I converted my sealed subwoofers to ported ones because I was unhappy with the somewhat weak bass character. The change was not surprising, less SPL and less group delay in the simulation program, but in practice the bass was deeper, I attribute this to the design of the speaker, not suitable for sealing according to the manufacturer, (despite the EBD that It indicated " suitable for both systems . " Many years ago I read a very explanatory article about how a speaker behaves in a room with the following title : " The box within the box " .
I'm old school, no EQ, (although I have the parametric one in the subwoofer amplifier with DSP) no REW, I use a lot of variety of damping distributed everywhere in the room, which has very irregular dimensions, luckily.
Hi Mark - I think I have to disagree a bit. Perhaps not in terms of the facts as you have laid them out, but in the conclusions.
I find it very feasible to use a simple 2nd order shelf filter to extend the bass response of a sealed box design. A 6 dB boost is usually a good starting point. And yes, this boost extends all the way down to infrasonic. But there is no real content down there anyway. When this is done correctly, the group delay above the shelf filter is is actually less then the plain sealed box... only below the shelf is it more. An increase in GD at 35 Hz is well worth a lowering of GD at 80 Hz...
I don't know why a person would use a high pass filter in the deep bass... am I missing something here? Maybe for a turntable perhaps. Nor would I recommend using a PEQ to boost the bass, because as you said, the bass rolloff becomes too steep and then it is no different than a vented box. But why would someone do this?
The vented box speaker will almost always be larger than the sealed+LT, often twice as big. In that same box, we could use 2 woofers, and thus overcome the theoretical +3 dB max SPL advantage of the vented box.
Maybe I am missing something... it would not be the first time 🙂 .
Hi Jim.
basic answer is excursion is most probably greater than you want at the low freq range, without a hpf.
Just cause it's sealed doesn't negate excursion issues, lord knows after EQ boost.
Use hornresp to sim, or try some old fashioned ways of measuring excursion. I mostly rely on hornresp
There's a lot more to say about bass-reflex vs sealed in general......regarding your other points/questions.
let me stick to the answer above before those ....thx
Last edited:
Hi hifijim,
Kindest regards,
M
Could you please comment on the "When it is done correctly . . ." statement?
Kindest regards,
M
Generalities may be helpful to understand things. But sometimes they rely on assumptions or axioms that does not apply for the specific use case.
For instance, in general a ported subwoofer has the potential to go deeper, and have more output in the octaves of interest, than a sealed box.
In general, a tapped horn also has the potential to have higher output in the lowest octaves than a ported box.
On the other hand, a sealed box with DSP and high power/efficiency amplification has the potential to be smaller, and go deep and loud enough, and limit the excursion. And with equalisation, whether it be via PEQ or LT, can reduce the group delay.
A lot depends on if one is using all the modern tools at their disposal, and what constraints or goals they are have (or the listener's/customer's preferences)
My deepest respect for Lawrence. I thought it is something only a DIYer would dare to attempt:
EIGHT ~11" woofers, dual opposed, in a vented configuration for bass below 125Hz (F6 19Hz).
This cabinet is 2 ft 2" wide, by 4 ft deep, the weight for a upright piano (760lbs). All in a passive crossover for a 5 way.
I'd design/build/buy it. But one man's singular vision may be another (wo)man's garish hallucination (?nightmare)
Reference:
https://vividaudio.com/downloads/Vivid_Audio-Moya_M1-e_brochure.pdf
For instance, in general a ported subwoofer has the potential to go deeper, and have more output in the octaves of interest, than a sealed box.
In general, a tapped horn also has the potential to have higher output in the lowest octaves than a ported box.
On the other hand, a sealed box with DSP and high power/efficiency amplification has the potential to be smaller, and go deep and loud enough, and limit the excursion. And with equalisation, whether it be via PEQ or LT, can reduce the group delay.
A lot depends on if one is using all the modern tools at their disposal, and what constraints or goals they are have (or the listener's/customer's preferences)
My deepest respect for Lawrence. I thought it is something only a DIYer would dare to attempt:
EIGHT ~11" woofers, dual opposed, in a vented configuration for bass below 125Hz (F6 19Hz).
This cabinet is 2 ft 2" wide, by 4 ft deep, the weight for a upright piano (760lbs). All in a passive crossover for a 5 way.
I'd design/build/buy it. But one man's singular vision may be another (wo)man's garish hallucination (?nightmare)
Reference:
https://vividaudio.com/downloads/Vivid_Audio-Moya_M1-e_brochure.pdf
Last edited:
As a person who used to own 2250 LP's and one of the last M97EraIV cartridges Shure made in Illinois, I have ripped the ribbons of a 10" woofer before making a $800 investment in 15" woofers. Walking across a wood floor makes a woofer dance an inch or more with a mag phono arm down on an LP. 1505-8kadt Xmax=0.9mm, deltapro-15a xmax=4.3 mm. Now that my local unemployed house painter has carried all the LP's cartridge and speakers off to his fence, I have no longer have a reason to invest in $$$ passive low pass filter parts; such as 450 uf series caps and 56 mH parallel coils. The LowPass filter in my disco mixer is ****, 3 db per octave and offensively audible on organ and piano records, much less popular music. I never used it.
