Why, he's still wrong. Bigger brakes are not the only way to make a car stop faster. Better tires, lighter car, better suspension, etc. can all do the same thing. In this example the fact that both scenario's are multi-faceted is the similarity. The pro's and con's of the various methods to accomplish a task are carefully weighed to provide a solution. I don't know of any situation where an engineer would make a subwoofer cone lighter with the goal of having a "faster" woofer, or better impulse response.
simon5 said:
Isn't it the subwoofer's problem? Remember you are not dealing with a linear "load". (which is where your reply falls apart.) And as auplater suggested (and as I have tried to numerous times) - its about distortion. Saying something is fast is synomous with saying something "tracks" the signal more accuratly or has less distortion (..though it is both "more" and "less" descriptive in saying it has less distortion) .
But lets get back to the very gross simplification above, and high x-max drivers generally.
1. is the cut-off (low pass crossover) infinite? No. This means that bandwidth is a bit more than a theoretical passband. While it might not matter with a theoretical passband - it likely will matter with a real world passband.
2. does the lowpass have distortion of its own? Depends on the crossover. Most do - particularly those reaching higher in order (the LR 4th is a good exception/compromise).
3. referring back to #1 - will more excursion likely be detrimental to a real world passband? Yup.
4. considering that drivers are not linear - where does most of this non-linearity arise from? The mechanical resonance.
5. will the mechanical resonance be more or less effected by an increase in excrusion? More - sometimes a LOT more.
6. when harmonic distortion is generated - does it just move "up" in freq. from the fundamental? Nope - it goes up AND down. Consider what this means for theory reply above.
7. as an extension of #6 - is there just that distortion? No, there is also distortion on the distortion (etc.). Consider what this does to the passband and perhaps you will realize why we typically look to the average of harmonic distortion for a given freq. when measuring a driver. (..though note that a non-averaged test is also good for spoting the genesis for the averaged distortion - perhaps allowing a better driver design.)
I could go on - but its a waste of my time.
My real world experience is that with a 15 inch driver with 33-34 mm one way excursion distortion is below the audible limit with music. This is at very high in room SPL. More than can be measured with a Rat Shack meter in car or home settings.
Also consider the limits of your ears. At what point do your ears start to compress the sound? Likely well below the point where your high X-max sub will start having audible distortion that is solely a product of it's high excursion.
You keep writing about a subs distortion - but my CD, receiver, amp sub combo keep playing very clean with both music and tones. Both in car and in home. One way I judge a subs ability to play clean is by setting the crossover between 80 and 100 Hz. If the sub is able to blend seemlessly with the kick drums, snare drum, piano, vocals, bass, guitar, etc. then the distortion is low enough to not interfere with my listening experience. When you have higher levels of distortion, the subs location becomes audible. It has more difficulty blending with the rest of the system, and stands out on it's own. I don't want that with my sound system.
Also consider the limits of your ears. At what point do your ears start to compress the sound? Likely well below the point where your high X-max sub will start having audible distortion that is solely a product of it's high excursion.
You keep writing about a subs distortion - but my CD, receiver, amp sub combo keep playing very clean with both music and tones. Both in car and in home. One way I judge a subs ability to play clean is by setting the crossover between 80 and 100 Hz. If the sub is able to blend seemlessly with the kick drums, snare drum, piano, vocals, bass, guitar, etc. then the distortion is low enough to not interfere with my listening experience. When you have higher levels of distortion, the subs location becomes audible. It has more difficulty blending with the rest of the system, and stands out on it's own. I don't want that with my sound system.
AudialDelta said:
Is it below the audible limit - and in what passband? Others have concluded differently. (..and it OK to disagree on this - it is subjective.)
I personally find that it makes more (not less) difference to have lower levels of distortion at sub-bass freq.s (say less than 40 Hz). This is where most of the sound is itself essentially distortion - i.e. reflected hall sound. Sideband decay of a low fundamental can also factor in here - but it is something I have a much greater threashold for hearing distorted.
In my real world situation I consider it below the audible limit because I have not heard it. Personally I am much more attuned to hearing distortion in the midrange. And theoretically humans are most sensitive in the midrange.
I guess you need to "not hear distortion" yourself to believe it. After not having it for 2-3 years now, I don't want it reintroduced in my personal systems. When trains go across the screen, water rushes by, etc. it should sound just like that, not like a movie or CD. Trains and plains should be able to pass by without drawing attention to the sub or it's enclosure.
