First of all I'd like to introduce myself to the board. I'm a long time lurker, and have enjoyed the wealth of information and discussions on the board. I'm a pro audio guy by trade, and a long time audiophile. There is sometimes some cross pollination between my work and hobby observations....
One of them is this whole notion of woofer "speed." Some will dismiss this term, but most I would say have experienced the phenomenon in one way or another. To my ear there is a huge difference in the perceived speed of lighter, low Xmax, pro audio woofers as opposed to the newish generation of high Xmax "subwoofer" drivers.
I've been running Magnepan 1.6's with Bag End ELF Subs for some time at home and have been quite happy with the setup. The Bag Ends use lightweight cone pro audio drivers, and walk all over the "slow and sloppy" sounding Dayton Titanic 15" Sub Kit(3cu ft sealed)I was using in terms of sound quality, if not output. The Bag Ends are a great compliment to the Maggies, but I've been curious about dipole/cardiod subs.
After perusing the board for some time, I decided to give it a go. I took a jigsaw out, and literally cut the back of the Titanic enclosure out. Now I have a U-frame.
Anyway, after some Spectrafoo analysis, and DSP tweaking, there is no doubt that I have better transient response, pitch definition, etc, than the Titanic in the sealed box--BUT--it still has a subtle quality of being "behind" the Maggies. The Bag Ends do not.
The cardioid pattern effect is interesting, as is the trembling of the room with the pulses of the U-Frame as opposed to the more punchy sound of the Bag End.
My next step is to try an Eminence Sigma 18 or similar in a U-Frame and compare the results. Does anyone have any thoughts on this? Mine are that it is a simple issue of trying to accelerate a heavy diaphragm too far....One that may be difficult to measure but not so difficult to hear.
Kind Regards,
Brett
One of them is this whole notion of woofer "speed." Some will dismiss this term, but most I would say have experienced the phenomenon in one way or another. To my ear there is a huge difference in the perceived speed of lighter, low Xmax, pro audio woofers as opposed to the newish generation of high Xmax "subwoofer" drivers.
I've been running Magnepan 1.6's with Bag End ELF Subs for some time at home and have been quite happy with the setup. The Bag Ends use lightweight cone pro audio drivers, and walk all over the "slow and sloppy" sounding Dayton Titanic 15" Sub Kit(3cu ft sealed)I was using in terms of sound quality, if not output. The Bag Ends are a great compliment to the Maggies, but I've been curious about dipole/cardiod subs.
After perusing the board for some time, I decided to give it a go. I took a jigsaw out, and literally cut the back of the Titanic enclosure out. Now I have a U-frame.
Anyway, after some Spectrafoo analysis, and DSP tweaking, there is no doubt that I have better transient response, pitch definition, etc, than the Titanic in the sealed box--BUT--it still has a subtle quality of being "behind" the Maggies. The Bag Ends do not.
The cardioid pattern effect is interesting, as is the trembling of the room with the pulses of the U-Frame as opposed to the more punchy sound of the Bag End.
My next step is to try an Eminence Sigma 18 or similar in a U-Frame and compare the results. Does anyone have any thoughts on this? Mine are that it is a simple issue of trying to accelerate a heavy diaphragm too far....One that may be difficult to measure but not so difficult to hear.
Kind Regards,
Brett
My background is similar to yours, i've been running concert rigs for the last 20 years so I know what you're talking about. The titanic series are, as you have noticed, mud motors (Qts: .49). Great for home theater "BOOM-n-rumble" not so good for music. Compared to the Selenium 15SWS800 15" (Qts: 0.37) the motor power to cone weight ratio is poor.
Yes I'm curious as to why people don't spend more time talking about this. The difference is quite startling. I'm a theatre guy so I'm looking for low frequency extension and absolute fidelity. Cost isn't really an object but size is often an issue. I've been thinking about ordering some of the Exodus Maelstroms, but am concerned with this being a problem with them. I've used the old Adire Maelstrom for years for LF effects w/Linkwitz Transform in sealed boxes with great success....But they were only 13mm Xmax, and still arguably not as tight or "quick" as the Bag Ends(which might have to do with them being a below resonance system unlike the LT). These other drivers are double that 13mm. Thats a loooong way. I've always wondered why Bag End used what are essentially Eminence Sigma's in their boxes, as they are ridiculously displacement limited...Maybe this is why. I'm thinking about setting up a U-Baffle or Flat Baffle test with the Titanic, Exodus, AE Dipole or IB series, in kind of a shootout, to get to the bottom of this. I was hoping to do the job with a couple of Exodus 18's or 21's per side. Maybe the answer is 6-Sigma's in some sort of space-saving sealed enclosure or some stacked W-bin ala Linkwitz.
