Miniature ESL

This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I would like to build the minimum possible, yet still decent, ESL. I read at Mark's site that the minimum is about 2 sq feet, so I guess that is the size I would like to build one at. I don't need a very loud speaker, just something to listen to music in my room. What are the cheapest possible parts I could use for pretty much everything? Thanks
You'll have to define "decent".

You can make an ESL 1" x 1", and it will produce sound, but it won't play loud, and it won't reproduce much below 5 or 6 kHz (unless you use it to make an earphone).

The transformers to step up the audio voltage are the biggest cost. It doesn't cost much more to make a 2' x 4' driver than it does to make a 6" x 1' driver.
You won't save much money by building a smaller pair of ESLs. You will save some, but I don't think that it is worth it. I suggest that you save up a little bit for a nice sized pair of ESLs. In the mean time I suggest that you read Roger Sanders book, ESL cookbook and start designing. Take your time and make a pair of speakers that you can be proud of.
Good luck
A while ago, I experimented with building electrostatic transducers. At the time I was thinking more along the lines of building a pair of earphones.

In any case, it is possible to make something that is both ludicrously cheap and sounds good. However, you're right for the most part -- it won't produce much below 500hz or so.

Here is a picture of some of the transducers I made:

<img src="">

The sheet metal is just standard lincaine that you can get at Menards or Home Depot. It's actually not very good for this sort of thing but it sounds fine for a cheap DIY experiment.

The spacer is cardboard taken from the back of a bunch of 8.5x11" notepads and cut with an exacto-knife.

The diaphragms are (yes, seriously) saran-wrap. Use the Best Yet brand or something that isn't as hardcore as actual cling-wrap, since you would like it to _not_ cling to your hands or other surfaces when you're working with it. The diaphragm coating material was just key-lock graphite lube from Home Depot, but you can get it at any hardware store. I just coated a single side using cotton balls.

Be sure to measure the resistance with a multimeter (make sure that the whole surface is fairly evenly coated).

The bias / audio supply pictured was built entirely from RadioShack parts. You can use their PA transformers (only a couple watts, so you can't drive these things _too_ loud). You could probably run them in series and get a little bit more output (however, I was already measuring about 300Vrms on the output). The bias was fixed at around 500V. I just used an isolating step-up transformer (to 240V, but my AC line runs high.. so it was more like 250V). From there, just use a standard voltage doubler and voila.

The cost to build all the transducers and all the electronic circuitry in the picture was under $50 and the large panel sounds sooo good for how cheaply it was done. If you were to couple it with a mid-bass, it would do very well as a tweeter in a small-room bookshelf system.

If you ask nicely, some plastics companies (DuPont included) will give you free samples of their various Mylar materials. I was able to get a notebook filled with 8.5 x 11" sheets of lots of different stuff, some of it much thinner than the stuff the ESL information exchange suggests.


My first test speakers were of the very quick and very dirty variety. We used metalized mylar for our membrane and a very poorly spun homemade transformer to step up the voltage for the first one. The spacer was some much too thick fiberglass that we had lying around. The stators were pieces of sheet metal with holes poorly drilled into it. The high voltage power supply was borrowed from the lab in which I work. We were very surprised to actually hear music coming out of the speaker when we fired it up. But, actually our transformer gave more output than our speaker. Not a joke! Our transformer was buzzing loudly enough so that one could hum along with the melody. It actually took us a while to convince ourselves that the speaker was making some sound, too.
This first test succeeded in convincing me that the darn things work, and I was soon investing in better quality components for a pair for myself.

I would like to know how to get samples of Mylar that you were talking about. Maybe if you can pass on a phone number or something.
I think DuPont has an actual Mylar website. I can't remember the address. On it, it mentions getting "free samples." It only takes a few minutes of searching from google if I recall correctly.

Benny -- The 8.5" x 11" panel could do "all but the bass" but my definition of "bass" might be different than yours. If I were going to cross it over with something, I'd most likely do it above 2-3kHz, but I think it would do just fine at about 500Hz or maybe even a little bit below. I built a smaller transducer using some 1mil Mylar and the bass response extended down to 100Hz audibly, but I'm guessing the F3 point was around more like 200Hz.

I've gotten a lot of volume out of an experimental 1 square foot panel that I was using to determine the effect of perforation size on the frequency response. If you want pretty good volume with some bass, build a 12" wide panel with one inch of spacers around the edges. Do not use a center spacer. Use about a 3/32" thick plexiglass spacer. I also suggest 1/8" to 3/16" diameter perforations in your metal. I used Barry Waldron's transformers and bias supply for my main panels. As I mentioned, these were prototypes for future panels.

In order to get the volume out of them you should insulate the steel with spray on automotive primer, which will allow you to use high bias voltages. Also, the power of the driving amplifier will affect your volume greatly. A 12 watt tube amp will hardly put out any sound, while my 150 watt SS amp drove the panel plenty loud.

It seem to me that you can get bass out of the panels by using a free moving area no greater than 10" wide. The center spacer which is often used to support the diaphragm seems to kill your excursion. On a large ESL panel, you need the center spacer for structural reasons, but on a small panel it's not necessary. If anyone has observed similar results, please post them here.

-Matthew Anker
Matthew Anker's Electrostatic Loudspeaker

Good luck!
Does anyone suggest a good supply of Mylar? I've always had dificulty finding Mylar that is wide enough for the speakers that I intend to contruct. I am thinking about building Mark Rehorst's stretching rig, but I would need Mylar that is a little wider than what McMasterCarr is pedeling. Perhaps a 30" wide piece?
I believe I got my 1/2 mil mylar from Professional Plastics. Just search for them on the web. The roll I got was about 30" wide I think. I haven't tried any of the Mylar from Barry Waldron, but this Mylar is very heat shrinkable.

Let me know if they are the wrong place, because my memory could be failing me now. I called so many places for mylar that I may have confused them with another company

This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.