Since the distortion is so low you need at least 2 drivers with big surface area to move much air with low excursion (low distortion)
3 high quality 10" woofers will work. like the SEAS https://audioxpress.com/article/test-bench-the-l26roy-10-subwoofer-from-seas
3 high quality 10" woofers will work. like the SEAS https://audioxpress.com/article/test-bench-the-l26roy-10-subwoofer-from-seas
Regarding cancelling 2nd order distortion by reversing one of the element...
Try to face them at the same polarity!
2dn order distortion is actually not that bad, and at low frequencies it enhances the subjective performance.
I just mentioned that distortion is bad when integrate the subs to the QUADS. For you who knows there is nothing new here, just pure science.
For you who are surprised or just ignore the fact that our ear is distorting up to 10% 2nd harmonic at low frequency. And that this "good" distortion is masking higher order distortions like the 3rd (less musical) component, there is plenty of interesting studies in this field.
Try to face them at the same polarity!
2dn order distortion is actually not that bad, and at low frequencies it enhances the subjective performance.
I just mentioned that distortion is bad when integrate the subs to the QUADS. For you who knows there is nothing new here, just pure science.
For you who are surprised or just ignore the fact that our ear is distorting up to 10% 2nd harmonic at low frequency. And that this "good" distortion is masking higher order distortions like the 3rd (less musical) component, there is plenty of interesting studies in this field.
If you do not demand earshattering levels and you use them for music, not home cinema, even less will do.
I have dipole subs with only one 12 inch driver per channel, in a modified H-frame. There is plenty of headroom for home use, and they are EQ'd down to 25 Hz.
I have these subs for over 20 years now, first with ESL57, later ESL63 and now Audiostatics. I did never measure max SPL of the system exactly but 95dB with all kinds of music is no problem for sure and that is quite loud in a home setting. More than I need anyway.
The driver is the (now obsolete) Peerless XLS12. Linkwitz used them in the past as well , I think in the Phoenix. Linear Xmax of a whopping 25 mm and 40 mm mechanical limit, that helps.
I've tested several woofers over the past few years. My preliminary finding is that if you can get away with a larger woofer like a single 15" do so. With a high efficiency 15" less excursion is necessary to produce equal volume (to move the same amount of air) compared with a 10". It's also a more efficient design in all regards -- a single large magnet moving a larger cone... For near field distance small woofers of 6"-10" are preferable as a smaller cone produces a more focused sound. The Visaton W200 and the Seas are excellent choices.
Any listening at over 10' and at reasonable volume 10" woofers become problematic in my experience. As is observed and reflected in linkwitzs LX521 subwoofer enclosure design, the increased excursion necessitated by smaller 10" woofers to achieve volume results in reflections, room interaction and ultimately a greater degree of audible resonances. In other words, a less excursive a woofer achieving equal output at a given frequency will produce less room and or enclosure related resonances due do to lower amplitude and more diffuse wave propagation. On the other hand, if listening in single point location, within a certain volume range, the focus of a smaller woofer may be advantageous.
One design which could potentially mitigate the issue of room reflection presented by smaller subwoofers is to design an open baffle bass enclosure where the rear of the woofer points toward the ceiling, or slightly canted forward toward the listener, in a compact curved folded horn of a ~45 degree radius where the front of the cone fires down and then forward out of the front of the enclosure towards the listener. The advantage here is that rather than dealing with rear wall reflections, one is afforded much more flexible positioning having presumably at least +8' of distance between the ~floor and ceiling for the woofer to dissipate its rear wave energy.
I have an Seas 10" LROY woofer on hand to experiment with. It's certainly an impressive design being powerful at 4ohms and ~18Bl. It's composed at high output as it is pistonic, has a robust cone and is thus low distortion. The heavy cone and voice coil is a limiting factor. Compared with a 55mms cone 98db sensitivity 15" woofer that I have on hand the Seas does not achieve equivalent speed and transient response. If using the the Seas for just sub bass frequencies, say 30/40hz and below, at normal listening levels, the Seas is maybe a better choice. For a more conventional approach of an 80hz crossover point, which is substantially different from 30hz, a low mass 15" is a far better choice. In other words, as a supplementary woofer with the Quads 57s or any other electrostat, I would look to a large driver with a light cone.
If low volume near field listening at a low crossover frequency is the plan then the Seas would be a good choice. Other brands that produce a similar product are Visaton, Wavecore, Faital Pro, and SB Acoustic//Audience, which seems to have reissued some of Peerless's older drivers.
All that said, the best way to achieve great fidelity is to preserve your ears.
