The Context

I am a hobbyist with no education in electrical engineering. I am working on a DIY project trying to make water resistant rechargeable bicycle lights. I am doing this as a hobby for fun and learning so it doesn't need to be commercially viable or to pay off. The device will be cast in silicone and will be completely covered so it needs wireless charging. Because of the size of it I cannot use any coils or wireless charging solutions I can buy, so I am attempting to make my own. As far I have come up with etching the charging coils in layers on copper foil that seems to work for me surprisingly well(as far at least):

Etched copper coil

The Problem

I understand I could improve the energy transfer by having many layers close to each other and the right AC frequency and I am trying to improve there, but as I understand having ferrite plates beneath the coils could improve things dramatically. Unfortunately the plates need to be very specific size and shape for my project and it is not practical to buy or order them. I am looking for a way to make something similar at home. My main question is: what are my options, if they exist at all? In this context it doesn't need to be perfect, it needs to be better than nothing.

Research as far and ideas

I have looked into making ferrite at home and found this: Make Your Own Ferrite to Improve Magnetic Fields, however on further reading about making ferrite I got the impression that crystals of iron oxides and a small amounts of some other metal need to be formed so I started to doubt that this really works without high pressure and temperature. Could someone confirm if this DIY method really works or is this nonsense?

I also have an idea to buy ferrite beads, crush them into powder and then cast my needed forms from that using some adhesive like silicone or PVA glue. Would that work?

Are there any other options?

  • \$\begingroup\$ I don't know if your coils are resonant. Resonance significantly helps transfer efficiency. \$\endgroup\$
    – Russell McMahon
    Jan 16, 2020 at 11:01
  • \$\begingroup\$ @RussellMcMahon could you maybe expand on how one would go about designing the coils in such a way that they are? \$\endgroup\$ Jan 16, 2020 at 12:31
  • \$\begingroup\$ "seems to work for me surprisingly well(as far at least):" what efficiency are you getting now? \$\endgroup\$ Jan 16, 2020 at 17:26
  • \$\begingroup\$ Oh no - I mean at this point I am really happy I manage to actually physically make the coil. :D Enjoying little things in life :D. However that means I am able to change its shape, the thickness/amount of turns in a layer, gaps between them and the amount of layers(so the total amount of turns). I think there is a chance it might work well because they are quite thin so the layers are going to be very close to each other. \$\endgroup\$ Jan 16, 2020 at 19:19

2 Answers 2


Are there any other options?

I would recommend you look at flexible ferrite sheets such as these: -

enter image description here

The relative permeability (\$\mu'\$) of a lot of them are good: -

enter image description here

If you are looking at 13.65 MHz (a standard) the purple graph indicates it has a relative permeability (\$\mu'\$) of over 100 with losses (\$\mu''\$) off the bottom of the graph.

Just google "ferrite sheets" and you'll find several different suppliers (MEC, Laird, Wurth, Ferricor) to whet your appetite.

Could someone confirm if this DIY method really works or is this nonsense?

I really have no idea about this but, in particular, I would be concerned about high frequency performance. 13.65 MHz is a standard frequency used for wireless charging and that might be ruled-out without the correct ferrite dust.

  • \$\begingroup\$ Thanks a lot! I will have a look at what I am able to buy where I live. I don't think I will have very high frequency. I thought it is usually around a few hundreds of kHz in chargers. That's what i find if I google 'wireless charging coil frequency': "The Qi specification calls for an AC frequency in the primary coil of between 110 and 205 kHz for the “low power” Qi chargers (up to 5 W) and 80 and 300 kHz (up to 120 W) for the “medium power” chargers." Isn't that adequate? \$\endgroup\$ Jan 16, 2020 at 10:53
  • 1
    \$\begingroup\$ I've designed one that operated at 600 kHz for a very bespoke application. It could pass up to 10 watts with ease but the load was only 5 watts so I didn't push it further. Operating in the hundreds of kHz is going to be beneficial for the power transmit frequency and the MOSFETs you will possibly use. I mentioned 13.65 MHz because it is used a lot but operating sub 1 MHz means you can use material M6 (\$\mu'\$ of over 200). \$\endgroup\$
    – Andy aka
    Jan 16, 2020 at 11:31
  • \$\begingroup\$ I haven't worked everything out, but I think I will not need that much power - it's a simple signal light, not for illuminating the road or anything, its going to have 190 mAh 3.7 V battery only and charging times do not need to be very fast. I am planning to use battery charger IC for 500mA. So as far as I understand things are looking good as far. Thanks a lot for the help! \$\endgroup\$ Jan 16, 2020 at 11:50
  • \$\begingroup\$ @MartynasŽiemys Is this question answered now? \$\endgroup\$
    – Andy aka
    Jan 20, 2020 at 20:22
  • 1
    \$\begingroup\$ Ferrite is generally 0.4 teslas across the board of all ferrite products. However, those sheets will never likely be used in a closed-magnetic situation so, the peak flux they will see should be minimal. \$\endgroup\$
    – Andy aka
    Jan 22, 2020 at 14:23

Ferrite is just ceramic tile. Order some over-size pieces, then get a tile cutter kit, or rent a tile saw?

  • \$\begingroup\$ I have a tile saw :D Awesome. I didn't think they come in big sizes, but yes, I could even use bigger plates for charging if I manage not to break them completely while sawing(I think that might be an issue, they are quite brittle). but that's an option worth exploring. Thanks. \$\endgroup\$ Jan 16, 2020 at 10:47
  • \$\begingroup\$ @MartynasŽiemys if you can saw ACROSS the surface, then you can saw a gouge, then do the score-and-snap trick. (Or even make a slot with dremel diamond wheel, then snap it.) More like glass-cutting. You ARE using high-freq resonance, right? Not just a simple broadband xfrmr? Resonant transformers (with capacitors across their coils) have immense coupling. That's why wireless power links always use them. Then, use litz wire, or double/triple-up the coil (or, stack several flat coils, wired in parallel.) For resonance, you need high-Q, low-R inductors \$\endgroup\$
    – wbeaty
    Jan 16, 2020 at 19:18
  • \$\begingroup\$ It is a hobby for me. I am a multimedia designer by profession and my knowledge of the subject is extremely limited at the moment so I am learning stuff as I go. I don't know what 'broadband xfrmr' means or what is considered to be high frequency, but I just learnt about the existence of resonance here so yes, I will aim to use that as well as I am able, but the frequency I am hoping to work for my purpose is only a couple of hundreds of kHz. \$\endgroup\$ Jan 16, 2020 at 19:33

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.