Subwoofer Efficiency is Mostly Irrelevant

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
These plots are very useful in understanding how each variable might influence passband SPL, and the behavior at and below resonance. But they don't quite address the original claims made in the title of the thread. For example, compared to a home subwoofer, a pro audio sub might have some or all of:
  • higher BL
  • higher Fs
  • lower Mms
  • higher Cms
The difficulty in making a sweeping general statement about one "class" of driver versus another is that drivers don't always neatly fit into well defined categories and new drivers (especially for pro audio) may be true outliers, e.g. iPAL line . In a nutshell, the designer should probably just consider each driver individually, both on its own merits and with its own shortcomings.

What I think IS very useful from this conversation is to make people aware of what to look for when doing driver comparisons. I think the concept of looking at the true power demand of a driver is rather new and when that driver will be used in an application where a lot of input power will be applied the difference in efficiency can show whether the driver and amp (power input) combo will truly meet the expected performance goals.
 
I see your point of efficiency.

Simply put, you are balancing a driver's deep F3 with its 1w/1m spl.


We can choose a high spl 1w/1m driver, but then the qts is low, thus the bass naturally rolls off at a higher frequency. That low qts would then need boosting for similar F3.

So either you add 6db of boost for 30hz F3 (with a low qts driver), or you pick a higher qts driver that doesn't need any boost that is 6db lower 1w/1m. Sort of a wash.


But the low qts driver would require a much smaller box, there is that...……...
 
Last edited:
I thought that for frequencies just below driver resonance peak the impedance was mainly inductive and above the resonance peak when the impedance was dropping, it then becomes capacitive, which is the opposite of what is said in the text above for Attachment#1 in post 40
(maybe a typo?)

Electrical characteristics of dynamic loudspeakers - Wikipedia

cheers,

You are confusing the capacitor Cas in the "analogous electrical circuit" with the actual electrical behavior and phase angle of the current thru the voice coil, which follows what you have described.
 
After receiving a few PM questions, I thought it might be helpful to post some of the trend plots I had generated a while back to illustrate how changing the 3 major electro-mechanical parameters of a woofer(mass, suspension compliance, and BL) affect TS parameters, mid-band sensitivity, and response for the same applied amplifier voltage. I grouped a few of the plots in what is hopefully a logical presentation.

Thank you bolserst, very helpful.

You and Charlie have me going back and re-measuring the impedance of my various sub builds, comparing the impedance peaks' widths vs specs.
 
Great discussion. I'd only like to add something that represents a great loss of quality performance in subs over the years: increasing driver power handling to crazy levels (which is what the driver Marketing Department urges on us*).

Why? If you agree that BL in drivers (or analogously in motorcycles) is the key parameter then making a driver which can handle 500 electrical watts means using really gross wire gauges in the VC. For a given magnet, that enlarges the gap a whole lot and makes BL shrink exponentially faster yet.**

So buying plain brute BL is the wise move.

B.
* OK, I am now listening to Vangelis's Mythodea (Jesse Norman and dozens of percusionists simulating the blast-off to the moon)
** someday humankind will re-invent electro-magnetic drivers now that solid-state rectifiers are cheap
 
Last edited:
I've been rounding up some different boxes with different drivers for my friend who is in car db scene. He has some contact to small importer which drivers i think is small batch from china.
And these drivers given parameters hornresp gives grievance. I've told to my friend i cant say how these drivers can be simulated if parameters can't be trusted.
I never look at given sensitivity.
I just throw given data to simulating software and check what box it would give wanted response.
I hope he won't buy these piece of shait subs anymore but sure many car subs seems to be louder than PA ones under 40-50hz.
My wet dream is that he would buy 2pc 18" ipals and i could desing box which goes to +100hz.

Last 4x12" subs on large vented peaked 152db 43hz or something but sounded so bad.

DB scene is so stupid
 
Last edited:
If they would measure the sensivity at the passband used of the driver, it would matter, but they don't so the published sensivity does not matter, but the real one does. And that is seen on a combo of factors like mentioned above. Those tell more than the publised sensivity number. But saying that the sensivity does not matter without context is ********.
 
Was looking for something else and found this thread.


my 2 cents.


Car subs have evolved over the past 20 years to take advantage of cheap power.



They currently trade power handling for small boxes that play low bass, per hoffman.


