A fun metal detector to find coins at the beach 
All the parts for the PC board, PC board, battery and winding wire $10.00  
including pack and post. You will also need an AM radio, base-board to 
mount the parts and a wooden handle approx 1.5 metres long. 

This project has not been called a GOLD! detector as this name has been left for the more complex detectors that actually discriminate been gold and other metals.  There is an enormous difference between detecting gold and ordinary metals (called base metals). Apart from the fact that gold is over 1000 times more expensive, its magnetic differences are such that we can produce a metal detector that will discriminate between metals, both ferrous and non ferrous, and GOLD! 
Gold detectors have come a long way in the past 15 years, especially during the rapid rise in gold prices, about 10 years ago. 
At that time, "GOLD!" was on everyone's lips and as its price soared, GOLD FEVER took over and fossickers by the thousands took to the countryside to try their luck. 

In areas where gold was found some 100 years ago in Australia, the country was dotted with prospectors combing the hills and flood-plains with gold detectors. 
Encouraged by reports of sizeable nuggets being discovered, buyers flocked to purchase gold detectors. Prospecting shops sprung up everywhere and offered detectors not much more complex that this model with an amplifier (the equivalent to the AM radio), for $299!  You may laugh, but when gold fever strikes, people do the craziest of things. 

The chance of picking up a nugget of gold is a million to one. This is because the ground where they are found is quite often filled with iron and other minerals that will affect the reading of electronic detecting equipment and reduce their sensitivity. To overcome this we must employ very sophisticated circuitry so that only the "signature" of gold is registered on the equipment. 

As you can imagine, detecting the difference between an aluminium ring-pull from a small nugget is an almost impossible task as ring-pulls are generally closer to the surface and swamp the minūte signature of any lumps of gold that may be buried deeper in the ground. 

Also the background effect of the minerals in the soil has to be cancelled and when you do this, you lose some of the sensitivity of the detector. The answer is to tune the equipment for the terrain you are covering so that it is at peak performance. This requires a fair degree of skill and that's why more advanced detectors are available on the market. 

To start you in this interesting field we have designed a very simple detector. It only requires a handful of components and an evening's work. 
This way you will learn electronics while being able to go out and find something valuable.
It's not only gold that's worth finding but a whole range of items including money, jewellery, metal objects and things that have been lost for 100 years or more. 

One of the best places to search is the beach. Lots of things are lost in the sand every year and it's very easy to scan the surface with a detector and dig them up. 
Because this project is very simple we have not called it a gold detector as it cannot discriminate between any of the base metals and gold. Instead, the word "gold detector" can only be introduced with a more elaborate model where some form of discrimination is available. 

We have called this design a "metal detector" as it lets you know when anything of a ferrous or non-ferrous nature is placed in the field of the coil. 

We will start the discussion when the conditions have settled down after a few cycles and the voltage on the base of the transistor is stable (fixed by the "holding" or "resisting" action of the 10n capacitor). 

The circuit is an oscillator and the way it keeps oscillating is due to positive feedback. This is the case with all oscillators and the component that provides the feedback is the 1n capacitor between 
the collector and emitter of the transistor. It may seem unusual that the transistor can be turned on via the emitter to keep it oscillating, but in fact it does not matter if the emitter or base receives a signal as the important factor is THE VOLTAGE DIFFERENCE between these two terminals. 

If the base is kept fixed and the emitter voltage is reduced, the transistor sees a higher voltage between the base and emitter and it is turned ON harder. If the voltage on the emitter increases, the transistor turns OFF as the difference between the two is reduced. 
This is exactly what happens in this circuit. The 1n capacitor between the collector and emitter influences the voltage on the emitter to turn the transistor on and off. It does this by constantly monitoring the voltage on the tuned circuit and passing the change to the emitter. 

In this project, the TUNED CIRCUIT is the parallel components consisting of the inductor (the search coil) and the 1n capacitor across it. This is called an LC circuit in which the L is the inductance of the inductor in Henries (or mH or uH) and C is the capacitance of the capacitor in Farads (or uF or nF or pF). 

We start when the transistor turns ON and allows a pulse of energy to enter the tuned circuit (later you will see how the transistor turns on). 

The pulse of energy (current) starts by trying to entering both the coil and capacitor. You would think the coil has the smallest resistance but the capacitor is uncharged and presents a theoretical zero resistance and begins to charge. When a small voltage appears across it, you would think the coil would become the least resistance as it consists of only a few turns of copper wire. 

But the wire is wound in a coil and forms an inductor (it has inductance). When a voltage is applied to it, the low resistance of the inductor allows a current to flow but this current produces magnetic flux that cuts the turns of the coil and produces a back-voltage that opposes the incoming current. It works like this: Suppose you supply 200mV to the coil. The back voltage it produces may be as high as 199mV and thus you only have 1mV with which to push current into the coil. 
If the resistance of the coil is 100milli-ohms, the current will be about 10mA. The capacitor will accept more than this and so it gets charged first.

As the voltage on the capacitor increases, it presents its voltage to the inductor and allows a current to flow (at a rate which the coil will accept) to produce magnetic flux. This flux is called electromagnetic lines of force and creates an expanding field. The capacitor cannot provide energy for very long and after a short time the current reduces and this causes the magnetic field to begin to collapse.

The collapsing magnetic field produces a voltage that is opposite to that originally supplied to it and the bottom of the coil becomes positive with respect to the top. 
If we think of the coil as being a tiny battery we see it adds its voltage to the 9v of the supply and the collector end of the coil becomes higher than 9v. 

This voltage is detected by the 1n feedback capacitor (between the collector and emitter) and it passes the voltage to the emitter where it increases the emitter voltage. The base of the transistor is kept stable and fixed by the holding action of the 10n capacitor and the transistor turns off slightly. This action continues and eventually the collector can be considered to be removed from the circuit so that it puts no load on the tuned circuit. When an inductor is not loaded like this, the collapsing magnetic field will produce maximum voltage. 

This is the case in the circuit above and as the magnetic field collapses, it produces a voltage (about 25v) that is considerably higher than that applied to it. This voltage is passed to the "C" component of the tuned circuit (the 1n capacitor connected across the coil) and the capacitor charges up. 

When all the magnetic flux has been converted to voltage the capacitor is charged and it begins to deliver this charge back to the coil. In the process, the voltage across the capacitor is reduced and this voltage is detected by the 1n capacitor across the collector emitter terminals of the transistor. The result is the voltage on the emitter is reduced and the transistor is turned on slightly to deliver a pulse of energy to the tuned circuit. 
This is when another pulse of energy is injected into the system and the cycle repeats. 
The frequency of the circuit is about 140kHz and is set by the inductance of the coil and the capacitor across it. 
When we place a piece of metal in the magnetic field of the coil, some of the lines of flux pass through the metal and are converted to an electric current called an EDDY CURRENT in the metal. 

This means we lose some of the magnetic flux and so there is less available to return to the coil when it begins to collapse. This means the reverse-voltage produced by the coil will be lower and so the capacitor will take less time to charge to its maximum value. Thus the transistor will be turned on sooner and so the frequency of the circuit increases. 

The flux produced by the coil is electromagnetic radiation identical to radio waves of the same frequency. If we place a radio near the coil and tune it to a harmonic, the two frequencies will "beat" together and produce a "quiet spot" on the radio. 
When a piece of metal enters the field of the coil, the frequency changes slightly and a low-frequency tone is emitted from the speaker. 
A shift in frequency of as little as a few hertz will be clearly heard and this is why the circuit is so effective. 
The sensitivity of the coil depends on making the circuit change frequency at the slightest insertion of a metal object. This requires operating the transistor at an amplitude that is not overdriving it, so that the slightest injection of a piece of metal into the field will alter the frequency. 

It is important to note that the AMPLITUDE of the waveform is also reduced when a piece of metal is introduced but the radio is not set up to detect this. Other metal detectors detect the drop in amplitude and later you will see how the two circuits compare. 

   Collapsing field produces a voltage into capacitor. 1n feedback capacitor sees this voltage  capacitor charges up and turns ON transistor via feedback capacitor  Transistor TOPS-UP capacitor and produces FULL magnetic field  Capacitor is fully CHARGED. Transistor is turned off  capacitor delivers charge to coil. Magnetic flux produced  capacitor half discharged  capacitor discharged  magnetic flux collapses and produces a voltage in the opposite direction  Capacitor half charged  capacitor fully charged - but in wrong direction to turn on transistor  capacitor delivers its charge to coil. Magnetic flux produced  capacitor half discharged  capacitor fully discharged - go to first frame   
1. Glide your "mouseover" the boxes above and study each frame.  2. Hold mouse on box for "description."  3.
Mouseover: for animation.