Dan Wiggins did a study on woofer speed that to my eyes had no faults that I could see. Not that great sample size. Mass had nothing to do with "woofer speed" if I recall correctly.
Here is a link to an earlier thread on the subject with a link to Wiggin's paper as well:
http://www.diyaudio.com/forums/subwoofers/156281-woofer-speed-inductance.html
My benchmark, the Bag End ELF System, is almost ruler flat from 8-90 Hz. It is the tightest, deepest, most articulate woofer system I have ever heard, and I have used them in bedrooms and 2000 seat theatres to the same effect.
Its perplexing--I'm not one to jump on ideas that are the not scientifically sound. I've just empirically seen this phenomenon too many times to dismiss it....Thanks for these other thread links, I'll have to check them out and see if there is anything that jumps out.
There are two parts of a speaker's transient response, how fast the cone starts, and how fast the cone stops when signal is applied and when removed.
In subwoofers, the amount of "ringing" after the signal stops can have a marked effect on perception of "tightness" and "speed". How fast the cone stops makes more of a difference in that perception than how fast it starts.
A speaker designed for a sealed cabinet depends on a small air mass providing resistance, putting that speaker in an open baffle will result in less cone control, "slow" (stopping) response.
Your in room experiments with dipole and sealed subs are mixing quite different response patterns with transient perceptions, taking them outdoors would make the differences easier to identify.
You are truly running a BG in 8Hz mode? If you are using it at more than ~80dB I would run a distortion test on it, and you will see why it sounds that way.
As for the sound difference between a pro and the dayton, it's all FR. EQ the bottom up on the pro, while puling out everything over 160Hz, and they will start to sound a lot more alike.
Last edited:
Weltersys,
That makes sense....The Adire Audio paper seems to tie all of this to inductance, which would corroborate with my observations. Agreed about the out of doors; I'm going to do some experiments outdoors when I can get out to my country house.
Perhaps bringing the Dipole into this discussion was a mistake(too many variables). What I can say is the Titanic MKIII in the sealed enclosure sounds very slow and muddy compared to the Bag End 18's. At first I thought it had something to do with resonance of that system, which the Bag End does not exhibit since we are operating below resonance. I thought that maybe the U-Frame setup would eliminate this, but it does not seem to change the perception of that woofer being slow.
I need to bite the bullet and order some AE IB or Dipole15's and see what we've got.
That makes sense....The Adire Audio paper seems to tie all of this to inductance, which would corroborate with my observations. Agreed about the out of doors; I'm going to do some experiments outdoors when I can get out to my country house.
Perhaps bringing the Dipole into this discussion was a mistake(too many variables). What I can say is the Titanic MKIII in the sealed enclosure sounds very slow and muddy compared to the Bag End 18's. At first I thought it had something to do with resonance of that system, which the Bag End does not exhibit since we are operating below resonance. I thought that maybe the U-Frame setup would eliminate this, but it does not seem to change the perception of that woofer being slow.
I need to bite the bullet and order some AE IB or Dipole15's and see what we've got.
?
That is some marketing shenanigans. The bag end sub uses a 18" driver, weighs 70lbs and costs over $1000 mspr($1500 powered). So it is no surprise it has more output than a $200 diy setup. The amount of output The bag end will produce from 8-20hz will be limited by the 6mm of x-max a sigmapro has and and no amount of processing will overcome that.
I'm running 4-18" drivers in a 13'x15' room at reasonable levels, so I'm pretty sure my distortion is pretty low with the BG. As for your second point, isn't what you suggest essentially what the integrator is doing, or my Linkwitz transform Adire Maelstrom subs are doing?
Ah, this old chestnut
I've heard speakers where the bass sounds "slow", speakers where the bass sounds "fast" and tight, (to use the commonly used terminology, which I don't particularly agree with) and I've read some ridiculous theories about what makes a fast woofer or what that even means.
There seems to be a pervading theory in some quarters that because smaller woofers have lighter cones, they can accelerate faster, and because of that they have "faster" bass. I read it all the time in speaker design articles and reviews, where they say they didn't choose a larger woofer (say one 10" over 2x 6") because they wanted fast bass. Hogwash!
That's like saying you want a smaller lighter car because the acceleration will be better than a larger heavier car. But no mention of horsepower ?? Power to weight ratio ?? Have they forgotten F=ma ?
