Push-pull and passive radiators

[AhmedF]

I have three questions about doing a passive radiator push-pull subwoofer, which I'll ask first and then follow with some setup information.

1) Is there any reason not to do it compared to having both of the drivers facing out or using a common enclosure?
2) Should the PR's also be push-pull, or does it not matter?
3) Are there disadvantages to having the two active woofers at 90 degrees from each other instead of on opposite walls or on the same baffle with one inverted.


After a deliberation and questions on a few DIY speaker sites, I've decided to upgrade my current subwoofer by doubling it up instead of scrapping it and using a different driver. The current subwoofer is a North Creek Music Systems Thunder (1), which uses a Peerless XLS 12 with the matching passive radiator I will power the system with two Outlaw Audio 2200 M-block amps (2) and EQ it with a Velodyne SMS-1 (3).

While I'm very happy with the sound quality of the Thunder, it did not have sufficient output in my new home's much larger living room.

I've been scouring this forum and the internet at large for multiple driver subwoofer cabinets, and I came upon an interesting page with information about push-pull subwoofers (4). I was thinking about building my new subwoofer as a push-pull with a woofer on the front and a woofer at the bottom firing into the enclosure, like the Miller & Kreisel (5) subwoofers, with a passive radiator on each side wall to cancel vibrations. However, in the six pages of threads on push-pull subwoofers in this forum, I couldn't find anything about combining push-pull drivers with passive radiators. So some questions:


Thanks,

FA

(1) North Creek Thunder subwoofer: http://www.northcreekmusic.com/Thunder/ThunderInfo.htm
(2) Outlaw Audio 2200 M-block amp: http://www.outlawaudio.com/products/2200.html
(3) Velodyne SMS-1: http://www.velodyne.com/velodyne/pr...mp;sid=676i995q
(4) Push-pull subwoofers (bottom of page) http://www.danmarx.org/audioinnovation/theories.html
(5) I looked up the Miller & Kreisel http://www.mksound.com/subwoofer.php

[FastEddy]

AhmedF: " 1) Is there any reason not to do it compared to having both of the drivers facing out or using a common enclosure? ..."

Don't do this: don't have the driver and the passive radiator on the same box face (same box plane), as one will be pushing when the other is pulling = cancelation of output = very little or no total output. (The bass sound is not very directional and will bleed right over to its adjacent speaker cone.) Dual active drivers = OK, as they both will push and pull in unison.

" 2) Should the PR's also be push-pull, or does it not matter? ..."

If by this, you mean that the PR is "spring loaded" or "stiffened" by other mechanical means, then, yes, it will matter ... as the excersion (movement) of the passive cone will be tightened up , changing its response ... this is all a matter of "tuning" the box, driver and passive to get the resonance you desire. (There are both types of passive radiators available = "spring loaded" and "unencumbered" .. and all push and pull to varying degrees, based on their "loading" = acoustic (sealed box) or mechanical or both.)

" 3) Are there disadvantages to having the two active woofers at 90 degrees from each other instead of on opposite walls or on the same baffle with one inverted. ..."

Only in that you may have "phasing problems" trying to keep all the frequency responses working together. Depending on frequency, some sound may get canceled (similar to above). IMOP, all active drivers should be on the same plane and passive radiators, box venting horns and, outlet baffels, vents, etc. on the "back side", etc. (For some interesting variations on this see: http://gr-research.com === very good kits, BTW ... higher freq middies and tweeters on different planes, bass drivers = not. GR also uses healthy spring loaded passives ... )

[AhmedF]

Originally posted by FastEddy
[B]AhmedF: " 1) Is there any reason not to do it compared to having both of the drivers facing out or using a common enclosure? ..."

Don't do this: don't have the driver and the passive radiator on the same box face (same box plane), as one will be pushing when the other is pulling = cancelation of output = very little or no total output. (The bass sound is not very directional and will bleed right over to its adjacent speaker cone.) Dual active drivers = OK, as they both will push and pull in unison.

The North Creek Thunder (see link above) goes lower than any commercial sub I've heard except the 18" Velodyne, and it has the woofer and PR on the same face.

What I'm considering for the new sub is somewhat different. One active driver would be on the front face. The other would be on the bottom, basket out, and wired out of phase. Like the M&K subwoofers. A single PR would be on each side wall of the enclosure.

Originally posted by FastEddy " 2) Should the PR's also be push-pull, or does it not matter? ..."

If by this, you mean that the PR is "spring loaded" or "stiffened" by other mechanical means, then, yes, it will matter ... as the excersion (movement) of the passive cone will be tightened up , changing its response ...

If I understand you right, my question is much simpler: would there be a reduction in distortion from having one PR facing cone-out and the other facing basket-out? I may not be able to do a PR facing out because that screw is pretty long and I don't want it to bludgeon my cat to death if he happens to be near the "outie" PR during a big bass note.

