I think that for LF, and especially for SW, the more cone area the better, although, as Sean says, there are other imnportant considerations. The large cone area is not to play louder. It is to play with the proper loudness without need for extreme cone excursion. A small subwoofer must be very loud one foot in front of it because the sound pressure must spread out to fill the whole room, with the SPL falling off rapidly with distance. If people used multiple 15" subwoofers (one for each speaker system) there would be less chatter about the difficulty of locating SW in the room. |
If it is a question of 2 smaller subs vs one larger one, I absolutely agree with Rives. The recommendation for large cone area assumes that cost is not the deciding factor (which I admit may be unrealistic unless you are building the subs yourself). Velodynes don't come cheap.
The loss of low frequency at low volume is mostly the result of human hearing characteristics, rather than actual reduction of SPL from the speaker system. Corrective action for this is the "Loudness" control that used to be a common feature of preamps. |
Sean...A speaker (driver) is most like a PM DC motor. Except for static friction, (if any) which causes a "deadband" for plus/minus a small voltage, the torque constant (KT) of such a DC motor is completely constant over its useful range. We used a servo motor with more than 400 inch oz of torque capability to measure gimbal bearing friction of about two inch oz, and gimbal unbalance to an accuracy of better than a tenth of an inch oz. Perhaps you are thinking of AC motors, which do not develop much torque at low rpm. And, BTW, the DC motors in the PWM gimbal servos that I worked with "changed direction" (when commanded) at 300Hz...not exactly tweeters, but certainly comparable to woofers.
Apart from broadband sensitivity, frequency response might vary with SPL. I can do that test too! |
Rives...I don't understand your comment about "inertia". The effect of cone mass does not have any threshold to be "broken". If the parameter were "friction" what you say would make sense, but a cone driver has no friction, just mass.
If sensitivity diminishes at low power level something other than inertia is at work. I, like every other audiophile, have the impression that this occurs, but I am of the opinion that it is the human ear sensitivity that is diminishing rather than the speaker output.
Because it is servo controlled, I would expect a Velodyne to perform well at low volume. If there were any nonlinearity of the driver at low SPL the servo would counteract it. As you suggest, Velodyne should have something to say about this. |
Sean...Other than compression at some high level of SPL, I can't think of a reason for a low SPL "knee" in driver sensitivity. Can you provide any further info/references?
Equalization that is SPL-dependent is another one of the tricks that the ubiquitous Behringer DEQ2496 can do. I haven't tried it yet, but I have hopes that it might provide the "Loudness" function which my preamp lacks. |
Sean...Mechanical losses are actually thermal in nature, like the electrical loss in the voice coil. In neither case is there any mechanism for a "threshold" to be overcome. But rather than debate, I will make some measurements. |
SOME DATA
Signal......SPL
..-67.......50
..-52.......55
..-45.......60
..-40.......65
..-35.......70
..-30.......75
..-25.......80
..-20.......85
..-15.......90
Both the electrical signal and the SPL were measured, in units of dB, using my two Behringer DEQ2496. The SPL at 60 and 80 were crosschecked with my Radio Shack meter, and agreed within 1dB. The signal was white noise from a DVD player (used for channel balance). The measurements were RMS. Because this bounces around a bit, some estimation was necessary to get an average value. I set the SPL using the preamp volume control, and then read the associated electrical signal. The speaker measured was one MG1.6 backed up with a subwoofer, and the mic was about 4 feet from the MG1.6. I could not go higher than 90 dB because my preamp volume control maxed out using the DVD player signal. 90 dB RMS is pretty loud, and there is no indication of compression.
A plot of the data shows that at low volume below 60dB the SPL does not increase as steeply as it does over the rest of the range. However, I do not think that this reflects the kind of low SPL inefficiency which Sean suggests, because the background noise of the room ranges between 45 and 55 dB, as a function of traffic on the road outside. This background noise is pulling up the SPL data for the lowest two points. Late tonight, when background noise is low, I will make some more measurements. |
Sean....I did not get to rerun the test last night when background noise was low. Maybe tonight. I am quite sure that the background noise is the reason that the plot becomes nonlinear at low level. By the way, the data looks too perfect over much of the range, and you might be suspicious, but that is exactly what I read from the instruments. It surprised me.
Per you suggestion I will run a test using only the dynamic cone subwoofer with a warble tone as a signal. This will cover the range 22.5 to 250. The warble will come from an Audio Control Richter Scale equalizer, and it is only as good as it is. One problem I anticipate here is that if I go to any loud SPL all the windows and doors are going to rattle, and screw up the measurements. |
MORE DATA
Signal..SPL
..NONE...40 to 55 background noise
..-80....42
..-75....46
..-70....50
..-65....53
..-60....58
..-55....63
..-50....68
..-45....73
..-40....78
..-35....83
..-30....88
..-25....93
..-20....98
..-15....103
..-10....108
The signal was white noise, as before, but limited to frequencies below 400 Hz, and reproduced by my subwoofer system. (Easy to do. I just muted the HF and adjusted the X/O frequency up to 400 Hz). From prior experience I know that the warble tone would have rattled things. The mic was positioned about one foot from the 15 inch driver.
This time I set the signal level using the preamp volume control, and read the resulting SPL. When reading the SPL for the range below about 60 dB I took the lowest of rms indications over about 30 seconds, which corresponded to a lull in the traffic (background noise). For the higher SPL readings I took the average, as before.
Plot the data and you will see that there is almost no suggestion of decreased sensitivity at low SPL, which I attribute to the greater care that I took to minimize error due to background noise. And anyway, the SPL range where the data is not perfectly linear (for whatever reason) is so low as to be almost inaudible, so it wouldn't matter anyway. I am particularly happy to see no compression for high SPL, which would be a worse problem.
All of this is for my speakers :-). Maybe yours are different :-( |
Although I like the idea of large LF drivers, I do not think that the 12 inch Velodyne can be described as a "toy". If space or budget constraints must the recognized the 12 inch units are a good choice.
About the 3 step equalization process...The second SW may null out a room resonance that the first SW saw, and corrected. So in the end that correction by the first SW is not needed, and the second-time-around equalization gets it right. |
Kenl....I suppose you could go back and forth forever adjusting the two Subwoofers. The suggestion from the Velodyne guy seems reasonable.
I am glad that I don't have to decide between two 12 inch and one 18 inch. I guess I would go with the two 12 inch.
FWIW, my subwoofer system consists of three 15 inch and three 12 inch drivers, in huge enclosures embedded in the wall. |
Sean...A loudspeaker driver is EXACTLY a linear version of a rotary PM DC motor, where force (torque) is proportional to applied voltage. The applied voltage is variable DC. What do you think would happen if you had a DC-coupled power amp (like my Kenwood LO7M) and applied a staircase voltage, 0.2, 0.4, 0.6, 0.8...1.0....etc? (The cone would move out in steps determined by the suspension elasticity).
The AC motor, induction or synchronous, develops low torque at low RPM basically because of inefficient
commutation. (A synchronous motor has zero torque when stopped, and requires either a mechanical spin, or auxillary windings to get it going. Sometimes, as with the motors used to spin gyro wheels of inertial instruments, a periodic overvoltage is applied to the motor so as to get it to "lock in" to synchronous opearation).
Your suggestion that my test hid the effect that you theorize is a cop out. Truth is my test would be INCAPABLE of hiding it. Do you suggest that I reported false data? Why don't you try it for yourself? |
Sean..."the manner in which the tests were conducted may not be revealing enough". The test SPL values went down into the inaudible range. What would you like?
Again I say, go make some measurements. I can see that, like St Thomas, you won't believe until you see it for yourself. |
Sean...What are we looking for? I thought it was a variation of sensitivity (dB/volt or dB/watt) as a function of SPL, particularly evidence of a reduction at low SPL. That's what I looked for. |
Gregm..."VERY difficult", perhaps, but to simulate a point source to the left or the right with a center speaker is downright impossible. I agree that the Velodyne 18 incher is a good product and will no doubt work well. |
Gregm...I'll see your two SW, and raise you one center SW. I also have SW drivers for the rears, but haven't got around to putting them in enclosures.
I think that every speaker system should be flat to near 20 Hz, and that usually means you need to add another driver, called a SW. |
