Low current in wireless electricity / Electromagnetic Radiation / Forums

Forums

4hv.org :: Forums :: Electromagnetic Radiation

« Previous topic | Next topic »

Low current in wireless electricity

1 2 3 next

Move Thread

LAN_403

atocp

Sun Jul 03 2011, 04:15PM

Registered Member #3983 Joined: Sun Jul 03 2011, 03:55PM
Location:
Posts: 11

Hi all, and thanks for looking at this post.

First off, I didn't do physics at school, so my in depth knowledge of physics fundamentals is slightly lacking. I am currently doing a Masters project, and a part of it involves me creating a wireless charging system for a vehicle.
At the moment, I am only trying to make a small scale version, and I will say within my project that it can 'scale up'. I don't really care if it cant, the people reading it don't really care about the phyics.

The problem: I have created a helical coil, 1mm diameter enamelled wire, 14 turns, 100mm diameter with 6mm spacing. I have created two of these coils, and calcualted their inductance to be ~13.5uH. The transmitter sends a sine wave at 750kHz at 12v. I can receive in the second coil at about 10-11v, but almost no current. The transmitter has a capacitor, and the receiver has 4 diodes to make a bridge rectifier and a capacitor with load on it from a decade box.
As I say, my physics is limited, so with the help of someone, we placed a 10ohm load and calculated the current to be around a few mA.

Can I ask why my current is so low? Is this down to the coil design and low inductance, or is there a way to fire a bigger current through it? We tried to use a mosfet, and it roasted, so we used two, and it didn't really work, still just a few milliamps.
Also, is there a way to calculate the current, before the rectifier, while it is still AC? I have looked through many google links, but some talk about RMS and Watts, others talk about kVA.

I have been looking at Marko's tutorial using a single coil and he is firing 24V through it and getting between 0.5 and 1 amps. I would love to get somewhere near that! And if possible, much higher.

Any help is much appreciated, as I have hit a wall in terms of this part of my project. As I say, it is for a vehicle so high power is required. http://www.haloipt.com/ have managed it, so Im not sure what to try next.

Thank you again,
Atocp

Forty

Sun Jul 03 2011, 05:00PM

Registered Member #3888 Joined: Sun May 15 2011, 09:50PM
Location: Erie, PA
Posts: 649

an LCF meter would be extremely helpful if you can find, buy, or make one. the differences between experimental and calculated theoretical can be a lot. you need the coils to have the same resonant frequency, or at least have one frequency be a harmonic of the other. what sort of distance are you trying to transmit this signal over? your coil sounds more like a spiral than a helix. i think you'd want to stick with the spiral (i believe that's what those wireless phone charger pads use.) I don't think the 6mm spacing is helping you at all. put the wires right next to each other, and if that means using more turns for a desired diameter, then you can up the capacitance in the LC circuit to maintain the desired frequency. using litz wire may also help, because at 750khz the skin effect definitely comes into play. litz wire is very easy to make yourself, just take many strands of smaller enameled wire and twist them together with a cordless drill. the strands need to stay insulated from each other along the length, but at the ends they can be soldered together into one wire. litz wire cuts down the apparent resistance of the wire by providing much more surface area for the high frequency signal to travel on (induction stove tops use a spiral of litz cable for wireless transfer of a lot of energy)
what i can't understand, is how you got assigned a project like this for your degree, but your degree never required physics courses lol.

atocp

Sun Jul 03 2011, 06:08PM

Registered Member #3983 Joined: Sun Jul 03 2011, 03:55PM
Location:
Posts: 11

Hi Forty,

Thank you for the very helpful advice. I have attached a photo to show the distances and the coils I have made. I am charging a scooter, so I would be looking at distances of around 20cm or less, I can design it to be less. I have made it this far because my degrees before didn't need physics. I made this topic idea myself (believe me, i've bitten off more than I can chew, but its too late to change), and the majority of the project is computer design based, as it is a digital engineering product design course. I need to produce a prototype however, so will buy a kids electric scooter and show it charging. It keeps the lecturers and markers happy. The wireless is only about 20% of the project, the rest has pretty much no physics.

Can I ask, why do the MIT, Intel, et al use frequencies at well into the MHZ range? I don't have access to a signal generator of that high, so I have to stick to below 1, so thought I'd go for 750kHz. They also use the helical design like I have, as opposed to the flat spiral design. Is there a reason for this? I understand that the diameter of the coil can have an effect, and helical means it is several of the same as opposed to several reducing circles in a spiral design, will this have a profound impact if I go to spiral?