I guess you need to "not hear distortion" yourself to believe it. After not having it for 2-3 years now, I don't want it reintroduced in my personal systems. When trains go across the screen, water rushes by, etc. it should sound just like that, not like a movie or CD. Trains and plains should be able to pass by without drawing attention to the sub or it's enclosure.
Hi again ScottG,
You are right, there's alot of problems in the real world...
But returning to your original sentence, you were talking about fast risetime and fast decay.
A subwoofer, assuming a good x-over is used, doesn't need to have a fast risetime AFAIK.
There's also those linear BL subwoofers that improve the non-linearities a bit.
Fast decay, I think it could be a good thing. The problem is decay is like 10% driver based and 90% box based. Just like a review of a PB10-ISD floating around the net, someone cut the decay time to a third of original by adding only wood clamps and a pile of books on the subwoofer...
BTW fast risetime is also affected alot by the box alignment, only Q=1 in a sealed box can follow the imput amplitude correctly, but does some other things wrong.
pinkmouse help me!
You are right, there's alot of problems in the real world...
But returning to your original sentence, you were talking about fast risetime and fast decay.
A subwoofer, assuming a good x-over is used, doesn't need to have a fast risetime AFAIK.
There's also those linear BL subwoofers that improve the non-linearities a bit.
Fast decay, I think it could be a good thing. The problem is decay is like 10% driver based and 90% box based. Just like a review of a PB10-ISD floating around the net, someone cut the decay time to a third of original by adding only wood clamps and a pile of books on the subwoofer...
BTW fast risetime is also affected alot by the box alignment, only Q=1 in a sealed box can follow the imput amplitude correctly, but does some other things wrong.
pinkmouse help me!
simon5 said:pinkmouse help me!
Well, all right then!
The problem, as I see it is not so much speed, as distortion. The issue seems to be that some are worried that a speaker can't move fast enough or have enough control for transients, not that it won't produce the fundamental sine wave. Correct?
Now, as I said in another post, we don't really need to worry about this. Any frequency above that of the crossover sine wave, must, by definition be crossed over to the next driver. No bass driver needs to move any faster than required to produce that basic sine wave.
Ok, if we take that as a given, then we only have to worry about the transient condition, when the cone first gets the signal, say going from no signal to the initial slope of the sine wave. Note, I'm not worried so much about the opposite situation, when the signal stops, as I would guess 90 percent of bass signals have some form of fade out, ( even reverb on synths will do the same thing, so you can't catch me out there!
), so the problem will be so much less.At this point, I have to admit my maths is very rusty, so rather than using equations, I'll try and explain in English.
So, if we assume that no cone can accelerate at an infinate rate, then the first part of the transition from stationary to movement will produce a rounding off at the bottom of the curve, and I think that this is what worries Scott, (if I'm wrong, please correct me).
Part two in a while, I think I need to draw a sketch...
pinkmouse said:
Well, all right then!
1. The problem, as I see it is not so much speed, as distortion. The issue seems to be that some are worried that a speaker can't move fast enough or have enough control for transients, not that it won't produce the fundamental sine wave. Correct?
2. Now, as I said in another post, we don't really need to worry about this. Any frequency above that of the crossover sine wave, must, by definition be crossed over to the next driver. No bass driver needs to move any faster than required to produce that basic sine wave.
3. Ok, if we take that as a given, then we only have to worry about the transient condition, when the cone first gets the signal, say going from no signal to the initial slope of the sine wave. Note, I'm not worried so much about the opposite situation, when the signal stops, as I would guess 90 percent of bass signals have some form of fade out, ( even reverb on synths will do the same thing, so you can't catch me out there!
4. At this point, I have to admit my maths is very rusty, so rather than using equations, I'll try and explain in English.
5. So, if we assume that no cone can accelerate at an infinate rate, then the first part of the transition from stationary to movement will produce a rounding off at the bottom of the curve, and I think that this is what worries Scott, (if I'm wrong, please correct me).
Part two in a while, I think I need to draw a sketch...
(note I included numbering for ease of reply.)
1. Correct. (..and the problem with fourier analysis was "time" based - i.e. music is effectivly aperiodic, not periodic - you did get that correct? ..It wasn't my intention to be obtuse.)
2. Not entirely correct - go back and read my previous numbered comments, but esentially - correct. (..note though that using the description "faster" could be problematic again - and be interpreted incorrectly. Better to add the term oscilation i.e. "faster" oscilation.)
