Category Archives: Main La Folia Site

The Aluminium L-Profiles (plate support)

You yourself will have to figure out the L-Profiles you’ll need and their specs.

Remember to include the aluminium profile’s width when you are figuring out the hole for your perforated plate.

Also take into account that this is the solution I found to the design problem I encountered when I chose to use a very thick wood plate as the frame for the loudspeaker.

Your requirements may be completely different, this is part of the fun. You design and adapt the project to your conditions.

I would now recommend going for a thinner wood plate, more like the original design. The air moved by the thin membrane and its own intrinsic weight does not justify using such a thick piece of wood, plus the plate itself adds to the weight and sturdiness of the frame that supports the membranes.

Once you have the magnets and membrane, choose the wood thickness so you can sandwich the magnets between the membrane and iron plate directly on the wooden frame.

Contact: Magnetostatic Loudspeakers

The Magnets

Magnet polarity

Figure 1

This, I feel, is the hardest part of the project (ON YOUR POCKET :-)), invest a large amount of money on a bunch of ceramic rectangles, for a project you still aren’t completely
sure will work…

On top of it all, the project gives you very little help with respect to the real life specifications they need to have. I had the “luck” that a Physics Instrumentalist friend of mine casually had 500 magnets of more or less the specs I needed lying around unused… after having callibrated some special astronomical CCD camera and was willing to give them to me for free…!

That they were a bit weak in magnetic force was no problem for me. I just planned to put the membrane closer to the magnets, 2mm instead of the project’s original 5 mm distance. I had two different types of magnets, one type was plastic, sort of the type you use for your refrigerator, 5.0 x 1.2 x 0.6 cm, of those I used 38 per loudspeaker (76 in total) for the tweeter section. And for the woofer section I used aprox. 150 magnets per speaker, of a ceramic type, 2.5 x 1.9 x 0.4 cm, of a much lower magnetic strength than the others. I recommend you find long and thin rectangular magnets, that follow the diagrammes specs with respect to the plane of magnetization (Fig. 1).

Figure 2

I have not been able to find magnet distributors in Chile…. but I have not tried too hard either. I hope you do find them where you live. The original project recommends rare earth magnets due to their strength and due to the fact that their magnetic fields are not strongly affected by the presence of other magnets close by. I will in the near future include more info on this subject.

Fig. 2 is a diagramme showing how to place the magnets on the perforated metal plate, notice the fact that the first row on one plate is the contrary of the other plate, this relates to the tweeter magnets too.

A diagram showing the magnets with respect to the membrane conductors

A list of materials needed.

The construction recipe

Contact: Magnetostatic Loudspeakers

Construction materials for the loudspeakers


  • Hammer
  • Screwdrivers
  • Drill
  • Ruler
  • Cardboard Knives
  • Multimeter
  • Pliers



  • Condensator 105 microF.
  • Coil, 0.95 mH.


  • Circular Saw
  • Oscilloscope
  • Tone Generator
  • Magnetic Field Meter
  • High Power Amplifier

Contact: Magnetostatic Loudspeakers

The Ribbon Speaker Membranes

woofer-bThis is the crucial part of the project, the “bass section” and “tweeter section’s membranes. I put them between quotation marks because they are neither, the cutting frequency is 1000 Hz (or is it 500 Hz…?), and the filter is of a very low slope, so they overlap a considerable amount with respect to the frequencies.

The membrane is at the moment the most time requiring part of the project, as it is designed. I will give, at the end, a series of possibilities as to how this could be shortened. It is
also the part that requires most care, patience and “abilities”, but these will develop during the building and design.

serpentiThe membrane consists of three parts, aluminium foil, a plastic support membrane and adhesive glue, that binds the two.

The plastic support in the original project was an oven polyester bag slit along the edges, giving you a strong, very much, non-stretchable film at a very low cost. The negative sides were the limited size, the crease at the middle, that can even have a discontinuity cut and the relative thickness of the film (and some oven bags are humidity sensitive to my horror!).