Any listening at over 10' and at reasonable volume 10" woofers become problematic in my experience. As is observed and reflected in linkwitzs LX521 subwoofer enclosure design, the increased excursion necessitated by smaller 10" woofers to achieve volume results in reflections, room interaction and ultimately a greater degree of audible resonances. In other words, a less excursive a woofer achieving equal output at a given frequency will produce less room and or enclosure related resonances due do to lower amplitude and more diffuse wave propagation. On the other hand, if listening in single point location, within a certain volume range, the focus of a smaller woofer may be advantageous.
One design which could potentially mitigate the issue of room reflection presented by smaller subwoofers is to design an open baffle bass enclosure where the rear of the woofer points toward the ceiling, or slightly canted forward toward the listener, in a compact curved folded horn of a ~45 degree radius where the front of the cone fires down and then forward out of the front of the enclosure towards the listener. The advantage here is that rather than dealing with rear wall reflections, one is afforded much more flexible positioning having presumably at least +8' of distance between the ~floor and ceiling for the woofer to dissipate its rear wave energy.
I have an Seas 10" LROY woofer on hand to experiment with. It's certainly an impressive design being powerful at 4ohms and ~18Bl. It's composed at high output as it is pistonic, has a robust cone and is thus low distortion. The heavy cone and voice coil is a limiting factor. Compared with a 55mms cone 98db sensitivity 15" woofer that I have on hand the Seas does not achieve equivalent speed and transient response. If using the the Seas for just sub bass frequencies, say 30/40hz and below, at normal listening levels, the Seas is maybe a better choice. For a more conventional approach of an 80hz crossover point, which is substantially different from 30hz, a low mass 15" is a far better choice. In other words, as a supplementary woofer with the Quads 57s or any other electrostat, I would look to a large driver with a light cone.
If low volume near field listening at a low crossover frequency is the plan then the Seas would be a good choice. Other brands that produce a similar product are Visaton, Wavecore, Faital Pro, and SB Acoustic//Audience, which seems to have reissued some of Peerless's older drivers.
All that said, the best way to achieve great fidelity is to preserve your ears.
Last edited:
It should be remembered that (sub)woofer speed and transient response is a totally red herring.
The crossover filter will send any fast, high-speed and transient signal components to the mid/high drivers.
What remains for the sub are the slow, docile, tame signal components.
Jan
The crossover filter will send any fast, high-speed and transient signal components to the mid/high drivers.
What remains for the sub are the slow, docile, tame signal components.
Jan
Oh, if only people could get this simple fact through their heads...
I believe the suggestion of "speed" in the bass comes from the higher frequencies produced by the same instruments that produce the bass, and some harmonics. This is mostly midrange territory. Low distortion of the sub so it doesn't produce higher harmonics, and good time coherence between sub and midrange is important here.
The typical "colouration" that is unavoidable in closed and vented box designs and which in essence is delayed and distorted release of the rear wave energy, is imho also very detrimental for this impression of speed.
Nothing beats an OB woofer here 🙂
The typical "colouration" that is unavoidable in closed and vented box designs and which in essence is delayed and distorted release of the rear wave energy, is imho also very detrimental for this impression of speed.
Nothing beats an OB woofer here 🙂
I also believe that the shortest possible transient decay, or a low amount of energy stored in the system, is also very important in a subwoofer.
Ported/tapped horns etc. are basically poorly damped resonant systems which have very long decay times compared with sealed or OB, blurring events in the temporal domain.
Much is made of waterfall plots of higher frequency drivers and the importance of quick decay, but this phenomenon is usually completely ignored in subs, where extended LF cutoff usually comes across as priority.
There is no transient signal in a sub. Therefore, there is no transient decay.
The signal going to the sub has no signal components above a few 100Hz.
A sub is never required to stop within a millisecond - that would require 1kHz signal components, which are not there.
These are routed to the midrange driver.
Jan
The signal going to the sub has no signal components above a few 100Hz.
A sub is never required to stop within a millisecond - that would require 1kHz signal components, which are not there.
These are routed to the midrange driver.
Jan
Perhaps better to look at it in terms of energy storage.
Closed/BR/PR/TL/horn systems all store energy in their cabinets in some way, and release that energy again at some point later in time (= the wrong time). OB's eliminate that problem altogether. Provided they have a reasonable low Q.
Closed/BR/PR/TL/horn systems all store energy in their cabinets in some way, and release that energy again at some point later in time (= the wrong time). OB's eliminate that problem altogether. Provided they have a reasonable low Q.
ESL eventually means constantly struggling for and with the bass response. Make your choice ...
Assuming
-Sd = 550 cm^2
- Xmax = 25mm
Then in free air, dipole max. SPL will be
78dB @ 25Hz
87dB @ 35Hz
96dB @ 50Hz
- The H-frame and in case of d=250mm accounts for +4dB
- The sub may (?) be placed on the floor: +6dB
88dB @ 25Hz
97dB @ 35Hz
106dB @ 50Hz
This is simulations. And it looks nice. In real life instead, as alway you will have to deal with annoying multifactor facts. As for me and in my "very best" time, I had 8 of these Peerless XXLS-12inch drivers built within 4 W-typed dipole enclosures. Despite the sheer overkill aspect of this setup, I did not really like the overall result when matched with ESL's. Sure, with all this volume displacement, I got some sub. I got sub in terms of frequency range, of course. But also sub in terms of overall rendition quality. Because of ESL's work in their's own class.