3. No.. we need to worry about leading and lagging conditions (i.e. rise-time and decay). They are inter-related subjectivly (the condition we are describing as "fast") and almost always objectivly. Most of the emphasis (what we look for in measurements) is directed to decay (as distortion - linear and non-linear). Rise-time however is rarely looked at beyond group dealy as phase relates to time.
The objective problem between the two is that the decay (and its amount) often delays and alters the rise in a music signal depending on the time between transients. (i.e. the reaction may be OK, but the re-reaction will likely be worse on short term transients.)
The subjective problem, (which is FAR more audible), could well be described as dynamic distortion (..and again the time between transients is very important). Here we have a transient signal (passband specific) that ..say.. SHOULD be down say 30db before then next rise - instead though its down perhaps 10 db. The problem then is the next rise is less "significant". The subjective effect has many different descriptors (with time as a modifier). In the short term that next rise or particular transient could sound slow, blunted, indistinct, etc.. An averaged response, (..averaged over the composition), would likely be described as compressed, loud, etc..
4. Heck my math is nearly non-existant. (unfortunetly it can be a much more precise language, but I seem to poor at languages in general - so I'm often sol when its used.)
5. Correct (modified by the above). Additionally, you need to be concerned about transient capability in relation to other drivers (call it quasi-group delay).
The problem then would be more than just time as a constant throught-out the group.. i.e. one driver actually seems "ahead" of another, but that one driver is only *occasionally* "behind" another.
This "sometimes lag" is often a problem with high x-max drivers. In otherwords when the transient is low in spl (relative to the average) and is activated in the passband where the high excursion driver operates - the sound seems OK. However (under that same general condition) when the spl increases for that transient up to or beyond the average then the high x-max driver seems to lag. -And yes, this is principally when some drivers are subjectivly deemed slow and others are deemed fast.- The classic example is a poor b@stard trying to "mate" a lowther in a BLH with something like a tumult-based subwoofer (in any configuration). Anytime any higher spl transient "hits" the tumult, the bass is deemed "slow". Of course it isn't that the driver is oscilating slowly - its that its transient capability is not that of a lowther in a BLH. (..and note to others - BLH means Back Loaded Horn - which reduces excursion.)
hmmm.. I spent way to much time on this.
http://www.worxaudio.com/product_desc_true.php?id=56
two of those in a box broke my will. Had more output that any single subwoofer I have heard, and the thing was -3 @ 20hz. The drivers by themselves are 95dB @ 1 watt, not as high as some PA speakers, but it offers far more output and depth than you normally can get out of a PA speaker.
We just shipped the first drivers last week, but they are not exactly cheap
two of those in a box broke my will. Had more output that any single subwoofer I have heard, and the thing was -3 @ 20hz. The drivers by themselves are 95dB @ 1 watt, not as high as some PA speakers, but it offers far more output and depth than you normally can get out of a PA speaker.
We just shipped the first drivers last week, but they are not exactly cheap
pinkmouse said:However, to give you a rest, I noticed an error, ( never post when you're half asleep), in my diagram above. Who can spot it?
I'll post again when we get the correct answer.
Looking at that red line appearing before the actual signal, I hereby officially congratulate you for inventing:
"The Prophetic Woofer"
Time to have a serious talk with your local patent office now, isn't it?
perhaps there is confusion about what this sine wave represents... remember... it's amplitude vertically and time horizontally...
it would seem that any presumed "slowness" in the response of the woofer might simply be represented by a "delay" in the onset of the sine wave... not necessarily a 'distortion" of its shape...
This would then be a simple "phase shift" relative to the input signal and have nothing to do with rise time, transient response, etc.
So, perhaps, we're talking about phase distortion, which IIRC, was addressed back 2 decades or so ago when stepped baffles were first introduced... and often can be compensated for in the overall transfer function analysis of the WHOLE system rather than its pieces/parts...
just my "capacitiy" for "inductive" reasoning... :-;
auplater
it would seem that any presumed "slowness" in the response of the woofer might simply be represented by a "delay" in the onset of the sine wave... not necessarily a 'distortion" of its shape...
This would then be a simple "phase shift" relative to the input signal and have nothing to do with rise time, transient response, etc.
So, perhaps, we're talking about phase distortion, which IIRC, was addressed back 2 decades or so ago when stepped baffles were first introduced... and often can be compensated for in the overall transfer function analysis of the WHOLE system rather than its pieces/parts...
just my "capacitiy" for "inductive" reasoning... :-;
auplater
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.