So i have contemplated the question of whether the supposedly-inherent limitations of ESL technology can be overcome and determined that they can.
Now i have a choice - either i try to build it myself, or somebody helps me ?
If i have to build it myself i will need some practical information on how to build ESLs - a cookbook ? i think somebody gave me a link to a book, but i don't remember where it is ? Help ?
Or if anybody wants to build the prototype super-ESL with me - that would be even better. Any takers ? I can't tell you what it is unless you are serious about building it. I actually have two designs depending on how ambitious you are. One is only slightly more difficult than regular ESL to build - and solves most of the problems. Another very complex to build but solves all problems.
Now i have a choice - either i try to build it myself, or somebody helps me ?
If i have to build it myself i will need some practical information on how to build ESLs - a cookbook ? i think somebody gave me a link to a book, but i don't remember where it is ? Help ?
Or if anybody wants to build the prototype super-ESL with me - that would be even better. Any takers ? I can't tell you what it is unless you are serious about building it. I actually have two designs depending on how ambitious you are. One is only slightly more difficult than regular ESL to build - and solves most of the problems. Another very complex to build but solves all problems.
we are all in the same boat here, what are your suggestions? I have posted my own, just to find out that every idea comes with a problem.
tried and tru is what i have been looking for, doing a test today on a theory, i wish you explain a little more information on your project, this stuff is not cheap, that is why i posted an alternate material. with that said, quote"I am shooting in the dark" maybe i will find my target. all i am trying to do is coat the panels, with something i have readily availabe, will know this afternoon.
tried and tru is what i have been looking for, doing a test today on a theory, i wish you explain a little more information on your project, this stuff is not cheap, that is why i posted an alternate material. with that said, quote"I am shooting in the dark" maybe i will find my target. all i am trying to do is coat the panels, with something i have readily availabe, will know this afternoon.
i specifically want to build a revolutionary product which i have invented.
can you explain what you mean by us being in the same boat ?
i have a hard time understanding the meaning of your post.
when i say i need a cookbook i don't mean that i will actually build a sample design. i mean i will take hints on construction methods and adopt them to my significantly different design.
Last edited:
Hi Borat,
Ok, here i am !
I offer you my over 20 years experience in building ESL. If you tell me what to build i will do it (as long its really something outperforming).
If you like to contact me: info@capaciti.de
Capaciti
Ok, here i am !
I offer you my over 20 years experience in building ESL. If you tell me what to build i will do it (as long its really something outperforming).
If you like to contact me: info@capaciti.de
Capaciti
ESL Cookbook
Hey Borat, I don't know for sure if this is the book someone else recommended, but I'm fairly sure it would be
BKAA20 - Electrostatic Loudspeaker Design Cookbook
And if anyone reads this and knows of a better book, please post, as (some) of the materials/methods are outdated.
Thanks, Steve
Hey Borat, I don't know for sure if this is the book someone else recommended, but I'm fairly sure it would be
BKAA20 - Electrostatic Loudspeaker Design Cookbook
And if anyone reads this and knows of a better book, please post, as (some) of the materials/methods are outdated.
Thanks, Steve
Last edited:
I always wondered about seriously pushing the dynamic limits of ESLs. Running large stator gaps and high voltages, using direct drive amplfier arrays with some kind of inverse square compensation for more linear high level LF drive.
Mavric,
I have deleted some of your posts. Please make your point clear. You are confusing the issue and not really helping.
Hi Andrew, I'm interested but any discussion should probably be done over in one of the older treads where the subject has been addressed. I wouldn't hold your breath for anyone to break out their new project in the works though. I would like to see the HV linear problem become more of a non-issue . A fairly new commercial OTL amp was mentioned awhile back, their site showed a second amp to be released with enough voltage to drive a 2.5mm stator to diaprahm spacing. Controlled and monitored with interface. I didn't e-mail them to ask if I should look forward to a review or even better, a dealer here in the US. Has anyone heard them in action or has Sound-Lab found the magic amp for their VHV stats and is ramping up production to truly be the King. Ha Ha
To Michael ( Capacity ):
The speaker that needs to be built essentially involves a panel which is unevenly energized. It will be energized fully at the center and partially at the edges.
The difference between the way the center of the panel and edges of the panel are energized will be in level only. There will not be any dedicated low-pass filtering for the edges and there will not be any time delay networks ( as in QUAD ESLs ).
Did you, perhaps, have something like this in mind yourself ?
Or would you consider this taking it in a new direction for you ?
The speaker that needs to be built essentially involves a panel which is unevenly energized. It will be energized fully at the center and partially at the edges.
The difference between the way the center of the panel and edges of the panel are energized will be in level only. There will not be any dedicated low-pass filtering for the edges and there will not be any time delay networks ( as in QUAD ESLs ).
Did you, perhaps, have something like this in mind yourself ?
Or would you consider this taking it in a new direction for you ?
Hi Borat,
this is a different approach compared to typical segmentation of flat panels, since there is no low-pass filtering for the outer areas.
let me ask two questions:
1. How will this design have comparable sensitivity to other panels, while a significant portion of the whole membrane area is limited in energy radiation ?
2. How will the spectral decay look like, while membrane areas will move in different level at same frequencies.
Capaciti
this is a different approach compared to typical segmentation of flat panels, since there is no low-pass filtering for the outer areas.
let me ask two questions:
1. How will this design have comparable sensitivity to other panels, while a significant portion of the whole membrane area is limited in energy radiation ?
2. How will the spectral decay look like, while membrane areas will move in different level at same frequencies.
Capaciti
Hi,
if I understand correctly, You suggest a stator with varying resistance, so that the classical discrete stepped segmentation becomes continous.