1. At Xmax = 25mm at low frequencies these Peerless drivers produced a lot (!) of wind and mechanical noise. This might be more or less the case for any driver, as Linwitz found out, this driver type having been one of the most suited in his experiments. Therefore, the real max. usable SPL is substantially lower if you do want rather to listen to clean bass than to the driver's noise and air turbulences. And so also the benefit as a deep frequency extender for an ESL will be lower than suggested by the raw Sd/Xmax-related SPL numbers.
2. Be aware that any addition of a magnetodynamic driver to an ESL will come along with two basically different transient response types. A light planar mylar membrane will not have the same dynamic and transient behavior than a heavy and complex bass cone assembly moving. Drive is finite ... This will mishmash any transient and/or phase response. So best for a coherent and deeper/louder bass response is not addding a dynamic driver, but might be addding a lot of ESP aera dedicated for bass instead. Your might place an extra bass-extending pair of ESL assemblies closer to your listening position, much closer than the main ESL stereo speaker pair for best SPL and least room modes stimulation.
3. Be aware, that especially for dipoles and for lower frequencies, the emmitted acoustical power is strongly related to the placement near walls. "Lower" relates to the ratio of wavelenght to boundary-dipole distance. Best compromize is an angle of 45° related to the next wall. Therefore, avoid having a dipole sub neatly orientated towards the listener for best results.
And now for the ESL57's need for sub? I would say none, or nearly none. You would not need much more bass potential along with an ESL57 (I also owned a pair of these). You basically need none, if you like to "simply" enjoy audio at home at reasonable levels. In this case, you will be fine with most of the available audio materials. But you will definitely not be able to show off a androgens-loaded TAIKO session in it's full fury dynamic to your stunned and jealous audio friends. No. Not even with a stacked pair of 4 ESL57s.
Actually, I listen to a sub-less pair of Quad63. I prefer the sublimely precise overall rendering to a belly-punching bass experience. If I really had the urge for more bass, then I would go, as mentionned above, for a pair of auxiliary bass ESL panels on both sides of my listening position. In any case I would avoid trying to match my ESLs with magnetodynamic subs. No more. "Match" comes closely before "mess" in my diytionary.
... And as for the Taiko-Session, I would better go to the concert hall. Even with a supersub, taiko may remain subsub.
Last edited:
While I tend to generally agree there is some issue with OB/dipole as well. The backwave will bounce against the wall behind the speaker, returns and merges with the front wave. At the wrong time too. I have some nice transmissionline bass speakers and they have a port which emits delayed backwave. But it is tightly controlled by the length of the line. There is also damping of specific frequency which works against the "transmissionline suckout". With OB/dipole it depends on placement and damping is generally poor or none. You also need lots of space with OB to allow deep bass. To close for wall means very little deep bass. In that case any bass (closed, reflex etc) is better than no bass.
You have missed my point; take for example a large ported enclosure. This is simply a poorly damped Helmholz resonator which is able to store energy, and when signal to the driver ceases, this energy has to be dissipated which takes far longer to happen than the settling time of the cone.
Take a look at the step response of a sealed vs. ported enclosure which illustrates this nicely - the aforementioned large ported enclosure can often be ringing more that 200ms after the signal has ceased.
As Jan Didden wrote. There is no need of the same transient response on a woofer as an electrostatic speaker since the play different frequencies.
2x10" woofers is a good suggestion since cone breakup is generally low, and the surface area is quite sufficient.
Breakup distortion is worse than a 2nd harmonic component (which sounds good...)
And to repeat what many many people has already explained is that if the woofer has a frequency response to 1kHz and you crossover at 100Hz there is not even phase shifts to be worried about.
Listen! That is my advice. Disconnect ESL57 or whatever you have and listen to the subs alone. Do you hear resonances? Breakup? Other bad noise? Whistle noise from pole went? Etc.. Generally more driver that share the load is better. All nasty distortion and breakup is highly depended on excursion.
25 years ago I built two towers with 6x15" cheapo (60 USD) woofers in each. They where tall and i suspended them with screws towards the ceiling. They stood rock solid... The sound was remarkable good! Even though they where dipole corrected the amplitude/excursion of each driver was quite low. Compared to a 6 times (360USD) more expensive driver you would never get that UMMPF.
As a sidenote... When i "burned them in" i connected all 12 drivers in series but reversed the polarity of every second speaker. So one was moving out and next one was moving in. That way they cancelled out each other. Then i plugged them in to the main grid 230 Volt 50Hz and let them run for 24 hours.
