Nifty AM Transmitter From Three Components

I always wondered how well those would work, and they're so simple I've never tried one, being on school science pages and videos.

I've seen the can oscillators programmed for broadcast band circuits, with the output fed to an RF final transistor stage that's then modulated.

That would be the best way to really do it, but it's cool the oscillator can be modulated without going out of lock or breaking up, since the audio is modulating the oscillator itself. Most of those oscillators are TTL too, if you could get CMOS versions, it should work even better because of the wider voltage variation CMOS can take.

... it's cool the oscillator can be modulated without going out of lock or breaking up, since the audio is modulating the oscillator itself. ...

_______

True-ish, but the percentage of modulation shown in those r-f waveforms was rather low -- which would produce a relatively poor audio S/N ratio at the receiver output compared to a fully-modulated, radiated AM signal having the same carrier power.

Recently I posted in another thread about my 13.560 MHz project.  It is the same basic idea using the 4 pin IC oscillator and transformer.

I followed it with an LC output coupler which gives a typical sine wave modulated output.

The modulation approaches 65 percent.  I plan to incorporate a modulated bufferoutput stage rather than modulate the oscillator itself.  That should improve it.

In all its glory, here is an ugly build of the simple 13.560 MHz transmitter.

Hard to see in this picture.  The two paralel LED's are used to bring the supply voltage down to around 3 volts from 5 volts.  There is an electrolytic cap across the transformer on the oscillator side to cut off some high end audio.

In the schematic you can see whats what.

The unmarked LC series network connects the oscillator output to the antenna.  The RF choke to ground rounded out the output changing it from positive going pules to a sine wave.  I used two LED's to supply enough current.  Just one LED looked a bit to bright.

And finally, here is a video showing the envelope pattern on a 'scope.  The audio you hear was playing on the shortwave radio at 13.560 MHz.  Keep in mind the audio chain was; audio broadcast over FM at 93.5 MHz to a pocket FM radio.  The pocket radio audio is fed to the 8 ohm side of a 70 volt audio line transformer.  The output side of the 70 volt audio line transformer is in series with the power connection to the oscillator.  I picked the wattage tap of the output side which gave the best symmetry.  The hand-held audio/video was captured with my SmartPhone.

Simple TX Video  You can see I'm actually getting more than the reported 65% modulation.  Granted this is a rough experiment but seems to work.

Next I'll couple the oscillator output to a buffer/output stage and modulate that instead of the oscillator itself.

I'm impressed with the modulation on just the oscillator can itself mram1500! It looks like 100 percent, with sharp peaks upwards, and hitting the baseline going down! Now you need to do a trapezoid test on that 🙂

I wonder how it can do it, if there's a capacitor inside on the oscillator section that's holding it up, I'd just think the PLL inside the can would be unstable with the changing audio voltage.

I think it will be better of course with an output transistor stage, and I've seen a schematic using a 5 volt oscillator can to directly drive an output MOSFET for 100 mW with an LM386 audio amp as a modulator. Easier to couple into an antenna coil too with a standard circuit.

Took some tinkering to get the output to be symetrical.  When operated at the suggested 5 volts, the upper peaks were squashed.  When operated below 3 volts, the downward peaks would cut off and upper peaks greatly exagerated.

The unmarked coil and cap were simply trial and error (emphasis on error:) until things came together.  The coil is just a few turns of hookup wire so I don't have a clue as to what inductance it is.  The cap is maybe one or two hundred pico-farads.

Maybe I'll try feeding the audio direct off the board instead of my "repeater" setup.

I don't think this oscillator can has a PLL in it.  Rather, probably just a crystal and IC type of oscillator.  The first one I had was accidentally destroyed when I plugged into the wrong (12 V) power buss.  Phssst...

Sounded good.  Wish I had one of those to try and start a station on SW.  Wonder how far it would go?

I don't know if it applies to mram1500's situation or not since I cannot view his video but there is a trap when using a scope to view the envelope of a modulated waveform.  If DC or audio frequency AC is coupled to the scope input along with RF then the observed pattern will be a composite of these signals and won't accurately show the RF envelope.  To get a true display of the envelope it is necessary that only signals at the RF frequency get coupled to the scope.  I wasted many hours debugging transmitters until I realized this.

Unless done properly the display from even a good transmitter will look non symmetrical and highly distorted when modulated. The simplest way to do this right is to place the scope probe near the antenna connection or clip it over the INSULATED antenna wire.  The very low parasitic coupling capacitance between the probe and the circuit will only allow the RF through.  If precise measurements of voltage are needed then the probe should be coupled to the circuit using a small capacitor which will pass only the RF signal.  Unfortunately, the value of this capacitor depends on circuit parameters but around 5 pF should work for RF above 1 MHz when using a 10X scope probe with 10 Mohm input R.  Even with this small cap some high frequency audio might get through.  It might be better to use a 1X probe with an R of 1 Mohm and calculate the C = 1/2piRf to set the corner frequency well above the audio range.

Neil

.

Although the scope was direct connected to the antenna, in the past I have simply connected it to a choke used as an RF pickup.

I'll have to try that and see how it looks.

Probably a few miles on 13.56 shortwave, a few milliwatts into a dipole can get out well.

 I have I have to see if I can get one of those 13.560 rigs fully assembled. It would be fun just to see what would happen. I've been on a good day you could actually shoot some DX.