Introduction: Paper Speakers

Picture of Paper Speakers

This project explores the technical, design, and aesthetic possibilities of 2-D, flexible audio speaker technology. The premise underlying this exploration is the idea that sound can be thought of as a physically immediate, transparent and embodied material. The end goal for me is the use of this material for my art practice: Sound Art, Installation, and Composition. This project includes not just research but also finished art pieces.

Step 1: How to Build a Paper Speaker

Picture of How to Build a Paper Speaker

Step 2: Some Topologies to Explore for Your Paper Speaker

Picture of Some Topologies to Explore for Your Paper Speaker

This figure shows the three basic topologies of circuit design for
paper speakers which I have explored. All three designs can generate a magnetic field capable of driving audio.

Spiral. Spiral designs are probably the most effective at driving audio, everything else being equal. However, spiral design has serious drawbacks: Only one circuit can be created. Once the signal arrives at the end of the spiral, the only effective way to continue with any serial connection would be to break away from the plane, or break into the surface of the plane.

Parallel. By using an open C like structure, rather than a closed spiral loop, it becomes possible to overcome the limitations of the Spiral pattern, in so far as serial connections within the plane can be achieved. The parallel structure also has the benefit, as can be seen in the acoustic drawing, of being highly flexible for expressive effect and figuration, etc.

Sparse. The sparse topology is actually either a spiral or parallel structure in disguise. Any kind of texture can be used to create such a structure by strategically breaking connections within the texture so that current will flow in a desired path. The benefits of a sparse topology are that the actual path of the circuitry can be completely hidden, freeing the surface from specific technical constraints. However, the sparse topology is likely the least effective of the three designs here, since the flow of current would be necessarily inexact.

Step 3: Some Examples of My Paper Speakers

Picture of Some Examples of My Paper Speakers

Step 4: An Alternative Way to Make Paper Speakers Using Visual Images

Picture of An Alternative Way to Make Paper Speakers Using Visual Images

It is also possible to explore photographic material as a basis for flat
audio circuitry design. Using commonly available rubberized ferrite magnetic sheets with a circuitry design maximized to respond to the vagaries of the anisotropic ferrite deposition on the flexible magnetic sheet, photographic half-tone strategies can be employed on the surface, such as in this audio speaker/image of William Burroughs

Step 5: Video of Some Working Speakers

I made some really large tapestries of these paper speakers (actually, these ones are on acetate not paper). I wrote some music for them . . Here they are at the Berkeley Art Museum


pachytrance (author)2015-12-29

I'll try

pachytrance (author)2015-12-29

I'll try

Dinas SuhadiP (author)2015-10-19

Form my observation, the frequency response of this "speaker" will be based on the material used as the backing. These copper foil pattern are acting like the driver coil in the conventional cone speaker. The pattern will determine the load or amount of current can pass through and create enough flux to collide with the permanent magnet nearby the "coil" which in return creates the vibrations which we called "sound".

From the video (which I believe taken using your phone), I realised that there are some drum-like sounds (lower mid range) and clear tinkling (treble range) suggesting that the acrylic you use is about 6mm thickness. Thinner acrylic with the same size will deliver better bass and lower mid range but not very good in the treble range.

I would use the acrylic with the wall mount screw post for the connector terminals and that way, with the space at the back, you can stick multiple strong strontium magnets at the back on the wall itself.

nancyjohns (author)2015-03-07

You called it "2-D" when NOTHING in this three-dimensional world can possibly be "2-D".

mk2001 (author)nancyjohns2015-06-23

You know what he means

Quintel11 made it! (author)2015-04-29


xxlauraxx (author)2015-02-18

These are so neat looking! Gorgeous.

Do you have a photo of a paper speaker attached to a sound source? I'd love to see/hear a video of it in action!

Jess Rowland (author)xxlauraxx2015-03-24

Hi Laura -

Thanks for the kind words - I added a link to a video - - j

hcoyle1 (author)2015-02-24

Would this work with conductive paint or is the copper foil/filament a requirement?

pantalone (author)hcoyle12015-02-24

Great question... That would make it a lot easier to experiment!

