Induction Charger Design

On a sunny day (Fri, 13 Nov 2009 08:02:02 -0500) it happened Wingsy wrote in :

Why not use 2 halves of a pot core?

----------------------- | ---- ---- | | | | | | | object 1 | | | | | | -- ------ -- -- ------ -- | | | | | | | | | | | | object 2 | ---- ---- | -----------------------

ed

I wonder if you might clarify the geometry a bit. As I understand it you've got two air coils and then just place one in the center of the other. Both along (let me call it the Z-axis). The Z-axis is the long axis of the device with the pick-up coil in the bottom.

So my silly thought is that you are wasting a lot of the magnetic field. You will want to concentrate it through the pickup coil. How about if the base has a "C" shaped piece of iron with the primary coil wrapped around it with field now along the X-axis, and then the pickup coil slides into the open part of the "C". You might even add a bit of iron to the pickup coil.

George H.

(I've never designed induction pickup coils...)

Run the frequency up to about 400 kHz. Make the coils resonant at that frequency (LC). Get the two coils in as close proximity as possible. Use ferrite cores if necessary to confine the flux path. Insure that there is NO metal, conductive plastic, PC clad, etc. in the vicinity of the coils that rob energy by eddy currents. Surrounding the device with the drive coil may be problematic in this regard. If all else fails put a couple of contacts on the base of the unit that pick up mating contacts at the bottom of the rectangular cavity. That's the way my cordless phone charger works, why re-invent the wheel?

No room. The space for the secondary is only .25" in one dimension. I'd think that a pot core of only .25" diameter would be so small as to be ineffective in coupling power over the distance needed. But it might work. I'll consider giving that a try. Thanks.

Correct.

I kinda like that, but it's similar to what I already tried (2 "C" cores with their ends adjacent to each other, with 1/8" separation.). It's certainly worth trying a different geometry. Thanks.

Well, obviously if I had my preference I'd simply use electrical contacts and forget all this induction stuff. Electrical contacts have been voted down since this pendant is in a waterproof case and is subject to be immersed in water at times (i.e., when taking a shower and wearing this thing around your neck).

And, I've tried frequencies up to 2 MHz, and for the coils that I've tried, anywhere from 10KHz to 100KHz is the sweet spot. The resonant secondary that I tried did have an iron core (not ferrite) and was resonant at around ... 25KHz if memory serves. Have yet to try a resonant primary. In fact I'm not going to like having to make either coil resonant since it would be an additional effort to tune the thing during manufacture. I probably have no choice about that because I think that is the key that will make this work at all.

Thanks for your suggestions.

From your dimensions I somehow cannot see what the worst case (longest) distance between the two coils is. Jan's suggestion to use pot core halves is a good one. Rod cores might be an advantage in some situations but not likely in yours since you don't seem to have the height for the receive coil.

In any case one key trick is to make the transmit side series-resonant. If not familiar with that concept study the topic "series-resonant converter". If this is a mass product making both sides resonant is probably not practical because you don't want to have a calibration procedure in production.

Oh, and watch FCC and other regs. You may have to confine yourself to shortwave ISM frequencies here and make sure the rectifier in the pendant doesn't spew too much in harmonics back out.

My "Philips" rechargeable electric toothbrush uses that method for charging the internal 3volt battery. Its about an inch in diameter and drops into a recess in the base unit. Essentially the whole base is the primary of the mains transformer (240v 50hz) with a hollow core. The battery is inside the core of the secondary winding. The brush just sits in the hole.

If it is the Sonicare series that's not quite how they work. The base transmits at 60kHz (AFAIR) and the toothbrush contains a large ferrite core with the receiving coil on there. The OP may need to run at a higher frequency, depending on size and weight limits. I don't think a pendant can be that heavy.

On a sunny day (Fri, 13 Nov 2009 10:35:37 -0500) it happened Wingsy wrote in :

Well, you can always use a light bulb and a photocell on the receiving side.

... and it'll get really hot. OTOH, there would be a "glow in the dark" feature. That won't likely receive an energy start label though :-)

TI has some new chips for this induction charge method. Probably app notes as well.

Cheers

-- Paul Hovnanian mailto: snipped-for-privacy@Hovnanian.com

------------------------------------------------------------------ Ask me about my vow of silence.

We have wireless pendants at work that are mounted in a case that have a solar panel in both the front and back of the device. The lights in the place generate enough energy to keep the 9v battery topped off between usage's. I don't remember the last time any one had to replace a battery and these get used everyday, many times.

A couple of unshielded drum-core inductors - off the shelf from Digikey - should work. Both should be capacitor resonated at the chosen frequency, a couple hundred KHz maybe. Get them as close as possible, parallel or along the same axis.

I've done this for electric meter reading, designed for use in countries where the power may be out a lot but the meters still need to be read inductively. Worked fine.

John

aked

"> In any case one key trick is to make the transmit side series- resonant."

Ahh, this makes sense. The energy can 'live' in the B field till it's sucked up by the reciever coil. As long as there aren't other hunks of conductor around!!

George H.

On a sunny day (Fri, 13 Nov 2009 15:02:49 -0800) it happened Joerg wrote in :

If you need 20 mW, 10% effciency for the photocell, 10% efficiency for the bulb, then total efficiency is 1% if you catch all light, makes a 2W bulb. Scale up to 200mW, 20 W CFL, not much heat. LEDs as light source are next, laser diodes... probably too dangerous. Or just leave it in the California sun(1) perhaps, if it shines.

Good try! But no cigar. That thing transmits power via the ISM band, the same band that my pendant & base uses. The base unit has a receiver for another pendant that is in use while the first pendant is in the charger. Their transmitter is 3 watts, and if in the same box as my base unit it would totally swamp the base's receiver.

Thanks for the suggestion though.

I like that. I'll talk to the enclosure manufacturer and see what they say about the effort it would take to include a solar cell into the pendant, while keeping it waterproof.

Thanks.