Inductive power transfer

[Cyberduke]

Hi Everyone, I have a project that I am working on that is pov display. These things are the internet full but I have my own ideas. I do not want to use a battery onboard or any sort of slip ring configuration. I want to use induction power transfer. I believe this might not be the best option, but it most certainly is the option that intrigues me the most, and it is something that I have always wanted to learn more about, so I choose to implement that.

 The thing I am not knowledgeable  enough about is this subject, I have read up about that and I think that I might have half an idea how that works. Here are some thoughts and plans that I have. Any criticism is welcome, it probably is that all this is a bunch of nonsense that I dreamed up since I could not find an easy and clear guide, I had to piece my understanding together from various reads.

1. The first thing that came to mind was that I could determine the inductance of my primary coil and then add a capacitor in series so that the resonant frequency of this "band pass filter" matches the 50Hz that I would be getting from my wall. Sadly the 50Hz I would have from my wall socket(yes I would step it down to something like 12v) being as low as it is, would require me to put in a rather large capacitor depending on the primary coil that I could wind.
2. I could create some sort of AC-AC (or even DC-AC) circuit that would change the frequency of the signal and therefore tune that into resonance to have maximum magnetic flux in the primary coil.

Any ideas and input would be appreciated.

[Gyro]

Welcome.

You're very unlikely to be able to transfer sufficient energy at 50Hz, you probably need to drive the primary from a higher frequency oscillator. One thing that can help you (a lot) it to create a good magnetic circuit between the primary and secondary. A way of doing this is to use two halves of a circular ferrite transformer core, wind the primary in one half and the secondary in the other and mount them on the spindle axis so that the primary half is fixed and the secondary half mounted in close proximity but free to rotate.

P.S. I'm thinking this type of core... http://www.ebay.co.uk/itm/FERROXCUBE-P14-8-3C81-FERRITE-CORE-P-3C81-/391174083015?hash=item5b13cad1c7:g:Q8MAAOxyx-BSa7zy  (Available in various sizes)

[Cyberduke]

Welcome.

Thank you, I am impressed by the amount of activity on the EEVblog forum.

You're very unlikely to be able to transfer sufficient energy at 50Hz, you probably need to drive the primary from a higher frequency oscillator.

I thought that was the case, guess I was in denial. Off reading up about how to do that

One thing that can help you (a lot) it to create a good magnetic circuit between the primary and secondary. A way of doing this is to use two halves of a circular ferrite transformer core, wind the primary in one half and the secondary in the other and mount them on the spindle axis so that the primary half is fixed and the secondary half mounted in close proximity but free to rotate.

I have thought about ways to reduce the reluctance of the magnetic circuit, but my current mechanical setup is going to make it a little tough. Added weight will definitely reduce the rpm of the motor which is very undesirable.

ps. Just checked that link, amazing idea. I think I should adjust my design to incorporate something like that.

[Gyro]

  It hopefully shouldn't add significant weight. Someone else might be along with a better idea. The higher the frequency, the less 'core' material you need, until none is needed, but of course the primary and secondary windings will always need to be concentric.

[Cyberduke]

Curious to see other ideas. This is a really nice project for me.
Valid points there, glad to see my understanding is on point, and yeah physics 101 right there.

[kulky64]

Why not use a rotary transformer found on VCR drum as in this project:
http://us.cactii.net/~bb/propclock/hardware.html
 

[Cyberduke]

Why not use a rotary transformer found on VCR drum as in this project:
http://us.cactii.net/~bb/propclock/hardware.html

Another great(possibly even easier) idea. Sourcing a broken VCR shouldn't be that hard.

ps. I am looking as oscillator circuits, But I have to admit that I am a little bit overwhelmed at all the circuits I find. Could anyone point me in a nice direction please?

[Gyro]

Hi Cyberduke,

Yes, kulky64's the VCR head is a neat idea and potentially gives you the brushless motor and decent bearings too. The transformer coupling is designed to give multiple signal couplings (one for each video head) so has a number of isolated magnetic circuits. I would have had reservations about the amount of power you could get across but by using all the windings it clearly works!

As far as the drive circuit, you could go discrete, or use an off-the-shelf DC-DC converter IC - to pick a random example from a quick search... http://www.ti.com/product/sn6505a/datasheet (just an example, there are many others). The magic search term is 'Isolated DC-DC converter', that gets you the 'transfomer' bit.

[Cyberduke]

Using a VCR head seems like a decent idea that already offers all I need onto one package.

The isolated DC-DC converter is also perfect, doing all the driving and stuff for me. Just one question though, never does it say anything about the specs of the transformer. Except obviously I can see that it should be a center tap transformer.  So that means that it should work with just about any? Up onto now the inductance of the transformer primary winding mattered a lot.

I have done a surface wise search and it seems finding a DC-DC converter IC here might take a little skill. I have searched all my usual sites, no luck, will be expanding the search when I get time.

 

[bktemp]

Instead of using a centre tapped transformer, you can also apply power at both ends by using an additional inductor. This reduces the efficiency a bit, because of the higher current flowing through the additional inductors, but it simplifies the transformer, especially for transformers where rewinding is difficult like the VCR head.
The inductance of the VCR head windings is rather low, therefore you need a high switching frequency (>100kHz).

[Gyro]

Looking at the project linked by kulky64, the Eagle schematics are there. It looks as if the author used a 555 timer driving a mosfet to feed the primary (prop_base top left) and a simple rectifier and regulator on the rotating part (prop_clock top right). It seems to work for him.

I've attached images of the the two Eagle schematic pages (...original author's copyright etc.) You'd do well to follow the link and study the article in full. 

P.S. It looks as if the motor drive is on the base schematic too.

[Cyberduke]

I was told by a more learned friend that a 555 won't work. Seems that he was wrong. Using a 555 seems like a very nice option too.

On the prop clock where he is filtering and regulating the input signal, why does he have soo many capacitors in parallel? Is it for some practical reason? since I see no theoretical reason therefore.

[Gyro]

The electrolytic and standard (probably 100nF) caps on the input to the regulator are to smooth and filter the rectifier output. The ones on the output are output filter on the regulator and decoupling caps on the other ICs, just all shown in one place.  Do a search for 'IC decopling'.

[Cyberduke]

Thank you, I think I have enough information to attempt it now. As soon as my schoolwork calm down I will start building.