# Transformer Induction coil (COIL CONSTRUCTION)

1. Feb 17, 2010

### jegues

Hello all,

I'm currently working on a project and I have a question in regards to creating an efficient induction coil for our circuit.

Without going into the specific details of the circuit(We aren't given the details anyways, just the circuit itself), it basically imitates a transformer.

We are given two VERY THIN strips of wire 15m long which we can then create two coils out of. A primary coil and a secondary coil. SEE FIGURE

When a ball passes through the coil(s) (either a metal ball or non-metal ball) the sensor gives a numerical output between 0 and 255 (or 1024 I don't recall off the top of my head) based on the strength of the sensed magnetic signal. This strength depends on how we "make" our coils, so this will also generate our "signal threshold" to determine whether the ball is metal or non-metal.

That being said, I have to be certain that the coils aren't going to change or deform, in turn changing our "signal threshold".

Now the heart of my question is, what is the most effective and efficient way to build this coil(s)?

Again, the wire is extremely thin, so if I was to wrap it around an object like a highlighter, and then remove the highlighter the wire doesn't hold its shape as well as it should and falls out of place.

Is there anything we could spray the coil with so that it will hold its shape, but at the same time not decrease the signal strength it generates?

Should we be looking to wrap the coils as close together and as tight together as possible?

Should we overlap the two coils?

Can we tape the coils without ruining the signal?

Thanks again!

#### Attached Files:

• ###### CoilSensor.JPG
File size:
28.6 KB
Views:
189
2. Feb 17, 2010

### famousken

Just use standard epoxy. wind your coils and then brush it on. Are you not allowed to use a coil form? depending on how large the coils need to be (Im guessing not very since its only 15M of thin wire) you could use a couple of empty spools that are used for solder, plumbers tape, etc.

3. Feb 17, 2010

### jegues

Will the epoxy affect the magnetic signal produced by the coils?

Also, what is the best way to produce our coils? Does it matter if they wire overlaps while wrapping?

Thanks again,

4. Feb 17, 2010

### vk6kro

You can wind the coils any way you like. It doesn't matter if the turns are wound on top of each other. The last turn can even be wrapped around the rest of the coil to hold it in place.

Epoxy should not affect the operation of the coil at normal frequencies (below 30 MHz).

The classic coil former is the cardboard tube inside toilet paper rolls.

Also, garden "pop-up" sprinklers have some nice plastic formers in them. The outer case can be very useful.

Last edited: Feb 17, 2010
5. Feb 17, 2010

### Mike_In_Plano

Hello VK,

It sounds like you want free standing coils. This is commonly done with bondable wire. The wire has a thin coating that is easily melted with heat or weak solvents (even IP alcohol as I recall). You can check for this stuff at mwswire.com under round magnet wire polyester or polyurethane bondable.

You'll need to hold it in place with something that doesn't want to bond. Delrin, teflon, or aluminum have all worked for me.

Best of Luck,

- Mike

6. Feb 17, 2010

### jegues

Any other ideas?

I think I'm going to end up using epoxy to get the wire to hold the coil shape. I might try using a "twist-tie" to hold the coil in place as well.

7. Feb 17, 2010

### Mike_In_Plano

Well, an old method was bees wax. I've used it for accelerometers and just generally holding stuff together temporarily. Then again, good ol' two part epoxy will hold it right.

8. Feb 17, 2010

### Mike_In_Plano

Is your metal ball magnetic, or just conductive?

9. Feb 17, 2010

### jegues

The two balls passing through both the secondary and primary coil are metal ball bearings, and marbles.

Apparently if the coils aren't wrapped properly or efficiently you won't produce a consistent signal and it's impossible to set the proper threshold to accurately determine whether the ball is metal or non-metal.

10. Feb 17, 2010

### Mike_In_Plano

I see. With a magnetic steel and a low excitation frequency, one coil will be coupled more tightly to the other as the ball falls in between. However, if you turn the frequency up too high, or use a ball that's conductive (but not magnetic), the ball will shield on coil from the other.

11. Feb 17, 2010

### Mike_In_Plano

Are you using a scope and a signal generator to demonstrate?

12. Feb 17, 2010

### Mike_In_Plano

Also, are you allowed to change the orientations of the coils?

13. Feb 17, 2010

### jegues

We are given a circuit board that is similar to that of a transformer.

We are given two strips of 15m long VERY THIN wire.

From these two strips of wire we are expected to make a primary coil and a secondary coil which are attached the circuit board in order to create an analog signal as a ball passes through the coil.

We are allowed to create the coil in any manner we chose, of course we want to do so in a manner that will create a strong and consistent signal as either a marble or metal ball bearing passes through it.

14. Feb 17, 2010

### Mike_In_Plano

This shows the feild structures for two orthogonal coils with the ball off center. That will reject a lot of back ground coupling, but give you two peaks as the ball rolls by.

#### Attached Files:

• ###### coils.jpg
File size:
35.3 KB
Views:
183
15. Feb 17, 2010

### jegues

I don't know if I would be able to position the two coils like that.

16. Feb 17, 2010

### Mike_In_Plano

Maybe with 1/2 inch PVC? The Excitation coil wrapped around the PVC while the Pick up is wound on a separate piece of PVC and epoxied so that it is centered over the Excitation coil. If your allowed to use a little extra circuitry, you can add a POT and trim out the alignment error.

17. Feb 17, 2010

### jegues

That sounds a little too complicated, I think the solution is simplier than that.

18. Feb 17, 2010

### Mike_In_Plano

You can do it either way, or even make an E shaped structure. I know most all of em'. But the sensitivity difference with orthogonal coils will be several orders of magnitude better, depending on how well you can align your coils.

Then again, if your doing this for a class, your prof would likely get suspicious if it were too advanced.

19. Feb 17, 2010

### jegues

I'm confused on how I would make the coils orthogonal to one another? Would be the balls simply pass by them then?

I need a better picture of just the physical coil setup, not the fields and such.

20. Feb 17, 2010

### Bob S

If neither coil is excited, a metal ball will not induce any signal in either coil unless the ball has a permanent magnetic moment.

Are you planning on exciting either coil (the "primary") with an ac frequency? What frequency and amplitude? Solid metal balls will reduce the inductance because of eddy currents (unless frequency is very low). A ferrite ball would increase the primary inductance. Any kind of metal ball will change the mutual inductance between the two coils. A dielectric ball (marble) will have no effect. Please be more specific on your design objective.

Bob S