When I made my pair of little panels in post #9 they sounded great.
I too had them hooked to the TV for a signal and I had no woofer connected at the time.
Even with them setting at an angle leaning against cheery wood TV cabinet on the floor (much less than optimum placing as it was just like a big dresser drawer cabinet) the quality was outstanding and 100 times more than what you get out of the TV's built speaker system that wasn't bad at all to begin with.
If you have never tried it then you won't know or will ever know, I got great results!!!
This was one of the reasons I have spent the last few years focusing on smaller panels than I have just duplicating a large panel design.
Although the results was so great that it did become annoying at times due to the poor acoustic's in your typical TV news room!
You could actually hear the rustling of the workers on the stage and behind the camera's.
But movies sounded great!!!
As suggested before I would use a open backed design if they have to be enclosed with a Flat ESL panel and not a curved one.
You could make a cover that is curved to go over them for aesthetic's.
Building a curve shaped panel comes with its own construction issues to begin with.
I have contemplated small cylindrical drivers in the past with radius of 2 to 6 inches and gave up on the idea because of the difficulty to apply the diaphragm evenly and uniformly around the frame while having enough strength for it not too collapse.
However you could build one with a slight curve rather easily with a 1/4" to maybe as high as 1/2" curvature depth for a 3" wide panel using my window screen stator method as it is quite flexible without cracking the stator coating.
For a panel of this size the curvature may actually be more beneficial compared to applying the technique to a much Larger panel size would be.
I know that it has been debated intensely so I am not trying to start another flame war on why or why not curved panels are any good.
Those whom know already know the differences.
The one thing that I can tell you is that in order to get any kind of decent performance out of a small panel, Is that you will have to learn to deal with much higher voltages to get any good efficiency out of them due to their lack of diaphragm surface area.
I have been using about 6.8KV of bias voltage and step-up transformer ratio of no less that about 1:160 and as high as 1:256!
I get efficiency's in the 88db to 91 db range with only a 4 or 5v peak input signal (2.83Vrms)
Expect to cross them over at a higher frequency as mentioned earlier as well.
This is not a bad thing as long as the woofer is of good quality, the magic will still be there.
I tried crossing them lower but it just didn't mate with the woofer properly.
This could have been caused by me trying to use some VST digital methods, But even then, frequency's above about 600Hz seemed to work the best.
I will try this with a real analog filter next time and see how it comes out.
Increasing the F/B path helps with this as well but it is mainly a function of the physics involved.
It takes a lot of excursion for a panel that size to perform well at frequency's below about 350Hz or so even though I did get mine down 200Hz okay.
That was the lowest frequency of operation as it used up every bit of its .072" to .075" D/S spacing and got even worse as it approached its resonate frequency of about 70Hz to 90Hz.
Evidence of clipping of the diaphragm into the stator was detected at this point but not so at about 300Hz and above.
Design for the lowest frequency that it has to produce as it is this frequency where it has its highest excursion and will be the loudest it can go when it is EQed to be flat.
At approximately 14 feet away you lose something like 9db to 10db of SPL compared to the level right at the diaphragm 1cm away from it.
I measured this in my last test and they were actually +110db 1cm from it!!!
This equated to my actual measurement of 100db at 1 meter at that particular level and about +105 db at 2 feet or so for my normal listening position.
I don't recall exactly what the input voltage was at that level but I think that it was around 20v to 30v peak and I may have also raised the bias to 9Kv or so in that test.
Raising the bias from 6.8KV added another 2-3db to the efficiency.
This is when I was reaching the limit of that design build and burned it with a bigger amp capable of a 56Vpeak swing.
So yes it can be done.
Is it practical, that is your call.
It works well as long has you keep your design limitations in mind and don't try to exceed them and then it will be reliable.
As far as going wireless or not, this is irrelevant.
The quality of your RF system will determine the overall quality in general.
Each driver is going to need its own amplifier and power sources.
And by the time you add the bias supply, step-up transformers, amps and their power supplies, This is all going to have to be in a separate package anyhow.
This however is where Class D amplifiers do shine.
I like the technology but I am not totally sold on the idea of their use for other than subwoofer use.
That is just a matter of my personal preference yours may be different.
Great quality Class D designs can be found, as some of TI's chip sets look very appetizing.
But I have not tried any of those methods yet, as I am strictly an analog person for now.
FWIW
jer
I too had them hooked to the TV for a signal and I had no woofer connected at the time.
Even with them setting at an angle leaning against cheery wood TV cabinet on the floor (much less than optimum placing as it was just like a big dresser drawer cabinet) the quality was outstanding and 100 times more than what you get out of the TV's built speaker system that wasn't bad at all to begin with.
