How does the inductor differ from the antenna?

[baconman71]

I've been trying to figure out how an inductor is different from an antenna. It's obvious that the inductor has more inductance than an antenna would, but let me try to explain it from my point of view. My current research project for school (lasting this year and into my senior year) is on the topic of resonant magnetic induction (like the MIT experiment). So back to my question, the inductors that I am trying to make will both create this changing magnetic field as well as a radiating EM waves. But I want to know how does the inductor differ from an antenna and what matters in figuring out equations for power transfer specifically? (such as impedance matching, Q factor, frequency, mutual inductance, size of the coils...those are just some that I thought of)

I'm not an expert like many of you are on the forum so any help would be FANTASTIC.

 

[vk6kro]

It is mostly a matter of dimensions.

A coil that is made by winding a spiral along a long thin object (like a fishing pole, for example) will radiate a lot better than one of the same inductance wound in layers on a former that was the size of a AA battery.

The critical difference is that the first inductor has substantial dimensions in one direction and it can have a voltage difference between the ends of the coil in that direction so that there are two voltages well separated in space that are out of phase with each other. This is what gives you the E component of the E/M wave.

The ideal antenna would be one with no spiral winding at all and it would be just a wire or a pipe, but then it would have to be longer than the spiral to be resonant. The actual length of the wire is important for an antenna, rather than its inductance.

The ideal antenna would also have as large a conductor as possible to minimise losses due to wire resistance and skin effect.

Hopefully the main source of loss can be optimised to be radiation from the antenna.

 

[sophiecentaur]

As said above, there is no difference in principle between the two. However, an inductor may produce a large magnetic field around it but the E field may be small and both fields are needed if you want to launch significant power in the form of an EM wave.

When you have two inductors, near each other, you don't want (need) to launch an EM wave into space. You need to 'match' the system so that as much power is transferred from one coil to the other as possible. You want as little EM power to be radiated as possible so your coils should not be 'good antennae'.
Unfortunately, this could get highly theoretical and too hard for you if you are still at School - you may well not have come across some things that 'we' take for granted.
Some aspects of this are fairly 'obvious' like the orientation of the coils and, if you can measure the inductance (or calculate it) then you can choose the right value of Capacitor to resonate it.
You ought to build your coils with the facility of 'tapping' onto them (connecting your input signal to the bottom few turns rather than straight across the coil), which will help you to match better and to get a higher Q. This needs to be easy to change so don't cover them up and be prepared to re-solder to different places.
If you want a satisfactory result from this exercise then you must be prepared to 'get in there' with construction and be prepared to do lots of modifications and re-runs. This is especially true if your theory is limited. Lots of systematic and intelligent use of the 'cut and try' method will allow you to optimise things.
There are a number of google hits on the topic. http://hackedgadgets.com/2011/01/13/diy-wireless-power-transfer/ The guy in the movie goes through all the theory and makes good DIY suggestions.

 

[baconman71]

Ok that makes sense. But then what are the most important factors in near field coupling? And what do you mean by matching? (impedance matching or matching of the actual coils) I want to test different coil shapes and different Q's in my experiment to see how everything works out but I just want to make sure that I can make the system efficient to where even the worst coils will produce a tiny amount of power. And I will play around with tapping into coils to see what happens.

Sorry for all the questions. There is a lot out there and I don't know what is important. :)

 

[rbj]

you can ask the same question regarding a capacitor and an antenna. either an inductor or a capacitor will generate fields outside of the device when driven by an AC voltage or current. the geometry of the device might minimize the field outside which is not what you want for an antenna.

 

[sophiecentaur]

Ok that makes sense. But then what are the most important factors in near field coupling? And what do you mean by matching? (impedance matching or matching of the actual coils) I want to test different coil shapes and different Q's in my experiment to see how everything works out but I just want to make sure that I can make the system efficient to where even the worst coils will produce a tiny amount of power. And I will play around with tapping into coils to see what happens.

Sorry for all the questions. There is a lot out there and I don't know what is important. :)

Did you give that link a good go? There's a lot in it and I'm sure it contains more than enough to answer your questions. You may need to force yourself to listen to all the comments but, there again, so did my students, too and they often needed to be reminded. haha.