There are many ways to explain how the circuit works and all of them are technically correct. 
Here's an animated way to describe the cycle. The transistor is only turned on when the voltage across the capacitor is negative on the bottom plate. This is when the voltage on the capacitor is added to the voltage of the supply and passed through the 1n capacitor across the transistor to reduce the voltage on the emitter. The transistor turns ON and delivers a short burst of energy to the TUNED CIRCUIT. 
The 4n7 on the emitter charges slightly during this action and the voltage on the emitter rises to turn the transistor OFF. 
The charge on the capacitor is passed to the coil and it produces expanding magnetic flux. The capacitor runs out of charge and the coil collapses. The collapsing magnetic flux produces a voltage in the opposite direction and this is passed to the capacitor. 
During this part of the cycle the voltage on the capacitor is not of the correct polarity to turn the transistor on and it remains OFF. 
This is the part of the cycle when very little load is placed on the tuned circuit and the voltage produced by the coil can be higher than the applied voltage. The capacitor then delivers its charge to the coil and the cycle repeats.  

All the parts fit on a small PC board with two wires from the coil and two from the battery. 

1  -  220R (red-red-brown-gold)
1  -  47k (yellow-purple-orange-gold)
2  - 1n greencaps (102)
1  -  4n7 greencap (472)
1  -  10n greencap (103)
1  -  47u electrolytic
1  -  BC 547 transistor
1  - slide switch
1  -  9v battery snap
1  -  9v battery
6.5m winding wire (gauge not critical)


The search coil is made by winding 16 turns around a circular object 12cm diameter. This can be a juice bottle or even a square object as the coil can be made circular afterwards. Use 4 pieces of sticky tape or electricians tape around the turns to keep them in place and glue the coil to the base board with silicon sealant. 

The base-board has a wooden handle screwed to it at an angle of 60°. You will also need a small transistor radio taped to the handle near the base so that it can pick up the field from the coil and detect when the frequency of the oscillator changes. The diagram below shows the best layout. 

Connect the battery and turn the transistor radio on. Tune across the dial and you get a number of spots where the radio will produce a whistle as a result of its local oscillator beating with the output of the coil of the detector. 

We got the best result at about 1400kHz and this is where the tone could be adjusted to a very low frequency. 
When the detector was swept over a 20¢ coin at about 10cm, the change in the tone could easily be detected. 
The frequency of the oscillator of the metal detector will change slightly as the battery voltage falls and as the temperature of the circuit increases on a hot day. 
This can be compensated by adjusting the frequency of the radio so that the tone is kept as low as possible. 
You are now ready to go out and try your luck. 

If you don't get a squeal from your radio after tuning the entire band, the fault will lie in the oscillator.
This could be due to the transistor not having sufficient gain to produce oscillation or some of the parts not soldered correctly. 
Try switching the circuit on and off quickly to spark it into action. If this doesn't work, check the wiring and make sure there are no shorts between the tracks. 

If you have taken too long to solder the transistor or used a very hot iron, it may be overheated and its gain will be reduced. This will prevent the oscillator starting up. Replace the transistor and take more care with soldering. 

The winding wire for the search coil is insulated with enamel to prevent the turns shorting against each other. But if you damage this coating by scraping or kinking the wire you may get two turns where the copper is touching each other. This will create a shorted turn and prevent the oscillator working. You must prevent any damaged sections touching each other. 

Do not wind tinned copper wire around the coil to hold the turns in place as this will create a shorted-turn and prevent the circuit from oscillating. 
It's best not to have any metal items near the coil as they will reduce its effectiveness. This includes nails and screws in the base-board. Metal objects that are away from the centre of the field are ok as they will have no effect. 

We have found the circuit to be extremely reliable and self-starting. If you are experiencing any difficulties, it's best to put another kit together kit as you may have damaged a capacitor or the transistor and these are extremely difficult to diagnose. 

We haven't found any valuable items with our detector but I hope you do.