I have chosen Mylar Film, something I was not able to find in Chile, but as I casually was in Buenos Aires, Argentina, I bought in the most unlikely shop, in a very adequate width of 12 micrometers (micra), 1 m. width and 15 m length, material for a big amount of loudspeakers… Some builders do not get to stuck up on the width or weight of the materials used on the membrane, with the argument that “the weight of the air moved by the membrane is so high that a few grams more or less on the membrane will be of no effect, I have no way of supporting or counterargument this, I chose the lighter and thinner materials, that give you a very flexible and non rattling membrane, a problem that arises when you have thick aluminium and a thick polyester material.

Here in Chile, and after I ripped the second membrane I was making (!!!%$$~@, booo hoooo), I was forced to find some more membrane material for the first full size prototype, so I looked
hard and found what can be a very good solution for those that can’t find Mylar film, stickers…. Stickers, in Chile at least, are mostly made of Mylar film with adhesive gum on the backside.

I found they sold the prime material for stickers in 1 x 1 meter sheets, just the right side for the project’s loudspeaker membrane, and then you need not worry about the sticking of the aluminium, the mylar already comes with it.

What about the aluminium foil on the polyester membrane? How do we get that and how do we stick it on? What width must it have? How do we cut it?, etc.

I’ll start with the glue. In the original project the aluminium foil they use is a foil used as thief alarm conductors for windows, a 30 micra wide foil, that included glue on it’s backside.

In my design, I was required to stick the aluminium foil on the Mylar, as some other designs do too. I recommend you use a 3M, Super 77 Spray adhesive, this permits a thin glue layer that is strong and durable. You can use other glues but remember to dilute them well, so you can spread them on easily, and thinly.

With respect to the foil, some have chosen to use aluminium foil, cut in long bands. I have chosen an easier route, I have used a special paper that includes a thin aluminium foil on waxed paper, here in Chile, this paper is used to pack chocolates here in Chile…. I do not know if this is universal… This paper has the characteristic that the foil comes off the paper if you
heat it and liquify the wax. Thus you can leave a very thin foil (aprox. 12 micra) on the polyester and eliminate the heavy paper.

How to place the membrane on the frame

This bi-layer permits an easier cutting of the membrane design. You cut the design directly on the foil, avoiding cutting through the base paper, thus the design is held together by it.

What about the cutting, you have to leave a 1 mm. space between each conductor! You may then say, “How are we to cut so precisely!”

Easy: You use a double sharp bladed “knife”, we call them in Chile, cardboard knives, sharp blades, wide and easy to handle, you put two together with a coin between, or something 1 mm. wide, scotch them together and start cutting! With care, obviously, but deep enough to cut well through the foil, once you do so it is easy to pull off the thin 1 mm. wide aluminum foil string from the waxed paper, you must be careful doing this, if the foil isn’t completely cut you may pull a piece off the conductor too, slowly, slowly, that’s the way.

In my opinion, this is the easiest way.


  • Membrane material: Mylar Film 12 micrometers (0.012 mm)
  • Alu-foil: approx. 12 micrometers (0.012 mm) Total width: 0.024mm
  • Membrane is not corrugated, just stretched out as is.

The original La Folia Magnetostatic Loudspeaker recipe (plus modifications)

In its original design, the loudspeakers are very
ineffective (70-72 dB SPL).

The most complicated part of the whole project is to get the materials and also depends on how much time you use experimenting with magnets,
the membranes, the MDF
wood frame
, the membrane design, etc.

The basic idea behind the project is a concept used by other very famous speakers in the Hi-Fi World: Stratshearn, Fostex,
Magnepan & mostly Apogee. It is a simplified version, in many aspects, of Apogee’s two-way speakers. Carver together with a friend of the original Danish writer of the project, Ole Thofte, have produced a speaker with the same principles as in the “La Folia” project.

Step by Step

1- Cut the wood pieces, one the mirror image of the other.

fastgori2- Resonance damping of the woofer hole of the MDF plates, just the top and bottom side of the hole, don’t choose a very thick sponge, it’ll be very hard to stretch the membrane if its to thick, and wrinkles will be an unavoidable part of your design, not so nice…

Cut the tweeter hole taking into consideration the magnet’s dimensions and the metal plates that cover the inner side of the hole, and so on.