As rbj says, you want a structure that has a strong local field but not one that radiates. This, as I look at it, can be achieved by having a large magnetic field (in a resonant coil) but small E field. Likewise, the receiving coil would have similar properties. In the movie, he only bothers with a small receive coil but I have seen setups with more or less identical constructions for both.
It would be handy for you to have an RF signal generator and a 'scope so that you can observe resonances and find an optimum frequency. Without those, you would be rather flying blind. 'Just' lighting an LED on a search coil should be fairly easy to do- it will be a matter of getting it 'really bright' and then measuring the volts that you get across a resistor, to find the power transferred.
As I remember, the movie suggests some design parameters and you should start with them. If this is a School project then you can copy quite flagrantly as long as you acknowledge your sources. You can't be expected to do it all from scratch. (In fact, no one ever would.)

 

[baconman71]

I just watched it and took notes on it. There is quite a bit of detail in the video which is fantastic. I have actually successfully made a set up of this process. It worked well as a primary attempt. But something that I have been trying to figure out is something with the kind of wire used at this frequency. He discusses the skin and proximity effect and talks about, at high frequencies, that the current goes to the outside of the wire. He also talks about thicker wire being better for less resistance and he also mentions hollow tubing as another possible kind of wiring. So if thicker wire is better for less resistance then why discuss something that is like a hollow tube and then why is Litz wire used? I have looked at other set ups of this idea and they have used big copper "tubing" as the primary inductor.



http://www.youtube.com/watch?v=LYqxH8lGznA&feature=related

and sorry again if the answer is actually blatant and I'm just not seeing it.

 

[sophiecentaur]

It depends on how much money you want to spend on this thing. If you see the size and tube gauge they use for MF transmitter coils then you will realist that it can be worth going Very Big. But not for you, I think. If you use a few MHz for your operation then ordinary 18gauge wire would be not to pricy and will hold itself in shape well. Yes, the skin effect means that current flows mostly near the surface and there are many factors to affect the Q. But there is a limit to how worth while it is to go in one direction of expense if something else is limiting the performance. Try a table top version first (as in the movie). You can easily establish the main factors that way and it is much easier to vary things at that scale. If you still have time and energy, double up on size and try again. You need to be realistic about your timescale and the last thing you want is to be rushing to write it all up at the end and produce an inadequate report because of last minute practical difficulties. Aim at a complete set of results ASAP and then you will have time to improve your system plus a fallback in the form of some existing data.
I've been involved in many projects at all levels and deadlines can take all the fun out of them if you're running too tight.

 

[baconman71]

Ok thank you very much. I thought it was interesting that they used hollow tubing. I had never heard of that before. But I do agree with you and I will make my experiment on a small scale.

Something else that I have been struggling with is the oscillator part. The video helped clear up quite a bit I must say about the process and what to look for in improving the circuit but what kind of an oscillator would be appropriate for this? I guess impedance matching would come into play here right?

 

[vk6kro]

Copper is relatively expensive and tubing uses less of it than solid rod.

So, since the RF current flows mainly on the outside of a conductor, it can save money to use tubing instead of solid rod.

None of this will matter to you since you will be using copper wire, not tubing.

Ideally, you should borrow an RF signal generator. This will be already calibrated for you.

Alternatively, you could make a Hartley oscillator. You can find out about these by using Google.
Click on "images" at the top of the screen and type "Hartley Oscillator".

The important point about Hartley oscillators is that they have a coil with a tapping on it. This is a connection somewhere on the coil other than the ends.
This coil could be one of the big coils in your experiment.
They allow oscillation over a wide range of frequencies.

As you can imagine, there is a lot of interest in this, and we get a lot of similar queries. Unfortunately, the advice is always the same. Power transfer at a distance is inefficient but within a room you can learn a lot about tuned circuits by doing such experiments.

Making such an oscillator is not difficult, but you will have the difficulty of knowing what frequency it is on or even if it is oscillating at all.