If a single sub is going to go from Point A to Point B in x-amount of time to produce y-sound, that takes a certain amount of energy. If you have eight subs moving same air, they do not require as much linear excursion. Which means that they can move faster. Play off your xmax vs your fs...
Hi Pogie,
I may be misunderstanding the point you are trying to make but this
As you said earlier move the same distance in the same time = same velocity.
Now reduce the cone movement below Xmax and you will have a reduced velocity. Less distance in the same time.
I may be misunderstanding the point you are trying to make but this
does not make sense.
As you said earlier move the same distance in the same time = same velocity.
Now reduce the cone movement below Xmax and you will have a reduced velocity. Less distance in the same time.
Okay - they can change direction faster.
Issac Newton - If you're shoving something big and heavy in a direction, it's going to take more time/energy to get it going, and also to stop it than for something smaller and lighter.
If you can get the same displacement (for the same effective volume) from multiple subs, they don't have to work as much. It'll sound sharper, more controlled. Because it is.
Issac Newton - If you're shoving something big and heavy in a direction, it's going to take more time/energy to get it going, and also to stop it than for something smaller and lighter.
If you can get the same displacement (for the same effective volume) from multiple subs, they don't have to work as much. It'll sound sharper, more controlled. Because it is.
...another thought...
a driver in a nonventilated enclosure and relatively low mass accelerates faster and changes direction faster...Two drivers in the same cabinet shall be identical (hard to find even if they are same production batch...), if put in individual nonvented enclosures, variations will be of minor effect compared to nonlinearities when put in a common enclosure.
A Eton 12" driver with Fs around 25Hz and only 90grams of mass (with 520sqcm conesurface) already has a good basis for a start...
Now give driver a spike and look how it reacts when coming to rest...
Get the point...?
a driver in a nonventilated enclosure and relatively low mass accelerates faster and changes direction faster...Two drivers in the same cabinet shall be identical (hard to find even if they are same production batch...), if put in individual nonvented enclosures, variations will be of minor effect compared to nonlinearities when put in a common enclosure.
A Eton 12" driver with Fs around 25Hz and only 90grams of mass (with 520sqcm conesurface) already has a good basis for a start...
Now give driver a spike and look how it reacts when coming to rest...
Get the point...?
“If a single sub is going to go from Point A to Point B in x-amount of time to produce y-sound, that takes a certain amount of energy. If you have eight subs moving same air, they do not require as much linear excursion. Which means that they can move faster. Play off your xmax vs. your fs...”
UUUHHHHMMMM – They require the same volume displacement to achieve a particular output level. Yes, it takes power to move a speaker, that’s why we amplifiers, class D or similar preferred for subs. And even for full range drivers with present day amplifier designs. If your speakers are moving faster, they must be moving further or at a higher frequency. Negative on both.
“If you can get the same displacement (for the same effective volume) from multiple subs, they don't have to work as much. It'll sound sharper, more controlled. Because it is.”
What’s the summed mass of your eight subs? Please write up the formula for work performed by a speaker to realize why you are silly.
“Now give driver a spike and look how it reacts when coming to rest...”
Low Le, low Qts, problem solved. Combine with a big amp. . .
“More cone surface (my room is about 450square feet), less excursion ergo less distortion at reasonable levels, depending on positioning better damping of radiation ergo less problems with room modes... and the list could go on and on.”
Remain in the linear operating range of the speaker. I have over 5 liters of linear volume displacement. Multiple drivers add up to more probability of response errors too. Can it be done effectively – yes. Will it take a lot longer and a bit of luck to set up – definitely.
I think this is what you are attempting to accomplish –
http://www.harman.com/wp/pdf/multsubs.pdf
The point being that tons of accurate, tight, clean - whatever adjective you like - sub bass is easily produced by a single driver. You’ve gone off the subject of the thread. The basics of physics are fundamental and can’t be overlooked. Speaker SPL is relative to volume displacement. There is no replacement for displacement. It’s simple and true.