3- Prepare the metal plate design, cut a piece either to fit the woofer hole or to be screwed directly to the back of your MDF Plate, you find out what is best for your particular case. The tweeter hole needs a piece on the back side too.

4- Put the magnets on in the desired design, all with due consideration to the conductor design you’ve chosen, the length of conductor you need, the resistance, the impedance (something I never considered, I just measured the Ohm resistance, and calculated it to be 3 ohm). If you’re not as unlucky as I to have very weak magnets, there should be no need to stick them on to the plate, their magnetic strength should be more than enough.

5- Design and create the membranes, more about that in the membrane section, and the possible improvements of the design. SWEAT, SWEAT, this is the tough part, so I feel…

6- The row of magnets in the tweeter section closest to the woofer must have the same polarity as the top row of magnets in the woofer section (important!!). The tweeter magnets are double, boosting
the magnetic field strength.

design7- The conductor design: This speaker system uses the collective principle. The current must flow in the same direction for all conductors in the same magnetic field.

In the case of the woofer conductor, it’s necessary to have an unbroken conductor, that covers the membrane, this way it moves as a single flat surface.

This design has two conductor strips, placed in series. The way the conductors transmits the current is always in the same direction and this creates a force that is either inward or outward, depending on the current direction with respect to the magnetic field. (In my project I used approx. 68 windings of the conductor). The total length of the aluminium conductor in the original project was 18 meters x 9 mm. (30 micrometers in width) and had a 3 ohm resistance.

tweeter-bThe tweeter conductor was originally built with four conductor coils, this makes the resistance larger, the conductor is 9 meters long and half the width, 4.5 mm, thus 3 ohm resistance.

The membrane should be corrugated, (now that I think about it I don’t know why…) thus the length shrinks aprox. 10%. Corrugating it will certainly avoid rattling and it tends to move too when the woofer is reproducing lots of bass.

8- Crossover and speaker connections: The woofer and tweeter are connected in parallel. Serially connected to the woofer you put a coil (0.95 mH), and serially connected to the tweeter a capacitor, 105 microFarad. The speakers impedance is approx. 3-4 ohms, very frequency linear.

9- The speaker feet, build them either out of the same wood, held by strong shelve angles or build a more sophisticated design.

10- Check if things are alright. Check out if there should be any short circuits with the multimeter, check if there are any missing connections too. the resistance measured at the speaker connections to the amplifier should be approx. 3,5 ohm. Now
connect the amp… See what can happen, if all is well
I assure you you’ll have a hard time erasing the wide and shiny
smile you’ll have on your face, once you’ve constructed such a beauty out of such lowly materials, you’ll need to do a number of adjustments to the sound balance, I had no spectrum analyzer, I just used my ears, the best instruments I have at my disposition, adjust the balance between the woofer and the tweeter by putting more or less resistance on the tweeter, this will crave either more or less current from the amp, thus having more or less treble, balancing the sound image accordingly, I know this is quite makeshift, but so is the
whole loudspeaker project!! :-).


My Personal Experience building the ribbon speakers

Once in Chile (1990), I remembered a Danish Loudspeaker

project I had not been able to do there. I asked my ex-girlfriend
in Copenhagen to send me as many La Folia articles as possible,
which she promptly did, thank you Helle.

I was a bit skeptical to begin with, I checked it out with diverse experts more related to the electronic environment
receiving contradicting opinions, I decided to follow the counsel given by a close friend, “just do it, don’t worry about the details, deal with them as you go along”. You can find a million reasons to not do it, if you want to, there are just as many reasons to do it too.

I started building mini models just to try out the principle and get an idea of it’s construction, my first model used a cardboard rectangle, kitchen plastic foil and chocolate paper aluminium foil (here in Chile, almost all chocolates are wrapped in a waxed paper covered with a thin aluminium foil, in Denmark they use this paper to cover their sandwiches) as a membrane, some spray adhesive glue and a few flat magnets, I checked the systems resistance and then I connected it to my always willing “Power Plant” NAD 3300 Amplifier, I assure you many others would have shorted with something like that  connected but no, out came some beautiful mids and highs, enough to stimulate a few more models of higher and higher difficulty level ending with two wood framed 30 by 20 cm full-tone models, with pedestals and all, these sounded awfully sweet, even with a bit of low frequencies, giving me a fleeting glimpse of what was to be the reward of my work, if I scraped enough money to get all the materials required.

Just as the Danish project documents said the most time requiring process was to get a hold of all you’d need to build your hi-fi dream speakers, in Denmark it’s not so big a problem, just follow the project instructions, but when you live on the other side of the world, you have to find your own sources, so will you, but I will give some help as to where you can find the right material.

There is also the need for a perforated iron plate.

More info on the membrane

Contact: Magnetostatic Loudspeakers

The perforated iron plate (holds the magnets in place)

The magnets do not float in mid-air, they are held by their own magnetic force on an iron plate, as the loudspeaker works as a dipole, emitting sound pressure in both directions, the plate has to be perforated, thus permitting the sound production and avoiding destructive interference.

The plate is cut into pieces, one big piece for the bass baffle, in my case 93 cm. in height, 29 cm in the top part and 34 cm. in the widest part, see diagramme. You also need a piece for the back and sides of the tweeter hole, thus a long 93 x 5.5 cm piece, with long pieces to cover the sides of the tweeter, also permitting the adherence of the tweeter magnets, 2 pieces of 93 by the width of the MDF plate
you buy per loudspeaker.

First of all, you can choose plates that range from paper thin to very thick. I chose 3mm. width, this was as wide as possible mostly with respect to my pocket ($$…) and wider than the original project (1.25 mm), 50% air (amount of perforations).

A thicker plate makes the construction more robust, with less propensity to vibrations, makes the loudspeaker all the more sturdy and would theoretically help to achieve a better overall sound.

The wider the plate, the tougher it is to find a place they can cut it. I
had a terrible hassle finding a pneumatic guillotine that could
cut the plate, on top of it all, the plate weighs a lot, a car is
a must, or taxis if they are affordable in your country. They normally
sell the plates as one piece, 2 x 1 meter, at least in metric countries,
you can make four back pieces and the rest of the magnet supports
with one plate, so get together and buy the plates as a pair.

I would recommend from my own experience, to give
the plate an antioxidant treatment, paint or varnish, so your loudspeaker has a nicer appearance and lasts longer. It is also good to get the edges polished after they are cut by the guillotine, they can be very sharp and dangerous to work with, if they are not dulled
a bit.

Material list

Construction recipe

Contact: Magnetostatic Loudspeakers

The Tweeters

The magnets are held on to the tweeter hole in the MDF frame by their own magnetic force, and the fact that you have long strips of metal plates covering the sides
of the hole (don’t you….???!!), thus the north pole of the
magnets are on one side and the other side holds the south
pole magnets.

This is done so the top row of magnets on the woofer section of that speaker must have the same polarity as the row of magnets in the tweeter section closest to the woofer section (sounds wierd but it’s simple, don’t worry on what exactly is the N and the S polarity, just assume that one is the N and the other is the S and work based upon that assumption.

When this is done, you are thus sure
that the speakers are working in fase.The magnets are arranged as shown in the figure. In my case, as in the Danish original project, two
magnets are placed on top of each other to generate a stronger magnetic field. The distance between the magnets on the same side is obvious, because they will not stay any closer than they want to (their magnetic field wants to).

The personal dimensions involved will depend on the material you use, just make the distances such as to have a 1 mm. distance between the tweeter membrane and the magnets, on each side.

This diagramme shows the returning conductors made out of the aluminium material , I chose to use wires and make the tweeter conductors not as wide as the original project. You experiment and find out.

A list of materials needed.
The construction recipe

Contact: Magnetostatic Loudspeakers