Coin recognition Project

Jeswin · Oct 25, 2011

Hello I am doing an university project on coin recognition. I have researched several methods such as eddy current and light gates, inductive coils etc. Could somebody guide me through this. I am not sure how to make a start.

Thank you

duffy · Oct 25, 2011

I work in coin-op. A "coin mech" discriminates coins based on size, weight, ferrous content and conductivity. The latter were traditionally done simply by rolling the coin past a strong magnet. If ferrous, it would stick. More subtly, a highly conductive coin (the copper core in a quarter) would cause the coin to slow down. After it passes the magnet, it hits a knife-edge. Coins rolling a little too fast fall far and are rejected, coins rolling more slowly fall short and are accepted.

A more modern electronic mech does the same tests with a coil. It operates an ac field using the coil as the reactive component in a simple relaxation oscillator. As the coin passes into the region of changing flux, it affects the oscillation in 2 ways: frequency and amplitude. If the frequency is lowered, this is due to the coin containing ferrous content (i.e. steel) and increasing the inductance of the coil. If the coin does not greatly change the frequency, but lowers the amplitude of the oscillations, this is caused by eddy current losses in a highly conductive metal like copper, silver, gold, aluminum, etc. which reduce the "Q" of the coil.

Jeswin · Feb 2, 2012

Hello,

I am doing a project titled 'coin recogntion' and the specification requires me the machine to identify all british coins. My research has shown me that inductive sensors are useful for differentiating metals with different characteristics. The first method I investigated was passing the coin through an air gap transformer and recording the change in voltage at the secondary; this test was very viable.

The other method involves running the coin across a single inductor; the eddy currents in on the coin should change the impedance of the inductor and if the impedance changes, the voltage accross it should change.

But the above phenomena was not very successful. I dont know if I am doing something wrong. I applied a range of freq(100Hz to 300Khz) and tried changing amplitude but nothing happens on the scope. Also does the reluctance value of the coil affect the eddy current response.

I know this is a long question but could someone kindly help me?? Thank you

ronsimpson · Feb 2, 2012

Jeswin said:
But the above phenomena was not very successful. I dont know if I am doing something wrong. I applied a range of freq(100Hz to 300Khz) and tried changing amplitude but nothing happens on the scope. Also does the reluctance value of the coil affect the eddy current response.

Have you tried a metal washer or slug. It maybe they react very different that a coin? (counter fit detector)

Brevor · Feb 2, 2012

I used to work for a company that made coin acceptors, The method you discribe is pretty much the way we did it. We used 3 inductive sensor circuits. One was an air gap like you discribe. The other 2
used a single inductor. I dont remember what frequency the air gap sensor ran at, but the 2 single inductor sensors ran at 350 Khz and 450 Khz. With the single inductor sensors we were measuring the frequency shift not the amplitude change.

Jeswin · Feb 2, 2012

Hi, Thanks for your reply. I really appreciate it.
My inductor is connected in series with a pull up resistor of very low value and the circuit is powered by a voltage source of set frequency. I presumed that the frequency cannot change since it is determined by the source. Would you be able to know if the coil was part of an oscillator circuit? I am also unsure if the reluctance of the inductor will have a MAJOR influence on the amplitude or frequency response.

Jeswin · Feb 2, 2012

I tried many shapes (metals and non-metals) but neither frequency or amplitude seems to change. The inductor is connected in a series circuit along with a pull up resistor. Would you be able to tell me if the dimensions of the inductor has a major effect on the response due to eddy current induced on the coin?? It has a core of 9mm and a length of 25mm with 226 turns(Single layer).

Jeswin · Feb 2, 2012

Coil Design for Eddy current inspection

Hi,

I am doing a project where on of the sensors are inductive and I am required to transmit a magnetic field on a metal plate and then use a receiver coil to 'read' the magnetic field generated by the eddy currents on the metal. I am not sure if the dimensions of the coil has a MAJOR effect on the eddy current strength. Could somebody tell me the effect on the magnetic field if I used a single layered or a multilayered coil. The multilayered coil will probably have a smaller magnetic path (length).

cowboybob · Feb 2, 2012

The greater your "transmitting" magnetic field strength, the greater the eddy currents' that will develop in the metal plate.

The multilayered coil will probably have a smaller magnetic path (length)

Not sure what your saying here.