They use a heavy cone to do this. it is essentially an isobaric pair in a single frame, huge motor and heavy cone.



Btw, the SA sub is an entry level sub by that manufacurer, their upper model 18s have motors of exceptional size with BL curves as flat as proverbial pancakes by comarison to the SA's nice parabola.


A typical user of these subs will drop them in about 4.5 cubes each and power them with anywhere from 1500 to 2500w a piece. they have phenomenal xmax, 25mm one way.



4th and 6th order enclosures are very common applications.


sundown makes an amp that will place 20000w into a 1 ohm load all day and they sell a number of subs that will take that all day as well.


the car scene is quite insane, and i love it. its like drag racing with sound, absolutely useless for anything but fun.
 
If you know of Stereo integrity, the sundown Owner and SI owner are great friends. Both very passionate about audio.



SI just released a 12 that will produce a flat output to 1200hz in a 1 cuft sealed box.


the 2 men are doing fantastic things with motor geometries that place amazing BL at their figertis without huge Le penalties.
 
…making a driver which can handle 500 electrical watts means using really gross wire gauges in the VC
Short term power handling may increase from greater thermal mass of individual wires or the total voice coil. But, in general, the long term power handling is determined largely by the surface area of the coil(ie large diameter=good), efficiency of the thermal conduction paths to the top-plate/pole-piece, and the ventilation or airflow paths built into the motor assembly.

… For a given magnet, that enlarges the gap a whole lot and makes BL shrink exponentially faster yet
Unlike magnetic tractive force, for the same magnet and gap area the flux density in the air gap varies linearly with the gap size. The magnetic circuit is analogous to a simple series electrical circuit containing a voltage source(magnet), wires(iron path), and resistor(reluctance of air gap). The magnitude of the current(flux) is determined almost entirely by the reluctance of the air gap which increases linearly with size. I agree it is desirable to keep the gap as tight as possible for increased B, but it doesn’t vanish in an exponential fashion as you increase the gap.

… buying plain brute BL is the wise move …
Mid-band and LF efficiency are both proportional to (BL^2/Re), so you can’t look at just BL.

Re = ρ * L /Aw where:
ρ = resistivity of metal used in wire
L = total wire length
Aw = area of individual wire

If the VC is similar in height to the top plate, a little algebraic manipulation can remove Re and L from the equation, showing efficiency is proportional to (B^2*Vw/ ρ); where Vw is the volume taken up by the VC winding. So for a given Vw, it wouldn’t matter what wire size you use to wind the coil, the efficiency will be the same. What will change is the sensitivity, or how much voltage you need to get a given SPL. Midband and LF sensitivity is proportional to (BL/Re). With efficiency set by B and Vw, wire size and number of layers can be selected to achieve a desirable (BL/Re) ratio such that target SPL can be attained using amplifiers available in the current marketplace. Huh, one thing I hadn't noticed before is that for a given gap size, increasing the diameter of the VC would increase Vw and thus efficiency.

In reality, things are "a bit" more complicated than this simplified view used to illustrate some of the first order relationships between efficiency, sensitivity, and motor parameters. In particular, as the length of the VC is extended beyond the gap to improve BL linearity with stroke, wire size will need to be increased to retain reasonable sensitivity. I'd wager this is the primary reason you see larger gauge wire in long-stroke subwoofers; not pursuit of power handling numbers...although that doesn't seem to stop the marketing department from bragging about it. Another thing to consider is that VC inductance, which increases as the square of the turns & diameter, reduces sensitivity when more turns of finer wire are used. Compromises will need to be made between motor parameters like VC length & diameter, wire gauge, number of layers, amount of shorting material taking up gap space, and performance parameters like Xmax, distortion, power handling, and of course efficiency and sensitivity(including inductance effects).
 

Attachments

  • Woofer_Motor_B.png
    Woofer_Motor_B.png
    20.3 KB · Views: 168
If you know of Stereo integrity, the sundown Owner and SI owner are great friends. Both very passionate about audio.



SI just released a 12 that will produce a flat output to 1200hz in a 1 cuft sealed box.


the 2 men are doing fantastic things with motor geometries that place amazing BL at their figertis without huge Le penalties.

Scott at Fi Car Audio is coming up with some great drivers too. How does 34mm one way Xmax grab you?
 
Last edited:
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.