- James
UUUHHHHMMMM – They require the same volume displacement to achieve a particular output level. Yes, it takes power to move a speaker, that’s why we amplifiers, class D or similar preferred for subs. And even for full range drivers with present day amplifier designs. If your speakers are moving faster, they must be moving further or at a higher frequency. Negative on both.
“If you can get the same displacement (for the same effective volume) from multiple subs, they don't have to work as much. It'll sound sharper, more controlled. Because it is.”
What’s the summed mass of your eight subs? Please write up the formula for work performed by a speaker to realize why you are silly.
“Now give driver a spike and look how it reacts when coming to rest...”
Low Le, low Qts, problem solved. Combine with a big amp. . .
“More cone surface (my room is about 450square feet), less excursion ergo less distortion at reasonable levels, depending on positioning better damping of radiation ergo less problems with room modes... and the list could go on and on.”
Remain in the linear operating range of the speaker. I have over 5 liters of linear volume displacement. Multiple drivers add up to more probability of response errors too. Can it be done effectively – yes. Will it take a lot longer and a bit of luck to set up – definitely.
I think this is what you are attempting to accomplish –
http://www.harman.com/wp/pdf/multsubs.pdf
The point being that tons of accurate, tight, clean - whatever adjective you like - sub bass is easily produced by a single driver. You’ve gone off the subject of the thread. The basics of physics are fundamental and can’t be overlooked. Speaker SPL is relative to volume displacement. There is no replacement for displacement. It’s simple and true.
- James
AudialDelta said:
I was going to reply to this, but it just takes to long..
I will say this though - when someone says "faster" they are NOT tallking about how quickly the driver pumps back and forth, otherwise they would simply say the driver had a more extended high freq. response. What they mean is that the driver has a faster rise-time and shorter decay (..or achieves this with lower excursion potentials).
ScottG said:
Same thing surely? The driver only has to move to reproduce a sine wave at the crossover frquency. If it's anything other than a sine wave, Fourier tells us that the waveform is made up of higher frequency components, which providing your crossover works, should be passed to the next driver/s. 🙂
http://www.mtxaudio.com/caraudio/products/subwoofers/jackHammer.cfm
Just thought I would add this little number to the list of sick and twisted.🙂
Just thought I would add this little number to the list of sick and twisted.🙂
Wouldn't it just be easier and lead to a better debate if you just told me yours? You know as well as I do, you can find differing definitions of just about anything on the web. 😉
He obviously didn't read what I wrote, or didn't understand it.
I mentioned low Le and low Qts for this specific reason.
Sometimes cones are made heavier for good reasons, such as extended low frequency response.
I don't know what he is using to calculate his woofers speed. But there is absolutely nothing slow about a Tumult.
I mentioned low Le and low Qts for this specific reason.
Sometimes cones are made heavier for good reasons, such as extended low frequency response.
I don't know what he is using to calculate his woofers speed. But there is absolutely nothing slow about a Tumult.
An interesting discussion...
...in particular about "bass speeds". So there are people claiming that there would be need for "faster" bass-drivers. I've since ever been stunned on how and why this "quality term" is demanded. For myself, I hereby officially confess that within decades of speaker design I have never found any single reason to care or even think about something like "bass speed", neither with my calculus nor actual designs. For an explanation on how my obviously erratic views on this topic are, I will want to explain how narrowminded my own considerations have been (and still are).
Looking at a given driver in combination with the according x-over as one bandwith-limited transducer-unit, I see an *overall* transfer function. Depending on the design principle (closed, vented, horn, etc. ) there usually are quite a couple of complex poles at both lower and upper end of this "unit's" transfer characteristic. In my neverending blindness with looking at frequency and time behaviour in particular, I have have not yet found a reason for caring whether a certain pole of said transfer function originates in the driver itself, enclosure or the x-over network. A simple example: In case I would want a 24dB low-frequency cutoff, I could in theory use a vented design or at the same time create a closed box with according additional 12dB/Oct electrical filtering. Of course power handling, excursion, distorsion, etc. will be very different, but I'm talking about the transfer function providing just that "speed" aka "impulse respose characteristiscs" here, as far as I (don't) understand?