Yes, a larger woofer has a heavier cone than a smaller woofer, but given similar sensitivity and Qes it also has a FAR more powerful motor that is equally able to accelerate the cone as necessary, in fact generally the motor becomes more powerful at a faster rate than the cone becomes heavier as drivers get bigger, thanks to larger voice coil and magnet structures.
Lets also not forget that a larger driver producing the same frequency at the same SPL doesn't need to accelerate the cone back and forth as fast as the smaller driver anyway, since less excursion is required. Less excursion at the same frequency means less peak velocity even though the acceleration force acting on the cone is greater.
Then you have the fact that the maximum "speed" of the woofer is inherently limited by the low pass filter that drives it, and sets a maximum on how "fast" it can move.
No, it's nothing to do with the weight of the cone, at least not directly in any simplistic "lighter is better" way.
I think it's pretty simple - with a couple of small caveats, a tapering down bass response (decreasing with increasing frequency) in the 30-150Hz range results in a "slow bass" sound, while a response that tapers up fairly monotonically results in a "fast bass" sound.
In fact a very "fast" bass response that some seem to strive for is technically incorrect as well, with a flat response being more correct, although it can sound pleasant on some types of music and if you have to choose between a response that slopes down with increasing frequency or one that slopes smoothly up, the latter is probably a "safer" choice, as an early gradual roll-off in the bass has better transient characteristics than a response that actually increases with lowering frequency then cuts off abruptly.
Anyone who has studied or measured room gain effects knows where I'm going with this - a speaker naively designed with a flat and extended bass response (down to say below 30Hz) is going to have far too much low end bass response below ~50Hz possibly on the order of 10-15dB in smaller rooms, and still a few dB even in quite large rooms.
Coupled with this is a tendency to position speakers a long way from room boundaries - a metre or more, and you can easily develop deep notches in the response in the 60-150 region. 10-15dB too much below 50Hz, 10dB or more notches in the mid/upper bass and you have a recipe for disaster.
Replace the big speaker with a bass response that is anechoically flat to a very low frequency with a much smaller one (probably with smaller woofers) with a lot higher cut-off frequency and/or early gradual roll off and by good luck or chance you may find that low bass below 50Hz is attenuated enough by the natural roll off of the speaker so that the total result is much more in balance between 30-150Hz.
Naive speaker designer or reviewer then concludes that the smaller woofers have "faster bass". In fact much the same result could have been achieved by equalizing the low frequency room gain out, leaving far more available dynamic range compared to the smaller speaker.
Moving the speakers to correct the mid/upper bass response as much as possible and then EQ'ing any remaining excessive bass rise below 50Hz can make a world of difference to the quality of the bass.
So what were the caveats I mentioned ? The first one is that the speaker itself doesn't have a very high Q (boom box) alignment that would lead to large peaks in the group delay response at low frequencies, this will give a soggy sounding bass no matter what you do to the overall frequency tilt from 30-150 Hz. How much is too much is a matter for debate of course, some people won't accept any form of bass reflex alignment for example, others will.
The second point is unequal dynamic range across the bass region from 30-150Hz, something that is extremely important with a main/subwoofer setup.
The correct frequency balance has to exist at high SPL levels as well as during small signal measurements.
So often I see something like a 12" subwoofer (or bigger) mated to a small skinny floor standing speaker with, if it's lucky, a couple of 6" "woofers", then crossed over somewhere between 50-80Hz. What's the problem with this ?
The problem is that the 12" subwoofer, even a mediocre one, has far more dynamic range below the subwoofer crossover frequency than the small main speaker has above the crossover region.
Even if everything was equalized for a well integrated and balanced response at low signal levels (no small challenge in itself) as soon as you play it loud the smaller drivers can't keep up - they suffer dynamic range compression in the upper/mid bass region, and the dynamic frequency balance shifts to being heavy on low bass again, so the bass loses its attack since the attack comes from the higher bass frequencies.
For this reason, although it wont please subwoofer fans I've always felt the best quality bass for music (rather than LFE effects) is to have the same drivers operating from the low cutoff frequency ~30Hz right up to at least 150Hz so that the entire bass range is reproduced by the same drivers.
By doing so you make it impossible for dynamic range compression to occur at higher bass frequencies before lower bass frequencies, since a driver of a single type and size will always (at least in a sealed box) start to have dynamic range compression happen first at the lower frequencies - thus tending towards a "faster" bass sound as it does compress, rather than a "slow" response you might see with a subwoofer and inadequately sized woofer in a main speaker.
(You also avoid introducing another crossover point along with phase shift and group delay, right in the middle of the bass range.)
