Hi,
if You can solve the problems of creating the complex 3D shape with lowest dimensional tolerance I'd say it'd be possible.
But other than for placing the device snuggly into a rooms corner I don't see any positive for such a shape.
As a corner is the worst place to place a planar driver, I guess no one has wasted more than a second to that idea.
A very crude Idea. I think that pretty much sums it up
jauu
Calvin
jauu
Calvin
if You can solve the problems of creating the complex 3D shape with lowest dimensional tolerance I'd say it'd be possible.
But other than for placing the device snuggly into a rooms corner I don't see any positive for such a shape.
As a corner is the worst place to place a planar driver, I guess no one has wasted more than a second to that idea.
A very crude Idea. I think that pretty much sums it up
jauu
Calvin
jauu
Calvin
A perfect 1/8 sphere in a corner acts like point source in 4 pi space. What's crude about that? Nothing. My question is if it can be built, 'cause I'm not familiar with this area of DIY.But other than for placing the device snuggly into a rooms corner I don't see any positive for such a shape.
As a corner is the worst place to place a planar driver, I guess no one has wasted more than a second to that idea.
A very crude Idea. I think that pretty much sums it up
Hi,
due to the ESL membrane beeing acoustically transparent and the impossibility to null out the early reflections, the sound will be crappy.
But then ... close to invisible and crappy sound ... maybe some Bose fans could like it.
Second, the idea of a point source as an ideal only applies to theoretical world thinking.
In a practical world and under real room conditions, it is neither doable nor something to wish for.
jauu
Calvin
due to the ESL membrane beeing acoustically transparent and the impossibility to null out the early reflections, the sound will be crappy.
But then ... close to invisible and crappy sound ... maybe some Bose fans could like it.
Second, the idea of a point source as an ideal only applies to theoretical world thinking.
In a practical world and under real room conditions, it is neither doable nor something to wish for.
jauu
Calvin
Oh, I see. I think that can be dealt with enough, might need to have an opening behind the driver, but doesn't really matter if it can't be built effectively.
I wish I had the admirable restraint Calvin shows in his replies to dumptruck.
The Dayton-Wright ESLs, where each panel was fashioned from 8 cells, formed a section of a sphere. Maybe like 1/20, dunno. Wright argued for the benefit of a point source... maybe just so he could claim a patent for commercial-exploitation reasons.
For sure, benefits to dispersion by splaying the cells a bit even if the theory is, well, theoretical.
I recall some chit-chat about absorbing rear waves from ESLs. I fashioned some meter-square sheets of fiberglass board to sit a few inches behind my meter-square speakers but didn't like the results. The consensus was that is harmed the sound. The benefits of dipole sound are hard to identify theoretically but are apparent to listeners' ears.
Ben
The Dayton-Wright ESLs, where each panel was fashioned from 8 cells, formed a section of a sphere. Maybe like 1/20, dunno. Wright argued for the benefit of a point source... maybe just so he could claim a patent for commercial-exploitation reasons.
For sure, benefits to dispersion by splaying the cells a bit even if the theory is, well, theoretical.
I recall some chit-chat about absorbing rear waves from ESLs. I fashioned some meter-square sheets of fiberglass board to sit a few inches behind my meter-square speakers but didn't like the results. The consensus was that is harmed the sound. The benefits of dipole sound are hard to identify theoretically but are apparent to listeners' ears.
Ben
.... since you ask, there is nothing offensive about your question. Very good issue to raise from a guy who tells us he knows nothing about "this [???] area of DIY" - might even be a good question from a more speculative ESL builder (which is why I answered with a wee fact albeit a tangent).
But your reply, and here I paraphrase, "Shut up, I only want you to answer my question...."....
Ben
Electrostatic driver construction is the area I meant.
Yeah, I guess so, sorry to be rude. Kinda seemed like I was the one being insulted, with no real response to my question, but instead:
"A very crude Idea"
"crappy sound ... maybe some Bose fans"
"no one has wasted more than a second to that idea"
Especially given that I didn't even describe any implementation.
It probably wouldn't be hard to build, Just very time consuming if you used many little cells.
But, I can tell you that have done much experimenting in trying to figure out how to dampen the back wave of my small ESL's.