If you have never tried it then you won't know or will ever know, I got great results!!!
This was one of the reasons I have spent the last few years focusing on smaller panels than I have just duplicating a large panel design.
Although the results was so great that it did become annoying at times due to the poor acoustic's in your typical TV news room!
You could actually hear the rustling of the workers on the stage and behind the camera's.
But movies sounded great!!!
As suggested before I would use a open backed design if they have to be enclosed with a Flat ESL panel and not a curved one.
You could make a cover that is curved to go over them for aesthetic's.
Building a curve shaped panel comes with its own construction issues to begin with.
I have contemplated small cylindrical drivers in the past with radius of 2 to 6 inches and gave up on the idea because of the difficulty to apply the diaphragm evenly and uniformly around the frame while having enough strength for it not too collapse.
However you could build one with a slight curve rather easily with a 1/4" to maybe as high as 1/2" curvature depth for a 3" wide panel using my window screen stator method as it is quite flexible without cracking the stator coating.
For a panel of this size the curvature may actually be more beneficial compared to applying the technique to a much Larger panel size would be.
I know that it has been debated intensely so I am not trying to start another flame war on why or why not curved panels are any good.
Those whom know already know the differences.
The one thing that I can tell you is that in order to get any kind of decent performance out of a small panel, Is that you will have to learn to deal with much higher voltages to get any good efficiency out of them due to their lack of diaphragm surface area.
I have been using about 6.8KV of bias voltage and step-up transformer ratio of no less that about 1:160 and as high as 1:256!
I get efficiency's in the 88db to 91 db range with only a 4 or 5v peak input signal (2.83Vrms)
Expect to cross them over at a higher frequency as mentioned earlier as well.
This is not a bad thing as long as the woofer is of good quality, the magic will still be there.
I tried crossing them lower but it just didn't mate with the woofer properly.
This could have been caused by me trying to use some VST digital methods, But even then, frequency's above about 600Hz seemed to work the best.
I will try this with a real analog filter next time and see how it comes out.
Increasing the F/B path helps with this as well but it is mainly a function of the physics involved.
It takes a lot of excursion for a panel that size to perform well at frequency's below about 350Hz or so even though I did get mine down 200Hz okay.
That was the lowest frequency of operation as it used up every bit of its .072" to .075" D/S spacing and got even worse as it approached its resonate frequency of about 70Hz to 90Hz.
Evidence of clipping of the diaphragm into the stator was detected at this point but not so at about 300Hz and above.
Design for the lowest frequency that it has to produce as it is this frequency where it has its highest excursion and will be the loudest it can go when it is EQed to be flat.
At approximately 14 feet away you lose something like 9db to 10db of SPL compared to the level right at the diaphragm 1cm away from it.
I measured this in my last test and they were actually +110db 1cm from it!!!
This equated to my actual measurement of 100db at 1 meter at that particular level and about +105 db at 2 feet or so for my normal listening position.
I don't recall exactly what the input voltage was at that level but I think that it was around 20v to 30v peak and I may have also raised the bias to 9Kv or so in that test.
Raising the bias from 6.8KV added another 2-3db to the efficiency.
This is when I was reaching the limit of that design build and burned it with a bigger amp capable of a 56Vpeak swing.
So yes it can be done.
Is it practical, that is your call.
It works well as long has you keep your design limitations in mind and don't try to exceed them and then it will be reliable.
As far as going wireless or not, this is irrelevant.
The quality of your RF system will determine the overall quality in general.
Each driver is going to need its own amplifier and power sources.
And by the time you add the bias supply, step-up transformers, amps and their power supplies, This is all going to have to be in a separate package anyhow.
This however is where Class D amplifiers do shine.
I like the technology but I am not totally sold on the idea of their use for other than subwoofer use.
That is just a matter of my personal preference yours may be different.
Great quality Class D designs can be found, as some of TI's chip sets look very appetizing.
But I have not tried any of those methods yet, as I am strictly an analog person for now.
FWIW
jer
I forgot to post a few pictures of the window screen stator I had mentioned in the above post.
The First two show its flexibility.
The third one shows a test I did of it withholding high voltage at 13.8Kv without any arc through, except to the very edge where the coating was thinner and not coated at all.
I had made this piece about Two years ago and I just took the First two pictures.
If I tried that test again today I am sure that it would pass the same as it did two years ago.
The last time I tested this piece was about 8 months ago and it passed the test then as well.
Its width is exactly 3 inches wide and 8 1/2 inches long with about 6 to 7 inches of a quality arc proven coated section.
Enjoy!!
jer
The First two show its flexibility.
The third one shows a test I did of it withholding high voltage at 13.8Kv without any arc through, except to the very edge where the coating was thinner and not coated at all.
I had made this piece about Two years ago and I just took the First two pictures.
If I tried that test again today I am sure that it would pass the same as it did two years ago.
The last time I tested this piece was about 8 months ago and it passed the test then as well.
Its width is exactly 3 inches wide and 8 1/2 inches long with about 6 to 7 inches of a quality arc proven coated section.
Enjoy!!
jer
Attachments
Last edited:
jer...is a littel fearless... but i would never speak for jar....
but we cant wate for others to push the limts....were all just here for sec.....
i have got great sound from doing mods others say are nuts....
The bias is a big part of the sound of any ESL panel.....i have found....i think as much as the setup tranx....an crossover....
but we cant wate for others to push the limts....were all just here for sec.....
i have got great sound from doing mods others say are nuts....
The bias is a big part of the sound of any ESL panel.....i have found....i think as much as the setup tranx....an crossover....
Ahhhhh,ha,ha,ha,ha,ha......It is all in the name DIY ESL Research!!!
There were many (besides myself) asking questions and no one seemed to know what or want to give the answers to them!
So, I did a little experimenting on my own !!!!
I grew up as a kid playing with old tube chassis, So HV is just second nature to me.
When I was young is used to wind tesla coils and mess with car ignition coils and such.
Then I made my own Transformer based on the Model-T type and that thing sure packed a whalop !!!
Later, I discovered the wonderful world of plasma from the a.c. output of a T.V. flyback transformer.
I used to make Plasma from a Car ignition coil hooked to the output of my Sunn Concert lead guitar amplifier.
I wish I could have had patent the idea of the plasma globe lamps as I discovered that at age 14 in 1977!!
I used to make Piezo speakers out of plate of glass and such.
Then I discovered the Popular Mechanics magazine.
When I got the issues that show cased the Magnapan's and the Quad 57 ESL that was all she wrote for me.....I was hooked!!
I spent all of the rest of my lifetime since researching on how to go about building these things.
My uncle is a HAM Radio operator and was one of the top Electrical engineers at Ford Motor Company and he had got me started with electronics at the age of three.
The rest is all history from there.
The one thing that my uncle alway told me is that, "You must design for at least 50% more if you can't go 100%, or else it won't be reliable and you will spend more time fixing it than for what it is worth".
I wanted to know exactly what the limit's are for ESL's and the materials used to build them and I wanted to use common off of the shelf items to prove how easily it could be done.
Knowing raising the voltages raises the level of sensitivity, I need to know how far I could go the the choice the materials I was using.
And every time you can double the bias without the thing arcing you gain 6db of free efficiency.
And going from your common 2.5Kv to 10Kv of bias voltage give you an incredible 12db more of gain!
But my materials stopped me from getting much past 10Kv on the last setup but I did get to 12Kv for a short time just to prove this.
This means that the amp only has to deliver 12db less voltage to perform at the same SPL.
1/16 the power!!
This means that you actually can get good performance with a smaller amplifier as well as a with small panel ESL.
I can't imagine what it would be like if these methods were applied to a large panel system as I have yet to build one!!
I guess you could say....."I am obsessed with it" !!!! He,he,he,he,he,he
jer
There were many (besides myself) asking questions and no one seemed to know what or want to give the answers to them!
So, I did a little experimenting on my own !!!!
I grew up as a kid playing with old tube chassis, So HV is just second nature to me.
When I was young is used to wind tesla coils and mess with car ignition coils and such.
Then I made my own Transformer based on the Model-T type and that thing sure packed a whalop !!!
Later, I discovered the wonderful world of plasma from the a.c. output of a T.V. flyback transformer.
I used to make Plasma from a Car ignition coil hooked to the output of my Sunn Concert lead guitar amplifier.
I wish I could have had patent the idea of the plasma globe lamps as I discovered that at age 14 in 1977!!
I used to make Piezo speakers out of plate of glass and such.
Then I discovered the Popular Mechanics magazine.
When I got the issues that show cased the Magnapan's and the Quad 57 ESL that was all she wrote for me.....I was hooked!!
I spent all of the rest of my lifetime since researching on how to go about building these things.
My uncle is a HAM Radio operator and was one of the top Electrical engineers at Ford Motor Company and he had got me started with electronics at the age of three.
The rest is all history from there.
The one thing that my uncle alway told me is that, "You must design for at least 50% more if you can't go 100%, or else it won't be reliable and you will spend more time fixing it than for what it is worth".
I wanted to know exactly what the limit's are for ESL's and the materials used to build them and I wanted to use common off of the shelf items to prove how easily it could be done.
Knowing raising the voltages raises the level of sensitivity, I need to know how far I could go the the choice the materials I was using.
