Dear friends,
I have gathered most of the materials needed to build a pair of ELS. I would like to hear from you regarding the spacer material and its thickness. I am thinking of using double sided tape as a spacer and for attaching the diaphragm to the stators. I have now 1 mm and 1.5 mm tapes. Which one should I use?
I will be making the stators from PCB. I will drill holes on them using my home made CNC machine. The stators will be around 40 cm x 50 cm. What hole size would be good - 2mm, 3mm, 4mm or other?
I would appreciate for all your help.
Wachara C.
I have gathered most of the materials needed to build a pair of ELS. I would like to hear from you regarding the spacer material and its thickness. I am thinking of using double sided tape as a spacer and for attaching the diaphragm to the stators. I have now 1 mm and 1.5 mm tapes. Which one should I use?
I will be making the stators from PCB. I will drill holes on them using my home made CNC machine. The stators will be around 40 cm x 50 cm. What hole size would be good - 2mm, 3mm, 4mm or other?
I would appreciate for all your help.
Wachara C.
For the spacers I would use PCB material as well (FR4 is perfect). Tape has trouble keeping the tension and it becomes very difficult if not impossible to disassemble the unit. You can glue the diaphragm to one of the stators and bolt the thing shut (drill some holes through the stator assembly where the spacers are located)
I think the size of 40x50 cm is a bit odd. Do you plan to use electrical segmentation?
The stator spacing depends on what you want to do with the panel. Is it full range or not? A spacing of 1mm should be a good start.
For the hole pattern, holes of 3mm configured like shown in the picture has proven to work rather nicely.
I think the size of 40x50 cm is a bit odd. Do you plan to use electrical segmentation?
The stator spacing depends on what you want to do with the panel. Is it full range or not? A spacing of 1mm should be a good start.
For the hole pattern, holes of 3mm configured like shown in the picture has proven to work rather nicely.
Attachments
Thanks Arend-jan for your quick answer.
I would eventually stack a few of the panels to make a bigger one - something like 4 of 40x50 per side.
So you don't think double sided tape is a good solution. But, I have found mylar to be very difficult to glue with anything. I have a difficult time to find a good epoxy glue here.
When playing loud music on them, would the spacing of only 1mm be a problem? Would the diaphragm ever hit the stator? Or should I put a spacer of 1mm at every 100mm spacing?
Wachara C.
I would eventually stack a few of the panels to make a bigger one - something like 4 of 40x50 per side.
So you don't think double sided tape is a good solution. But, I have found mylar to be very difficult to glue with anything. I have a difficult time to find a good epoxy glue here.
When playing loud music on them, would the spacing of only 1mm be a problem? Would the diaphragm ever hit the stator? Or should I put a spacer of 1mm at every 100mm spacing?
Wachara C.
Yes you should definitely have multiple spacers otherwise the panel will not be stable. Should be fun to experiment with.
Epoxy glue is crap. You are better of with a contact cement (neoprene or water based) like they use for veneer (wood) glueing.
I know there are people that like to use tape, but with a high tension panel this is likely to slacken over time. Personally, I would not recommend it.
Epoxy glue is crap. You are better of with a contact cement (neoprene or water based) like they use for veneer (wood) glueing.
I know there are people that like to use tape, but with a high tension panel this is likely to slacken over time. Personally, I would not recommend it.
Yes, but they are not all equal. You will have to test some to find which work for you. Use a very thin layer of glue.
The best glue I was able to find is 3M 4693 but it is troublesome to get outside the US.
The best glue I was able to find is 3M 4693 but it is troublesome to get outside the US.
I would recomment to make a few smaller prototypes before building a larger one. you will find out that there are quite many problems to solve.
For relatively wide panel like yours , I would use slightly higher d/s spacing (probably 1,5 - 2 mm). But it all depends on what lowest frequencies you want to reproduce , what will be drive voltage, etc.
For relatively wide panel like yours , I would use slightly higher d/s spacing (probably 1,5 - 2 mm). But it all depends on what lowest frequencies you want to reproduce , what will be drive voltage, etc.
arend-jan said:
Like many, I also started out with epoxy glue many years ago. What I did not like about it is that it gives a relatively thick glue film, it does not really stick to the mylar, and most of all that it is a real mess to remove it from the stators. Perhaps with the solvent that it would work better, but I did not try that (did not even know you can dilute epoxy glue). That's why I went out to search for something different. I must have tried at least a dozen different glues.
What I want from the glue is:
1. a good bond
2. a thin film
3. easy to remove if you have to start over again.
4. no time pressure to attach the film
5. if possible, no health issues
With super glue and primer I did not get a nice fill of the gap and it's nasty stuff to work with (irritates the eyes and respiration). Most other glues that I tried give a thick and uneven glue film, also not good. I tried some proprietary glues for ESLs but they gave me time pressure because the glue had already dried up before I got the whole stator covered.
With the 3M 4693, I get a really thin glue line (almost invisible) and a very good and instant bond. No cure time, just rub over the film once (with my stretcher I have the stator below the film so I can see and manipulate the glue line) and you can cut it loose. I only put the glue on the stator, not on the film.
The instant bond is also an advantage if you want to build as many panels as I do.
The film is not hard to remove if you have to do it all over again, and because the glue line is so thin you only sand it a bit with a piece of paper and you are ready to put on a new diaphragm.
Another glue that works well is LePage Press-Tite Green, a water based contact cement. But the 3M is my favourite.
What I want from the glue is:
1. a good bond
2. a thin film
3. easy to remove if you have to start over again.
4. no time pressure to attach the film
5. if possible, no health issues
With super glue and primer I did not get a nice fill of the gap and it's nasty stuff to work with (irritates the eyes and respiration). Most other glues that I tried give a thick and uneven glue film, also not good. I tried some proprietary glues for ESLs but they gave me time pressure because the glue had already dried up before I got the whole stator covered.
With the 3M 4693, I get a really thin glue line (almost invisible) and a very good and instant bond. No cure time, just rub over the film once (with my stretcher I have the stator below the film so I can see and manipulate the glue line) and you can cut it loose. I only put the glue on the stator, not on the film.
The instant bond is also an advantage if you want to build as many panels as I do.
The film is not hard to remove if you have to do it all over again, and because the glue line is so thin you only sand it a bit with a piece of paper and you are ready to put on a new diaphragm.
Another glue that works well is LePage Press-Tite Green, a water based contact cement. But the 3M is my favourite.
Dear Arend-Jan and Bazukas,
Thanks again for all your advises. May I know what spacer thickness you are using and what is the size of your panel? What is your bias voltage?
I am looking to build a full range ESL. What is the low frequency response I can get if I were to use 1mm, 1.5mm or 2mm spacers? I understand that with a bigger gap between spacers and stators, the diaphragm can vibrate more aggressively and therefore the low frequency would be better. Do I understand it correctly? I have got a 6 micron mylar film which I will be using as diaphragm. Do you think it will be good enough?
I appreciate very much for all your comments.
Wachara C.
Thanks again for all your advises. May I know what spacer thickness you are using and what is the size of your panel? What is your bias voltage?
I am looking to build a full range ESL. What is the low frequency response I can get if I were to use 1mm, 1.5mm or 2mm spacers? I understand that with a bigger gap between spacers and stators, the diaphragm can vibrate more aggressively and therefore the low frequency would be better. Do I understand it correctly? I have got a 6 micron mylar film which I will be using as diaphragm. Do you think it will be good enough?
I appreciate very much for all your comments.
Wachara C.
A full range panel is very hard to do well. There are many factors, too many in fact to explain in a single post. I agree with bazukaz, it is a good idea to build a small panel first and experiment with it so you get to solve the basic problems and get a feeling for diaphragm tension, voltages etc.
If you do a search there have been many posts on the subject. Sorry but there are no single best values, it's all compromises.
If you do a search there have been many posts on the subject. Sorry but there are no single best values, it's all compromises.
Hi Arend-jan,
I have actually made a set of electrostatic headphone already. So, I'm not totally new with building ESL. I am gathering as much information as possible - so that, hopefully, I will make it right the first time.
It is my goal to finally make a full range one. I know it won't be easy.
Wachara C.
I have actually made a set of electrostatic headphone already. So, I'm not totally new with building ESL. I am gathering as much information as possible - so that, hopefully, I will make it right the first time.
It is my goal to finally make a full range one. I know it won't be easy.
Wachara C.
chinsettawong said:
Hole size of 2-4 mm should be more or less ok, but keep in mind that you will need quite much open area(somewhere around 30 per cent or more).
I don't know how your CNC home-made machine looks like, but I think that making stators properly in this way should be very difficult. Is this machine accurate enough to drill so many holes close to each other in a such precise manner ? How will you insulate stators to prevent arching ?
So wide panel will be very directional, do you plan to use segments to reduce this effect ?
Hi Bazukas,
My CNC machine is very accurate. I would say error is within 0.01 mm. Drilling holes like these should be very easy. I have done it with my headphone's stators.
I will use lacquer paint to insulate my copper side of the stator. Since the other side is fiberglass, I don't think I need to insulate it.
When you say wide panel, how wide do you consider it a wide one? I really have no idea how directional it will be. I think I will just build 2 panels to try out first and see how it goes.
What is the dimension of your ESL?
Wachara C.
My CNC machine is very accurate. I would say error is within 0.01 mm. Drilling holes like these should be very easy. I have done it with my headphone's stators.
I will use lacquer paint to insulate my copper side of the stator. Since the other side is fiberglass, I don't think I need to insulate it.
When you say wide panel, how wide do you consider it a wide one? I really have no idea how directional it will be. I think I will just build 2 panels to try out first and see how it goes.
What is the dimension of your ESL?
Wachara C.
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
chinsettawong said:
6 micron will do the job, 4 micron will perform even slightly better but for the time being you are safe with 6 micron.
For full range you'll need at least a 2 mm spacing. You have to think about damping the fundamental resonance. Small spacings will force you to control the resonance.
Beaming is one of the biggest problems of esls. It occurs when the wavelength of the emitted sound is aroud the same or smaller than the width of the mylar.
Hi MJ Dijkstra,
Thanks for your information.
I do not quite understand your mentioning of damping the fundamental resonance. Could you please explain what it means and how to make the small spacings to to control it? Do you mean to place a little space between panels?
About beaming issue, do you have any idea how to improve it?
Wachara C.
Thanks for your information.
I do not quite understand your mentioning of damping the fundamental resonance. Could you please explain what it means and how to make the small spacings to to control it? Do you mean to place a little space between panels?
About beaming issue, do you have any idea how to improve it?
Wachara C.
chinsettawong said:
The fundamental response can be as large as 10 dB or even more. It makes bass ill-defined. By damping the resonance you'll get more control and as excursion decreases smaller spacings can be used. In case stators made of perforated sheets you could consider gluing a fine mesh or cloth upon the stator. As the cloth is fixed it will act as a true air resistance. ( a speaker grill cloth wrapped around the esl will simply move away and is less effective)
Beaming can be reduced by electrically segment the stators; simple thing in case of a wire-stator, not in perforated sheet.
Actually, if you're using PCB as the stator material it should not be too difficult to do the electrical segmentation. You must etch or cut the copper conductor so that you end up with several vertical strips or segments of varying width. For example you could have one narrow (3 cm?) segment that handles all frequencies, including the highest ones; a second segment that's wider (maybe 15 cm?) whose bandwidth is limited to those frequencies below 2 kHz; and the remaining wide segment would be limited to only the lowest frequencies.
The usual way to attenuate the high frequencies is to put a resistor in series with the ESL panel so that you have an RC circuit.
Few
The usual way to attenuate the high frequencies is to put a resistor in series with the ESL panel so that you have an RC circuit.
Few
Hi Few,
Thanks for your input. Do you mean that I should cut the copper on a PCB into 3 different zones? Or do you mean I should do 3 different pieces with different width on each of them?
Wachara C.
Thanks for your input. Do you mean that I should cut the copper on a PCB into 3 different zones? Or do you mean I should do 3 different pieces with different width on each of them?
Wachara C.
MJ Dijkstra said:
Hi MJ Dijkstra,
I have given it a thought about your comment of "beaming can be reduced by electrically segment the stators; simple thing in case of a wire stator". What if I CNC my stator to imitate the lines like those on a wire stator. Instead of drilling the PCB with holes, I can cut the lines of 2 or 3 mm running the whole length of the PCB easily. In that case I can also cut out some copper and make it into different zones as you and Few suggested. Do you see any advantage of doing it this way?
Wachara C.
The advantage of wire does not come from the orientation but the fact that you can hook each wire up with series resistors to create the electrical segmentation.
To get the same effect with the circuit board you can etch the copper to create electrically separated zones (strips). I would create vertical strips and the high frequency strip should be 3cm wide or less. If you want to do it really nice you would put the smallest strip in the middle, then flanked with two wider strips on both sides (for the middle/low frequencies) and so on.
To get the same effect with the circuit board you can etch the copper to create electrically separated zones (strips). I would create vertical strips and the high frequency strip should be 3cm wide or less. If you want to do it really nice you would put the smallest strip in the middle, then flanked with two wider strips on both sides (for the middle/low frequencies) and so on.
Attachments
- Status
- Not open for further replies.