Learn about the transformer feedback oscillator with the:


Piezo Siren Circuit

The Piezo Siren circuit is a simple arrangement driving a piezo diaphragm.

A very simple circuit to show
how a transformer feedback
oscillator works.

Getting the circuit to work will depend on a number
of factors including selecting the correct 10mH
choke and piezo. You may have to try various
types and swap the feedback winding to achieve
oscillation. Read the article before doing anything.

Step by Step

No kit is available for this as we are only interested in describing how the circuit works. But to get the most out of article you should put it together using parts from your junk box.

The piezo diaphragm can come from a music card and if you have two or three diaphragms, they can be added in parallel-as shown on the diagram to see how they affect the frequency of the output. The only component you will have to make is the "transformer."

This is a standard 10mH inductor with 80 turns of enamelled wire wound around it to create the feedback winding. The inductor

can be found in a number of our kits such the light alarm, battery backed piezo siren or purchased as a separate component from a large electronics store.

Even though the circuit is very simple, it introduces a number of important features and that's why we have decided to cover every aspect in detail.


The circuit uses a transistor to drive a piezo and when we discuss the operation of a transistor we are describing TRANSISTOR ACTION. This is the action of a transistor amplifying the current entering the base and allowing a larger current to flow through the collector-emitter circuit. TRANSISTOR ACTION is simply another name for TRANSISTOR AMPLIFICATION.

The circuit also demonstrates TRANSFORMER ACTION (the component across the piezo diaphragm is essentially an inductor with an over-wind to supply positive feedback to the base of the transistor).

These two windings demonstrate TRANSFORMER ACTION (the action of a waveform in one winding being passed to another winding. The waveform is passed via electro-magnetism - a magnetic field - and this is magnetism produced by electrical current.)

The inductor and piezo diaphragm in parallel form an arrangement called a PARALLEL RESONANT CIRCUIT.

The overall circuit forms an oscillator called a TRANSFORMER FEEDBACK OSCILLATOR.

This gives us four topics to discuss. We will discuss them in a non-technical way so you don't get lost in technicalities.


Transistor action is simply the amplifying action of the transistor. A transistor amplifies the current entering the base and causes a higher current to flow in the collector-emitter circuit. The ratio of these two is the gain of the transistor and is generally about 100, however the gain can range from 20 to 400 or more, depending on the type of transistor and the value of the surrounding circuit components.

Our circuit turns on when the transistor receives current into the base via the 47k base-bias resistor.

The transistor amplifies the current about 100 times and allows the higher current to flow in the collector-emitter circuit. This is called TRANSISTOR ACTION or TRANSISTOR AMPLIFICATION.

The base bias resistor is designed to partially turn the transistor ON and this causes a medium amount of current to flow through the collector-emitter circuit.


Connected to the collector is a coil of wire called an inductor. The wire is wound on a ferrite core. Ferrite is an iron material in which the particles of iron are surrounded by an insulating material so that the iron particles are magnetically separate.

This allows each particle to form very small magnetic dipoles and prevents currents called EDDY CURRENTS from passing through the material.

When current passes through the winding it produces EXPANDING MAGNETIC FLUX and this flux cuts the turns of the winding we have added, called the feedback winding.

This feedback winding is connected to the base of the transistor so that the current it produces is ADDED to the current supplied by the base-bias resistor and this causes a greater current to flow in the base of the transistor.

This causes the transistor to turn on more with the result that a higher current flows through the inductor.

The current continues to increase until the transistor is fully turned on and at this point in the cycle the flux is a maximum but it is NOT EXPANDING. Only expanding flux cuts the turns of the feedback winding and produces a current in it. Stationary magnetic flux does not produce (induce) a current in the over-wind and thus the current produced by the feedback winding ceases.

This causes less current to flow in the base of the transistor and the transistor turns off a slight amount (don't forget, the base-bias resistor is still providing a small amount of turn-on current).

The current through the inductor reduces and the flux begins to reduce. This causes a collapsing magnetic field to be present and this flux cuts the feedback winding to produce a current in it of opposite polarity.

The feedback current opposes the current supplied by the base-bias resistor and the base sees a lower current.

This causes less current to flow in the base and the transistor turns off more.

This action continues to run around the circuit until the transistor is fully turned off.

At this point the current supplied by the base bias resistor takes over to start the cycle over again.

The number of turns on the feedback winding is worked out by winding sufficient turns to get the circuit to work then removing a few at a time until the circuit fails to work. A few turns are then added to guarantee operation. The energy from the feedback winding must be enough to maintain oscillation.


