I'm looking into making some ESLs but I'm on a very tight budget and I'm wondering what the cheapest I could make a hybrid system. I won't need a crossover because I can use the one from my current system.
I could use the sub from my main system also as I will be using only one at a time (though maybe I could use one pair as surrounds)
Also how small do you think I could make them? My space is pretty small so I need to have them as small as possible. I would guess that square would be the best shape (actually circular, but thats tricky to make) for dealing with the phase cancelation but I don't know too much about it.
I could use the sub from my main system also as I will be using only one at a time (though maybe I could use one pair as surrounds)
Also how small do you think I could make them? My space is pretty small so I need to have them as small as possible. I would guess that square would be the best shape (actually circular, but thats tricky to make) for dealing with the phase cancelation but I don't know too much about it.
You can make them any size you want too!
I use wire mesh because of cost reason but it does have its issues to over come.
Most use pereforated metal and is probably the easiest and most fool proof method and cost won't be so bad if you are just planing on build some smaller panels since it sound like you have most every thing else.
Here are some pictures from this morning of my little panels that I finally got running again after quite some time.
As well as some preliminary test data.
These have been EQ'd to be flat as well, using a Mackie 32-8 mixing boards filters.
enjoy !!
jer 🙂
I use wire mesh because of cost reason but it does have its issues to over come.
Most use pereforated metal and is probably the easiest and most fool proof method and cost won't be so bad if you are just planing on build some smaller panels since it sound like you have most every thing else.
Here are some pictures from this morning of my little panels that I finally got running again after quite some time.
As well as some preliminary test data.
These have been EQ'd to be flat as well, using a Mackie 32-8 mixing boards filters.
enjoy !!
jer 🙂
Attachments
Last edited:
Idk where the most recent post went because I got the email about it but its not here haha
I did in fact mean sick haha, keep in mind I'm 16, we have our very own language 😉
Urban Dictionary: sick
I did in fact mean sick haha, keep in mind I'm 16, we have our very own language 😉
Urban Dictionary: sick
Hi,
the cheapest You can build is probabely a wire stator like the Audiostatic.
A plastic frame, 1/10-1/5 inch thick, of pvc or similar, that holds a bunch of parallel cross linking rods to which the wire is attched. The rods may be rectangular pieces of plastic, or metal. The wire may be single stranded H05-VU or H07-VU, a pvc insulated wire for main household purposes (as such easy sourceable and cheap). I got the plastic for my first ESL for free (some leftover pieces they couldn´t use for anything else any more). Bought the roll of wire and the aluminium rods in a DIY-shop, but You may get even that for free from a scrapyard. Glued the diaphragm with double sided sticky carpet tape from TESA to the frame. Took a lot of effort and time to straighten the wires but that didn´t count budgetwise. The most costly parts were the audio trannies, but with a segmented wire stator You can even try a bunch of cheap and simple EI-power transformers (I used a sixpack of 10VA, 6V/230V). For biasing the membrane a cascade is ok. Lowcost parts may be found in a surplus store.
jauu
Calvin
the cheapest You can build is probabely a wire stator like the Audiostatic.
A plastic frame, 1/10-1/5 inch thick, of pvc or similar, that holds a bunch of parallel cross linking rods to which the wire is attched. The rods may be rectangular pieces of plastic, or metal. The wire may be single stranded H05-VU or H07-VU, a pvc insulated wire for main household purposes (as such easy sourceable and cheap). I got the plastic for my first ESL for free (some leftover pieces they couldn´t use for anything else any more). Bought the roll of wire and the aluminium rods in a DIY-shop, but You may get even that for free from a scrapyard. Glued the diaphragm with double sided sticky carpet tape from TESA to the frame. Took a lot of effort and time to straighten the wires but that didn´t count budgetwise. The most costly parts were the audio trannies, but with a segmented wire stator You can even try a bunch of cheap and simple EI-power transformers (I used a sixpack of 10VA, 6V/230V). For biasing the membrane a cascade is ok. Lowcost parts may be found in a surplus store.
jauu
Calvin
Attachments
Last edited:
Hi,
some data.
Stator 1: 1440x240mm | frame 30x5mm PVC | 40pcs. of Wire H07-VU (diam. 2.65mm) spaced 4.25mm apart | d/s 2.35mm | 4.5µm Hostaphan diaphragm | membrane-coating after my glue formula (cheap, transparent, non-toxic, easy to mix) | Grafit33 as sprayed on bias contact strip | 10x2mm rectangular aluminum rods as cross linking rods | household glues like TESA or UHU to glue the frame structure together | TESA double sided sticky carpet tape to glue the membrane to the frame | 5 segments (3 electrical) arranged symmetrically around the vertcal axis
Stator 2: basically the same, apart from 10x10 square aluminum tubes as cross linking rods | Household aluminum foil as bias contact strip |
Stator Mini: 790x130mm | frame 18*4mm unfilled PVC | 20pcs. of Wire H07-VU (diam. 2.65mm) | 5x5mm massive steel rods as cross linking rods | Household aluminum foil as bias contact strip | 3 segments (2 electrical) arranged symmetrically around the vertical axis
jauu
Calvin
some data.
Stator 1: 1440x240mm | frame 30x5mm PVC | 40pcs. of Wire H07-VU (diam. 2.65mm) spaced 4.25mm apart | d/s 2.35mm | 4.5µm Hostaphan diaphragm | membrane-coating after my glue formula (cheap, transparent, non-toxic, easy to mix) | Grafit33 as sprayed on bias contact strip | 10x2mm rectangular aluminum rods as cross linking rods | household glues like TESA or UHU to glue the frame structure together | TESA double sided sticky carpet tape to glue the membrane to the frame | 5 segments (3 electrical) arranged symmetrically around the vertcal axis
Stator 2: basically the same, apart from 10x10 square aluminum tubes as cross linking rods | Household aluminum foil as bias contact strip |
Stator Mini: 790x130mm | frame 18*4mm unfilled PVC | 20pcs. of Wire H07-VU (diam. 2.65mm) | 5x5mm massive steel rods as cross linking rods | Household aluminum foil as bias contact strip | 3 segments (2 electrical) arranged symmetrically around the vertical axis
jauu
Calvin
Here are a few more methods,
this method similar to what Calvin had described and is the construction idea that I had borrowed from,
http://www.diyaudio.com/forums/planars-exotics/203990-esl-woofer-anybody-game-2.html#post2862780
http://www.diyaudio.com/forums/plan...nstruct-cube-louver-acoustat.html#post2150040
I may try this method using some Kynar covered wire wrap wire for some small panels.
I had recently found a source of it for $30 to $40 for a 1000' spool.
There are a few larger threads dedicated to this method of construction.
Here are some TIG welding wire stators that I have done.
They just need to be coated with some polyurathene or some clear acrylic enamel to be finished.
I had just found a cheap source of Clear acrylic for $1.69 for a can.
I am planning another panel of the same size as the one I posted using TIG rod.
Because I have plenty of it and found I difficult ( although not impossible) to seal my old panels to handle the extremely high bias voltages that I am aiming for.
Right now they run nicely at 7.5kv to 8kv of bias voltage.
http://www.diyaudio.com/forums/planars-exotics/203990-esl-woofer-anybody-game.html#post2857124
I like building the smaller panels and the sound just as good as a larger one (quality wise).
It actually takes 2 or 3 of those panels to match the efficiency of a small woofer using a very high transformation ratio (1:255 to 1:500).
I am getting very close to it though using higher than normal bias and driving voltages.
Using I very high transformation ratio demands a lot from the amplifier and warrants a custom wound unit as the stray capacitance of the transformer is in the order of 10 times that of the panel.
My little panels have a capacitance of 35pf to 60pf and my transformer has 600pf.
This with the very high transformation ratio that I am using brings my impedance down to the 1ohm or less range at 20Khz, and, only 1/10 of the power is being used to create the sound.
So surface area is the key.
Build them as big as you can get away with and you should have no problems driving them using the common methods described in these threads.
http://www.diyaudio.com/forums/plan...nstruct-cube-louver-acoustat.html#post2141539
One of my very First panels was has a 8" X 22" diagphram and it performed nicely For a smaller sized panel.
I have not ran the thing since I had built it in 2003.
I have to build some new frames for the Diagphram and they will be in working order again.
I will be doing this some time very soon so I will post the results once I get them going again.
They used wire mesh as well and spray paint as the coating.
But if you use this method stay away from any paints that have pigment in it mostly white (titanium Dioxide),Use one color coat if you must and then use a clear for the rest.
And red oxide primers are okay as they have a high content of talc and this is good the rest use titanium dioxide so don't use those.
I have found that Titanium dioxide conducts quite well at high voltages.
There are pictures of those posted also somewhere.
http://www.diyaudio.com/forums/plan...truct-cube-louver-acoustat-2.html#post2162588
When I started I had no job and no money so I became obsessed with how cheaply can I build these things and you would be surprised in all of the commonly available materials that you can build these out of.
Here are a few more must read threads,
http://www.diyaudio.com/forums/planars-exotics/166280-jazzmans-new-stat-panels.html#post2174937
http://www.diyaudio.com/forums/planars-exotics/166039-esls-planars-etc-me.html#post2170693
http://www.diyaudio.com/forums/planars-exotics/162127-different-ideas-esl-panels.html#post2099310
http://www.diyaudio.com/forums/planars-exotics/165495-start-finish-esl-hybrid.html#post2160924
diyAudio - Search Results
http://www.diyaudio.com/forums/plan...e-know-us-source-esl-toroids.html#post1927233
Your biggest investment will be the Bias Supply and Step up Transformers.
But with all of the help of the talented and crafty DIYer's in these threads some very viable and extremely affordable resources have been revealed.
Good luck !!!!
jer 🙂
P.S. If you have any questions at all just ask there is always someone to have an answer.
this method similar to what Calvin had described and is the construction idea that I had borrowed from,
http://www.diyaudio.com/forums/planars-exotics/203990-esl-woofer-anybody-game-2.html#post2862780
http://www.diyaudio.com/forums/plan...nstruct-cube-louver-acoustat.html#post2150040
I may try this method using some Kynar covered wire wrap wire for some small panels.
I had recently found a source of it for $30 to $40 for a 1000' spool.
There are a few larger threads dedicated to this method of construction.
Here are some TIG welding wire stators that I have done.
They just need to be coated with some polyurathene or some clear acrylic enamel to be finished.
I had just found a cheap source of Clear acrylic for $1.69 for a can.
I am planning another panel of the same size as the one I posted using TIG rod.
Because I have plenty of it and found I difficult ( although not impossible) to seal my old panels to handle the extremely high bias voltages that I am aiming for.
Right now they run nicely at 7.5kv to 8kv of bias voltage.
http://www.diyaudio.com/forums/planars-exotics/203990-esl-woofer-anybody-game.html#post2857124
I like building the smaller panels and the sound just as good as a larger one (quality wise).
It actually takes 2 or 3 of those panels to match the efficiency of a small woofer using a very high transformation ratio (1:255 to 1:500).
I am getting very close to it though using higher than normal bias and driving voltages.
Using I very high transformation ratio demands a lot from the amplifier and warrants a custom wound unit as the stray capacitance of the transformer is in the order of 10 times that of the panel.
My little panels have a capacitance of 35pf to 60pf and my transformer has 600pf.
This with the very high transformation ratio that I am using brings my impedance down to the 1ohm or less range at 20Khz, and, only 1/10 of the power is being used to create the sound.
So surface area is the key.
Build them as big as you can get away with and you should have no problems driving them using the common methods described in these threads.
http://www.diyaudio.com/forums/plan...nstruct-cube-louver-acoustat.html#post2141539
One of my very First panels was has a 8" X 22" diagphram and it performed nicely For a smaller sized panel.
I have not ran the thing since I had built it in 2003.
I have to build some new frames for the Diagphram and they will be in working order again.
I will be doing this some time very soon so I will post the results once I get them going again.
They used wire mesh as well and spray paint as the coating.
But if you use this method stay away from any paints that have pigment in it mostly white (titanium Dioxide),Use one color coat if you must and then use a clear for the rest.
And red oxide primers are okay as they have a high content of talc and this is good the rest use titanium dioxide so don't use those.
I have found that Titanium dioxide conducts quite well at high voltages.
There are pictures of those posted also somewhere.
http://www.diyaudio.com/forums/plan...truct-cube-louver-acoustat-2.html#post2162588
When I started I had no job and no money so I became obsessed with how cheaply can I build these things and you would be surprised in all of the commonly available materials that you can build these out of.
Here are a few more must read threads,
http://www.diyaudio.com/forums/planars-exotics/166280-jazzmans-new-stat-panels.html#post2174937
http://www.diyaudio.com/forums/planars-exotics/166039-esls-planars-etc-me.html#post2170693
http://www.diyaudio.com/forums/planars-exotics/162127-different-ideas-esl-panels.html#post2099310
http://www.diyaudio.com/forums/planars-exotics/165495-start-finish-esl-hybrid.html#post2160924
diyAudio - Search Results
http://www.diyaudio.com/forums/plan...e-know-us-source-esl-toroids.html#post1927233
Your biggest investment will be the Bias Supply and Step up Transformers.
But with all of the help of the talented and crafty DIYer's in these threads some very viable and extremely affordable resources have been revealed.
Good luck !!!!
jer 🙂
P.S. If you have any questions at all just ask there is always someone to have an answer.
