Woofer Speed
Hi 7V,
The test results seemed pretty clear, and Adire isn't the only source for this.
If mass is increased, as in the test Adire performed, without increasing delay, then there is no significant measureable difference in a heavier cone vs a lighter cone in determining "fast bass". There are obviously other mechanical considerations relating to mass, but not bass speed.
If inductance is increased, a delay occurs that is measureable. That's fairly conclusive as well.
Tom Danley has also stated a similar position.
Am I missing something here?
Tim
Hi 7V,
The test results seemed pretty clear, and Adire isn't the only source for this.
If mass is increased, as in the test Adire performed, without increasing delay, then there is no significant measureable difference in a heavier cone vs a lighter cone in determining "fast bass". There are obviously other mechanical considerations relating to mass, but not bass speed.
If inductance is increased, a delay occurs that is measureable. That's fairly conclusive as well.
Tom Danley has also stated a similar position.
Am I missing something here?
Tim
Re: Woofer Speed
Do you think that the added-mass bass would sound as fast? I believe that jan tilburg was on the money when he talked about 'more bass' sounding subjectively slower. He was talking about room resonances. How about driver resonances?
The graphs do indicate that the initial impulse is relatively unaltered by adding this mass. The attack seems ok, but not the decay.
"The high frequency - the transient part of transient response - is reduced! The inductance cut the high frequency extension - mass did not!"
Now, I would argue from the graphs that the added-mass curve does, in fact, show the high frequency diminishing. However, it's no surprise that you get this with an inductor.
Of course we're all missing something here. We're not discussing these issues because we know all the answers. We're discussing this so we can figure things out and learn.
Let's assume that the measurements are correct and repeatable. Comparing the red and blue graphs, it seems to me that the added-mass graph shows a driver that is less controlled and has more resonances. Look at the troughs in the frequency response curve at 50Hz and 400Hz. Do these correspond with the overshoots in the impulse test at >1.9ms?Tim Moorman said:
Do you think that the added-mass bass would sound as fast? I believe that jan tilburg was on the money when he talked about 'more bass' sounding subjectively slower. He was talking about room resonances. How about driver resonances?
The graphs do indicate that the initial impulse is relatively unaltered by adding this mass. The attack seems ok, but not the decay.
I believe that speaker designers have a name for when you add inductance to a bass coil - a low pass filter.
"The high frequency - the transient part of transient response - is reduced! The inductance cut the high frequency extension - mass did not!"
Now, I would argue from the graphs that the added-mass curve does, in fact, show the high frequency diminishing. However, it's no surprise that you get this with an inductor.
Of course we're all missing something here. We're not discussing these issues because we know all the answers. We're discussing this so we can figure things out and learn.
Woofer Speed
7V
Yes, added inductance does form the equivalent of a low pass filter, as Dan Wiggins (or his partner) stated.
Extend these results to their logical conclusion on the other end of the spectrum. High frequency extension is then limited, to some extent, by the inductance of the VC, not the mass of the tweeter.
W. Marshall Leach found that you could reliably predict the result of the "lossy inductance" of the VC on frequency response.
Obviously, the inclination is to make a voice coil as robust as possible, increase the "L" in BL, for pro use, particularly. But, from a subjective point of view, most people seem to agree that those drivers with some attention given to lowering inductance, through multiple copper shorting coils and the like, sound better.
The Lambda woofers, when still being made, were good examples. It was not uncommon for a 15" Lambda driver with the full treatment to exhibit very low VC inductance and get rave reviews.
Agree, too, that a full compliment of unmolested high freq components will make a woofer sound more articulate, but it must carry through the crossover in phase, on a multiway or sub/mains system.
As I see it, the box dominates response in the last octave, and slightly overdamped response will yield better sounding bass.
Low bass (below 30 Hz) with very little distortion or box resonance, just sounds very low (test signal) and not very "fast" without the high frequency components present.
Above the box, low inductance will allow a more extended response with probably lower distortion. Look at the Klippel results - aside from BL curves and suspension numbers ( big aside), inductance is the another big source of problems.
Tim
7V
Yes, added inductance does form the equivalent of a low pass filter, as Dan Wiggins (or his partner) stated.
Extend these results to their logical conclusion on the other end of the spectrum. High frequency extension is then limited, to some extent, by the inductance of the VC, not the mass of the tweeter.