One final note about bass reflex designs - a sealed box as mentioned above, always compresses the most at the lowest frequencies first, but a bass reflex design can actually produce more SPL at the box tuning than it can in the octave above that, which is not what we want when we try to avoid frequency dependant compression where the higher frequencies are compressed more than the lower ones.
In a poorly designed bass reflex design this can be an issue - especially if the port is tuned very low for a relatively small driver, to try to extend the F3 unreasonably low for the size of driver, T/S modelling not withstanding.
Although it might have a flat modelled response, at higher volume levels (not that high actually, with a small driver) the speaker will have frequency dependant compression where it can produce far more output around the box tuning frequency than it can above it - end result, "slow" bass again, lacking impact.
This can be minimized by using a driver that is large enough to not rely too much on the loading of the port to achieve the desired SPL, and not choosing an alignment that puts the box tuning a long way below the frequencies that the same driver in an equalized sealed box could produce adequately by itself.
(In other words shape the respose with the bass reflex alignment, but make sure the driver by itself has enough dynamic range in a sealed actively equalized box for the frequencies it's covering, and consider the improvement in maximum SPL of the bass reflex as "extra headroom" rather than depending on it to reach normal playback levels)
Some of the best sounding articulate bass I've heard has been from relatively large drivers being operated fairly high in frequency, for example 12" being used up to ~250Hz, so it's all about good design and the right target response, not some magical characteristic of "small light" woofers.
That's my thoughts anyway, now to don my fire protective suit and sit back
I've heard speakers where the bass sounds "slow", speakers where the bass sounds "fast" and tight, (to use the commonly used terminology, which I don't particularly agree with) and I've read some ridiculous theories about what makes a fast woofer or what that even means.
There seems to be a pervading theory in some quarters that because smaller woofers have lighter cones, they can accelerate faster, and because of that they have "faster" bass. I read it all the time in speaker design articles and reviews, where they say they didn't choose a larger woofer (say one 10" over 2x 6") because they wanted fast bass. Hogwash!
That's like saying you want a smaller lighter car because the acceleration will be better than a larger heavier car. But no mention of horsepower ?? Power to weight ratio ?? Have they forgotten F=ma ?
Yes, a larger woofer has a heavier cone than a smaller woofer, but given similar sensitivity and Qes it also has a FAR more powerful motor that is equally able to accelerate the cone as necessary, in fact generally the motor becomes more powerful at a faster rate than the cone becomes heavier as drivers get bigger, thanks to larger voice coil and magnet structures.
Lets also not forget that a larger driver producing the same frequency at the same SPL doesn't need to accelerate the cone back and forth as fast as the smaller driver anyway, since less excursion is required. Less excursion at the same frequency means less peak velocity even though the acceleration force acting on the cone is greater.
Then you have the fact that the maximum "speed" of the woofer is inherently limited by the low pass filter that drives it, and sets a maximum on how "fast" it can move.
No, it's nothing to do with the weight of the cone, at least not directly in any simplistic "lighter is better" way.
I think it's pretty simple - with a couple of small caveats, a tapering down bass response (decreasing with increasing frequency) in the 30-150Hz range results in a "slow bass" sound, while a response that tapers up fairly monotonically results in a "fast bass" sound.
In fact a very "fast" bass response that some seem to strive for is technically incorrect as well, with a flat response being more correct, although it can sound pleasant on some types of music and if you have to choose between a response that slopes down with increasing frequency or one that slopes smoothly up, the latter is probably a "safer" choice, as an early gradual roll-off in the bass has better transient characteristics than a response that actually increases with lowering frequency then cuts off abruptly.
Anyone who has studied or measured room gain effects knows where I'm going with this - a speaker naively designed with a flat and extended bass response (down to say below 30Hz) is going to have far too much low end bass response below ~50Hz possibly on the order of 10-15dB in smaller rooms, and still a few dB even in quite large rooms.
Coupled with this is a tendency to position speakers a long way from room boundaries - a metre or more, and you can easily develop deep notches in the response in the 60-150 region. 10-15dB too much below 50Hz, 10dB or more notches in the mid/upper bass and you have a recipe for disaster.
Replace the big speaker with a bass response that is anechoically flat to a very low frequency with a much smaller one (probably with smaller woofers) with a lot higher cut-off frequency and/or early gradual roll off and by good luck or chance you may find that low bass below 50Hz is attenuated enough by the natural roll off of the speaker so that the total result is much more in balance between 30-150Hz.