I will look at making some litz once I have upped the current slightly. I have to pay for everything, so the less wire I use the better at the moment.

Thank you again for your help. Help like this gets names printed in dissertations

atocp

Forty

Sun Jul 03 2011, 08:24PM

Registered Member #3888 Joined: Sun May 15 2011, 09:50PM
Location: Erie, PA
Posts: 649

ah the picture helps.
higher frequency probably just travels distances better. for 20cm i'm sure your ~750khz will do.
the skin depth of copper at that frequency is about 0.076mm, so roughly 72% of your 1mm dia wire isn't conducting a thing. you don't need to buy anything. just find a good size wall wart (step down power adapter plugs) or some other transformer with lots of really fine wire on it. set up two clamps or something on either side of a room, carefully wrap the wire back and forth between the clamps (stick the winding on a screwdriver and just walk back and forth.) when you've got the thickness you want, or run out of wire, hold the bundle on one end while you snip it off the clamp, then tie the bundle into a knot and clamp it back down to something. go snip the other end, set the bundle down. go back to the clamped end and pull the bundle as you walk back to the other side, then tie the loose end in a knot, stick it in a drill, and twist it until the wire starts pulling back. (probably a complicated description for something really simple, the knotting and pulling business is just to try to make all the lengths equal, so that it twists equally)

atocp

Sun Jul 03 2011, 11:11PM

Registered Member #3983 Joined: Sun Jul 03 2011, 03:55PM
Location:
Posts: 11

Hi Forty,

thanks again for the help, and the instructions for making the litz wire. i think i might try the power drill technique. i can get a reel of 0.125mm for about a tenner, at its 2200m length.

I calculated the frequency to be 750 so I attached an oscilliscope on the recieving coil in order to figure out where the maximum was and it ended up being around 710kHz. Thats where I get the maximum transfer of power.

Another couple of questions that have been thrown up...Will litz wire have the same calculations for figuring out the capacitance of the coil and the inductance? Do you have a way I can measure the current on the other side? Any ammeters I have access to will only go to 50 or 60Hz if checking AC?

I am going to try this tomorrow when I go into university. Hopefull I will have some success.

Thank you again,
atocp

Sulaiman

Sun Jul 03 2011, 11:15PM

Registered Member #162 Joined: Mon Feb 13 2006, 10:25AM
Location: United Kingdom
Posts: 3141

The reason that you are not getting a good power throughput is
when you put a load on the receiver coil it causes a magnetic field exactly in opposition to the transmitter field which 'repels' the magnetic flux linkage between the two coils.
I suggest that you get two "E" shaped ferrite cores
wind the transmit coil on the center limb of one piece
and receive coil on the center limb of the other
when the two "E" pieces are brought together they will form a transformer for efficient wireless power transfer.
A circular 'pot' core or two halves of an ETD-type core would be good too.
If you use air-core coils there will definitely be high 'leakage inductance' causing poor power transfer and heating by induction of nearby metalic parts.

atocp

Mon Jul 04 2011, 11:04AM

Registered Member #3983 Joined: Sun Jul 03 2011, 03:55PM
Location:
Posts: 11

Hi Sulaiman,

Thanks for your advice on the cores. I was looking into that before, but decided not to go there as it would add quite a lot of weight to my design. Can I ask, when you use cores, the magnetic field is stronger along the ferrite, however, when the air gap appears, the magnetic field bulges and becomes much weaker. As this is a wireless transfer system, will the gap not cause this to happen, and therefore greatly reduce the effectiveness of using a core?
Also, if I change my design to a flat spiral design, do I just try to use a really shallow ferrite plate to back the coil, or can I no longer use a core?

Many thanks for your help, I'm in uni just now, so I'm going to try and make another couple of coils today.

Thanks,
atocp

Sulaiman

Mon Jul 04 2011, 09:29PM

Registered Member #162 Joined: Mon Feb 13 2006, 10:25AM
Location: United Kingdom
Posts: 3141

For charging electric vehicles I assume (!) that the 'wireless' part is not for remote charging but rather for electrical isolation between the vehicle and the power source, so if ferrite cores are used they could have quite a small airgap (plastic-gap) for isolation purposes, nickel-zinc ferrite being of such high resistivity would probably not need an intentional airgap even.

