Hi I am planning to build some electrostatic bass midbass panels which can operate from ~80hz to 1000hz. My goal is to emulate or improve upon the sound of Quad 57 bass midbass panels and to have some drivers which I can test against other types of drivers in various configurations in a modular capacity.
A couple notes and goals: I have owned various electrostats and panels including Quad 57s, Acoustat Xs with servo amps, KLH 9s, and extensively auditioned Soundlabs, other Quads, Apogee ribbons, etc. The Quad 57s which were restored by Wayne Piquet, had the best bass and mids of any speaker I have heard.
I have been experimenting with low mass 15" cones for this same application and I figure even a relatively thick mylar based electrostatic driver will be lower mass and more performative than any 15" woofer.
My goal is to basically replicate Peter Walker and Wayne Piquet's formula while optimizing for maximum SPL and headroom possible from an electrostatic panel. What thickness of mylar or similar, high or low modulus, conductive coating material, and stator gap/xmax is optimal for a bass midbass electrostatic driver of this nature?
A couple notes and goals: I have owned various electrostats and panels including Quad 57s, Acoustat Xs with servo amps, KLH 9s, and extensively auditioned Soundlabs, other Quads, Apogee ribbons, etc. The Quad 57s which were restored by Wayne Piquet, had the best bass and mids of any speaker I have heard.
I have been experimenting with low mass 15" cones for this same application and I figure even a relatively thick mylar based electrostatic driver will be lower mass and more performative than any 15" woofer.
My goal is to basically replicate Peter Walker and Wayne Piquet's formula while optimizing for maximum SPL and headroom possible from an electrostatic panel. What thickness of mylar or similar, high or low modulus, conductive coating material, and stator gap/xmax is optimal for a bass midbass electrostatic driver of this nature?
Last edited:
Hi,
A ESLs field of reign begins >>100Hz. Bass is not their chocolate side.
Electrostatic bass always suffers from small excursion capability, low force-per-area, and the almost non-existance of damping.
This asks for larger membrane areas required than for dynamic drivers and results in high-Q resonce/One-note-bass.
They are only to find in open baffle style speakers, as cabinets would be either unsuitable or very large.
Also open baffle designs suffer from serious phase cancellation.
On the other hand, the special dipolar distribution character and room interaction is what defines the sound of these kind of basses.
Imho and experience a dynamic dipolar bass is superior to an ESL bass in almost any aspect, especially dynamics, tonality and size.
Mass doesn't play a negative role in this frequency range but a positive.
The low mass makes an open baffle style mounting rather a neccessity than a choice, and almost rules out any other type of possibly more desirable cabinet style.
What makes the 57's mids so good imho is the missing of the all to often to find lower-mid suckout (due to phase cancellation) that plagues especially panels with 'transparent optics'.
Using a good wideband panel crossed over not too low and adding a dynamic dipole bass seems to be a more promising concept.
jauu
Calvin
A ESLs field of reign begins >>100Hz. Bass is not their chocolate side.
Electrostatic bass always suffers from small excursion capability, low force-per-area, and the almost non-existance of damping.
This asks for larger membrane areas required than for dynamic drivers and results in high-Q resonce/One-note-bass.
They are only to find in open baffle style speakers, as cabinets would be either unsuitable or very large.
Also open baffle designs suffer from serious phase cancellation.
On the other hand, the special dipolar distribution character and room interaction is what defines the sound of these kind of basses.
Imho and experience a dynamic dipolar bass is superior to an ESL bass in almost any aspect, especially dynamics, tonality and size.
Yes, for lower mass, but No for performance.
Mass doesn't play a negative role in this frequency range but a positive.
The low mass makes an open baffle style mounting rather a neccessity than a choice, and almost rules out any other type of possibly more desirable cabinet style.
What makes the 57's mids so good imho is the missing of the all to often to find lower-mid suckout (due to phase cancellation) that plagues especially panels with 'transparent optics'.
Using a good wideband panel crossed over not too low and adding a dynamic dipole bass seems to be a more promising concept.
jauu
Calvin
Being someone with a long history with OB bass and ESL panels, I couldn't agree more with the wise words of Calvin.
An ESL panel has a fundamental resonance with a typical Q of 5-15. Depending on panel size and tension it will usually be in the 40-100 hz range. You simply can't expect good bass reproduction from such a system. Better don't use the panel anywhere near it's resonance but cross over to a dynamic OB woofer two octaves above it. That will also reduce excursion, increase max spl dramatically and allow for lower stepup ratio in the transformer, making things a lot easier overall.
Such a panel used with a lower frequency limit of 150-200 hz is very well capable of playing up all the way to 20khz. What is your motivation to use it only up to 3khz, requiring an additional cross over point?
An ESL panel has a fundamental resonance with a typical Q of 5-15. Depending on panel size and tension it will usually be in the 40-100 hz range. You simply can't expect good bass reproduction from such a system. Better don't use the panel anywhere near it's resonance but cross over to a dynamic OB woofer two octaves above it. That will also reduce excursion, increase max spl dramatically and allow for lower stepup ratio in the transformer, making things a lot easier overall.
Such a panel used with a lower frequency limit of 150-200 hz is very well capable of playing up all the way to 20khz. What is your motivation to use it only up to 3khz, requiring an additional cross over point?
Last edited:
Just a curious question: What is the max excursion of the membrane of an ESL like Quad 57 or 63 ?
That's in the same league as Magnepan excursion-wise. They are not ESLs, but the max output in the bass for a given diaphragm size would be comparable. And sadly quite useless below 100 Hz.....
I agree with the general limitations in bass as mentioned above. Quads have a fair amount of damping wads though. That lowers Q. I remember reading somewhere that Q is around 2. Dont expect it to be as tight as a well designed dynamic system. Everything is relative. I can enjoy an audiostatic + sub. I get used to the bass. Until putting my dynamic TL bass aside......
Thickness of mylar is not critical. Tensioning it properly is most important. You can increase spacing as audiostatic did, but even in their models using double the amount of step up transformers, efficiency is low. This requires high output amps. Quad was designed for lower output tube amps.
Thickness of mylar is not critical. Tensioning it properly is most important. You can increase spacing as audiostatic did, but even in their models using double the amount of step up transformers, efficiency is low. This requires high output amps. Quad was designed for lower output tube amps.
Q is 2.5 for the 63 according to the old publications by PW. For the 57 I don't know, probably much higher looking at the construction.
Imho the heavy damping of the 63 comes at a high price and has a profound negative impact on overall clarity and reproduction of detail. I consider the 63 a good speaker but the sound quality is by no means even close to what ESL is capable of. All of this can be avoided by not trying to use the ESL for deep bass.
Imho the heavy damping of the 63 comes at a high price and has a profound negative impact on overall clarity and reproduction of detail. I consider the 63 a good speaker but the sound quality is by no means even close to what ESL is capable of. All of this can be avoided by not trying to use the ESL for deep bass.
I could not disagree more @maudio
Where in the ESL-63 is there a "heavy damping"? You mean the mesh on the rear stator?
The air gap in the ESL-57 bass panel is 2,4 mm just like the 63. Because of the curved mounting of the 57 panels however, there may be other limitations there.
The distance to the stator is much bigger in the 57 bass panels, you have to add 2 mm of PVC on each side. Hence the higher bias voltage.
The air gap in the high panel of the ESL-57 is only 0,5 mm, with also a much lower bias voltage. Same goes for stator distance here, just 0,5 mm.
The ESL-57 is flawed in many ways and the 63 out performs it on all counts.
I invite all Dutch here to come and listen what a Quad ESL-63 can do in bass with the right coating applied. Scary from time to time. Just PM me.
Less high in the 63 compared to the 57, I don't think so. It beams much less, yes that is correct, so the experience it quite different.
Where in the ESL-63 is there a "heavy damping"? You mean the mesh on the rear stator?
The air gap in the ESL-57 bass panel is 2,4 mm just like the 63. Because of the curved mounting of the 57 panels however, there may be other limitations there.
The distance to the stator is much bigger in the 57 bass panels, you have to add 2 mm of PVC on each side. Hence the higher bias voltage.
The air gap in the high panel of the ESL-57 is only 0,5 mm, with also a much lower bias voltage. Same goes for stator distance here, just 0,5 mm.
The ESL-57 is flawed in many ways and the 63 out performs it on all counts.
I invite all Dutch here to come and listen what a Quad ESL-63 can do in bass with the right coating applied. Scary from time to time. Just PM me.
Less high in the 63 compared to the 57, I don't think so. It beams much less, yes that is correct, so the experience it quite different.
Getting good bass out of a planar is tricky. The high Q is dealt with by sectioning the bass panel to distribute resonances. Magnepan uses tuning "dots". The left and right speakers are tuned differently. This results in multiple resonances that overlap in a prescribed manner.
Ed
Ed
I am interested in designing a driver that can be integrated in a midfield speaker and with a constant directivity horn and compression driver. The system is active and features independent crossover and amplifier channels for each section. I find the advantages of a good compression driver in this implementation significant. No other driver I know of can push high frequencies over distance like a compression driver thus mitigating hf rolloff that otherwise occurs due to acoustical impedance. That said I also intend to experiment with es panels for hf reproduction.
Maudio's observation mimics mine. 63s and later iterations have always sounded muffled and lifeless compared with 57s in my experience. I presume this is related to the pseudo coaxial multi panel configuration of the 57s compared with the full range array or single panel designs of the 63s and later iterations. While perhaps less elegant than a full range single panel speaker, 57s have always sounds better to me.
Last edited:
What can I say? Maybe you have never heard a real good rebuild 63.
57 lacks an octave of bass, has quite some irritating panel resonance, only 15 watt (very small step-up transformer), extreme beaming, etc. etc.
I have both and listen to Quad ESL for 40+ years now. It take the 63 over the 57 every day and twice on Sundays. YMMV
In brief summary I have tested various 15s 12s and smaller crossed over at 1000hz over the past month or so -- 15G40, 15P8FEN, 15MI100, 15W500, GPA515-8C, 15PR400, SB Bianco, 285-2000II, etc to name a few. The finding in every instance is that the lower mass driver is always more performative than the rest. The lowest mass I have found is the 15W500. I am content with these and the 15MI100. I am hoping to come up with an alternative to test against these for midrange purposes. Another variety of midrange of interest is mid compression drivers like the AXI2050 which sounded a bit dull in my first testing of it, DCM50, HMF200, Monacor, 850, 950 etc. I designed my speakers to be modular to be able to test various configurations as new ideas emerge or new drivers become available.
I prefer a properly implemented open baffle to any closed or ported cabinet, subs excluded. The 15MI100 renders far more detail than the 15G40 both in bass and mids for example. No cabinet will change this. I presume 2000w rather than the 500w amps I'm using would benefit the 15G40 more than the 15MI100 though for my purposes that does not apply.
Last edited:
Lots is a matter of taste. The 57 may be flawed but the 63 isnt perfect either. Lots of electronics, coils made of miles of wire, a step like approach of what should be a smoothly curved wavefront. Some people may be densitive to electronic flaws, others may be irritated by resonance. Pick your flaws. Personally I like the mid/highs of audiostatic type speakers and find the quad 63 midrange heavy, but that is just me.
I am highly irritated by resonances. Those who aren't are tone deaf as far as I'm concerned and enjoy music for the thump factor, which is fine. While ludicrous I am fond of the Magico M6s as they are very non resonant. I think there are simpler more elegant and smarter designs one could implement to achieve the same conclusion however. I can't say I have ever attempted to listen for electronic flaws such as the ones you describe. I agree insertion loss and group delay are real though in my limited understanding of the subject, I presume the issue applies to low level signals like that out of a low impedance moving coil cartridge more so than an amplified signal going into speakers. That said impedance load, presented by all of the factors you mention and others is the critical factor at every point.
I agree the 63s are too mid heavy. They sound sluggish. Have you had an opportunity to test Audiostatic speakers against the few other niche electrostatic speakers or planar ribbon speakers available today -- ER Audio, Sanders, etc? I have some ER Audio Minipanels and associated electronics which I will get to testing sooner than later.
I agree the 63s are too mid heavy. They sound sluggish. Have you had an opportunity to test Audiostatic speakers against the few other niche electrostatic speakers or planar ribbon speakers available today -- ER Audio, Sanders, etc? I have some ER Audio Minipanels and associated electronics which I will get to testing sooner than later.
You also need a dissipative coating in the range of 10^8 or 10^9
It is doable... Im looking for factory treated film thickness is of less importance 12-25um would work fine.
If someone finds a good source i know that many DIYers would buy. Maybe a group buy?
It is doable... Im looking for factory treated film thickness is of less importance 12-25um would work fine.
If someone finds a good source i know that many DIYers would buy. Maybe a group buy?
Years ago, I was in a factory that treats film (they were mostly doing nichrome coating). I'd guess their baled scrap pile and scrap room of rolls at that time each contained more film than has been used by the entirety of DIY electrostatic speaker building. I'm not saying it's impossible, but getting the attention of a typical film converter is not going to be easy, unless they also happen to be an audio enthusiast and/or they already run something you can repurpose.
Last edited: