Tragus Implants and You (but also me)

I’ve been putting off making my first install because it felt like I didn’t really have a use case yet and I had a lot to learn before it was viable, but then I saw Rich Lee’s magnets, and then Vicarious’ setup, and I knew I had to have them. That said, I’m a little daunted by what doesn’t quite seem to be working and was wondering if anyone has had any real success with amplifiers yet. The problems seem to be:

-Middling fidelity. That’s kind of unfortunate, but it is what it is. The fact is the poles on any injectable magnets are going to sit up-down rather than in-out, so I’m not sure if that affects the direction of movement or not. I’m guessing it does, and that the fact that any sound comes through at all is a result of the embedding into the skin. Tape, glue, any fixative probably won’t be able to accurately reproduce the effect for experimentation, which is why I want to do this.

-low volume vs power consumption. Again physics is not on our side, but I’ve been doing as much research as I can into voice coils and solenoids. The shape of the coil matters as much to power efficiency as range, so I have some ideas to improve fidelity and power consumption at once. I think the necklace is a nice thought, but I don’t think it’s going to end up being the most effective shape. in the rim of a beanie, on the other hand, 1) allows a rounder shape, 2) gives a place to keep the electronics, 3) brings the range way down, and 4) is less of an obvious article of clothing. I also have a few concepts: either of a high collar with a folded coil (which I found a patent for) like the main character V of Cyberpunk 2077 wears, a jacket with embedded “epaulets” with a small coil under each ear next to the neck, or some kind of vertical coil that can be attached to a seat headrest in a car and send, for example, GPS directions to only the driver.

My last thought is that maybe the answer isn’t to try to move the magnet we have now even more, maybe the answer is more magnets. Not next to each other per se, but is it possible that one could be placed directly under the earlobe, behind the jawbone? There’s not a lot of fat there and the skin moves more than the tragus, but they should be sufficiently far apart to not interfere with each other and cause pain.

I’m still a little way out from being able to much of any research into any of these ideas, but I’m very interested to hear if you think I’m on the right track or especially if you have any experience or information that I might be missing.

1 Like

I’ve been working on this for a few months now. I’m busy today, but I’ll get back to you soon. You’re right about the necklace being too far away, and the hat brim being a good choice.

1 Like

I’m excited to hear! I really thought this would be a bigger thing to more people but the roadblocks seem to put a damper on it.

So let me start out by saying that I’ve been approaching this from an engineering standpoint. The difficulty is that the behavior of fields is pretty high order stuff. Locating and understanding the actual math behind it is rather difficult. That being said, we can feel this problem out and verify by experimentation.

I’ve seen commercial units advertised to help cheat on tests. They use a 9V battery which supposedly lasts about 3hrs. They have a necklace coil. The issue is that those units really only work when the tiny magnet is against your eardrum, so we’re not going to get the same results with a tragus implant because of all the fleshy padding muffling the sound.

Personally, I don’t think think Rich Lee got very good performance out of his concealed necklace unit, and that was using an 18V power source (two 9V in series) and a pretty sizable amp. He might have pulled off a very low mosquito buzz volume level. This was adequate for his purposes, but I want to be able to hear music and sounds with higher volume and fidelity. We’re going to have to get creative about the design. So far I have tried the following:

This 3W Bluethooth Amplifier
Worked well, but the Pk-Pk amplitude of the output was 1.1V, which was very weak with higher resistance coils.

This 160W digital amplifier
Has worked out very well. The Pk-Pk amplitude is directly related to the DC power source, which can be anywhere 8-26V. I’ve been using a DC power supply to run it at 18V.


  • Cannibalized voice coil

    I took this from an existing speaker. It worked pretty well, but was only perceptible up to 3cm away. Best results when the axis of symmetry was pointing toward my ear canal (perpendicular to the poles of the magnet).
    Specs: Diameter = 13.8mm, Height = 3mm, Inductance = 59μH, DC Resistance = 9Ω

  • Self supported copper coil
    For the first try I used 7 turns of 18AWG enameled copper wire. Worked terribly and I have since dismantled it so no pics.

  • 3D printed coil form

    Started with 38AWG enameled copper wire so that I could achieve more windings in a smaller space. I got poor results with this coil, but I noticed that when I held my magnet implant againts the copper things improved. Decided to rewind it with some 18AWG steel wire in there and got slightly better range/fidelity.
    Specs: Copper Gauge = 38AWG, Coil Width = 16.5cm, Coil Height = 20cm, Inductance = 51μH, DC Resistance = 17.6Ω

  • Steel coil form

    These tests led me to try a thick (~3mm OD) steel ring as the coil form. I wrapped the ring in kapton tape to provide a sticky base for the copper to adhere to. I used a middling gauge copper this time to cut down on the resistance. This worked, but not significantly better than the 3D printed coil form. I might try this again with more windings, it was just so labor intensive I gave up early.
    Specs: Copper Gauge = 28AWG, Coil Width = 16.5cm, Coil Height = 20cm, Inductance = 76.5μH, DC Resistance = 1.9Ω

  • Discreet Inductor

    Given that the large diameter coil necklaces don’t seem to work at a distance of more than a few cm, I decided to try other coil shapes. I used this discreet inductor the other day. It is 100μH, and has a DC Resistance of less than 1Ω. It worked okay, but got pretty hot. I’m going to do more testing.

I’m going to keep trying other things, but I pretty much think we have to do a complete redesign, or do a hat-brim coil like you mentioned. One important note when designing the coils is that DC Resistance is not the whole story when it comes to how much current the coil is going to pull from the amplifier. What you really need is Impedance at the target frequency.


You said that the best performance is when the voice coil was perpendicular to your magnet, so that must mean that regardless of polar orientation the magnet will vibrate away from the coil, correct?

I never got to take a physics course, I was on the AP Chem track, so a lot of this math is over my head, but I’m trying. What I’m understanding is that the coil can be tuned/built to resonate for a particular frequency, right? So, for example, I listen to a lot of audiobooks. Vocal ranges are between 85-255Hz. For the purposes of telephony, 300-3400Hz is reserved. So could we target resonance around the middle of that and see improvement in the whole range?

And what does resonance get for us? does it mean efficiency, or volume increase? I guess they more or less go hand in hand.

E: where can you find any information or sources for buying discrete coils or calculating any of the relevant metrics? I can’t search anything without coming up with induction heating coils for metalworking.

does anyone know why tragus implants are safe while we have learnt sensing magnets in the fingertips are not?

sensing magnets in the fingertips are not

What do you mean? They’re not effectively different to the best of my knowledge. In fact they shouldn’t be different from any other implant, they all go in the same bioglass. Caveat being fingertips are much more prone to crush forces and being used, and therefore more likely to break. They just have different use cases. Also the xG3 is much larger than the typical fingertip, so less likely to break

Nevermind, I see what you’re talking about now. It pretty much comes down to mechanical wear over time and use. An encapsulated magnet should not have the same issue as a coated magnet, but is not small enough to serve the same function.

1 Like

The shape of the magnet affects the shape of it’s field. Here’s how the xG3s field looks when it’s installed vertically in your tragus:
Amal described it best when he says that magnetic fields are like bubbles. When you send a signal through a magnetic field, it’s like vibrating the bubble.

Here’s how the magnetic field created by that voice coil looks:
It worked better when the voice coil flux line arrows were pointed perpendicular to the flux line arrows of the xG3. It’s possible this was because the sides of the xG3 field present a more uniform, less turbulent area of the magnetic field to propogate the vibrations. This won’t make as much if a difference for a necklace sized coil, because the flux lines don’t really exist at the center of a coil that large.

If we tune the coil to the correct frequency range, more of the power delivered by the amplifier will be translated into a large/strong magnetic field (inductive), and less of the power will be wasted as heat (resistive).

Electronics Tutorials
All About Circuits
Octopart (for sourcing parts)

Ah, i thought the tragus magnets were the same as the sensing ones (coated not encapsulated)
Good to know, thanks!

1 Like

Just ordered one of those commercial units. I’ll test out how it works and measure the antenna properties where it arrives.

1 Like

This is definitely thinking out loud, but I’m also having trouble using the right terms to find what I want to know.

We’ve been focusing a lot on optimizing coil shape, and while that is important, I think there’s a way we could improve the circuitry. Since resonance can be tuned using trimmers, how could we use the audio input to modulate the resonance frequency on the fly? As in an analog choke, or a varicap/tuning diode, then split the audio signal to the coil and the variable capacitor to keep the resonant frequency at whatever the audio frequency is.