Spiral wound inductors with flat plates would work, but not as well as a 'pot core' shape.
Any non-magnetic gap must cause leakage inductance.

Dr. Slack

Tue Jul 05 2011, 05:59AM

Registered Member #72 Joined: Thu Feb 09 2006, 08:29AM
Location: UK St. Albans
Posts: 1659

The two problems are that the primary and the secondary are in different places (well, duh, of course they are, it's a way of getting power from A to B), and there's all that air around.

In a conventional transformer, you have very high primary inductance due to the core, and very low leakage inductance, due to both the core and the closeness of the windings to each other. As a result, it's easy to transfer large amounts of power without much thought to managing these inductances.

In a wireless power transformer, the primary inductance is very low, and the leakage is very high. The first shunts your source, unless you resonate it. The second isolates your secondary, unless you resonate it. Unless you make it your number one job to handle these extreme inductances that are working against you, your power transfer will be feeble. Basically, you need to push the geometry back towards a conventional transformer, then resonate out the remaining inductances.

The three things to do are ...

a) minimise the length of the the path that the flux has to take around both windings to reduce the leakage inductance, so use flat pancake coils and place them as close together as possible

b) replace as much of the air in that path as possible with ferrite to reduce the leakage inductance further, at least a ferrite backing to the coils, and preferrably a ferrite centre and side legs as well, leaving only the essential gap betweent hem as air. If you can wind the coils onto E or pot cores, and place them *really* close together, seperated by only a plastic shim or something to say "look, I have an air-gap", then so much the better.

c) tune the coils a la DRSSTC to resonate out the remaining leakage inductance. This means you need to operate at resonant frequency, so either a very precise control of parameters, or a feedback tuner for the transmitter.

Switch mode power supply controllers are starting to appear that use this regime, mainly so that the transformer can be built with a large seperation between primary and secondary for electrical isolation (the same reason that TC's use large seperation). Try Googling "LLC resonant converter", you'll get manufacturer's App notes which should give you all the design equations, and components, you'll need.

To get good power transfer is not going to be as straightforward as you thought, maybe not easy enough for just 20% of the project.

atocp

Tue Jul 05 2011, 03:35PM

Registered Member #3983 Joined: Sun Jul 03 2011, 03:55PM
Location:
Posts: 11

Hi there,

Thank you both for your extra advice. I feel I might need to describe my problem slightly more. I am going to be charging a vehicle wirelessly, see HaloIPT for the car version. When I say I'm going to be charging a vehicle, for the purposes of this project, it will only be an electric kids scooter 250W motor, small lithium ion battery pack etc, not a moped as my project design will state.
For this, I rekon I will purchase a small kids scooter, and attach a wireless coil to the underside of the scooter. The charger for this scooter uses a 12v 600mA output supply to the batteries, so I need to get somewhere in that region with my coils.

I have made two coils of litz wire. I now have 2x 43 turn, 0.125mm wire (x8 strands) 100mm diameter, which is giving me an inductance of around 360uH. I am using 150pF capacitors, and I am able to successfully measure an input of 12V, but i'm not quite sure how to measue the output, as its way over 12v because of peak to peak. When I put a 10ohm load on it, it drops to pretty much 0. It is tuned to be around the 750kHz mark, or thereabouts. Its an old analogue signal generator and im not 100% the exact frequency. On another note, my last coils were able to transfer enough to power two LED's, however, these new coils don't seem to manage it

I think my voltage efficiency transfer is quite good, but is very low current, not the 600mA I am requiring.

The air gap will be around <20cm. Im not worried about efficiency too much, but if I can get my current up high enough towards 0.5A I will be able to successfully demonstrate to my lecturers etc its operation. I am going to change my coil design, but only once I have sufficient proof that I can get decent current.

I refer to Marko's design of wireless power -> he can manage to send between 0.5 and 1A and 24v. This is with one copper pipe and a copper pipe for the reciever. Should I be doing this instead of winding coils? He also used two transistors, which didn't help me in my last coil design, but I will probably be implementing on Thursday to see if that makes a difference. I think I might send a little message to him to ask him what his input current was.

Again, many thanks for the advice. I have attached todays photo of my progress. This is the 2 small coils litz wire coils 45mm apart. I am now going to go and try and figure out how to actually calculate what current is being transferred, and see if Google has the answer.

1 2 3 next

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob