I have been designing rear loaded and compound horns for several years now and experimenting with different drivers and observing what makes a driver well suited to a horn enclosure. The conclution I have come to is that the most important qualities are a lightweight diaphram with excellent control and damping characteristics. Which is what brought me to electrostatics. The diaphram is nearly weightless, and stiffness isn't really even an issue because the stators control the entire surface evenly. I can't help but wonder what could be done with a small circular electrostatic say 3" - 6" in diameter in a compound horn enclosure. I guess the biggest thing I'm wondering right now is how much can a electrostatic diaphram move before it gets distorted and how would one measure or determine the T/S parameters or whatever the equivalent information of such a speaker. Anyone with any information or suggestions please pitch in. I welcome information from everyone, but please don't tell me it can't be done and enumberate all the reasons it's a stupid idea. I tend to get defensive and upset by naysayers and have been banned for a few days more than once for telling people to shut up. Please feel free to contribute any facts or ideas you might have but if your goal is to tell me I'm wasting my time, don't waste yours.
I been wanting to horn load my little panel too se how well it would work.
It has a 9.75" X 3.25" diagphram and gets extremely loud.
I run it at 5kv to 7kv bias with about 25kv p-p across the stators (maxed out).
So I have been trying figure out a horn design for it.
I don't think it would be pheasible for a wide panel but a small width one might work well.
I think it would be worth giving it a try. jer
It has a 9.75" X 3.25" diagphram and gets extremely loud.
I run it at 5kv to 7kv bias with about 25kv p-p across the stators (maxed out).
So I have been trying figure out a horn design for it.
I don't think it would be pheasible for a wide panel but a small width one might work well.
I think it would be worth giving it a try. jer
A horn loaded ESL is unlikely to work for reasons discussed towards the end of this thread I started.http://www.diyaudio.com/forums/plan...ostatic-amt.html?highlight=Electrostatic+AMT? Basically ES forces are too weak and the membrane motion simply stalls when loaded by anything other than free air. I wish it were otherwise, but the laws of physics are uncompromising.
Keith
Keith
Yes, I have been thinking about that myself.
As I have run it to issues when trying to just dampen the backwave with some fiberglass insulation and I lost a bit of detail on the high if I got to close with it as it was loading up the diagphram.
But that was in mono and might not be noticeable to much in stereo.
But if I can come up with a size and shape it would be very easy to try as I already have the drivers and driver system made.
I think that it might actualy work good for a midrange system,But I don't think it will be loud enough for PA use though.
Yet my little driver is quite powerful when running at maxed out voltages not like your average esl running at a 2kv bias.
jer
As I have run it to issues when trying to just dampen the backwave with some fiberglass insulation and I lost a bit of detail on the high if I got to close with it as it was loading up the diagphram.
But that was in mono and might not be noticeable to much in stereo.
But if I can come up with a size and shape it would be very easy to try as I already have the drivers and driver system made.
I think that it might actualy work good for a midrange system,But I don't think it will be loud enough for PA use though.
Yet my little driver is quite powerful when running at maxed out voltages not like your average esl running at a 2kv bias.
jer
Last edited:
It would be cool to see some experiments on this.
It could be done in several different ways I think.
Beveridge have the acoustic lens and an enclosure, Maybe one could do something similar modelled as a FLH?
It could be anything from a very narrow panel to a wide one?
A hornloaded line source sounds like fun.
Or you could keep the dipole characteristics and build it symmetrical front/back? I think I've actually seen some experiments on this but doing it without a chamber the size will get out of hand pretty quick if you widen the panel.
To bad if the laws of physica are working against us, it would've been interesting to see what could be done.
It could be done in several different ways I think.
Beveridge have the acoustic lens and an enclosure, Maybe one could do something similar modelled as a FLH?
It could be anything from a very narrow panel to a wide one?
A hornloaded line source sounds like fun.
Or you could keep the dipole characteristics and build it symmetrical front/back? I think I've actually seen some experiments on this but doing it without a chamber the size will get out of hand pretty quick if you widen the panel.
To bad if the laws of physica are working against us, it would've been interesting to see what could be done.
There is a way around the apparent inability of an ESL to drive a horn. The idea was published in an AES paper by Josef Merhaut.
AES E-Library Horn-Loaded Electrostatic Loudspeaker
Basically, a prisimatic waveguide is used to transfer the resistive impedance at the throat of the horn to a larger area ESL. The benefits are a purely resistive airload on the ESL diaphragm over the whole working range of the horn. The benefits are a very predictable flat response and no pronounced diaphragm resonance normally caused by the reactive air-mass loading on the diaphragm.
A few details were posted here when discussing diaphragm resonance:
http://www.diyaudio.com/forums/plan...m-resonance-change-hv-bias-3.html#post2392743
He also received a patent(US3590169) for the idea, although it doesn't contain all the technical information provided in the AES paper.
ELECTROSTATIC HORN LOUDSPEAKER - Google Patents
Last edited:
I was thinking of making a round panel as small as possible to achieve the desired frequency extension(80Hz-20Khz would be perfect). I like the idea of loading both sides of the panel but not symetricly but rather a front horn for high frequencies and a rear horn for low frequencies (crossed over somewhere between 200-500Hz).
The other thing I was thinking about was coupling them with a tube amp and instead of putting an output transformer on it to bring down the voltage and another transformer on the speaker to bring it back up again you could direct couple it to the speaker and even use a single ended amp and the tube bias could be the diaphram bias...maybe, just thinking out loud, er... in type.
The other thing I was thinking about was coupling them with a tube amp and instead of putting an output transformer on it to bring down the voltage and another transformer on the speaker to bring it back up again you could direct couple it to the speaker and even use a single ended amp and the tube bias could be the diaphram bias...maybe, just thinking out loud, er... in type.
Last edited:
Yes I was think of some sort of asymeterical design but don't know which method yet as that will take some experimentation.
Yes you can couple the panel directly to a tube amp as it has been done ,BUT,It takes some serious voltages to get to a very large SPL.
When I pushed my panel past its limit (to the point of ignition from an arc over through a mounting bolt) I was applying in the order of 25kv p-p across the stators.
With music the average person could not be in the same room and even I could not stay in the same room with a steady sine tone.
As I am quite used to it being loud.
Sadly my SPL meter does not work otherwise I would have a figure for you to post.
jer
Yes you can couple the panel directly to a tube amp as it has been done ,BUT,It takes some serious voltages to get to a very large SPL.
When I pushed my panel past its limit (to the point of ignition from an arc over through a mounting bolt) I was applying in the order of 25kv p-p across the stators.
With music the average person could not be in the same room and even I could not stay in the same room with a steady sine tone.
As I am quite used to it being loud.
Sadly my SPL meter does not work otherwise I would have a figure for you to post.
jer
I have actualy wittnessed clipping at th lower frequency's using a microphone and a scope.
From the diagphram hitting the stator.
You can get quite a bit of excursion.
I tried to measure it but it is very difficult to do so.
But since I have had slapping of the diagphram to the stator should give you some kind of idea.
I also had I slight bit of it on my 7.75" wide panels using a d/s of .070" to .080".
But using a d/s of around .1" to .11" it pretty much disappeared but I never got the chance to drive them hard properly.
As I haven't ran them since 2003 when I built them.
All I need to do is rebuild some new frames and bolt them back together.
Remember that this is at very low frequency's as I was running them fullrange.
If they are crossed over at 300hz to even 500hz like most do these problems most likely would not occur at all.
jer
From the diagphram hitting the stator.
You can get quite a bit of excursion.
I tried to measure it but it is very difficult to do so.
But since I have had slapping of the diagphram to the stator should give you some kind of idea.
I also had I slight bit of it on my 7.75" wide panels using a d/s of .070" to .080".
But using a d/s of around .1" to .11" it pretty much disappeared but I never got the chance to drive them hard properly.
