Building a Simple AM Transmitter

Thread Starter

rananthan3

Joined Sep 27, 2018
9
Amplitude modulation first allowed transmission of voice wirelessly using continuous wave (CW) transmission. Before CW transmission, we had damped wave transmissions which are oscillations that died out like a pendulum. Think of damped oscillations as not trying to keep swinging on a swing or not cranking your grandfather clock. Eventually, the oscillations cease. Continuous oscillations are like giving the pendulum a push to keep it going, pushing a friend on a swing or cranking your grandfather clock; they keep oscillating so long as energy is added to the system. An example of a damped oscillator is a spark gap transmitter. A continuous wave oscillator example is the Alexanderson alternator which was the device used to broadcast the first AM signal.

What is modulation? Modulation means that we are changing the signal. When you talk to someone, your voice is transmitted at certain frequencies that can only travel as far as you can be heard. In order to send a voice signal further, we have to modulate the signal to radio frequencies. The two types of modulation are AM and FM.

Now that we have a solid understanding of Amplitude Modulation, let’s see how it is implemented. We can model the circuit below as having five components: 1) source 2) amplifier 3) filter 4) oscillator 5) antenna. Theoretically the capacitors could be called a sixth component but their main purpose is for something known as AC coupling. The basic premise behind coupling is to remove DC voltage which sounds like noise at the output of the amplifier. The source in this case is a 3.5 mm stereo audio jack. The two stereo pins are tied together by being on the same row and the ground pin is grounded to the negative terminal of the battery (9V). The amplifier is no simple device. You can find the datasheet here. I found that the examples in the data sheet, specifically the AM radio power amplifier schematic, work well. If you want the details of how the LM386 was designed, here you go. The Zobel network and more specifically Boucherot cell helps to reduce noise if there were a loudspeaker at the output signal before it is fed into the crystal oscillator. You can get rid of this filter. Finally, the AM signal is sent through the antenna which is just a long wire.

After assembling the circuit, plugging in a source, and carefully tuning your radio to 1 MHz or 100 X 10 KHz, you can hear an audio signal being broadcast wirelessly. Here is the schematic I followed without the filter.



Schematic.JPG




20181016_075821.jpg 20181016_075821.jpg
 

MisterBill2

Joined Jan 23, 2018
18,167
I see the circuit posted and it will never work. The reason is that the small oscillator module is a system that requires a constant supply voltage and it delivers a constant output amplitude and frequency. Read the manufacture's specification to understand that. But that same oscillator module can drive an amplifier whose output amplitude is varied by modulating the supply voltage.

But you need to know that in order to legally transmit a useful signal at that frequency, which is in the middle of the Broadcast band, you need a license, at least in the USA. In other countries the requirements may vary.
 

DickCappels

Joined Aug 21, 2008
10,153
If you stay below 100 milliwatts power to the oscillator, have a single element antenna no longer than 3 meters, and do not interfere with licensed stations you can operate license-free under Title 47, Part 15 of the Code of Federal Regulations in the United States.
 

Thread Starter

rananthan3

Joined Sep 27, 2018
9
@Dodgydave No questions yet. @MisterBill2 Thanks for the heads up. The transmitter worked but I could not find the specifications for the oscillator. @DickCappels I measured the voltage and current at Pin 4 of the oscillator. The power = (4.24V) * (0.6A) = 2.544W unless my measurements or calculations are wrong. I guess I need to use a lower voltage battery.

Thanks for your replies.
 

BR-549

Joined Sep 22, 2013
4,928
I have never taken a "can" apart. I couldn't find info on that can. I would have thought......the first thing connected to the input(inside the can)....would be a voltage regulator.

Very interesting.
 

MisterBill2

Joined Jan 23, 2018
18,167
None more amazed than I am that it even sort of worked. To make the system work correctly, use that module to drive an NPN transistor, or an IRF510 or IRF511 FET with a tuned circuit resonant at the same frequency as the collector (or drain) load. Then modulate the supply voltage for that device, which will take a small amount of power. Now you have a classic AM transmitter! Perhaps Wendy will draw the circuit, for hose who need help visualizing a circuit.
 

MisterBill2

Joined Jan 23, 2018
18,167
It does not seem rational that the oscillator module would be drawing 600mA. That is not reasonable. 60 mA would be on the high side of believable at the rated supply voltage. But it may have a shunt zener to protect it from over voltages, although I am not aware of such modules including that feature. And in addition, the LM386 amplifier is not able to deliver 2.5 watts. So there was some sort of error somehow, or else some very abnormal operation someplace.
As for opening the module, only do that with failed ones that have no value beyond scrap, since opening it will certainly cause it to fail.
 

wayneh

Joined Sep 9, 2010
17,496
I see the circuit posted and it will never work. The reason is that the small oscillator module is a system that requires a constant supply voltage and it delivers a constant output amplitude and frequency.
Not quite right. Frequency yes, but the output amplitude depends on the supply voltage, which can vary quite a bit. So if you modulate the supply voltage, you get AM out.
 

MisterBill2

Joined Jan 23, 2018
18,167
Not quite right. Frequency yes, but the output amplitude depends on the supply voltage, which can vary quite a bit. So if you modulate the supply voltage, you get AM out.
OK, when I say "never work" I am thinking "in the manner that it was designed to work." The manufacturers will probably not claim that their little module will meet any of the published specifications unless it is supplied with the specified voltage.
 

wayneh

Joined Sep 9, 2010
17,496
OK, when I say "never work" I am thinking "in the manner that it was designed to work." The manufacturers will probably not claim that their little module will meet any of the published specifications unless it is supplied with the specified voltage.
Maybe. The specified voltage range is 5±0.5V. I have no idea what the signal modulation voltage is in my project, or this one. I suppose it probably exceeds those limits. But this thing produces a short range, low quality signal (a square wave!) by design. A little deviation in frequency is not much of a concern.
 

John_2016

Joined Nov 23, 2016
55
Hi rananthan3

1.- would you please be so kind to clarify what receiver did you use to ascertain what you wrote above

.. you can hear an audio signal being broadcast wirelessly ..

so when you say it 'worked' because you see the LM consuming, it has to 'work' on the right direction.

2.- What was the distance from the circuit? where you in the same room?
 

MisterBill2

Joined Jan 23, 2018
18,167
One real concern about using a square wave is the harmonics, which a square wave has lots of harmonics, with the result that it is also putting out a signal on those frequencies. Those are mostly odd multiples of the square wave frequency, so you would be transmitting on 3, 5, and 7 Mhz, with the 5 and 7 frequencies being in the amateur radio bands, or at least very close to them. THAT is the reason for using a reasonably high "Q" tuned circuit in an amplifier stage. Suppression of harmonics is quite important for most transmitters.
 

MrChips

Joined Oct 2, 2009
30,707
Tested this idea, first with the output of an AF function generator connected directly to the oscillator power pin.
Next, I took audio output from mp3 player into the base of a PNP transistor as the high-side driver into the oscillator power.

Both signals were picked up clearly on an AM receiver.

I examined the RF spectrum of the unmodulated 1MHz signal on an RF spectrum analyzer and it extended beyond 160MHz, at every 1MHz.
 

MisterBill2

Joined Jan 23, 2018
18,167
Tested this idea, first with the output of an AF function generator connected directly to the oscillator power pin.
Next, I took audio output from mp3 player into the base of a PNP transistor as the high-side driver into the oscillator power.

Both signals were picked up clearly on an AM receiver.

I examined the RF spectrum of the unmodulated 1MHz signal on an RF spectrum analyzer and it extended beyond 160MHz, at every 1MHz.
Undoubtedly that was a square wave device
 
Top