Last edited:
The transformers and power supplies shown on my blog page below are about as inexpensive as it gets: jazzman-esl-page.blogspot.com/
Wow thanks for all the info guys! When I get back home ill go over all those links and stuff.
Thanks Charlie for that, I actually read your website from start to finish before I made this thread. I didn't realize that kit was for sale, I may have to take you up on it.
And Calvin thanks for your info and pics. I'll compare the two methods to see how much I could save and how much longer and harder it would be
Thanks Charlie for that, I actually read your website from start to finish before I made this thread. I didn't realize that kit was for sale, I may have to take you up on it.
And Calvin thanks for your info and pics. I'll compare the two methods to see how much I could save and how much longer and harder it would be
Actually, I don't sell ESL's or components-- just sharing the info and bargains I found while researching and building my own ESL speakers. For example, the trannys and power supplies I use can be assembled for just $125; which is about as cheap as can be, I'd say.
Last edited:
Oh sorry I misunderstood. You should consider compiling those components and selling kits. From my own experience buying components you could probably get much better prices if you bought in bulk and I bet people on this forum would buy it.
Alright maybe this is a crazy idea but...
What if I made a ton of small panels and placed then at angles to each other in like a horizontal octogon or something and then built some sort of ported enclosure for them. I was wondering how they would feel about being put in an enclosure to mitigate phase cancellation. Also how might that be wired?
A few things I would be worried about would be problems with the mylar being dampened by the enclosure, problems at the membrane's resonant frequency, and low sensitivity.
I've seen curved panels but I've always wondered how they could be made, would a full(or almost full) circle be made for 360 degree horizontal dispersion?
What if I made a ton of small panels and placed then at angles to each other in like a horizontal octogon or something and then built some sort of ported enclosure for them. I was wondering how they would feel about being put in an enclosure to mitigate phase cancellation. Also how might that be wired?
A few things I would be worried about would be problems with the mylar being dampened by the enclosure, problems at the membrane's resonant frequency, and low sensitivity.
I've seen curved panels but I've always wondered how they could be made, would a full(or almost full) circle be made for 360 degree horizontal dispersion?
This Has been done before.
I have a copy of an article that was published in the Audio Amateur called "A High Efficiency Electrostatic Loud Speaker system" it also described a tube output D-D push pull Class A amplifier that was powered from a rather low voltage (1500v)at about 35 to 50 watts per tube.
It used many panels about the size of my little panel stacked up and formed into an arc with about 7 degree's of angle per adjacent panel.
Some one two years ago had posted a picture of such a panel that was built that he had acquired.
I don't know if he ever got it back operational though.
The panels all must be constructed identically in order for it to work properly.
They also used a small D/S of around .040" (40mils) in order to increase the efficiency.
This is fine for a Hybrid system but does nothing for a good bass extension.
They where claiming a +110 db SPL level in which was quite phenomenal for the day and age (mid to early 70's).
The frames of each panel panel was built up of Plexiglas and different thicknesses of acetate.
The stator was made out of perforated steel or aluminium such as Lincain as described in the article.
It used the spacer dot method to help support the diagphram in such a small D/S spacing.
It is an excellent article and I still have a copy of it buried in my stacks of papers.
I would always refer to it as I studied up on construction techniques of ESL's.
Curved panels can be made as there have been many discussions on this if that is what you prefer.
Some have been successful at it providing that you have the bending tools to make the panels consistent.
Google up "An interview with Rodger Sanders" and you will understand why he (The inventor of the curved ESL) does not prefer them.
This is a matter subjectivity and personal preference (I prefer the flat panel as well).
Putting the panels in an enclosure has a whole another set of issues and has been discussed as well.
I have done some experiments whit angling several panels and this works but is a trail and error process to get them just right with a minimal amount of lobing.
I have tried to dampen the back wave a bit and had found the results less than optimum as the back wave pressure seems to limit the movement of the diagphram taking away from its high end Finesse.
But is some thing that should be explore in more detail.
The biggest issue about putting them in an enclosure is that any reflections that occur come straight through the front of the diagphram coloring the sound just like the common box speaker does ruining the magic sound that ESL's have.
Again this is a matter of subjectivity and may work for some.
jer 🙂
I have a copy of an article that was published in the Audio Amateur called "A High Efficiency Electrostatic Loud Speaker system" it also described a tube output D-D push pull Class A amplifier that was powered from a rather low voltage (1500v)at about 35 to 50 watts per tube.
It used many panels about the size of my little panel stacked up and formed into an arc with about 7 degree's of angle per adjacent panel.
Some one two years ago had posted a picture of such a panel that was built that he had acquired.