W. Marshall Leach found that you could reliably predict the result of the "lossy inductance" of the VC on frequency response.
Obviously, the inclination is to make a voice coil as robust as possible, increase the "L" in BL, for pro use, particularly. But, from a subjective point of view, most people seem to agree that those drivers with some attention given to lowering inductance, through multiple copper shorting coils and the like, sound better.
The Lambda woofers, when still being made, were good examples. It was not uncommon for a 15" Lambda driver with the full treatment to exhibit very low VC inductance and get rave reviews.
Agree, too, that a full compliment of unmolested high freq components will make a woofer sound more articulate, but it must carry through the crossover in phase, on a multiway or sub/mains system.
As I see it, the box dominates response in the last octave, and slightly overdamped response will yield better sounding bass.
Low bass (below 30 Hz) with very little distortion or box resonance, just sounds very low (test signal) and not very "fast" without the high frequency components present.
Above the box, low inductance will allow a more extended response with probably lower distortion. Look at the Klippel results - aside from BL curves and suspension numbers ( big aside), inductance is the another big source of problems.
Tim
Re: Re: Inductance
I dunno, how many drivers change mass as they move/operate? Remember, a constant is just a scalar. It does NOT affect how the current moves through the driver, just how much acceleration you get out for a given current in.
And, of course, above Fs a driver's acceleration is proportional to SPL, not to bandwidth (drivers are constant acceleration devices in their passband - between Fs and the inductive rolloff, or breakup modes, on the high end).
I think you're confusing acceleration with transient response for a driver; it's not the same thing. Rate of change of acceleration is what you want, not the raw acceleration itself.
I've been questioned about this dozens of times over the last year or two, even had people from around the world make measurements... What's claimed (and also stated by the likes of Danley, Toole, Geddes) is correct. You can do the measurements yourself (as we did): measure a driver's bandwidth (which, from the Fourier Transform IS the transient response), then add a bunch of mass to it, and some additional inductance. Which one loses high frequency output?
Mass affects SPL/effeciency only. It does not affect transient response.
Dan Wiggins
Adire Audio
7V said:
I dunno, how many drivers change mass as they move/operate? Remember, a constant is just a scalar. It does NOT affect how the current moves through the driver, just how much acceleration you get out for a given current in.
And, of course, above Fs a driver's acceleration is proportional to SPL, not to bandwidth (drivers are constant acceleration devices in their passband - between Fs and the inductive rolloff, or breakup modes, on the high end).
I think you're confusing acceleration with transient response for a driver; it's not the same thing. Rate of change of acceleration is what you want, not the raw acceleration itself.
I've been questioned about this dozens of times over the last year or two, even had people from around the world make measurements... What's claimed (and also stated by the likes of Danley, Toole, Geddes) is correct. You can do the measurements yourself (as we did): measure a driver's bandwidth (which, from the Fourier Transform IS the transient response), then add a bunch of mass to it, and some additional inductance. Which one loses high frequency output?
Mass affects SPL/effeciency only. It does not affect transient response.
Dan Wiggins
Adire Audio
Just ran across an interesting article on the subject of "Fast" bass. http://www.soundstage.com/maxdb/maxdb061999.htm
Forgetting any discussion of bandwidth, which driver/alignment most closely tracks a wave of fixed frequency but varying amplitude?
Bass is nothing more than a low frequency signal whose amplitude is modulated by a beat frequency (what musicians might know as rhythmn).😀
In fact. that's exactly how the earliest synth-basses worked. Admitedly they sound nothing like drums, but drums have very complex harmonic patterns compared to string or helmholtz instruments.
If the amplitude of that signal cannot be modulated in time with the beat, it will smear out the rhythmn, which I (as a musician/physicist) would call 'slow' bass.
Bass is nothing more than a low frequency signal whose amplitude is modulated by a beat frequency (what musicians might know as rhythmn).😀
In fact. that's exactly how the earliest synth-basses worked. Admitedly they sound nothing like drums, but drums have very complex harmonic patterns compared to string or helmholtz instruments.
If the amplitude of that signal cannot be modulated in time with the beat, it will smear out the rhythmn, which I (as a musician/physicist) would call 'slow' bass.