As I haven't ran them since 2003 when I built them.
All I need to do is rebuild some new frames and bolt them back together.
Remember that this is at very low frequency's as I was running them fullrange.
If they are crossed over at 300hz to even 500hz like most do these problems most likely would not occur at all.
jer
I had done some static test and the forces are large enough to suck the diagphram compleatly to one stator if they are not balanced or if the diagphram tension is not high enough on either size.
I had this problem on the bigger ones until I got everthing just right.
Then they worked great once I did.
jer
I had this problem on the bigger ones until I got everthing just right.
Then they worked great once I did.
jer
I could be wrong but it seem that would be true of a large panel but perhaps not so much for a smaller one. I was considering something on the order of an 8"(20cm) diameter circular panel. If i could get 1mm of movement the bass extension to 80hz shouldn't be a problem, right? I guess I don't understand what determines coupling efficiency.
I am just wondering if it might widen the dispersion pattern and help cut down the beaming effect using a smaller width panel.
That it my only curiousity.
Like I said earlier I already have the driver and drive electronics and all I need is a shape and size.
As a simple horn can be easily constructed very cheaply for use at 300hz and/or above.
I am excluding driver parameters and going by air displacment only.
I am new to horn design so it is virgin grounds for me.
But it is an experiment that would not cost me anything to find out.
At best it would cut back on the low frequency dipole cancelations and hoping that the horn flare won't cause any diffractions in the wave front as it travels through the horn.
Won't know if it works unles you try it!
I already know that it is not pheasible with a wider diagphram.
I have seen a project where it was tried.
jer
That it my only curiousity.
Like I said earlier I already have the driver and drive electronics and all I need is a shape and size.
As a simple horn can be easily constructed very cheaply for use at 300hz and/or above.
I am excluding driver parameters and going by air displacment only.
I am new to horn design so it is virgin grounds for me.
But it is an experiment that would not cost me anything to find out.
At best it would cut back on the low frequency dipole cancelations and hoping that the horn flare won't cause any diffractions in the wave front as it travels through the horn.
Won't know if it works unles you try it!
I already know that it is not pheasible with a wider diagphram.
I have seen a project where it was tried.
jer
You can avoid this by incorperating a phase plug suspended in the center of the horn.
Hi,
No. The ratio of diaphragm mass to air load mass is independent of size.
1mm of movement ? You don't understand electrostatics at all it seems.
(Well at least their basic limitations, Quads with 1mm would kick ****.
Problem is maximum excursion determines efficiency, 1mm its very low.)
Electrostatics have very high coupling efficiency, diaphragm to air, it
is not the issue. The air load on a electrostatic diaphragm exceeds
its mass, not remotely the case for a moving coil driver, hence in
the latter case an acoustic transformer, i.e. a horn, helps.
rgds, sreten.
Think of a normal driver moving in air, the air load although crucial
is relatively small and efficiency is low. Then consider the same
driver in a much heavier gas, such that the gas dominates.
Horn loading works for the former, but not the latter.
Last edited:
Okay that was helpful info, but then it would seem to me that electrostatics should be very efficient but it seems that the opposite is what I've heard. Why is it that electrostatics are known for their almost problematic inefficiency? I guess my goal as as self-appointed audio engineer is to find the absolute best way to recreate an audio signal. I have been dedicating every spare minute of my time for the last 12 years to this very cause. There are many schools of thought all of which seem to have valid points and I try to give equal consideration to them all in a quest for the truth. from what I've read and experienced the ultimate path seems to stem from minimizing sources of distortion and compression. The best way to do this seems to be through ultimate efficiency and the path to ultimate efficiency seems to be minimizing weight and maximizing motor strength. I have so far been trying to achieve this through horn loading high Q speakers with light weight cones. Electrostatics should achieve these same goals but they seem to be famous for low efficiency. It seems like a paradox, the only reason I can think of for this is the losses due to the transformers necissary due to the high voltages necissary. Am I missing something? I'm sure I am, what is it?
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.