Also thank you for introducing me to octopart, seeing that’s like the first time I walked into the LEGO store


I’m not an audio engineer, so I’m not sure how to do that. My understanding is that most of the signal optimization is handled by the amplifier circuit, and the voice coil is just a passive load which the signal is acting upon. I think we need to overcome the physical hurdles of getting the sound to the magnet implant, then we can worry about optimizing the circuitry to increase sound quality

Here’s the unit I bought

It runs off of two 9V batteries. The amplifier unit is very tiny. It works very well with the included 2mm x 1mm magnets when they’re all the way up in your ear. Surprisingly good sound quality when the coil was up against my ear. Could still make out words when it was tucked away under my shirt in a noisy environment.

I could feel the vibrations pretty strongly with my finger magnet, but could hear absolutely nothing with my tragus xG3. I think the problem might be the sheer mass of the xG3. The tiny magnets are where it’s at.

I’ll want to play with it while it’s functional a bit more, before I take it apart. I’ll let you know more technical specs later like the coil Inductance and the amplifier circuit configuration.

Was actually able to hear it with my tragus xG3 in a quieter room, but I had to press it up against it. Not better results than my steel core coil and 160W amp. I’ll have to try that setup with the tiny magnets and compare.

The coil was very well constructed. Inductance was 3.63mH. here’s the board. It’s just a very simple LM386 amplifier running off 18VDC


I’ve been doing some thinking and very basic experimentation, and I had something I was wondering if it would be possible for you to try: could you get a pair of foam earplugs, the kind you roll to insert, and see what it sounds like with them in? If you can get good sound with them in then I’m wondering if anatomy is the problem. Is behind the ear a viable implant location? Direct skull contact, no cartilage? I noticed that using a basic motor I could hear both a little louder and a higher and wider range in that location.

Greetings people im a newbie here and this thread has caught my attention. Being partially deaf and a hearing aid user (currently use Siemens aquarious if anybody is interested, older stlye aids but preform well in my humid work environment and are easily adjustable for fine tuning by myself with a hi-pro usb and laptop, aswell as Siemens a mini-tek and transmitters for bluetooth streaming directly from phone and other devices)
So as you can imagine the concept of implantable hearing really excites me.

I’m very curious about the suitability of the trageaus as an implant sight and possibility of maybe jaw or skull being better locations

I think back to an experience I had a few years back that perplexed me for many weeks. Whilst at home and alone I often never wear my aids and I started experience a rumbling sound similar to inaudible voices when I was sat on the far end of sofa with one arm outstretched. For weeks this sensation would repeat itself in exactly the same spot and position and for weeks it baffled me.

What was it i thought? Is my hearing maracelusly returning? Has a physic power awakened in me? am I hearing voices from beyond the grave? Belive me when I say every possibility I could think of rational or not was debated in my mind and still no closer to an answer was I.

The most curious thing was if I moved my arm the sound would come and go. And this got me thinking I was experienceng some form of bone conduction that in a specific position my hole body was serving as a giant antenna and vibrating sound directly into my ears.

Eventually I discovered that an old bluetooth speaker I seldom used had been accidentally turn on, probably by my cat and was lying speaker side down under the sofa. Inaudible to me all that time the sound waves from a radio talk show had been vibrating themselves across the floor up the sofa and amplified in my body.

So I guess the point of this story is that wil bone conduction positioning is crucial, where would the optimun location for the xG3? Has anybody tried just attaching magnets at various locations around the skull and seen how they work?

Apologies for the long post but im genuinely fascinated about this and the implications it could have for us deaf folks

Working on it


Ok so this has served very little than my curiosity, but I decided to disassemble an old aid with t-coil capability( morden aids tend not to have it anymore) I belive the small red coil I removed is the used for induction loop ditection. I doubt this is of any use to you but I like taking things apart. Side note neck loops where always a bit dodgy for sound quality despite all the tech going on inside the aids, would often have to hold loop up to jaw to get clarity the big ones built into public buildings are a lot better. I’m curious as to the results of implants in a building situation :thinking:

1 Like

I’m looking at the red piece but I can’t quite tell, is it wrapped in a coil? Is it solid, and if so is it magnetized on its own?