Thanks, nice article. Perhaps a little oversimplified to be 100% accurate/true in parts, but nice reading anyway!Wright said:
Re: Re: Re: Inductance
In your test you added 28.5 grams to the dust cap. What you are saying is that you could add more mass and the SPL would reduce until finally, with enough mass, the output would be zero. Just before this point the shape of the impulse response (and hence frequency response) would be identical to the zero added mass measurement. Is that right?
Interestingly, the 'SoundStage' article listed by Wright says "In fact, bass speed is virtually 100% a function of how ideally the midrange and woofer are integrated." Another view, another simplification.
Dan, you aim to keep the inductance of the bass driver as low as possible. What's your philosophy when it comes to the crossover with the mid driver (or tweeter)? Do you run the bass driver as high as possible, without a crossover and put the high pass point of the mid (or tweeter) to match the natural roll-off of the bass driver?
Interesting game this speaker design. 😉
No driver changes its mass as it operates but aren't we talking about comparing two drivers with different masses?DanWiggins said:
In your test you added 28.5 grams to the dust cap. What you are saying is that you could add more mass and the SPL would reduce until finally, with enough mass, the output would be zero. Just before this point the shape of the impulse response (and hence frequency response) would be identical to the zero added mass measurement. Is that right?
Interestingly, the 'SoundStage' article listed by Wright says "In fact, bass speed is virtually 100% a function of how ideally the midrange and woofer are integrated." Another view, another simplification.
Dan, you aim to keep the inductance of the bass driver as low as possible. What's your philosophy when it comes to the crossover with the mid driver (or tweeter)? Do you run the bass driver as high as possible, without a crossover and put the high pass point of the mid (or tweeter) to match the natural roll-off of the bass driver?
Interesting game this speaker design. 😉
Inductance is reflected in the impedance. At frequencies where inductance is high you need less amps to keep the cone moving, but you will therefore have also less control.
Inductance depends on the speed at wich the coil moves through the magnetic field. So there is one more advantage of bigger woofers. At a certain spl the bigger woofer will have a lower speed and so a lower inductance, which results in more control.
But it is also a matter of balance since the inductance voltage is highest at that point of the sinus where the voltage provided by the amplifier is zero. Where the cone changes direction the amplifiervoltage has a maximum and induction is zero. That is just what you need.
So that also means its influence is less than you might expect from Re and Res values.
difficult!
Inductance depends on the speed at wich the coil moves through the magnetic field. So there is one more advantage of bigger woofers. At a certain spl the bigger woofer will have a lower speed and so a lower inductance, which results in more control.
But it is also a matter of balance since the inductance voltage is highest at that point of the sinus where the voltage provided by the amplifier is zero. Where the cone changes direction the amplifiervoltage has a maximum and induction is zero. That is just what you need.
So that also means its influence is less than you might expect from Re and Res values.
difficult!
Steve,
The Soundstage article is more about perceived speed, not actual measured transient response. The two really aren't related at all... Perceived speed of a system is much more about frequency balance (cf. Gabrielsson, Tolve) than just objectively measured transient response.
Then there's the whole "accuracy/speed/quality" thing. Objectively, accuracy can be measured via FR, THD, IMD, etc. Subjectively, throw those "standards" out the window (cf. Geddes, Lee).
I try to keep the two definitely seperate. Transient response does not imply "speed' of bass, and vice versa. In fact, because drivers are constant acceleration devices, the "fastest" woofer is the one that plays the lowest the loudest. For instance, since acceleration is constant with frequency, that means velocity - technically the "speed" of the woofer - will double for every halving of frequency (assuming the frequency response is flat). Thus if you want to get technically correct, woofer "speed" is more about which driver plays low AND loud, not the higher output.
Subjective "speed" is much more about things not transient response related.
Dan Wiggins
Adire Audio
The Soundstage article is more about perceived speed, not actual measured transient response. The two really aren't related at all... Perceived speed of a system is much more about frequency balance (cf. Gabrielsson, Tolve) than just objectively measured transient response.
Then there's the whole "accuracy/speed/quality" thing. Objectively, accuracy can be measured via FR, THD, IMD, etc. Subjectively, throw those "standards" out the window (cf. Geddes, Lee).