If anything (even the fuzzy side of fiberglass insulation batte) gets within less than a foot of the diaphragm, you lose all of the detail and high end finesse that ESL's are known for.
Every apparatus that I could get to even remotely work without degrading the sound was very large and was no different than just using proper room wall treatments behind them to get the effect I was after, with the exception of using baffles to effectively lengthen the front to back path only slightly increasing my low end extension by about 3db.
Again increasing the size of a rather small setup.
I tried taking three and four of my panels and setting them up side by side.
At first I had voltage hogging issues with the panels as they were all tied in parallel due to them all not being exactly perfect and identical.
After solving that issue I can't imagine one trying to assemble 20 or 30 or even 40 or more little cells and expecting them to be all exactly equal no matter how hard you tried while making them.
Although it is not impossible as my last pair ( the Black Powder coated ones that I burned up) were very well matched and exactly equal within each other.
Once I got the multiple panels running the most of an angle I could get between each adjacent panel without hearing any lobes and dips in the Horizontal plane was just a few degrees, about 3-5 degree or so and maybe as high as 7 degrees but that was pushing it.
That setup only benefited from the increased surface area and width and little if any to the already fairly wide horizontal dispersion of the small width panels (3" to 4.5").
Therefore that is an increase of 10 degrees for the setup for three panels as I get about 30 degrees of dispersion with one panel.
This would be calling for a very large array if one were trying to build any part of a sphere.
I did try using a piece of glass directly behind the ESL at various angle in order to try to widen the dispersion and it did seem to work okay as long as I kept its distance from the diaphragm less than about 1.25" at one end to avoid any comb filtering effects due to the delay of the reflected rear wave mixing with the front wave.
But, After a while I removed the reflector and it just had a sounded better without it, but the difference was not huge.
I may explore this again later sometime and do some real tests as it is a take on the Beam splitter idea and it does work okay and i would like to see if maybe other shapes will work better as well.
That test was done with a mono setup so there was no way for me to tell if there was any benefits to using such a configuration.
As mentioned I don't think that you will benefit from such a setup in a small room as it will just increase the amount of early reflection with many more angles added to the final listening area and will smear the overall stereo image.
I have that problem now with my Current stack of speakers.
Where the Full range cabinets that are on top are more into the corners and without any wall treatments left becomes right and vice versa for the higher frequency's by the time the sound reaches my listening area 14' away and that is using some plain ole' dynamic drivers.
Bass frequency's accumulate in the corners and this too may cause issue the the ESL's diaphragms.
I have one set of my woofers (4x10's) facing the corners of my room and this does work quite nicely for the low frequency's but I have my doubts that it would work well for the high's using a fairly large array, this is much different than using a small dome as a point source placed in a corner.
Without using any segmentation the array will become very directional at the higher frequency's just as a planar does as a function of its size to the frequency that it is trying to produce.
FWIW
jer
But, I can tell you that have done much experimenting in trying to figure out how to dampen the back wave of my small ESL's.
If anything (even the fuzzy side of fiberglass insulation batte) gets within less than a foot of the diaphragm, you lose all of the detail and high end finesse that ESL's are known for.
Every apparatus that I could get to even remotely work without degrading the sound was very large and was no different than just using proper room wall treatments behind them to get the effect I was after, with the exception of using baffles to effectively lengthen the front to back path only slightly increasing my low end extension by about 3db.
Again increasing the size of a rather small setup.
I tried taking three and four of my panels and setting them up side by side.
At first I had voltage hogging issues with the panels as they were all tied in parallel due to them all not being exactly perfect and identical.
After solving that issue I can't imagine one trying to assemble 20 or 30 or even 40 or more little cells and expecting them to be all exactly equal no matter how hard you tried while making them.
Although it is not impossible as my last pair ( the Black Powder coated ones that I burned up) were very well matched and exactly equal within each other.
Once I got the multiple panels running the most of an angle I could get between each adjacent panel without hearing any lobes and dips in the Horizontal plane was just a few degrees, about 3-5 degree or so and maybe as high as 7 degrees but that was pushing it.
That setup only benefited from the increased surface area and width and little if any to the already fairly wide horizontal dispersion of the small width panels (3" to 4.5").
Therefore that is an increase of 10 degrees for the setup for three panels as I get about 30 degrees of dispersion with one panel.
This would be calling for a very large array if one were trying to build any part of a sphere.
I did try using a piece of glass directly behind the ESL at various angle in order to try to widen the dispersion and it did seem to work okay as long as I kept its distance from the diaphragm less than about 1.25" at one end to avoid any comb filtering effects due to the delay of the reflected rear wave mixing with the front wave.
But, After a while I removed the reflector and it just had a sounded better without it, but the difference was not huge.
I may explore this again later sometime and do some real tests as it is a take on the Beam splitter idea and it does work okay and i would like to see if maybe other shapes will work better as well.
That test was done with a mono setup so there was no way for me to tell if there was any benefits to using such a configuration.
As mentioned I don't think that you will benefit from such a setup in a small room as it will just increase the amount of early reflection with many more angles added to the final listening area and will smear the overall stereo image.
I have that problem now with my Current stack of speakers.
Where the Full range cabinets that are on top are more into the corners and without any wall treatments left becomes right and vice versa for the higher frequency's by the time the sound reaches my listening area 14' away and that is using some plain ole' dynamic drivers.
Bass frequency's accumulate in the corners and this too may cause issue the the ESL's diaphragms.
I have one set of my woofers (4x10's) facing the corners of my room and this does work quite nicely for the low frequency's but I have my doubts that it would work well for the high's using a fairly large array, this is much different than using a small dome as a point source placed in a corner.
Without using any segmentation the array will become very directional at the higher frequency's just as a planar does as a function of its size to the frequency that it is trying to produce.
FWIW
jer
Thanks for describing your experience. I guess I should clarify what I mean by "1/8" sphere (Calvin got it, I think). I mean this piece of sphere surface, nor more no less:
Looking for anything from impractically large to impractically small for hifi usefulness. This would be for experimental purposes only (as in empty test rooms and data, not fiddling around listening to weird speakers), where coming as close as possible to a real pulsating 1/8-sphere surface is important, but bandwidth, SPL, and many types of distortion are not.
Looking for anything from impractically large to impractically small for hifi usefulness. This would be for experimental purposes only (as in empty test rooms and data, not fiddling around listening to weird speakers), where coming as close as possible to a real pulsating 1/8-sphere surface is important, but bandwidth, SPL, and many types of distortion are not.
Last edited:
The idea of curving a panel in both vertical & horizontal planes isn't new. It has been done using multiple small square panels to segment the panel's area allowing it to be curved in both directions, albeit not using one continuous stator... perhaps best example is the Crown ES-224, but the WAMM, and many others have done the same.
Here's a picture of my ES-224s... maybe not as elegant as a single panel, but a practical, and effective solution none the less.
Here's a picture of my ES-224s... maybe not as elegant as a single panel, but a practical, and effective solution none the less.
Last edited:
Yes, I know exactly what an 1/8 sphere means.
I am not trying to detour you from trying your experiment as I am always up for something new.
I was just trying to state what would be involved in trying to create one using ESL technologies.
It would be a very large array to start with.
I have thought about doing something similar myself for use in outdoor venues.
You would definitely have to create a highly efficient panel with lots of excursion to make it worthwhile to get down to at least the 200hz to 300hz range, the start of what I like to call the critical midrange, as most call it mid bass.
jer
I am not trying to detour you from trying your experiment as I am always up for something new.
I was just trying to state what would be involved in trying to create one using ESL technologies.
It would be a very large array to start with.
I have thought about doing something similar myself for use in outdoor venues.
You would definitely have to create a highly efficient panel with lots of excursion to make it worthwhile to get down to at least the 200hz to 300hz range, the start of what I like to call the critical midrange, as most call it mid bass.
jer
This is the kind of answer I was looking for. Why do you think that? What about just a few pieces that are actually curved? The problem apart from creating the curve is even tension, correct?
To be useful for real hifi, yeah, it would have to get pretty low, aside from some vague possibility of very creative in-wall integration with other driver. This question is NOT to that end, though; I just want to measure what happens if you actually get this type of radiation in a room. My main plan is actually just making a little trihedral waveguide for a dome tweeter that meets it at a 1/8"-sphere cross section, but looking for other possibilities (plus, they don't tend to be half-spheres, or any particular part of a sphere, right? That's kind of a problem).