Jess Rowland (author)pantalone2015-03-24

Aluminum foil from your local supermarket will also work : )

(but you can't solder to aluminum . . )-


Jess Rowland (author)hcoyle12015-03-24

Yep, a bunch of potential materials could work as long as they have enough conductivity . . most conductive inks and paints right now have too much resistivity . . but I hope that changes soon.

Jess Rowland (author)2015-02-20

Hi - Thanks for the concern. There are a couple reasons I made the sheet that way (after many long hours of research . . .). 1. The fields do not cancel out because they extend only locally within the hexagon structure (they are minute fields of low strength individually) - and 2. The magnets localized to each hexagonal structure are polarized consistently to be in-phase across the entire sheet (i.e north up for clockwise and north down for counter-clockwise). This prevents phase cancellation. There is not intended to be a net field across the whole sheet, only locally for each of the many magnet interaction across the array. Hope that helps clarify. j

dddddd (author)Jess Rowland2015-02-27

Very eye-catching and original, and it can be done with any tessellating shape (triangles, squares), and really, any shapes at all.

What is the resistance of the 16 element speaker whose picture you posted?

Gorgeous build. Very original. Visually evocative of the physics involved.

Jess Rowland (author)dddddd2015-03-24

Hi - thanks - to answer your question, the resistance for all of these structures is very low - around 1 ohm at best. I use resistors in line often to keep my amplifiers from working too hard, or sometimes design the foil to create higher resistance.

jim.katz.3 (author)2015-02-25

The first examples look a lot like the security anti-theft stickers used in retail stores. Could a bunch of those be connected for this project?

Jess Rowland (author)jim.katz.32015-03-24

heh - cool idea - thanks for sharing.

hol640alb (author)2015-02-28

first of all, For you who doubt theese topologies would work, just Putin a strong magnet next to a single speaker wire and jou will notice that the wire shakes in a way that cerates sound.

Second, please share a video of your speaker in work! it would be nice to se!

Goodnslo (author)hol640alb2015-03-18

I agree. I want to see a video.

Jess Rowland (author)Goodnslo2015-03-24

. . on the way - j

goldenskyhook (author)2015-03-09

For those of you going on about the hexagons being "clockwise" or not, they are not spirals! They are grids.

I still don't get what you mean in step 5! HOW do you "apply" the magnet? Do you just stick it to the paper with tape or glue? My sense is that it needs to be slightly apart from the actual paper and coil to leave room for vibration? HELP?

Hi - yes, your intuition is correct - a little space to allow for excursion of the material (vibration) really helps, but it's also a pretty robust phenomenon. You could just place the magnet directy.

ser_pez (author)2015-03-24

Do you think these would work with copper tape?

Jess Rowland (author)ser_pez2015-03-24

Yep! That's a fun and easy way to explore the space . .

Phoghat (author)2015-03-19

Fooled around with some paper plate speakers, but these look quite elegant

Espasio (author)2015-03-15

а где магниты?

sergey.lourie (author)2015-02-27

How did you solve the problem of paper having zero magnetic properties? AFAIK, for a speaker to function, you need a coil for elecromagnetic induction, and a magnet that will move in the inducted magnetic field causing the media vibrations. Of course you can think of magnetostrictive/magnetoelastic effect, but paper, AFAIR, doesn't show any of the such...

bapgill (author)sergey.lourie2015-03-09

One easy solution to the problem you are describing is to buy some disk magnets and fix them to a sheet of stiff cardboard 5 mm or approx 1/4 inch behind the geometric centres of the parallel C sections of the foil. That is locate them directly behind centre of 'the hexagon of each cell. They are not expensive. Do a google to

Hope this helps.

tjfoth (author)sergey.lourie2015-02-28

The author stated that he put a magnet near the coils. Most speakers are made by attaching a coil to paper and having that coil go around a magnet. Effectively, that is what is says he has done here.

tjfoth (author)tjfoth2015-02-28

Please see bullet 5 in step 1.

