# Why are audio transformers required?

I was looking at simple AM radio schematics and came across this one:

As you can see it has an audio transformer. I'm still attempting to grasp the concept of impedance and such, so can someone explain why this is necessary? And how would such a transformer even work to change the impedance? What would be the difference if you didn't use an audio transformer?

• The schematic is quite meaningless! Dec 14, 2011 at 19:32
• @Leon Heller - It is unusual - looks like what you see with eg ZN414 receivers where power feed and audio are combined. And RF too in extreme cases :-). Dec 14, 2011 at 21:34
• is that supposed to be a receiver or a transmitter? looks more like a transmitter to me. Dec 14, 2011 at 21:50
• I've added the link to where I found that schematic. I didn't realize it would be such an odd schematic since it was suppose to be a learning resource Dec 14, 2011 at 22:10
• Definitely not a good circuit to understand the typical use of audio transformers. Mar 12, 2014 at 16:53

This is a bit of a tricky circuit.

Normally, a transformer produces a voltage ratio matching its turns ratio, a current ratio that is the inverse of its turns ratio, and therefore an impedance ratio that is the square of its turns ratio.

Now in this circuit, we have a mystery box which is most likely a square wave clock oscillator. By appearance, the transformer secondary is being used to couple the audio as an A/C "ripple" on top of it's power supply, in the hopes that this will produce some AM modulation of the output.

It's not entirely clear that the transformer is being correctly applied; without knowing the output impedance of what the jack is plugged into or the beyond-data-sheet properties of the oscillator, we can really only speculate if the transfer is best the way shown, turned around the other way, substituted with a 1:1, etc. Likely this is a "pragmatic" circuit as much as a "calculated optimal" one.

It's possible that the use of a transformer at all may be primarily to provide isolation between the circuits. Powering through a small series resistor with a capacitor to couple in the audio could be another option, though perhaps less efficient.

There are two additional problems which merit some thought before building this:

1) The oscillator probably isn't rated for a 9v supply. Most want 5v, or 3.3 or perhaps today something even lower. It's not clear that the DC resistance of the secondary will drop the supply voltage enough under this small load to be within the limits.

2) The oscillator is going to output a square wave, which is rich in harmonics. Without a low pass filter to round the square wave to a perfect sine wave, this will not only transmit at 1 MHz as intended, but also at 3, 5, 7, 9, 11, etc MHz, potentially up into places where such spurious emission produces harmful interference (for example, 7 MHz + the audio frequency would land in the morse code allocation of the 40m ham band, where trying to receive extremely weak signals is common and interference detested) . Needless to say, there are regulation about spectral purity for various transmitter power levels.

• Transmitting square waves directly is U-G-L-Y. Even a simple passive RLC LC filter would improve things a lot. Dec 14, 2011 at 21:43

From the very limited schematic you show, it looks like the transformer is being used to add the audio signal to the 9V supply. There are other ways that could have been done, but a transformer is one way. It also looks like they wanted to low impedance signal from a much higher impedance "line level" signal.

However, it is wrong to infer that audio transformers are always required just because this circuit used one. In some cases audio transformers are used to get immunity from common mode noise, but there are other ways to couple audio signals too. As should be expected, there are various tradeoffs.

If you try to interpret the circuit as a very simple AM transmitter, then the transformer is acting as a 'modulation transformer'. Essentially, without it, the supply to the oscillator is steady DC, so the amplitude of the RF out of the oscillator is constant. Now push an audio signal into that transformer, and the supply that the oscillator sees goes up and down with the audio, and so the amplitude of the output goes up and down as well. Instant AM modulation.

Fwiw, transformers change impedance because they change voltage and current ratios. If the secondary has higher voltage than the primary, then it will also have lower current, and the ratio of voltage to current in the secondary (i.e., the impedance) will be higher than the equivalent ratio in the primary. In a transformer with a 1:1 turns ratio, the voltages and currents are all the same, so in that situation, impedance isn't changed at all.

So as far as that goes, in this circuit, I don't believe it matters a whole lot about impedance matching, as all the transformer is doing is allowing the audio signal to be imposed on the power input to the oscillator, while isolating the audio source from the battery voltage. I would bet the circuit isn't too picky about the particular turns ratio. As a matter of fact, the percentage modulation would depend on the turns ratio, so you might experiment with different transformers if the signal seems weak.

You need a transformer because oscillators use audio jack's power and they need at least 3 volts to work. (3-12 V) But audio output of any non-amplified sound player is nearly 1 volt. You can't run a oscillator with 1 volt.

Audio transformers are power supplies of AM radio transmitters. If you can't find a transformer you can use an amplifier, too. I tried to connect a PC fan to an amplified sound source. Those fans need at least 5 volts to work. And my fan worked!

By running the Vss supply through the secondary coil of a transformer whose primary coil is connected to an audio input source such as a microphone, the actual voltage supplied to the oscillator will fluctuate based on the variations in the input signal.

Because crystal oscillators are very stable, these voltage variations won’t affect the frequency generated by the oscillator, but they will affect the voltage of the oscillator output. Thus, the audio input signal will be reflected as voltage changes in the oscillator’s output signal.