As far as the circuit is concerned, a piezo is a capacitor of about 10n to 20n. The inductor and capacitor form a tuned circuit or PARALLEL RESONANT CIRCUIT and if the piezo is removed, the circuit will not operate (see below for the waveform produced when the capacitor is removed). A tuned circuit consisting of a coil and capacitor produces a waveform very similar to a sinewave when energy is supplied to them. The transistor plays no part in the shape of the waveform, it merely turns on at the correct instant to supply the pulse of energy.

Piezo diaphragms give the loudest output at about 3kHz to 5kHz and two or three can be added in parallel across the inductor to alter the frequency and bring

it near the 3-5khz range. This will give the loudest output from all the piezos.


A circuit oscillates when its input receives an IN PHASE signal from the output. That is: a signal that ADDS to the input signal. This is called a positive feedback signal.

Oscillation is sometimes called instability and this always occurs when positive feedback is present.

The feedback winding in our circuit must be wired so that positive feedback occurs. If the circuit does not oscillate when first switched on. simply reverse the feedback winding and oscillation will Occur.

This type of circuit is called an Armstrong oscillator (ref: Understanding Oscillators by Barry Davis P 94.)

The values that set the frequency of oscillation are the inductance of the inductor and the type of piezo diaphragm.


We said we were not going to get into any complex discussions in this article and I hope you have followed us up to now.

If I said you can buy hundreds of different 10mH chokes, you may start to wonder if you will ever get this project working.

Fortunately most of the chokes will work in this circuit but just in case you get an unusual one, I want to mention why some may not work.

The two we tried had a DC resistance of 4 ohms and 180 ohms and it would be reasonable to assume anything in between will be suitable. However you can some 10mH chokes outside this range and their operation will be unknown. Especially chokes with very low resistances.

Why can the resistance vary so much how is the value of inductance determined?

A particular value of inductance such as 10mH can be obtained by winding a few turns of very thick wire around a large core or many turns of thinner wire around a smaller core or lots of turns of very thin wire around a very small core.

Each of these will create a 10mH choke and give the same result in a "test circuit".

What do we mean by the same result? Each will create the same amount of magnetic flux in the core of the inductor when a known current flows.

But when these inductors are used in our circuit they will behave differently because of the different DC resistances. (In the laboratory, a different voltage will applied to each to achieve the same current flow due to the different resistance values). We have a fixed voltage 3v for our circuit and so a medium resistance coil is required.

If the resistance is too low, the transistor will have difficulty passing sufficient current to generate the necessary flux. If the coil resistance is too high, the supply voltage will be insufficient to pass the required current.


Firstly, the waveform appearing across the piezo diaphragm is generated by the parallel combination of the coil and piezo. We have learnt that the piezo is seen by the circuit as a capacitor and this means the coil and piezo form a parallel tuned circuit.

The transistor plays no part in the generation of this waveform.

The transistor merely delivers a small amount of energy to the resonant circuit. The coil and capacitor will then continue to create a waveform for a number of cycles by passing energy from the coil to the capacitor and back again. But each time a small amount of the energy is lost by the piezo when it distorts the diaphragm in the process of making a sound and the waveform would eventually die away.

Some of the energy is also lost when it is passed to the feedback winding and injected into the base of the transistor to turn it on.

The job of the transistor is to inject a small amount of energy into the inductor /capacitor combination at the correct portion of each cycle so that the waveform is maintained.

The frequency of the waveform is determined by the value of the coil and capacitor (piezo). The transistor or the 47k base-bias resistor has no effect on determining the frequency. If the capacitor (the piezo) is removed, the frequency will increase and change shape into a very spiky waveform.

The timing-effect or "controlling effect" of the capacitor has been lost and the frequency is now determined by the speed with which the coil can produce magnetic flux and the time for the transistor to turn on.

The amplitude of the waveform will also increase from 3v to 60v and this could damage the transistor. In fact the 60v we measured across the coil represents the zener voltage of the transistor as the transistor is 'zenering' when the capacitor is removed. That's why a simple circuit such as this can damage a transistor when the piezo is removed.

This is a completely different set of conditions and will not be covered in this discussion.

One of the important features of the capacitor (the piezo) is the "controlling effect" on the frequency. The capacitor takes time to charge and discharge and this determine the frequency of the waveform.

As more piezos are added in parallel across the coil, the frequency reduces and as it comes into the range 3kHz to 5kHz, the output of each piezo increases.

Before leaving this discussion we will go over the twelve step-by-step diagrams on the previous pages. They explain a very important concept called TRANSFORMER ACTION and if you have built the circuit you will be able to carry out a simple experiment to show the effect of removing turns from the feedback winding.