I don't know if he ever got it back operational though.
The panels all must be constructed identically in order for it to work properly.
They also used a small D/S of around .040" (40mils) in order to increase the efficiency.
This is fine for a Hybrid system but does nothing for a good bass extension.
They where claiming a +110 db SPL level in which was quite phenomenal for the day and age (mid to early 70's).
The frames of each panel panel was built up of Plexiglas and different thicknesses of acetate.
The stator was made out of perforated steel or aluminium such as Lincain as described in the article.
It used the spacer dot method to help support the diagphram in such a small D/S spacing.
It is an excellent article and I still have a copy of it buried in my stacks of papers.
I would always refer to it as I studied up on construction techniques of ESL's.
Curved panels can be made as there have been many discussions on this if that is what you prefer.
Some have been successful at it providing that you have the bending tools to make the panels consistent.
Google up "An interview with Rodger Sanders" and you will understand why he (The inventor of the curved ESL) does not prefer them.
This is a matter subjectivity and personal preference (I prefer the flat panel as well).
Putting the panels in an enclosure has a whole another set of issues and has been discussed as well.
I have done some experiments whit angling several panels and this works but is a trail and error process to get them just right with a minimal amount of lobing.
I have tried to dampen the back wave a bit and had found the results less than optimum as the back wave pressure seems to limit the movement of the diagphram taking away from its high end Finesse.
But is some thing that should be explore in more detail.
The biggest issue about putting them in an enclosure is that any reflections that occur come straight through the front of the diagphram coloring the sound just like the common box speaker does ruining the magic sound that ESL's have.
Again this is a matter of subjectivity and may work for some.
jer 🙂
I found the video and watched it. He made some interesting points I'll have to consider. Thanks for the info
I was planning on making a standard tweeter line array to go along with my main full range line arrays to help out the comb filtering but I was thinking of instead making a ESL tweeter, crossed over at around 1000 Hz and going up to 20000 Hz. I was going to give it at least a minor curve just to match my other speakers so the transition isn't so noticeable. I was thinking of making it around 2ft by 6 inches with a minor curve, I would probably get the biggest tube I could find at home depot and use that as a form, or even frame if I went with the wire method.
Do you guys think this pairing could work? From what I've read higher range ESLs are quite a bit easier because they can be made with a smaller spacing ie higher sensitivity and I wouldn't have to worry about the backwave, also I could use a thinner membrane. I was thinking that having them above that range would make dialogue and everything go to the esl which would undoubtedly sound better than my traditional towers. I'm biamping it anyways so integrating it into the system would be easy, I would just need an amp that could handle the capacitive load at high frequencies
I was planning on making a standard tweeter line array to go along with my main full range line arrays to help out the comb filtering but I was thinking of instead making a ESL tweeter, crossed over at around 1000 Hz and going up to 20000 Hz. I was going to give it at least a minor curve just to match my other speakers so the transition isn't so noticeable. I was thinking of making it around 2ft by 6 inches with a minor curve, I would probably get the biggest tube I could find at home depot and use that as a form, or even frame if I went with the wire method.
Do you guys think this pairing could work? From what I've read higher range ESLs are quite a bit easier because they can be made with a smaller spacing ie higher sensitivity and I wouldn't have to worry about the backwave, also I could use a thinner membrane. I was thinking that having them above that range would make dialogue and everything go to the esl which would undoubtedly sound better than my traditional towers. I'm biamping it anyways so integrating it into the system would be easy, I would just need an amp that could handle the capacitive load at high frequencies
That is exactly my plans.
Right now I am tweaking my little panels to get the most I can out of them.
Then I will build two 8 foot long ones ( two 4 foot panels each ).
In a quick test earlier they were down -3db at 400hz from 600hz for 3.25" wide diagphram (4.75 total width).
I am shooting for the 200hz to 300hz range so a 4.5" to 5.5" range should do it.
Here are few pictures of the woofer system and a few of a micro ESl that I made out of a couple of old credit cards as a headphone driver.
I only made one so far,But it does sound go though.
It runs on a 500V Bias.
jer 🙂
P.S. I am only using the 4 12's and the 2 10's are facing the corner behind each of the stack of 12's.
Right now I am tweaking my little panels to get the most I can out of them.
Then I will build two 8 foot long ones ( two 4 foot panels each ).
In a quick test earlier they were down -3db at 400hz from 600hz for 3.25" wide diagphram (4.75 total width).
I am shooting for the 200hz to 300hz range so a 4.5" to 5.5" range should do it.
Here are few pictures of the woofer system and a few of a micro ESl that I made out of a couple of old credit cards as a headphone driver.
I only made one so far,But it does sound go though.
It runs on a 500V Bias.
jer 🙂
P.S. I am only using the 4 12's and the 2 10's are facing the corner behind each of the stack of 12's.
Attachments
Last edited:
Hi,
the mentioned video should be taken with a grain of salt.
Its a marketing video. As such it makes use of generalizations for explanations.
While they give the listener a certain basic idea about the issue they may not be precise and sometimes they may even be plainly false.
For example does a Transmissionline woofer not solve the problem of integration of panel and woofer. A TL is in general no improvement over a CB or BR box, since it doesn´t adress the real problem -which is the difference in distribution characters between panel and woofer. In fact may a TL just be regarded as a special case of a BR-box or Helmholtz-resonator with a monopolar distribution character over its working range.
Another aspect regards the explanation of the need of high directivity.
He doesn´t distinguish between early and late reflections and he doesn´t loose a word about the impracticality of the extreme directivity of wide panels (talking of deviations in amplitude response of 10dB occuring within just 1° of change of listening position!). Its not without good reason, that basically any other manufacturer of speakers cares about the distribution character of their speakers and that laser-like beaming speakers are the absolute minority.
The issue of distribution may be dealt with by electrical segmentation within a single panel, or mechanical segmentation using a couple of tilted panels.
The panels will still be highly directive, but at least You don´t need to pinpoint nail Your head fixed like a mummy.
Since the effort of building a multitude of smaller panels is higher a single larger cell may be preferrable and I´d opt for a single larger panel.
After my experience the outstanding sonic qualities of an ESL require certain minimum membrane areas and sufficiently low crossover frequencies. If smaller, than the ESL sounds just like any ordinary dynamic mid or tweeter. If sized big enough and crossed over preferably below 200Hz the panel can show all its superiority over dynamic drivers.
Going lower in crossover raises problems due to the needed amount of equalization and excursion (reduced dynamic range, increased voltage and power demands). Going higher and the crossing over is situated right in the most sensitive region of hearing. Keep the frequency range of the human voice free from crossovers.
The choice of stator construction also influences the demands You put on the amplifier. R.Sanders was totally wrong when he sayd that transistor amplifiers all sound the same on ESLs but different with dynamic speakers (in the early days...nowadays he´s offering his own amps). Just the opposite is true, simply because most amps are not designed to cope with ESL-loads, with electrically segmented stators beeing the less tricky and non-segmented stators presenting a difficult load. Since the vast majority of amps works with global feedback and in class AB push-pull and is not designed to handle large phaseshifts of the load (hence large amounts of complex powers) they simply suck and sound crap. More often than not a bad sounding ESL-presentation is due to the limits of the electronics.
jauu
Calvin
ps. The curved ESL was probabely invented and patented 5 minutes after invention of the principle itself...some time before R.Sanders wrote about it.
Mr. P.Walker´s text contain hints on curved panels and there´s at latest a patent from 1956 describing the construction such panels.
the mentioned video should be taken with a grain of salt.
Its a marketing video. As such it makes use of generalizations for explanations.
While they give the listener a certain basic idea about the issue they may not be precise and sometimes they may even be plainly false.
For example does a Transmissionline woofer not solve the problem of integration of panel and woofer. A TL is in general no improvement over a CB or BR box, since it doesn´t adress the real problem -which is the difference in distribution characters between panel and woofer. In fact may a TL just be regarded as a special case of a BR-box or Helmholtz-resonator with a monopolar distribution character over its working range.
Another aspect regards the explanation of the need of high directivity.
He doesn´t distinguish between early and late reflections and he doesn´t loose a word about the impracticality of the extreme directivity of wide panels (talking of deviations in amplitude response of 10dB occuring within just 1° of change of listening position!). Its not without good reason, that basically any other manufacturer of speakers cares about the distribution character of their speakers and that laser-like beaming speakers are the absolute minority.
The issue of distribution may be dealt with by electrical segmentation within a single panel, or mechanical segmentation using a couple of tilted panels.
The panels will still be highly directive, but at least You don´t need to pinpoint nail Your head fixed like a mummy.
Since the effort of building a multitude of smaller panels is higher a single larger cell may be preferrable and I´d opt for a single larger panel.
After my experience the outstanding sonic qualities of an ESL require certain minimum membrane areas and sufficiently low crossover frequencies. If smaller, than the ESL sounds just like any ordinary dynamic mid or tweeter. If sized big enough and crossed over preferably below 200Hz the panel can show all its superiority over dynamic drivers.