Naive speaker designer or reviewer then concludes that the smaller woofers have "faster bass". In fact much the same result could have been achieved by equalizing the low frequency room gain out, leaving far more available dynamic range compared to the smaller speaker.
Moving the speakers to correct the mid/upper bass response as much as possible and then EQ'ing any remaining excessive bass rise below 50Hz can make a world of difference to the quality of the bass.
So what were the caveats I mentioned ? The first one is that the speaker itself doesn't have a very high Q (boom box) alignment that would lead to large peaks in the group delay response at low frequencies, this will give a soggy sounding bass no matter what you do to the overall frequency tilt from 30-150 Hz. How much is too much is a matter for debate of course, some people won't accept any form of bass reflex alignment for example, others will.
The second point is unequal dynamic range across the bass region from 30-150Hz, something that is extremely important with a main/subwoofer setup.
The correct frequency balance has to exist at high SPL levels as well as during small signal measurements.
So often I see something like a 12" subwoofer (or bigger) mated to a small skinny floor standing speaker with, if it's lucky, a couple of 6" "woofers", then crossed over somewhere between 50-80Hz. What's the problem with this ?
The problem is that the 12" subwoofer, even a mediocre one, has far more dynamic range below the subwoofer crossover frequency than the small main speaker has above the crossover region.
Even if everything was equalized for a well integrated and balanced response at low signal levels (no small challenge in itself) as soon as you play it loud the smaller drivers can't keep up - they suffer dynamic range compression in the upper/mid bass region, and the dynamic frequency balance shifts to being heavy on low bass again, so the bass loses its attack since the attack comes from the higher bass frequencies.
For this reason, although it wont please subwoofer fans I've always felt the best quality bass for music (rather than LFE effects) is to have the same drivers operating from the low cutoff frequency ~30Hz right up to at least 150Hz so that the entire bass range is reproduced by the same drivers.
By doing so you make it impossible for dynamic range compression to occur at higher bass frequencies before lower bass frequencies, since a driver of a single type and size will always (at least in a sealed box) start to have dynamic range compression happen first at the lower frequencies - thus tending towards a "faster" bass sound as it does compress, rather than a "slow" response you might see with a subwoofer and inadequately sized woofer in a main speaker.
(You also avoid introducing another crossover point along with phase shift and group delay, right in the middle of the bass range.)
One final note about bass reflex designs - a sealed box as mentioned above, always compresses the most at the lowest frequencies first, but a bass reflex design can actually produce more SPL at the box tuning than it can in the octave above that, which is not what we want when we try to avoid frequency dependant compression where the higher frequencies are compressed more than the lower ones.
In a poorly designed bass reflex design this can be an issue - especially if the port is tuned very low for a relatively small driver, to try to extend the F3 unreasonably low for the size of driver, T/S modelling not withstanding.
Although it might have a flat modelled response, at higher volume levels (not that high actually, with a small driver) the speaker will have frequency dependant compression where it can produce far more output around the box tuning frequency than it can above it - end result, "slow" bass again, lacking impact.
This can be minimized by using a driver that is large enough to not rely too much on the loading of the port to achieve the desired SPL, and not choosing an alignment that puts the box tuning a long way below the frequencies that the same driver in an equalized sealed box could produce adequately by itself.
(In other words shape the respose with the bass reflex alignment, but make sure the driver by itself has enough dynamic range in a sealed actively equalized box for the frequencies it's covering, and consider the improvement in maximum SPL of the bass reflex as "extra headroom" rather than depending on it to reach normal playback levels)
Some of the best sounding articulate bass I've heard has been from relatively large drivers being operated fairly high in frequency, for example 12" being used up to ~250Hz, so it's all about good design and the right target response, not some magical characteristic of "small light" woofers.
That's my thoughts anyway, now to don my fire protective suit and sit back
Last edited:
Mr. Doom,
The Dayton will actually put out more very low frequency energy than the Bag End will....There is nothing special about the Bag End drivers or enclosure; and completely agree with your assertion about Xmax and processing. My question is not one of output, it is of sound quality. With the Bag Ends, you just have to add more cones to get the output you want, which is fine with me. I've heard through the grapevine that they will use 24-18" units to mix some of these bass heavy feature films. We regularly use at least 8-18" drivers to reproduce/reinforce orchestral music in a 1000 seat theatre.
OK this is totally unscientific....See this video below. It may not sound sloppy but it sure looks sloppy:
YouTube - Maelstrom-X 21" Excursion Video Playing Clubbed to Death
YouTube - Maelstrom-X 21" Excursion Video Playing Clubbed to Death
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.