Simply loosen the turns and slip some of them off the top of the inductor. A point will be reached when the circuit ceases to operate. This is the CRITICAL VALUE and by adding a few turns onto the inductor the oscillator will start up again.

When designing a circuit like this, you must add a few extra turns so that the circuit is guaranteed to start-up every time.

As you remove the turns you are supplying the transistor with less energy and a point is reached where the transistor does not have sufficient gain to allow enough current to flow through the coil and create the required amount of magnetic flux for the feedback winding.



The first frame shows the transistor turns on when current enters the base via the 47k resistor.

The transistor amplifies this current and allows about 100-200 times the current to flow in the collector-emitter circuit.


This current also flows through the coil and produces magnetic flux. The flux lines are shown on the diagram as upward lines but this is only diagrammatical.

We have shown upwards lines to represent expanding flux and downwards lines to represent collapsing flux. The lines are actually imaginary but the important point is the strength of the magnetic field is INCREASING.


The turns of the feedback winding are wound over the main coil and the expanding flux cuts the turns to produce a voltage. It also produces a current but the size of the voltage and current is not important at this stage. The thing that is important is the POLARITY OF THE VOLTAGE.

The feedback winding must be wired so that the voltage ADDs to the voltage from the 47k resistor. When the voltage ADDS, more current will flow into the base of the transistor. This causes the transistor to turn ON more.


When the transistor turns on more, a higher current flows through the coil and this produces more flux.


The higher flux produces more voltage (and current) in the feedback winding and a cyclic action results with the transistor turning on more and more.


A point is reached where the transistor is turned on FULLY and the magnetic flux produced by the coil is a MAXIMUM but it is NOT EXPANDlNG and the feedback winding does not see any moving flux. Therefore it does not produce any voltage or current AT ALL.


This causes the transistor to turn off almost completely - to the level provided by the 47k resistor.

The current in the coil falls to almost zero and the magnetic flux cannot be maintained.

The result is the magnetic flux collapses. In other words it becomes REDUCING MAGNETIC FLUX and this flux cuts the turns of the feedback winding. (The flux we are talking about is in the ferrite core of the inductor.)


We can consider this reducing magnetic flux to be REVERSE MAGNETIC FLUX as it produces a voltage and current in the feedback winding that is of opposite polarity to that in frame 3. The current produced by the feedback winding is opposite to that supplied by the 47k resistor and thus the transistor turns OFF completely.


The magnetic field continues to collapse until all the magnetic energy from the core has been converted to electrical energy in both windings.


At this point in the cycle two things happen.

  1. The electrical energy produced by the main winding produces a very high voltage (with very low current) and this voltage is passed to the capacitor (the piezo) where it distorts the diaphragm see Piezo Diaphragm notes on page 60.
  2. The energy produced by the feedback winding keeps the transistor OFF.


The magnetic flux contained in the ferrite core of the inductor is completely converted to electrical energy in the two windings.


When all the magnetic flux is converted to electrical energy, the reverse voltage produced by the feedback winding ceases and the transistor gets turned on again via the 47k resistor to start the next cycle.

This action repeats 5,000 times per second and the piezo diaphragm bends and "dishes" to produce the annoying sound we know as a "piezo."


This simple circuit highlights the complex nature of a transformer. Whenever you see a transformer in a circuit you will not be able to work out what it is doing unless you know the parameters of the components driving it.

The only way to get these parameters is to see the waveforms on a CRO (Cathode Ray Oscilloscope).

The size and shape of the waveform will depend on many factors: The size of the transformer, the number of turns, the core material, the number of windings, the loading on the transformer, the frequency of operation, and many other things.

Transformers are very handy devices and will always find their place in modern circuits as they can convert a voltage to a higher or lower value, transform current to a higher or lower value, modify waveforms, invert pulses and a dozen other things.

Three things we have covered in this discussion are:

  1. A waveform (or signal or voltage or current) is passed from one winding to another via magnetic flux. In other words the signal in one winding produces magnetic flux and this flux cuts the turns of the other winding and produces a waveform (or signal) in that winding.
  2. The transfer of a waveform only occurs when the signal in the first winding is changing.

There is one thing that changes the waveform considerably. If the driving circuit is switched oft abruptly, the magnetic flux contained in the core of the transformer collapses and cuts the turns

of each winding and produces a waveform that can be very high in value.

This is sometimes called the FLYBACK EFFECT such as used in a flyback circuit with a flyback transformer.

This is why transformers are such fascinating devices.,

Main Index
KIts And Projects
Basic Theory
Return to the top

This article was taken from Talking Electronics