agis68 (author)2015-03-06

the best experimental instructable about sound speakers ever....may be the winner


gfwhell (author)2015-03-03

This is a very interesting concept. What result could be expected if you constructed "two" using printed circuit diaphragms and placed them "back to back" series connected, the space between them could be "close and sealed" acting as a spring .They would be attracted and repelled by the action of AC current thus dispensing with the permanent magnet

bpark1000 (author)2015-02-25

You make no mention whatsoever of setting up permanent magnets for the currents in the traces to interact with. To be effective, you need a strong magnetic field parallel to the plane of the membrane, and the reversals (poles) of the field need to match up with traces carrying current in the opposite direction, in order for all of the currents to push the membrane in the same direction, perpendicular to the plane of the membrane. For example, if you choose spirals, the spirals should not spiral all the way to the center, and adjacent spirals need to carry currents in the opposite direction (counter-clockwise versus clockwise). The magnet poles would be set to lie directly under the centers of the spirals, adjacent poles being opposite.

tjfoth (author)bpark10002015-02-28

Please see bullet 5 in step 1.

okenssy (author)2015-02-26

How about Gold Leaf instead of copper filling or led filling?

monishgogri (author)2015-02-26

would a pencil drawing of the spirals do, as the graphite in the pencil lead conducts electricity! much easier prototyping!

rroglio (author)2015-02-26

Hy! Great job...

Can you post a working video?


actimm (author)2015-02-25

It occurs to me you should be able to use plate glass for this as well. I wonder how it would sound. Perhaps a stained glass speaker?

mledón díaz (author)2015-02-25

That's a very creative idea yet i'd also like to see it working

dexterelu (author)2015-02-25

I'm assuming any pattern can be achieved by laser-printing and etching the copper. Is there any way you could add a video of your designs?

ahorn8 (author)2015-02-24

Do you have any template for etching the designs in copper? This is an interesting idea for a project but without sharing the complex design I wonder how its a complete instructable.

frostystones (author)2015-02-24

my dad did some R&D at Magna Int. involving a rear window of a car as a speaker. although it did work, there were too many other complications to make it a feasible for the open market.

jsolterbeck (author)2015-02-24

Love it! Need more!

pantalone (author)2015-02-24

Brilliant and beautiful! As with the other commentors, I'd love to hear a sample of the audio these produce. I'd also like to know what method you used to cut the copper foil so precisely, and how you arrived at the hexagonal structure. What were some of the other shapes you tried (besides spiral) and how did they compare? I think a longer instructable would be a great idea!

yzerbeat (author)2015-02-24

That looks amzing, a video with a little soundclip would be nice!

Salvagione (author)yzerbeat2015-02-24


lfoss (author)2015-02-24

woodNfish (author)2015-02-24

I actually doubt your parallel and sparse topologies actually produce any sound based on electrical theory. Every electrical current through a wire creates a magnetic field and the spiral or coil, to be more correct, was designed to create a more intense magnetic field. Your parallel design does nothing more than reduce the wire resistance by creating more paths in parallel and your random, or "sparse" design simply creates a lot of shorts for the current to follow from one polarity to the other.

Your Burroughs image is interesting, but looks to me like it would operate more as a heating element than anything else. You never did say in your instructable if any of your experiments actually worked, and I doubt they do.

tiorbinist (author)2015-02-24

An excellent introduction to the idea that speakers don't have to have round voice-coils wrapped around magnets... but it would have been nice if you had shared your techniques for cutting the thin copper sheet. There are a number of possibilities: milling via CNC, etching with an acid, etc. Obviously, if you are cutting it with an X-acto knife, this isn't going to be useful to the majority of us (who mostly cut ourselves with X-acto knives).

Do you have any calculations you do to determine how much output you will get? Perhaps some guidance on how thick and wide the copper must be to support the current flow? Some consideration to impedance (at least at 1Khz with a 1v signal) would be helpful, since it has a bearing on how useful the sound is. (One example of this would be if one configuration produces more volume than another; the quality of the resulting sound might also be important.)

You obviously do very neat and attractive work, but I'd really like to see an instructable show how to do it, rather than just what was done.

lnorooz (author)2015-02-24

Did you use a laser cutter on these designs? Or are you just that talented? :)

About This Instructable




More by Jess Rowland:Paper Speakers
Add instructable to: