Bottom limit for driver FA resonance?

[bentoronto]

I'm a big fan of sealed boxes and horns with sealed boxes behind the driver. But their low frequency performance is limited by the resonance of the driver.

Villchur made the AR-1 around 1954 with 2 cu. ft. and a driver with a resonance, just trying to remember, of maybe 12 Hz in the fresh air. Anybody recall if that was what the driver resonance was? I think the system resonance rose to maybe 35 of so and decreased not too fast below; seemed quite profound on organ music, maybe even today. The cone was made from some kind of heavy industrial-stiff 1/8 cardboard and a very compliant suspension. Great speaker. I have one sitting around... waiting for the Antique Road Show to come to Toronto for a dollar appraisal (kidding).

I've sometimes wondered what are the practical obstacles to making drivers with a 5 Hz free-air driver resonance? And/or designing sealed box speaker systems for them? What's the bottom limit to driver resonance?

(I hope nobody hijacks this into a thread of about why this-or-that enclosure is better than a sealed box.)

[MaVo]

Driver resonance doesnt limit the response. Simulate your system and look at the graph. There is allways output below fc.

[Iain McNeill]

On a very fundamental level... you get low resonance by increasing the mass of the cone and by reducing the compliance of the surround (the spring).

Unfortunately doing both these things reduces the control you have on the movement of the cone therefore distortion starts to become a problem.

[Pikefish]

Increasing the mass lowers free air resonance, and lowers sensitivity. Also, INCREASING the compliance lowers Fs, while at the same time increasing sensitivity, but the mechanical damping goes down too, increasing some distortion. Higher Sd for given cms & mms also leads to a lower Fs, but there's a whole raft of reasons they don't make those.

A very low Fs driver might have high mass, and have a higly compliant (soft) support structure, but this will also cause it to sag massively, probably even lean sideways when mounted side-firing, causing all manner of excitable wobbles, maybe even so much that the VC hits things in the gap.
The remedy for this then seems to be a stiffer or another spider/surround, which brings down the compliance and brings up Fs, and decreases sensitivity even further. Argh!

This discussion is the domain of rotary subs, they go to "0Hz" (just blowing air), anyone put one of these in a TL?

[bentoronto]

Appreciate the explanations. But still not clear why "... a car will never go faster than 17 mph..." when it is just a matter of incremental technology or novel parameterization.

Surely somebody ought to be able to make (and prepare for shipping) a 5 Hz, down-facing only driver, even with a fairly conventional format? Certainly folks who packaged them into a powered, sealed, equalized, flat-to-18-Hz, fed-backed sub-woofer system.

Anybody know the Villchur AR-1 free-air resonance?

[sreten]

Hi,

Driver Fs depends on the cone mass and the suspension.
Its Vas (equivalent box volume) depends on the suspension
stiffness and the cone area.

Very low Fs = very high Vas for the driver size / cone mass.

Very low Fs requires low suspension stiffness that once
you put it in a sealed box makes hardly any difference.

That is Fbox is dominated by Vbox and Vas hardly matters.

Very high Vas and low Fs makes a vented box not a good idea
at all because it will be very susceptable to subsonic overload.

For sealed boxes ideally Vbox should be < 1/3 of Vas.

e.g. for Qbox = 0.7,
Qdriver = 0.4, Vbox = ~ 1/2 Vas
Qdriver = 0.3, Vbox = ~ 1/4 Vas
Qdriver = 0.2, Vbox = ~ 1/12 Vas

However many drivers used sealed have highish Q,
(used sealed because Q not good for vented)
e.g. for Qbox = 0.7,
Qdriver = 0.5, Vbox = ~ Vas
Qdriver = 0.6, Vbox = ~ 2 x Vas

In the lower Qts cases Vbox << Vas, and massively lowering
Fs and Qts by increasing Vas will not make any real difference.

As resonance mass / compliance is a square function I'm
not sure you are aware just how floppy 5Hz would be.

Take a typical 12" sub with Fs ~ 20Hz and cone mass ~ 250g.
To get to 5Hz you need to remove 15/16 of the suspension,
(or add 3.75kg / 8.25lb to the cone mass .....)

I'd say the practical limit is around Qdriver = 0.3.
Lowering Fs and Qts beyond this will not make much
difference except to the numbers on a spec sheet.

/sreten.

[myhrrhleine]

No limit.
Just engineering issues.
lower Fs means lower sensitivity and more HP to drive it.
also greater xmax to produce the sound at equal sound levels

[bentoronto]

Anybody have the engineering sophistication to critique the assumptions (both those that are explicit as well as those that are implicit) and the pessimistic conclusions of sreten?

Or should we conclude with sreten that moving free-air resonance below, say 10 Hz is presently inconceivable?

[sreten]

Hi,

Pessimistic ?
That sort of implies I have chosen "the facts" that suit me.

Going below 10Hz is not inconceivable, it is pointless unless
the required Fbox is also extremely low, say less than 20 Hz.

Equalised flat to 18Hz, and flat to 18 Hz are very different things.

For Fbox = 35Hz as you mention, going below ~ 15Hz is pointless.
You can take that 15Hz driver and remove 3/4 of its suspension.
Its now a 7.5Hz driver.
In the same box Fbox is still ~ 35Hz (obviously slightly lower).

/sreten.

[bossobass]

Originally posted by bentoronto


Anybody know the Villchur AR-1 free-air resonance?

The AR-1 woofer was built with a cast-aluminum frame and Alnico-5 magnet (9.3 lb. magnetic circuit), a 2-inch heavy-copper voice coil, double-wound on a bronze bobbin, with approximately one-half inch overhang in the gap. The free-air resonance of the early woofers was approximately 14-15 Hz, and later ones were closer to 17 Hz. Mounted in the 1.7 cu. ft. AR-1/AR-3 enclosure, the system resonance rose to the optimum 43 Hz, +/- 3 Hz.


Bosso