Last edited:
The main reason why a large array would be required is do to the nature of the mechanics of the operation of the diaphragm it self.
Sure you could build a cell that is only 1" square but you would be limited in how loud you can go at the lowest frequency's due to excursion reasons that are limited because of the mechanics of the device.
Next the diaphragm doesn't always act like a perfect piston it typically has its most displacement in the center and the displacement gets less as you get closer to the point were it is mounted.
I have read many discussions about this in these very threads purposing ways to possibly correct this.
But, I haven't seen in data on any such experiments except for one in some magazine done on a microscale chip level for ESL transducers for stuff like cell phones and mems devices and such.
Something like a dome is rigid and will move uniformly in one direction as a perfect piston.
You need a diaphragm that will move equally in all directions at the same time.
The only way to get even close to this using a very thin material that has no rigid properties.
And, to support it in segments and the smaller the segments are the more uniform the action will be.
Still you need a diaphragm that will move equally in all directions at the same time.
The only thing that I can think of to get very close if not perfect to such an action is to use a Plasma driver.
This is why plasma speakers are so highly sought after and sound so good if implemented properly.
The poor examples I have seen on youtube are because of sadly designed driving electronics.
Due to the required power to create Plasma in large displacement it is only feasible for mid to high frequency's.
Imagine if you could control the amount of power that it takes to produce the displacement to make low rumble from a nice crack of lighting!!
The displacement of Plasma is linear to the amount of power applied and the presure of the displacement is uniform in all directions as well (hence point source)if it is produced off the end of a wire or needle.
If it is produced as a long arc then it has a uniform cylindrical wavefront emitted uniformly in all directions.
This is all of course assuming that you are emitting into a free space where there are no reflections.
Calculating what is happening in a room is not extremely hard to do, But it does take some time to understand what is happening, I know it took me a while to understand it.
When I had my recording studio going I would spend hours playing with my signal generator sweeping it up and down and walking around the room listening for the room Nodes (peaks and valleys).
It was a rather good sized room of 13.5' x 24' with lots of absorption for high frequency's .
In the room I have now (18' x 11') I have no sound treatment yet and I can hear frequency's completely disappear and re-appear with just a slight turn of my head!!
jer
Sure you could build a cell that is only 1" square but you would be limited in how loud you can go at the lowest frequency's due to excursion reasons that are limited because of the mechanics of the device.
Next the diaphragm doesn't always act like a perfect piston it typically has its most displacement in the center and the displacement gets less as you get closer to the point were it is mounted.
I have read many discussions about this in these very threads purposing ways to possibly correct this.
But, I haven't seen in data on any such experiments except for one in some magazine done on a microscale chip level for ESL transducers for stuff like cell phones and mems devices and such.
Something like a dome is rigid and will move uniformly in one direction as a perfect piston.
You need a diaphragm that will move equally in all directions at the same time.
The only way to get even close to this using a very thin material that has no rigid properties.
And, to support it in segments and the smaller the segments are the more uniform the action will be.
Still you need a diaphragm that will move equally in all directions at the same time.
The only thing that I can think of to get very close if not perfect to such an action is to use a Plasma driver.
This is why plasma speakers are so highly sought after and sound so good if implemented properly.
The poor examples I have seen on youtube are because of sadly designed driving electronics.
Due to the required power to create Plasma in large displacement it is only feasible for mid to high frequency's.
Imagine if you could control the amount of power that it takes to produce the displacement to make low rumble from a nice crack of lighting!!
The displacement of Plasma is linear to the amount of power applied and the presure of the displacement is uniform in all directions as well (hence point source)if it is produced off the end of a wire or needle.
If it is produced as a long arc then it has a uniform cylindrical wavefront emitted uniformly in all directions.
This is all of course assuming that you are emitting into a free space where there are no reflections.
Calculating what is happening in a room is not extremely hard to do, But it does take some time to understand what is happening, I know it took me a while to understand it.
When I had my recording studio going I would spend hours playing with my signal generator sweeping it up and down and walking around the room listening for the room Nodes (peaks and valleys).
It was a rather good sized room of 13.5' x 24' with lots of absorption for high frequency's .
In the room I have now (18' x 11') I have no sound treatment yet and I can hear frequency's completely disappear and re-appear with just a slight turn of my head!!
jer
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.