I try to keep the two definitely seperate. Transient response does not imply "speed' of bass, and vice versa. In fact, because drivers are constant acceleration devices, the "fastest" woofer is the one that plays the lowest the loudest. For instance, since acceleration is constant with frequency, that means velocity - technically the "speed" of the woofer - will double for every halving of frequency (assuming the frequency response is flat). Thus if you want to get technically correct, woofer "speed" is more about which driver plays low AND loud, not the higher output.
Subjective "speed" is much more about things not transient response related.
Dan Wiggins
Adire Audio
Just a question (and possibly one answered before, but haven't seen it).
Given that the heavier the cone, the greater the force needed to accelerate it, might not that greater force be exciting the enclosure more? It seems to me that enclosures often play a surprisingly large part in the overall sound of a speaker, and exciting the enclosure more must have more effect than exciting it less. And all the mass of the enclosure would take a long time to settle down, perhaps giving a slow impression.
Given that the heavier the cone, the greater the force needed to accelerate it, might not that greater force be exciting the enclosure more? It seems to me that enclosures often play a surprisingly large part in the overall sound of a speaker, and exciting the enclosure more must have more effect than exciting it less. And all the mass of the enclosure would take a long time to settle down, perhaps giving a slow impression.
No, it's the movement of the cone that excites the enclosure, irrespective of the force driving it.Nat Eddy said:
Your other point is correct, I think. A resonant enclosure will contribute to a bass that sounds slow. All resonances will, whether they're in the enclosure, room or driver.
Dan, would this be 'perceived speed'? These resonances will also show up in the impulse testing. Transient response?
The 'fastest' bass I know of comes from motional-feedback closed loop subwoofers.
Next closest is a sealed alignment close to 0.6 Qtc.
Anything that uses a Helmholtz resonator just doesn't cut it.
IMHO it has nothing to do with the bandwidth of the driver, and more to do with the ability to match the amplitude decay curve of a bass note, so that the notes are as clearly differentiated by the system as is possible, given the original recording.
The decay rate is affected by the filter properites, according to the equivalent circuit theory. An equivalent filter which rings obviously cannot match the correct decay rate. Adding a resonator to the circuit again introduces a different decay rate to the original signal, around that resonance.
These principles have supposedly been known in active sonar systems for years, where the ability to turn the sonar on and off quickly can be rather important to the crew of the boat using it😀
Next closest is a sealed alignment close to 0.6 Qtc.
Anything that uses a Helmholtz resonator just doesn't cut it.
IMHO it has nothing to do with the bandwidth of the driver, and more to do with the ability to match the amplitude decay curve of a bass note, so that the notes are as clearly differentiated by the system as is possible, given the original recording.
The decay rate is affected by the filter properites, according to the equivalent circuit theory. An equivalent filter which rings obviously cannot match the correct decay rate. Adding a resonator to the circuit again introduces a different decay rate to the original signal, around that resonance.
These principles have supposedly been known in active sonar systems for years, where the ability to turn the sonar on and off quickly can be rather important to the crew of the boat using it😀
My bass is faster than your bass, and don't you forget it !
With my EPOS ES11 speakers, the bass under 50Hz. is so incredibly fast, I can't even hear it.
Now that's fast.
So I added a DIY sealed low Qtc Adire Audio Tempest subwoofer.
But then the bass under 50Hz. wasn't so fast any more,
but at least I could hear it: I had room resonances that took a long time to attenuate to inaudibility. Maybe two full minutes or more ! Okay, maybe not two minutes, but a long time.
So I added a Behringer Feedback Destroyer to cut down those pesky bass peaks so they faded to inaudibility about as fast as
the other bass frequencies.
But then I noticed that certain frequencies caused dishes in the kitchen to rattle, and walls resonated at other frequencies,
and windows at other frequencies.
So I tried to quiet all the room noises ... but then when I played the bass loud, the wife kept bellowing: "turn it down" ... and the dog howled ... so I finally disconnected my subwoofer and now my bass is once again so fast ... I can't even hear it.
Richard BassNut Greene
My motto: The only fast bass ... is no bass.
PS: A professional bass player told me he prefers slow resonant bass sound quality. He said fast bass is for synthesizer players and other losers - not real musicians like him. So if a real professional musician doesn't like fast bass, why should we audiophiles want fast bass?😀
With my EPOS ES11 speakers, the bass under 50Hz. is so incredibly fast, I can't even hear it.