Now for the low-pass transfer behaviour at upper cut off. Again, the "unit" will show the combination (in fact: multiplication) of the driver's behaviour and the crossover network, probably baffle step too. In case that bass-driver is unequalized, the overall response behaviour will result from the (then dominant) electrical filtering. With this filtering usually already had including frequency and phase characteristics of the driver & baffle, *any* change in woofer rise time would just mistune the crossover frequency & slope. Bummer.
A more improved design could for instance use the intrinsic HF-cutoff of any driver by shifting that already existing cutoff to lower frequencies. (Think of a kind of Linkwitz-Riley transform for *upper* cutoff, and don't forget eventual baffle step with that) This method usually providing a neat & clean acoustic slope of 12dB/Oct at the new (shifted) cutoff, for steeper slopes additional 6 or 12dB/Oct may be added up.
Nevertheless, in whatever way you are combining a driver with an electrical network, the *overall transfer function* will define impulse response and thus rise time. Rise time of acoustical output is strictly correlated to filter transfer functions & cutoff frequencies. Change that woofer "speed" and your targeted filter-type will change, or your cutoff frequency, or both. X-over "mistuned" again. Bummer Nr. 2.
So one of my conclusions was: since I will have to limit upper bandwith of my woofers anyway, poor me has to find a way of significantly slowing down my woofers. And with any woofers I have ever worked on in my life
I have found that all of these were way too "fast" without appropriate filtering.
"Driver speed?" I still don't get it...
Thanx for reading & kind regards
redunzelizer
...in particular about "bass speeds". So there are people claiming that there would be need for "faster" bass-drivers. I've since ever been stunned on how and why this "quality term" is demanded. For myself, I hereby officially confess that within decades of speaker design I have never found any single reason to care or even think about something like "bass speed", neither with my calculus nor actual designs. For an explanation on how my obviously erratic views on this topic are, I will want to explain how narrowminded my own considerations have been (and still are).
Looking at a given driver in combination with the according x-over as one bandwith-limited transducer-unit, I see an *overall* transfer function. Depending on the design principle (closed, vented, horn, etc. ) there usually are quite a couple of complex poles at both lower and upper end of this "unit's" transfer characteristic. In my neverending blindness with looking at frequency and time behaviour in particular, I have have not yet found a reason for caring whether a certain pole of said transfer function originates in the driver itself, enclosure or the x-over network. A simple example: In case I would want a 24dB low-frequency cutoff, I could in theory use a vented design or at the same time create a closed box with according additional 12dB/Oct electrical filtering. Of course power handling, excursion, distorsion, etc. will be very different, but I'm talking about the transfer function providing just that "speed" aka "impulse respose characteristiscs" here, as far as I (don't) understand?
Now for the low-pass transfer behaviour at upper cut off. Again, the "unit" will show the combination (in fact: multiplication) of the driver's behaviour and the crossover network, probably baffle step too. In case that bass-driver is unequalized, the overall response behaviour will result from the (then dominant) electrical filtering. With this filtering usually already had including frequency and phase characteristics of the driver & baffle, *any* change in woofer rise time would just mistune the crossover frequency & slope. Bummer.
A more improved design could for instance use the intrinsic HF-cutoff of any driver by shifting that already existing cutoff to lower frequencies. (Think of a kind of Linkwitz-Riley transform for *upper* cutoff, and don't forget eventual baffle step with that) This method usually providing a neat & clean acoustic slope of 12dB/Oct at the new (shifted) cutoff, for steeper slopes additional 6 or 12dB/Oct may be added up.
Nevertheless, in whatever way you are combining a driver with an electrical network, the *overall transfer function* will define impulse response and thus rise time. Rise time of acoustical output is strictly correlated to filter transfer functions & cutoff frequencies. Change that woofer "speed" and your targeted filter-type will change, or your cutoff frequency, or both. X-over "mistuned" again. Bummer Nr. 2.
So one of my conclusions was: since I will have to limit upper bandwith of my woofers anyway, poor me has to find a way of significantly slowing down my woofers. And with any woofers I have ever worked on in my life
I have found that all of these were way too "fast" without appropriate filtering.
"Driver speed?" I still don't get it...