Going lower in crossover raises problems due to the needed amount of equalization and excursion (reduced dynamic range, increased voltage and power demands). Going higher and the crossing over is situated right in the most sensitive region of hearing. Keep the frequency range of the human voice free from crossovers.
The choice of stator construction also influences the demands You put on the amplifier. R.Sanders was totally wrong when he sayd that transistor amplifiers all sound the same on ESLs but different with dynamic speakers (in the early days...nowadays he´s offering his own amps). Just the opposite is true, simply because most amps are not designed to cope with ESL-loads, with electrically segmented stators beeing the less tricky and non-segmented stators presenting a difficult load. Since the vast majority of amps works with global feedback and in class AB push-pull and is not designed to handle large phaseshifts of the load (hence large amounts of complex powers) they simply suck and sound crap. More often than not a bad sounding ESL-presentation is due to the limits of the electronics.
jauu
Calvin
ps. The curved ESL was probabely invented and patented 5 minutes after invention of the principle itself...some time before R.Sanders wrote about it.
Mr. P.Walker´s text contain hints on curved panels and there´s at latest a patent from 1956 describing the construction such panels.
Very Good,Calvin !
I would have to quite agree with points that you have mentioned.
I had referred the video because these are the type of things that must be considered when designing a system, as I had mentioned they are very subjective.
For one to claim that one is better than the other may not work in all situations.
I would like to try a curved panel one day as I have heard the ML's and know what they have to offer.
I just don't have the means too make them at the moment.
The narrower panels are less directive than a wider panel.
This becomes a fine balance and compromise in lowest crossover frequency, panel width, diagphram displacement (xmax) and driving voltages.
In any case how they are placed in the room and how the room is setup plays the biggest factor as with any speaker.
Although I have not yet made a TL woofer system I have not had any problems getting them to meld using my more common BR woofer system.
Most say to use a sealed enclosure for the best sound, But I prefer a properly tuned BR as it just sound more full and natural to me.
Lately, I seem to fancy a smaller volume box tuned quite low as with my 8" sub that I used when I First hooked it up ( it is picture sitting on top of my 2 X 10" cabinets).
That woofer system gives a nice clean transient response with about 1db to 2db more on the low bass extension (due to the port) without sounding boomy at all.
Quite impressive for a little 8" driver as it has enough xmax should I want more using a sub bass boost filter.
As room gain alone does a nice job by itself.
They will be the basis of my next sub system using 2 or 3 per channel.
With the setup in the second picture in post two,Once I got the panel set just right, it disappeared and it worked in unison with the little BR 5.25" woofer.
And, Even more so when it was properly EQed to flat as it brought back that magical moment of why I was content to be listening even in mono for all of the summer of two years ago.
Yes the electronics will make or break the system and must be thought out properly.
The smaller panel is nice but it poses a whole another set of design factors.
The 1# being a smaller surface area as Calvin had mentioned.
To make up the difference one must raise the driving voltages.
This means raising the transformation ratio but doing this lowers the impedance seen by the amplifier quite considerably.
For instance yesterday I got my little panel to do nicely at about 98db at .5m and 93db or so at 1m using test tones at 500hz to 2khz with a 10v peak signal into the transformer.
BUT, To do this I had to raise my transformation ratio to 1:330 using a 7 turn primary on my cores.
With the high capacitance of the transformers (at 600pf) my impedance at 500hz was about 4.5ohms and by the time I got to 2khz it was about 1.125ohms.
Little amps don't like this to good.
Much less with .125 ohms at 20khz for this example.
So, the most I could get out of it musically was about 76db to 80db in my listening range distance.
The amp could only produce about 2v peak and no more than 3v to 4v peaks before the overload protection would shut it down.
This is why surface area is so important.
A properly design transformer build would make much better results in this type of situation.
By using a larger panel the efficiency is increased and such high transformation ratio is not required, making things much easier on the amplifier and the loss factor is then greatly reduced from my example of 10:1 to about 1/2 to 1/4 or better.
Just a few more things to consider.
jer 🙂
I would have to quite agree with points that you have mentioned.
I had referred the video because these are the type of things that must be considered when designing a system, as I had mentioned they are very subjective.
For one to claim that one is better than the other may not work in all situations.
I would like to try a curved panel one day as I have heard the ML's and know what they have to offer.
I just don't have the means too make them at the moment.
The narrower panels are less directive than a wider panel.
This becomes a fine balance and compromise in lowest crossover frequency, panel width, diagphram displacement (xmax) and driving voltages.
In any case how they are placed in the room and how the room is setup plays the biggest factor as with any speaker.
Although I have not yet made a TL woofer system I have not had any problems getting them to meld using my more common BR woofer system.
Most say to use a sealed enclosure for the best sound, But I prefer a properly tuned BR as it just sound more full and natural to me.
Lately, I seem to fancy a smaller volume box tuned quite low as with my 8" sub that I used when I First hooked it up ( it is picture sitting on top of my 2 X 10" cabinets).
That woofer system gives a nice clean transient response with about 1db to 2db more on the low bass extension (due to the port) without sounding boomy at all.
Quite impressive for a little 8" driver as it has enough xmax should I want more using a sub bass boost filter.
As room gain alone does a nice job by itself.
They will be the basis of my next sub system using 2 or 3 per channel.
With the setup in the second picture in post two,Once I got the panel set just right, it disappeared and it worked in unison with the little BR 5.25" woofer.
And, Even more so when it was properly EQed to flat as it brought back that magical moment of why I was content to be listening even in mono for all of the summer of two years ago.
Yes the electronics will make or break the system and must be thought out properly.
The smaller panel is nice but it poses a whole another set of design factors.
The 1# being a smaller surface area as Calvin had mentioned.
To make up the difference one must raise the driving voltages.
This means raising the transformation ratio but doing this lowers the impedance seen by the amplifier quite considerably.
For instance yesterday I got my little panel to do nicely at about 98db at .5m and 93db or so at 1m using test tones at 500hz to 2khz with a 10v peak signal into the transformer.
BUT, To do this I had to raise my transformation ratio to 1:330 using a 7 turn primary on my cores.
With the high capacitance of the transformers (at 600pf) my impedance at 500hz was about 4.5ohms and by the time I got to 2khz it was about 1.125ohms.
Little amps don't like this to good.
Much less with .125 ohms at 20khz for this example.
So, the most I could get out of it musically was about 76db to 80db in my listening range distance.
The amp could only produce about 2v peak and no more than 3v to 4v peaks before the overload protection would shut it down.
This is why surface area is so important.
A properly design transformer build would make much better results in this type of situation.
By using a larger panel the efficiency is increased and such high transformation ratio is not required, making things much easier on the amplifier and the loss factor is then greatly reduced from my example of 10:1 to about 1/2 to 1/4 or better.
Just a few more things to consider.
jer 🙂
Last edited:
Alright so I went into design mode during school and came up with a quick rendering. I didn't have time for anything but the stator but you can get the general idea...
What do you guys think about mixing it with THESE speakers above 1000 Hz actively crossed over. I think it could be a good match because those speakers are also line(ish) sources and are pretty flat up to 1000 Hz when some comb filtering stuff starts.
I was thinking because of the higher crossover point I could have a smaller D/S and thin membrane as well as pretty small panels without having to worry about phase cancelations.
I was want to make the ESLs about the same size as the woofers so I went with (tentatively) 4.5" by 18" for 81 in^2 area each. I was planning for putting a horizontal tape spacer each 4.5 inches to make the diaphragm conform to the curve a bit better and still have a good D/S: spacer ratio.
I was thinking 6 micron mylar but if you think I could go less I would love the extra sensitivity
The dispersion from these I imagine would be about 45 degrees (with 43 degrees of an arc radiating sound) which would match the woofers pretty well. What do you guys think?
This is my first design so feel free to point out any issues you see
Thanks!
What do you guys think about mixing it with THESE speakers above 1000 Hz actively crossed over. I think it could be a good match because those speakers are also line(ish) sources and are pretty flat up to 1000 Hz when some comb filtering stuff starts.
I was thinking because of the higher crossover point I could have a smaller D/S and thin membrane as well as pretty small panels without having to worry about phase cancelations.
I was want to make the ESLs about the same size as the woofers so I went with (tentatively) 4.5" by 18" for 81 in^2 area each. I was planning for putting a horizontal tape spacer each 4.5 inches to make the diaphragm conform to the curve a bit better and still have a good D/S: spacer ratio.
I was thinking 6 micron mylar but if you think I could go less I would love the extra sensitivity
The dispersion from these I imagine would be about 45 degrees (with 43 degrees of an arc radiating sound) which would match the woofers pretty well. What do you guys think?
This is my first design so feel free to point out any issues you see
Thanks!
Attachments
I think that 1000 Hz is in the human voice frequency range and as Calvin and others mentioned, you probably shouldn't want to have any crossover in that range.
Well considering that its an active crossover I can just mess around with it and find what works best so I'm not too worried about that. What do you think about the panel design? Is the radius to small or is it doable
- Status
- Not open for further replies.