Now that's fast.
So I added a DIY sealed low Qtc Adire Audio Tempest subwoofer.
But then the bass under 50Hz. wasn't so fast any more,
but at least I could hear it: I had room resonances that took a long time to attenuate to inaudibility. Maybe two full minutes or more ! Okay, maybe not two minutes, but a long time.
So I added a Behringer Feedback Destroyer to cut down those pesky bass peaks so they faded to inaudibility about as fast as
the other bass frequencies.
But then I noticed that certain frequencies caused dishes in the kitchen to rattle, and walls resonated at other frequencies,
and windows at other frequencies.
So I tried to quiet all the room noises ... but then when I played the bass loud, the wife kept bellowing: "turn it down" ... and the dog howled ... so I finally disconnected my subwoofer and now my bass is once again so fast ... I can't even hear it.
Richard BassNut Greene
My motto: The only fast bass ... is no bass.
PS: A professional bass player told me he prefers slow resonant bass sound quality. He said fast bass is for synthesizer players and other losers - not real musicians like him. So if a real professional musician doesn't like fast bass, why should we audiophiles want fast bass?😀
Good One!
Dear No Bass,
Gotta kick out of that one.
My dog howls at sirens on TV/films on the HT rig. Can't figure out which to get rid of ... the tweeter or the dog.
Tim-The magic is in the midrange- Moorman
Dear No Bass,
Gotta kick out of that one.
My dog howls at sirens on TV/films on the HT rig. Can't figure out which to get rid of ... the tweeter or the dog.
Tim-The magic is in the midrange- Moorman
Re: My bass is faster than your bass, and don't you forget it !
Really, the only way that we can judge is to compare with the sound of an acoustic double bass, bass drum or earthquake. 🙂
We're getting into areas of terminology here. My understanding is that, however 'fast' the bass is, it won't be faster than the instrument itself. A good bass should reproduce the sound of the bass instrument without delay, overhang or blurring.Richard Greene said:
Really, the only way that we can judge is to compare with the sound of an acoustic double bass, bass drum or earthquake. 🙂
Thanks for the response, Steve. If I understand you, you are suggesting that the motor force driving the cone isn't really an issue with enclosure excitation -- its the actual sound emitted by the driver, and then what ever helmholz resonances are excited by that sound. But what I was inquireing about was a slightly different issue. Driver components and the interface bewteen the driver and the enclosure are not perfectly rigid or perfectly damped, and it seems to me that the more mass being accelerated, the greater the forces operating on all the parts, hence more resonances. I am making two asumptions here -- that the mass of air moved by the cone is significantly less than the mass of any cone (otherwise there might not be too much relative difference in masses being accelerated), and that the heavier the cone, the greater the mass of the coils, magnet and etc also. Now these parts being heavier are going to resonate at lower frequencies than would lighter parts, and the heavier all the parts of the driver, the better the impedence match to the (much) heavier enclosure, so the easier it is for parasitic resonances to excite the enclosure.
Does this make sense, and even if it does, does it actually have an effect?
Does this make sense, and even if it does, does it actually have an effect?
Wright said:
Or the same article here
http://yu-ra.tripod.com/fbvssb.htm
and some more
http://yu-ra.tripod.com/index_topics.htm
Yes, Nat. I think that you make a valid point here and one that I had not really considered before.Nat Eddy said:
The greater the moving mass of the driver, the greater the normal reaction to that mass and, hence, cabinet vibration. Whether we should lose sleep over this, I don't know. I doubt it but I haven't tested cabinet vibrations using cones with different masses - I'll add it to my "To Do" list and report back around 2010. 🙂
I do consider that the speaker designer's remit includes drive units, cabinets and room and this includes both airborne resonances and mechanical vibration. So the way we mount the drivers in the cabinets, the vibrational behaviour of the cabinet itself and the mechanical interface between speaker, stand and floor are all part of the process.
Incidentally, the magnets don't move so, in theory, the heavier the mass of the magnet, the more the vibrations could be damped - but let's not go there as I have no idea whether it would be significant or not.
Also, we can't say that heavier parts will necessarily vibrate at lower frequencies. Look at a pendulum. It's the length that determines its frequency, not the mass.
- Status
- Not open for further replies.