There are still used US built ported boxes available flat to 54 hz and -10db at 40, in the standard production test setup, an open non-echo chamber. My flat wall placement of ported 15" woofers boosts 40 hz to -4 db from 54 hz level, which effect may go down to 25 hz my white noise test indicates. (measurement polluted by a diesel sewer pump 1 mile away). Active sealed subwoofers now available visibly come from factories I find offensive. I may build a 53-27 hz sub some day from a Lab15 if eminence does not lay off all their KY workers.
Aren't we forgetting a bit, that KEF chose those smaller drivers to fit them close to the Coax, control dispersion and in general make a sleek design?
I very rarely hear a ported speaker sound good in the bass. It is as if I always hear that "woombf" from the port, providing some kind of annoyance in the rest of the upper spectrum. Most commercial speakers I listen to, have that elevated tweeter.... just 3-5dB more than the midrange - are older people with money for these, always a bit more deaf?
Weird part for me, is that extra tweeter response, has very little energy. Most of the details and quality of sound perception, is in many cases at the integration of the tweeter/midrange/cross-over - way lower in frequency. Most of these speakers have a "free" firing tweeter, which "lights" up the room, creating the sensation of too much high frequency - IMO. A waveguide mostly does this much, much better.
A lot of people should turn their buying decisions around. Find out how loud you want to play, hearing loss, music taste, budget and personal DIY skills. Then work backwards from there, rather than falling in love with one specific driver, brand, storytelling or other person's opinions.
I chose a Coax from KEF, because it combines well engineered technology that has very little breakup in the midrange, extended smooth tweeter, packed in a point source that takes up minimum space and has a waveguided tweeter. This fits with my preference for a good listening experience and by going DIY, I could choose bigger woofers - in my case a dual WO24P in 75L closed box with added subwoofers.
When closed, I added a first order HP at around 60-70Hz, to relieve them from the lowest bass and avoid excessive phase-shift, making it easier to integrate subwoofers.
Their commercial versions, are all to "tweety" and ported.... not my taste.
Most expensive speakers have pretty nice top-end, but bass.... noooo. Either, way too boomy or not there at all. A tiny simple DSP and directly powered active woofer.... woosh... so many issued solved.
Would be fun to revisit a set of these old B&W 605 S2, with just the simplest of DSP added to the woofers.
https://audio-reviews.com/speaker-pair/speaker-pair-bw-dm-605-s2-1029
I very rarely hear a ported speaker sound good in the bass. It is as if I always hear that "woombf" from the port, providing some kind of annoyance in the rest of the upper spectrum. Most commercial speakers I listen to, have that elevated tweeter.... just 3-5dB more than the midrange - are older people with money for these, always a bit more deaf?
Weird part for me, is that extra tweeter response, has very little energy. Most of the details and quality of sound perception, is in many cases at the integration of the tweeter/midrange/cross-over - way lower in frequency. Most of these speakers have a "free" firing tweeter, which "lights" up the room, creating the sensation of too much high frequency - IMO. A waveguide mostly does this much, much better.
A lot of people should turn their buying decisions around. Find out how loud you want to play, hearing loss, music taste, budget and personal DIY skills. Then work backwards from there, rather than falling in love with one specific driver, brand, storytelling or other person's opinions.
I chose a Coax from KEF, because it combines well engineered technology that has very little breakup in the midrange, extended smooth tweeter, packed in a point source that takes up minimum space and has a waveguided tweeter. This fits with my preference for a good listening experience and by going DIY, I could choose bigger woofers - in my case a dual WO24P in 75L closed box with added subwoofers.
When closed, I added a first order HP at around 60-70Hz, to relieve them from the lowest bass and avoid excessive phase-shift, making it easier to integrate subwoofers.
Their commercial versions, are all to "tweety" and ported.... not my taste.
Most expensive speakers have pretty nice top-end, but bass.... noooo. Either, way too boomy or not there at all. A tiny simple DSP and directly powered active woofer.... woosh... so many issued solved.
Would be fun to revisit a set of these old B&W 605 S2, with just the simplest of DSP added to the woofers.
https://audio-reviews.com/speaker-pair/speaker-pair-bw-dm-605-s2-1029
D
Deleted member 375592
IFAIK, in a vented design, below f tuning the sub and the port work in phase (4th order, 4*90=0), i.e. it becomes an open baffle, and the excursions may be ... very high even if the subsonic input is lowish. So it's recommended to add a sharp elliptic high-pass tuned a bit below f Helmolthz. Many people, including myself, use the same plate amplifier from an old ported subwoofer for closed-box DIY sub.
I must respectfully disagree. The energy storage and release of a resonant system makes LF sound reproduction quite 'blurred' to my ear compared with sealed. I feel that step response is important.
The 603S2 is a 6 series floorstander introduced ~2020. It's on clear out for a price less than one could DIY..
Put large round-overs around the tweeter, rework the 3 way (twin vented 6.5" + 5" midrange + 1" tweeter) crossover…
edit: removed link to Bezos' bookstore.
Last edited:
That totally depends how you design the things.
If you're not careful, port compression and also other side effects from the port are real. Not to talk about the fact that the volume needs to be pretty big to not gave a gigantic long port.
Btw. Not able to do the last low octave?
Never used anything bigger than 15 inch I assume?
@tktran303 - true. But the 605 s2 is more than 20 years old, so just an example.... B&W did not age well, since the Vivid designer left the party - IMO 😉
Same transfer function= same step response= same impulse response= same stored energy. That is how it works.
But in reality it is near impossible to get exactly the same transfer functions, because with the vented system, there is a second radiating element, which i.m.o. is extremely difficult to control. Furthermore, the in-room summing of both radiating elements is a complex issue. Thus simple summing of a NF measurement of the woofer does +NF measurement of the vent does not telle the whole story, not to mention the pipe resonances of the vent. You need a rather advanced simulator to realistically model such a system in room. In stereo, we have 4 sources.
Many years ago I did extensive trials in order to damp the most annoying resonances of a simple vented floor stander. By the time I killed all the spurious resonances, I ended up with a quasi sealed alignment.
Just for the sake of completeness: sealed system is resonant as well. It's just lower order, thus group delay is smaller.
We are starting to get into the discussion where people use a slightly different meaning of the word "resonant system".
Anything that is at least a 2nd order system (or higher) has a resonance frequency as well as Q-factor.
So in a system with a very low Q-factor, technically there is still a resonance frequency, it's only extremely dampened, but there is a resonance frequency nonetheless.
However, we wouldn't call such a system a system that resonates well.
So from that perspective, I can see why people wouldn't call anything lower Q=0.5 a resonant system.
But it leads to confusion.
It's not that important, since it all just translates to the frequency response anyway.
And a bit of difference in group-delay.
Hi, coffee break philosophy:
34Hz would have 10m wavelength, 30'. This means, that before the cone has even reached max excursion, let alone kept on resonating, there is already early reflections at listening spot superimposing into a null or a peak. I think this stuff is moot for domestic room stereo, the bass boxes can be almost anything sensible and the room would still dominate sound at listening position. I think Low Q, or low resonanse, bass box just has higher probability to sound better given small room bass problems (peaks).
However, a low frequency (high excursion assuming nice frequency response) sound from a woofer would distort the whole bandwidth of the woofer. If it's two way speaker this "problem" extends all the way through midrange to high mids and even beyond. So better use separate bass system and mid system to avoid this. Making a split between bass and mids is opportunity to somehow deal with the room as well.
So I see distortion is not separate issue but tied to the whole system, and when the whole system is optimized for all kinds of things, I'm quite sure distortion isn't on the table anymore given the transducers are not cheapest possible, but something sensible, use good enough transducers, design good enough system. Conversely, if one has distortion problem with a system, well, perhaps it has other problems too and should be thought over.
While it's an interesting topic I don't think it's something to waste time too much on, as it really doesn't seem to be an issue to waste time on. Parroting? for sure, but anyone can come up with same conclusion experimenting with systems and doing some logic.
34Hz would have 10m wavelength, 30'. This means, that before the cone has even reached max excursion, let alone kept on resonating, there is already early reflections at listening spot superimposing into a null or a peak. I think this stuff is moot for domestic room stereo, the bass boxes can be almost anything sensible and the room would still dominate sound at listening position. I think Low Q, or low resonanse, bass box just has higher probability to sound better given small room bass problems (peaks).
However, a low frequency (high excursion assuming nice frequency response) sound from a woofer would distort the whole bandwidth of the woofer. If it's two way speaker this "problem" extends all the way through midrange to high mids and even beyond. So better use separate bass system and mid system to avoid this. Making a split between bass and mids is opportunity to somehow deal with the room as well.
So I see distortion is not separate issue but tied to the whole system, and when the whole system is optimized for all kinds of things, I'm quite sure distortion isn't on the table anymore given the transducers are not cheapest possible, but something sensible, use good enough transducers, design good enough system. Conversely, if one has distortion problem with a system, well, perhaps it has other problems too and should be thought over.
While it's an interesting topic I don't think it's something to waste time too much on, as it really doesn't seem to be an issue to waste time on. Parroting? for sure, but anyone can come up with same conclusion experimenting with systems and doing some logic.
Last edited:
We are talking about IMD here.
Or in other words, the same BL(x), Cms(x) and Le(x) story all over again.
There are two possible solutions to this.
Spend a lot of effort (= money) into a driver that does this super well.
Or like you said, split the low-end and the rest of the system up.
Which I think is by far the better approach, since like you said we have the opportunity to deal with room modes.
Technically this can be done with a woofer that does low as well as mid part, but it's a lot more tricky and not possible with some serious thinking of how to implement this in some custom DSP filter blocks.
Another benefit, is that we can get much cheaper drivers, since they have to be only good at one part, instead of the entire freq range.