And every time you can double the bias without the thing arcing you gain 6db of free efficiency.
And going from your common 2.5Kv to 10Kv of bias voltage give you an incredible 12db more of gain!
But my materials stopped me from getting much past 10Kv on the last setup but I did get to 12Kv for a short time just to prove this.
This means that the amp only has to deliver 12db less voltage to perform at the same SPL.
1/16 the power!!
This means that you actually can get good performance with a smaller amplifier as well as a with small panel ESL.
I can't imagine what it would be like if these methods were applied to a large panel system as I have yet to build one!!
I guess you could say....."I am obsessed with it" !!!! He,he,he,he,he,he
jer
Last edited:
Yes YEs...you are on the right track.....never change.....it all here for us....were not here for it!.....what ever it IS......people get hungup... on it good a nuf..maybe not..............this work would not be the same with out you!
This ESL sound is well.....Canbe Mindblowing...
thanks jer............
This ESL sound is well.....Canbe Mindblowing...
thanks jer............
I too am Guilty of it!!!
If it weren't for seeing your build Charlie and about using the power toroid's it would have taken me even longer to get back in to this stuff.....But not to long.
Once I got bitten when I powered up my very first panels in 2003 I knew there was no turning back at that point!!
I too can feel the anguish as it it took me about 18 years until I made my first set in 2003 from the time I found the Roger Sanders and David Hermeyer articles in a Audio Amatuer Speaker Builder book I found at the local library in 1985!!
I searched the web for every little bit of info I could find including plasma speakers since around 1997, before then.
The TAC website is what really got me going at the time and then DIYaudio came along as it was just starting out as well.
jer
If it weren't for seeing your build Charlie and about using the power toroid's it would have taken me even longer to get back in to this stuff.....But not to long.
Once I got bitten when I powered up my very first panels in 2003 I knew there was no turning back at that point!!
I too can feel the anguish as it it took me about 18 years until I made my first set in 2003 from the time I found the Roger Sanders and David Hermeyer articles in a Audio Amatuer Speaker Builder book I found at the local library in 1985!!
I searched the web for every little bit of info I could find including plasma speakers since around 1997, before then.
The TAC website is what really got me going at the time and then DIYaudio came along as it was just starting out as well.
jer
If it really has to be a bar (not curved) one could do a 6 speaker bessel array (with only one stator connected for the 1/2-weighted sections)
A Paul Kemble web page - speaker bessel arrays.
with a ribbon in the center for better dispersion on top.
BTW, I still find that experimenting with RTR panels would reduce the anger for the case that it leads to nothing.
A Paul Kemble web page - speaker bessel arrays.
with a ribbon in the center for better dispersion on top.
BTW, I still find that experimenting with RTR panels would reduce the anger for the case that it leads to nothing.
I have known of the Bessel arrangement for quite some time but I have never tried it.
When I built my 2x3-6.5" woofer box in 1994 I had considered adding a few more to try it out when I noticed the many lobes it produced at the higher frequency's.
But I never did.
At the time (mid 1980's) it was mentioned in one of my big books but it didn't go into any great detail as it was a new thing and it was kept a bit under the radar because it had just been patented (or pending) then.
I still have that book here somewhere.
Through the years later I still have had a difficult time trying to find any more detailed info on such a system or any examples of ones whom tried it showing working arrays that they have actually built.
I have found nothing so far.
I have seen the link that was posted but it was about all that I could find even for just a few years ago.
But, It wouldn't surprize me any, if it is actually being used in big Stadium PA systems.
I had thought about giving this a try with many small ESL panels but it would be quite very time consuming to build enough panels and I think that it would only be beneficial for a very very large array for whatever type of driver one should decide to use.
jer
When I built my 2x3-6.5" woofer box in 1994 I had considered adding a few more to try it out when I noticed the many lobes it produced at the higher frequency's.
But I never did.
At the time (mid 1980's) it was mentioned in one of my big books but it didn't go into any great detail as it was a new thing and it was kept a bit under the radar because it had just been patented (or pending) then.
I still have that book here somewhere.
Through the years later I still have had a difficult time trying to find any more detailed info on such a system or any examples of ones whom tried it showing working arrays that they have actually built.
I have found nothing so far.
I have seen the link that was posted but it was about all that I could find even for just a few years ago.
But, It wouldn't surprize me any, if it is actually being used in big Stadium PA systems.
I had thought about giving this a try with many small ESL panels but it would be quite very time consuming to build enough panels and I think that it would only be beneficial for a very very large array for whatever type of driver one should decide to use.
jer
I just found Don Keele's paper on Bessel Array's !!!
6012 - Effective Performance of Bessel Arrays
jer
6012 - Effective Performance of Bessel Arrays
jer
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.