AM Modulator - Optical Aproach

Months ago I bought this DIY AM radio receiver kit from Banggood. I have assembled it. (How to do this I intended to describe in separate Instructable) Even without any tuning, there was possible to catch some radio stations, but I tried to reach its best performance by adjusting the resonant circuits. The radio was playing better and was receiving more stations, but the frequencies of the receiving stations shown by the variable capacitor wheel were not corresponding to their real value. I have found that even the receiver works, it is not trimmed with the correct settings. Possibly it has different intermediate frequency instead the standard 455 KHz. I decided to make a simple AM frequency generator to trim all resonant circuits in the proper way. You can find a lot of circuits of such generators in Internet. Most of them contain some internal oscillators with embedded different number of switchable coils or capacitors, RF (radio frequency) mixers and other different radio circuits. I decided to go in more simple way – to use a simple AM modulator and as input to apply the signals generated by two external signal generators, which I had available. The first one is based on the MAX038 chip. I have written this instructable about it. I wanted to use this as RF frequency source. The second generator used in this project is also a DIY kit based on the XR2206 chip. It is very easy to solder and works fine. Another nice alternative could be this. I used it as low frequency generator. It was providing the AM modulating signal.

Again...- In the Internet you can find a lot of circuits of AM modulators, but I wanted to use some new approach – my idea was to modulate somehow the gain of a single stage RF amplifier. As a base circuit I have taken a single stage common-emitter amplifier with emitter degeneration. The schematics of the amplifier is presented on the picture. Its gain can be presented in the form:

A=-R1/R0

- the sign “-” is put to show the inversion of the signal polarity, but in our case it does not matter. To change the amplifier gain and thus to invoke amplitude modulation I decided to modulate the value of the resistor in the emitter chain R0. Reducing it value will increase the gain and vice versa. To be able to modulate its value, I decided to use LDR (light dependent resistor), combined with a white LED.

To join both devices in a single part,

I used a thermal shrinkable tube black color to isolate the photosensitive resistor from the ambient light. Further, I have found that even one layer of plastic tube is not enough fully to stop the light, and I inserted the join in a second one. Using multi-meter I measured the dark resistance of the LDR. After that I took a potentiometer of 47KOhm in series with 1KOhm resistor, connected it in series with the LED and applied 5V supply to this circuit. Turning the potentiometer I was controlling the resistance of the LDR. It was changing from 4.1KOhm to 300Ohm.

I wanted to have total gain of the AM modulator ~ 1.5. I have chosen a collector resistor (R1) 5.1KOhm. Then, I would need to have ~3KOhm for R0. I turned the potentiometer until I measured this value of the LDR, I dissembled the circuit, and measured the value of the serial connected potentiometer and resistor – it was around 35 KOhm. I decided to use 33KOhm standard resistor value device. At this value the LDR resistance became 2.88KOhm. Now the values of other two resistors R2 and R3 had to be defined They are used for proper biasing of the amplifier. To be able to set the biasing correct, first the Beta (current gain) of the transistor Q1 must be known. I have measured to be 118. I used common purpose low power silicon NPN BJT device.

The next step I to chose the collector current. I have choose it to be 0.5mA. This defines the DC output voltage of the amplifier to be close to the middle value of the supply voltage, allowing it the maximum output swing. The voltage potential at the collector node is calculated by the formula:

Vc=Vdd-(Ic*R1)=5V-(0.5mA*5.1K)=2.45V.

With Beta=118 the base current is Ib=Ic/Beta= 0.5mA/118=4.24uA (where Ic is the collector current)

The emitter current is sum of both currents: Ie=0.504mA

The potential at the emitter node is calculated as : Ve=Ie*R0=0.504mA*2.88KOhm=1.45V

For Vce remains ~ 1V.

The potential at base is calculated as Vb=Vr0+Vbe= 1.45V+0.7V=2.15V (here I put Vbe=0.7V - standard for Si BJT. For Ge it is 0.6)

To bias the amplifier correctly the
current flowing through the resistor divider must be times higher than the base current. I choose 10 times. ….

In this way Ir2 = 9* Ib=9*4.24uA=38.2uA

R2 = Vb/Ir2 ~ 56 KOhm

R3=(Vdd-Vb)/Ir3 ~ 68 KOhm.

I did not have these values in my
resistors wallet, and I have taken R3=33Kohm, R2=27KOhm – their ratio is the same as those calculated.

Finally I added a source follower loaded with 1KOhm resistor. It is used to reduce the output resistance of the AM modulator and to isolate the amplifier transistor from the load.

The whole circuit with added emitter follower is presented on the picture above.

As PCB I used a piece of perfoboard.

At first I have soldered the power supply circuit based on the 7805 voltage regulator.

AT the input I put 47uF capacitor – each higher value could work, at the output I put capacitor bank ( the same capacitor as at the input+100nF ceramic one). After that I soldered the self-made optocoupler and the pre-biasing resistor for the LED. I have supplied the board and I have measured again the resistance of the LDR.

It can be seen on the picture - it is 2.88KOhm.

After that I have soldered all other parts of the AM modulator. Here you can see the measured DC values at the collector node.

The small difference comparing the calculated value is caused from not exactly defined Vbe of the transistor (taken 700 instead measured 670mV), error in the Beta measurement ( measured by collector current 100uA, but used at 0.5mA – the BJT Beta depends in some way on the current passing through the device.; resistor values spread errors...etc.

For the RF input I put a BNC connector. At the output I soldered a piece of thin coax cable. All cables I fixed to the PCB with hot glue.

I have connected both signal generators (see the picture of my setup). To observe the signal I have used a self-made oscilloscope based on the Jyetech kit DSO068. It is a nice toy – contains also signal generator inside. (Such redundancy – I have 3 signal generators on my desk !)I could use also this, which I described in this instructable, but I did not have it at home in this moment.

The MAX038 generator I used for RF frequency (the modulated one) – I could change up til 20 MHz. The XR2206 I used with fixed low frequency sine output. I have changing only the amplitude, what in result changed the depth of the modulation.

A capture of the oscilloscope screen shows a picture of the AM signal observed at the modulator output.

As conclusion – this modulator can be used for tuning of different AM stages. It is not fully linear, but for adjusting of resonant circuits, this is not such important. The AM modulator can be used also for FM circuits in some different way. Only RF frequency from the MAX038 generator is applied. The low frequency input is left floating. In this mode the modulator works as linear RF amplifier.

The trick is to apply the low frequency signal at the input FM of the MAX038 generator. (input FADC of MAX038 chip). In this way the generator produces FM signal and it is only amplified by the AM modulator. Of course in this configuration, if no amplification is needed, the AM modulator can be omitted.

Thank you for your attention.