This has already been patented. The patent describes a stator that has resistive features. The resistance varies over the statorßs surface and in doing so the frequency range the membrane works in can be controlled in a way that high frequencies are gradually attenuated.
This technique doesn´t solve any more problem than classical discrete segmentation solves.
So, where´s the beef?
Apart from some diffuse words....which parameters could be influenced so positively by Your claimed-revolutionary new design?
If it is just the different segmentation I´d be so sorry, but then You´d missed the victory party by several dozen years. 🙄
jauu
Calvin
if I understand correctly, You suggest a stator with varying resistance, so that the classical discrete stepped segmentation becomes continous.
This has already been patented. The patent describes a stator that has resistive features. The resistance varies over the statorßs surface and in doing so the frequency range the membrane works in can be controlled in a way that high frequencies are gradually attenuated.
This technique doesn´t solve any more problem than classical discrete segmentation solves.
So, where´s the beef?
Apart from some diffuse words....which parameters could be influenced so positively by Your claimed-revolutionary new design?
If it is just the different segmentation I´d be so sorry, but then You´d missed the victory party by several dozen years. 🙄
jauu
Calvin
I can't say any more on the subject here that i already said.
I need to work some issues out with Capacity first.
been there, theorised about it, thought some more about it and didn't see any major benefit to in.
I also considered a variable stator to stator spacing but this would have it's own issues.
I have finally settled on a narrow vertically curved panel where the distance at all places on the panel is approximately the same distance from the listening position - give or take to stop too small a sweet spot - bearing in mind these will be used as part of an open baffle line array.
the idea of changing diaphram resistance either vertically or horizontally is a nice one but i would guess pretty hard to consistantly get accurate.
OK Capacity bailed out.
He thought the cost / performance ratio of my design wasn't good.
I will post a detailed description of the design later.
He thought the cost / performance ratio of my design wasn't good.
I will post a detailed description of the design later.
Last edited:
like what ?
PS: next time when you theorize write about it. because as far as i am concerned - it is still my idea 🙂
Last edited:
I haven't given up, just changed direction slightly through the use of alternate methods.
The one of the biggest issues facing electrostatics builders is the capability to produce low frequencies without resorting to enormous panels and or very high step up voltages which in turn leads to an increase in diaphram to stator spacing for both anti arcing and excursion puposes. This of course leads to the need for even higher voltages and round and round we go.
I thought I could possibly solve was the change of stator to diphram spacing across the width of the panel and with a variation in voltage to the stator - in the case a wired design. This would allow the outer wires to be gradually driven by a higher voltage and also with a lower crossover frequency. This of course would mean that the entire panel is driven but not selectively and what would occur is that the inner parts of the diaphram would hit the stator due to the higher voltage on the other wires.
It is widely accepted that in order to effectively meet the demands of both high (ish) efficiency and reasonable panel size we have to make sacrifices.
In my case i'm going with an electrostatic tweeter covering the range from around 1k up - narrow and tall to be precise 10cm wide 2m tall. This will be mated to an open baffle line array of 20 mids per side for duties down to around 70 or 80hz final crossover point has not been determined yet. This in turn will cross to either dual 20hz tapped horns or one tapped for cinema and a second standard horn to 30hz for bass duties.
Moneys tight, times are hard but ill get there, i have most of what i need including 2 boxes wth 64 midrange drives, 2 15 inch subs, a pair of very high voltage stepup transformers capable of handling 5kv each (not that ill need that much) and the rest is readily available as time allows.
I have considered curving the vertical array in order to minimise comb filtering but i dont this this should be a problem due to crossover frequencies. Also due to the relatively narrow aspect of the electrostatic panel and the fact that the baffle is cylindrical at both sides i should get decent off axis performance and less cancelation due to waves meeting out of phase.
Don't get me wrong i love electrostats in fact I have 3 pairs of stax electrostatic cans including the orignal omega II's - the finest sound i've ever heard.
The one of the biggest issues facing electrostatics builders is the capability to produce low frequencies without resorting to enormous panels and or very high step up voltages which in turn leads to an increase in diaphram to stator spacing for both anti arcing and excursion puposes. This of course leads to the need for even higher voltages and round and round we go.
I thought I could possibly solve was the change of stator to diphram spacing across the width of the panel and with a variation in voltage to the stator - in the case a wired design. This would allow the outer wires to be gradually driven by a higher voltage and also with a lower crossover frequency. This of course would mean that the entire panel is driven but not selectively and what would occur is that the inner parts of the diaphram would hit the stator due to the higher voltage on the other wires.
It is widely accepted that in order to effectively meet the demands of both high (ish) efficiency and reasonable panel size we have to make sacrifices.
In my case i'm going with an electrostatic tweeter covering the range from around 1k up - narrow and tall to be precise 10cm wide 2m tall. This will be mated to an open baffle line array of 20 mids per side for duties down to around 70 or 80hz final crossover point has not been determined yet. This in turn will cross to either dual 20hz tapped horns or one tapped for cinema and a second standard horn to 30hz for bass duties.
Moneys tight, times are hard but ill get there, i have most of what i need including 2 boxes wth 64 midrange drives, 2 15 inch subs, a pair of very high voltage stepup transformers capable of handling 5kv each (not that ill need that much) and the rest is readily available as time allows.
I have considered curving the vertical array in order to minimise comb filtering but i dont this this should be a problem due to crossover frequencies. Also due to the relatively narrow aspect of the electrostatic panel and the fact that the baffle is cylindrical at both sides i should get decent off axis performance and less cancelation due to waves meeting out of phase.
Don't get me wrong i love electrostats in fact I have 3 pairs of stax electrostatic cans including the orignal omega II's - the finest sound i've ever heard.
- Status
- Not open for further replies.