[Cal Weldon]

Don't confuse the drone cone with an active one. There isn't a cancellation as the drone is reactive not active. The direction of the basket is immaterial. There is no push or pull as the drone is capable of neither.

Think of a passive radiator system as a ported system and you'll do fine. If you feel that having the port on the same face as the active is OK then having the drone there is fine also.

I don't know about having two drivers at 90 degrees wired out of phase though.

[MJL21193]

Originally posted by Cal Weldon
Don't confuse the drone cone with an active one. There isn't a cancellation as the drone is reactive not active. The direction of the basket is immaterial. There is no push or pull as the drone is capable of neither.

Dernit! Beat me to it.

[Svante]

There will be racation forces on the cabinet from using a drone cone. The cone is accelerated, and the force comes from the airpressure inside the box.

So, it might be an idea to use dual slaves, one on each side of the box, if one wants to cancel cabinet vibrations.

Also, it might be an idea to turn one of them inside out, in order to cancel even order distortion in the suspension. This is not completely sure though, there is a slight chance that the suspension nonlinearity is either so small it does not matter, or that it can be in opposite phase to other nonlinearities in the system.

[jdgonko]

-----
Don't do this: don't have the driver and the passive radiator on the same box face (same box plane), as one will be pushing when the other is pulling = cancelation of output = very little or no total output. (The bass sound is not very directional and will bleed right over to its adjacent speaker cone.) Dual active drivers = OK, as they both will push and pull in unison.
-----
I definitely agree. I have two Infinity Basslink subs in my 98 Jeep GC vertically mounted with their passive radiators up against the rear bench seat (active driver firing towards the back hatch). Goal: eliminate the rear wave from cancelling the front one and lengthening the total wave path (plus there's space for a full-size spare, haha). See their website and photos of the unit to learn of their size and design.

I just wrote a sinewave program (C++) that generates any frequency per channel (L, R, L&R) that I want (quick and dirty coding plus cheaper than a true "test" disc). Adjustable phase, frequency sweeping, and repetitive beats (kick-drum simulation) to be added later, heh. But, so far, it is extremely accurate from what audio programs have measured from the WAV files. Attenuation per channel, a user input, in dB is spot-on.

I generated two second test tones from 120 to 10Hz in 10Hz decrements at -1.0dBFS. Burned a disc and tried it out. The sound was smooth until 50 to 40Hz where it peaked sharply (i.e. 3-4 times louder). Then, 30Hz much quieter than 60Hz (i.e. where the latter's response still seemed flat). 20Hz was nearly nonexistent. 10Hz? Yeah...right...

My rookie thought is that the design has a natural peak and then steep cutoff... And, is why I'm designing my own subs now.

[bdubs767]

just tried my first PR setup the other day and I must say I am very pleased.

[BAM]

The passive radiator doesn't vibrate in phase with the driver, or fully 180 degrees out of phase with the driver -- instead, at the tuning frequency, it vibrates 90 degrees out of phase with the driver. Any phase cancellation with the active driver is not affected by the location on the box where the passive radiator is placed. The wavelengths involved (even up around 200 Hz) are so long that any change in location of the drivers on the subwoofer box will have practically no effect on the sub's response.

I know, there are some subwoofers in the pro audio market that direct their sound forward using a set of rear-facing drivers (the "cardioid dipole" variant), but those require significant electrical delay on the rear-facing drivers in order to produce any meaningful phase shift.

Bottom line - it doesn't matter where you put the drivers in your subwoofer enclosure.

[bear]

As noted, the PR is equivalent to the "port" but acts more like a sealed box below the functional range of the PR, vs. an open box for the port.

There is nothing much to be gained in "distortion reduction" from reversing the basket of either the active or PRs. Other distortion sources dominate by orders of magnitude here.

What is gained is a reduction or increase in the internal volume of the box! If ur close to the hairy edge in acceptable box volume, it may make a difference - or if you want to impress the public you also might turn your driver outwards...

The position of the drivers in the cabinet and their relationships to each other absolutely will make a difference. But not for the reasons you think. What does happen is that the pressures inside the cabinet are not even - they vary with nodes and antinodes wrt frequency. They're not laminar or equi-pressure by any stretch of the imagination. The suspension of the PR and the drivers has low resistance in one axis and extremely high resistance in all others, so the effects of these node/anti-nodes are not apparent and any motion appears linear externally.

Does this matter? Probably not in most cases. But in some cases you can blow out a surround over time if things are just right for creating the right conditions at a given freq...

In terms of symmetry, it seems that drivers front and back, and PRs two sides is a safer bet than one at 90 deg... also the offset from gravity can not be ignored, especially with a high excursion (soft mechanical suspension) speaker since it will offset due to sag in short order...

_-_-bear

PS. at center tuning freq, it has best be 180 out of phase wrt the rear radiation of the driver, ie. in phase in the front of the PR... thus the traditional name for a port "phase inverter"...