Thanx for reading & kind regards
redunzelizer
pinkmouse said:
The problem I'm having is with "time". Say 20 cycles per second. I'm not concerned with the ability to cycle 20 times in a second (..time as a period), rather I'm concerned with the ability to start that cycle and then stop it as quickly as possible after that second. There is a relationship there, but it isn't the same.
Now then consider a theoretical wideband driver with a legitimat flat freq. response from 20 Hz to 20 kHz. Furthermore lets say it has the ability to start and stop at 20 kHz within a tenth of a milisecond. Now - does this mean that the driver will start and stop this quickly at 20 Hz?
Does that help?
I've wondered about this "speed" debate for over a decade... just doesn't make any sense. Used to argue in rec.audio.highend for hours on just this subject 10 years ago...
sigh...
perhaps those attributing "tightness", "fast risetime", etc. are confusing harmonic distortion by products with the quality of the sound...
"flabby", "loose", "slow"...are all subjective terms no amount of debate can quantify them... whereas frequency response, distortion spectra, power handling, etc. are quantifiable and can be discussed.
Just my 2 cents
auplater
HomeBuilt BG's dipoles
http://www.diyaudio.com/forums/attachment.php?s=&postid=642400&stamp=1116116806
sigh...
perhaps those attributing "tightness", "fast risetime", etc. are confusing harmonic distortion by products with the quality of the sound...
"flabby", "loose", "slow"...are all subjective terms no amount of debate can quantify them... whereas frequency response, distortion spectra, power handling, etc. are quantifiable and can be discussed.
Just my 2 cents
auplater
HomeBuilt BG's dipoles
http://www.diyaudio.com/forums/attachment.php?s=&postid=642400&stamp=1116116806
lessee... 20 Khz = 20000 cps
1 sec / 20000 cycles = .00005 sec / cycle or .05 milliseconds... your hypothetical driver won't meet the spec you're proposing....
1 sec / 20 hz = .05 sec / cycle or 50 milliseconds....
starting and stopping don't mean anything wrt speed of the driver... by definition it has to respond as above without producing distortion artifacts...
auplater
1 sec / 20000 cycles = .00005 sec / cycle or .05 milliseconds... your hypothetical driver won't meet the spec you're proposing....
1 sec / 20 hz = .05 sec / cycle or 50 milliseconds....
starting and stopping don't mean anything wrt speed of the driver... by definition it has to respond as above without producing distortion artifacts...
auplater
I think what we may be thinking about, in somewhat violent agreement, is dwell time... The time when the sub essentially is not effective, because it is accelerating or braking, and how "far" it requires to accelerate or brake.
AudialDelta said:
I'm very familiar with these measurements as well as the Adire paper you're mentioning.
Quote from above paper: "So if you want faster transient response, ignore that moving mass parameter that some manufacturers push - look at the inductance! And if they don't list the inductance, ask yourself why - is there something they don't want to show? Inductance is the key to driver transient response - ask for it when transient response comes up!"
Really?
So, plain driver transient response *rise time without crossover* or baffle effects is considered a usefull criteria? I highly doubt that. From my point of view the voice coil inductances on larger woofers are in general way too low. But at the same time that vc-inductor is usually very-nonlinear too, so *adding* another "good" inductor right to the driver (even before thinking about any crossover) will both linearize the resulting overall inductance and change frequency/time behaviour. In detail: that inductance will provide one low-pass pole within the overall (driver + x-over) transfer function *anyway*, so why not let it impact at the desired frequencies firsthand, and not at some frequency in the nowherelands of a (then badly) controlled passband.
Low VC-inductance? Not that bad in the end, but for quite different reasons. You will decide yourself what for...
While Adire has well earned reputation for building great drivers, with suspecting some others of irritational marketing, they do just the same on that one: irritational marketing. They could have been a tad more honest...
just my two milliHenries (lacking)...
redunzelizer
That sounds similar to "don't lighten your car to go faster - just throw a bigger engine in it." This'll work to a point... the point where Newton takes the wheel, and you don't make the turn because the dang car weighs more than the control mechanism can steer (tires included).
- Status
- Not open for further replies.