They survived!
Fs was adjusted down a couple of Hz. 😎
600W in total but only 50W on each driver... NEVER EVER TRY THIS YOURSELF!!
Main power is lethal!
And the room they where playing in was an industrial where no one without proper knowledge had access.
2x10" woofers is a good suggestion since cone breakup is generally low, and the surface area is quite sufficient.
Breakup distortion is worse than a 2nd harmonic component (which sounds good...)
And to repeat what many many people has already explained is that if the woofer has a frequency response to 1kHz and you crossover at 100Hz there is not even phase shifts to be worried about.
Listen! That is my advice. Disconnect ESL57 or whatever you have and listen to the subs alone. Do you hear resonances? Breakup? Other bad noise? Whistle noise from pole went? Etc.. Generally more driver that share the load is better. All nasty distortion and breakup is highly depended on excursion.
25 years ago I built two towers with 6x15" cheapo (60 USD) woofers in each. They where tall and i suspended them with screws towards the ceiling. They stood rock solid... The sound was remarkable good! Even though they where dipole corrected the amplitude/excursion of each driver was quite low. Compared to a 6 times (360USD) more expensive driver you would never get that UMMPF.
As a sidenote... When i "burned them in" i connected all 12 drivers in series but reversed the polarity of every second speaker. So one was moving out and next one was moving in. That way they cancelled out each other. Then i plugged them in to the main grid 230 Volt 50Hz and let them run for 24 hours.
They survived!
Fs was adjusted down a couple of Hz. 😎
600W in total but only 50W on each driver... NEVER EVER TRY THIS YOURSELF!!
Main power is lethal!
And the room they where playing in was an industrial where no one without proper knowledge had access.
At the risk of upsetting anyone I have owned 57's for 30 years (tried 63's but found them too 'sterile') and I really don't get this obsession with bass. I have recordings I won't listen to simply because the studio over-ramped the bass and it kind of rules out any popular music produced since the millenium, but since I replaced the resistors and capacitors in my input transformers the bass is noticeably louder on the one I've done, and in proper condition I can't see that any more bass might enhance the experience of the music I listen to at the volumes I choose with my 15watt valve amplifier.
I guess, at the end of the day, it depends what you're using the 57's for. I realised in the 1990's that they're not a speaker for heavy rock or indeed anything that wants to be really loud, but luckily my music tastes, whilst eclectic, are largely jazz, female vocal & prog rock of the Pink Floyd variety (Dave Gilmour loved Quad by the way).
I suspect there's always been a difference in 'preference' between us Brits and our cousins from the USA, much like cars (perhaps automobiles?), size and raw power were not always the name of the game in the UK and a similar 'preference difference' seems to have always been evident with our choices of high-end hi-fi equipment. But if you own a pair of ESL 57's I'm probably preaching to the converted anyway.
I started listening to and loving my 57's with an unrestored 33/303 combo in the early 1990's - it sounded great and is still working well today. I then treated myself to a tired II / 22 valve combo, which when refurbished changed my listening experience completely. Listening to the 57's with the combination of valve amps they were originally designed to compliment (but only when all the components are working to original tolerances) is a magical experience with the right music (Miles Davies or Pink Floyd's Wish you were here on vinyl) - you won't want more bass!
I guess, at the end of the day, it depends what you're using the 57's for. I realised in the 1990's that they're not a speaker for heavy rock or indeed anything that wants to be really loud, but luckily my music tastes, whilst eclectic, are largely jazz, female vocal & prog rock of the Pink Floyd variety (Dave Gilmour loved Quad by the way).
I suspect there's always been a difference in 'preference' between us Brits and our cousins from the USA, much like cars (perhaps automobiles?), size and raw power were not always the name of the game in the UK and a similar 'preference difference' seems to have always been evident with our choices of high-end hi-fi equipment. But if you own a pair of ESL 57's I'm probably preaching to the converted anyway.
I started listening to and loving my 57's with an unrestored 33/303 combo in the early 1990's - it sounded great and is still working well today. I then treated myself to a tired II / 22 valve combo, which when refurbished changed my listening experience completely. Listening to the 57's with the combination of valve amps they were originally designed to compliment (but only when all the components are working to original tolerances) is a magical experience with the right music (Miles Davies or Pink Floyd's Wish you were here on vinyl) - you won't want more bass!
Mark Levinson thought dual 24” (610mm) woofers were appropriate in the HQD system.
(One seen here…)
(One seen here…)
This is true at the resonance frequency. But at the crossover frequency the ported box will act as a sealed box.
Dipole is better anyway... since what we want is less resonances from the room. And a dipole is only creating standing waves in one direction. While a box speaker is pressurizing the whole room.
If your room is big enough so that you can crate a space behind your dipole woofer >4 meters. And you got the money... then go dipole!