Multiple layered coils, as I understand the term, create stronger magnetic fields and, therefore, induce stronger eddies.

Jeswin · Feb 2, 2012

Sorry about the misunderstanding. I meant if I were to multilayered coils, the length of the coil would be smaller. Do I have to consider any capacitance effects between the layers?? Thanks

cowboybob · Feb 2, 2012

I'm assuming you mean between the layers of the wire making up the inducing magnetic coil, not the layer between the coil and the plate where the eddies are being induced.

It's my understanding that capacitive issues in the wires of a parallel wound multilayered magnetic coil tend to have a neutral (or perhaps better, a neutralized) effect on E-fields, which means that there should be no significant effect of the magnetic field(s) produced.

Parasitic oscillations might be a problem if your driver frequencies approach the resonance (or any of the harmonics) of the coil.

The shortening of the mag field is, to me, more of a question determining the overall physical dimensions of the test bed.

davenn · Feb 2, 2012

Size and or weight of the coin is also often used in coin slot systems

cheers
Dave

cowboybob · Feb 3, 2012

machine to identify all british coins[

(My emphasis)

davenn hit on an important issue.

If the coin lacks any significant amount of a ferrous metal, the induction technique won't work well, or at all depending.

In other words, to detect ALL coins is going to take at least one additional, non-induction detection technique.

BobW · Feb 3, 2012

Jeswin said:
Hi, Thanks for your reply. I really appreciate it.
My inductor is connected in series with a pull up resistor of very low value and the circuit is powered by a voltage source of set frequency. I presumed that the frequency cannot change since it is determined by the source. Would you be able to know if the coil was part of an oscillator circuit? I am also unsure if the reluctance of the inductor will have a MAJOR influence on the amplitude or frequency response.

The low value of pull-up resistor is probably loading down the inductor so much that it doesn't produce any significant change. I suggest using the inductor in combination with a capacitor as an LC oscillator, and as Brevor suggested, measure the frequency shift.

Brevor · Feb 3, 2012

Would you be able to know if the coil was part of an oscillator circuit? I am also unsure if the reluctance of the inductor will have a MAJOR influence on the amplitude or frequency response.

Yes the coil was in the LC network of a colpitts oscillator circuit.

Jeswin · Feb 3, 2012

Reluctance

Hi,

Does anybody know if the magnetic field generated by the coil is affected by the reluctance? Is it better to have a short coil with multiple layers or a long single layered coil?

DerStrom8 · Feb 3, 2012

I haven't heard "reluctance" used in this context. I'm not sure if you're using the right word

Perhaps you mean "inductance"?

Jeswin · Feb 3, 2012

Hi thanks for your reply,

I actually meant reluctance; this quantity is dependant on the dimensions of the coil( length and core diameter). I am working on a project called 'coin recognition' and one of the methods I am exercising to identify coins is to have a coil induce eddy currents on the surface of the coin and use a receiver coil to 'read' the magnetic field on induced by the eddies on the coin.

I am unsure if the strength of eddy currents induced on the coin is related to the dimensions of the coil.

DerStrom8 · Feb 3, 2012

Okay, I just did a little bit of research and found magnetic reluctance. A quick skim seemed to make sense.

I suggest you try a google search to start. You'd be surprised to find how many good answers are out there. If you don't find anything, then check back and let us know. We'll see what we can do for you.

Regards

duffy · Feb 3, 2012

Hi Jeswin - the field coils used in "coin mechs" are short with multiple layers. What you are looking for is actually a drop in the energy returned to the coil - more conductive coins (i.e. silver, copper) convert more of the induced field to eddy currents, and then to heat, where it is then lost from the system. Less conductive coins do not couple as effectively with the field, and so more energy is returned to the coil. A drop in frequency indicates ferrous content, as this will increase the inductance of the coil. Size is important, of course, because a larger copper coin will cut through more of the field than a smaller one. Sizing coins is handled independently of the sense coil in coin mechs.

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Coin recognition Project

New Member

Well-Known Member

New Member

Well-Known Member

Member

New Member

New Member

New Member

Well-Known Member

New Member

Well-Known Member

Active Member

Well-Known Member

Active Member

Member

New Member

Super Moderator

New Member

Super Moderator

Well-Known Member

Similar threads

New Articles From Microcontroller Tips