory that less vibration/loss is important, I completely agree with.
These are normally expensive speakers with really solid construction, but if one were to make a little devil sub, say two (or four?) serious drivers in a sealed box not bigger than a beer case, it would say the balance serves a real purpose.


I i ever get them finished, I´ll be more than happy to share some measurements. The nice one is 4x 8" from Seas , and the mean one two 18" PA woofers. Nothing fancy I think, but I figured a party sub might be fun.


  • savoy4.jpg
    37.5 KB · Views: 1,588
If drivers face into a manifold then compression loading can be incorporated into the design too at the expense of a larger box to add the manifold volume.

Polk had some dual driver sub designs facing out on two sides that sound very good.

Another advantage (or maybe disadvantage depending on the room) is that drives can be mounted close to room corners like some Allison designs.
pull/pull = push/push

If your thinking of just mounting the drivers cone in, its still push/push (bipole).
Normally only done if the dimensions are a problem. The box net volume is reduced since the driver doesnt take up any space inside. Not very high WAF on those :p

If the drivers(2 or more) are mounted inside the enclosure, its a three chamber band pass encosure. Band pass designs are used when you want to maximice SPL. With the compression chamber limiting xmax and the added efficiency we can get very loud small boxes. Even with poor drivers and materials.
The cost is poor low-frequency extension and narrow bandwidth..... or poor efficiency if its tuned for LFE.

Good for high output car stereo or if your company is called Bose. Not for Hi-Fi.
Joined 2001
Paid Member
The cancelling of the inertial forces is the main advantage of the design

If the bass drivers are tightly coupled magnet to magnet then one driver actively cancels the inertial forces of the other dramatically reducing the load put into the box, reducing the chances of box vibrations.

We built a pair of push-push subs with SDX10. Part of the effort was to see how much push-push reduced the box load. The box is made of 15mm BB plywood with a simple cross-brace scheme. I can pick up the box with one hand with no drivers, it is a big lift with drivers (i usually get help). To get the box to start ringing i have to turn things up near 10dB louder than i would ever use it. And only the top starts going. And that could be due to amp clipping as the frequencies detected were higher than the woofers pass-band.


We also often use this technique for helper woofers in FASTs.

Do note that at the frequencies involved any box — even with one driver, is an omnipole. Push-push may look like it would be a bipole but it is really an omnipole.

Joined 2008
Paid Member
Thanks, Dave

Thanks Dave, for your explanation and clarification.

There are several variations on the theme, but once you live with a system incorporating some fashion of vibration control aka reaction-forced cancelling,
all other bass systems sound like time smear/bass blurr.


  • reaction-forced.jpg
    355.5 KB · Views: 995
  • Like
Reactions: 1 user
It works either way. With magnet to cone you get reduced 2nd order harmonic distortion due to the hysteresis cancellation. Force cancellation comes from masses moving in opposite direction. Regardless of orientation the mass is the same.
Interesting, I've not considered that distortion could also be reduced in this way. Tight magnet to magnet coupling can actually be very difficult in practice, try it ;). I did manage it once with a pair of speakers I wasn't too worried about damaging and the vibration reduction was dramatic, I would be concerned though about distortion of the basket if the force of the coupling came from the usual mounting method. The only fool proof way might be to clamp or even glue the magnets together.
Last edited:
Joined 2001
Paid Member
Direct vrs indirect force transfer.

The Ripole is a cherry picked example. COmpact size, many small panels make the enclosure more rigid than most.

And there is no reason it can’t be push-push.


Direct connection thru the mounting bolts. Too tight a fit to have room for magnet to magnet.



  • ripole-woof.png
    38.6 KB · Views: 4,283
  • Like
Reactions: 1 user
But one has to worry about it increasing 3rd and less 2nd than 3rd has shown to less sonically pleasing than the other way around.


LOL! Me n' DJK [RIP] use to go round n'round on this! He firmly believed that less was less = good, no matter that the 'bad' one was elevated in amplitude, but then at the SPLs he worked with, doubt it mattered.

Regardless, at the high efficiency levels I'm used to, no way I'm going to do that.

It does beg the question if it would matter with these nowadays much more common very low efficiency drivers.