HOME Schematics Discover Forum Hobby Corner Dave's Circuits Electronic Resources Book Corner Contact Info


  Experimenter's Corner

  Previous Issues

Low Power AM Transmitter
By: Dave Johnson

A while back, I got an email from someone who drove around in a restored 1960s car.  He wanted to be able to leave the old carís original AM radio intact but he wanted to play music from his Ipod MP3 player through the carís radio. In other words, he wanted a

radio adapter, similar to those often used in cars today, which send the audio information to a carís FM radio. But he wanted to transmit the information on AM instead.  I didnít have a ready solution for this at the time but Iíve been thinking about it.  I thought it would be an interesting project.

There are a lot of AM radio transmitter circuits posted on the Internet.  Many use a few transistors as audio Amps and oscillators and are designed as low power AM transmitters, feeding a wire antenna.  The audio signal feeds some of the transistor bias circuits, causing the Amplitude of the RF signal to be modulated.  But, for this application, only a very short range is needed.  I might try and come up with my own design.  I think the antenna for this thing would require some careful thought and perhaps some experiments.  The transmitter does not need much power, since the distance between the transmitter and the carís AM radio receiver would only be a few feet. Still, some experiments would be necessary to see just how much transmitter power would be needed and what type of antenna works the best.  I would like to keep the overall transmitter package compact.  Maybe a ferrite rod antenna could be used so there would be no a need for any antenna wire.  I could also tap into the carís 12v power though the cigarette lighter.  That way I would not have to worry about a battery pack.  After some experiments perhaps I could then decide if the transmitter could be battery powered.

AM radio from the 1960ís

The US AM radio band ranges from about 540KHz to 1.6MHz.  Many AM dials also dip down to about 530KHz.  The low frequencies might be a good target since few radio stations transmit at that end.  However, 530KHz is often used for road condition alerts and weather reports.  Every region of the country will have different clear AM channels, so perhaps some transmitter frequency adjustment or selection will be needed. 

AM Radio Dial

The best way to insure a rock solid transmitter frequency is to use a quartz crystal.  However, a quick review of what crystal frequencies are available from the Digikey catalog was disappointing.  One 4.43MHz crystal, when divided down by 8, would yield a frequency of 553KHz.  That might work.  Also, ceramic resonators, although not as accurate as a crystal, are available in 540KHz, 640KHz, 800KHz, 1MHz and 1.22MHz frequencies.  A selector switch could be used to pick one of those.  I also could use an LC oscillator and vary either the capacitor or the inductor to change the frequency.  This is a bit less desirable, since one would not know what frequency was ďdialed inĒ unless the adjustment dial could be calibrated.  Or you could tune the carís AM radio to a clear channel first, and then adjust the transmitter frequency to match.  That would work.  Unless the car is being driven across country, once the transmitter frequency was set, it should be OK from then on.
A RC type oscillator could be configured to run at the right frequency but it would tend to drift due to temperature and supply voltage changes.  A more stable RC oscillator might use the LTC1799 from Linear Technology. This deviceís frequency can be set with a single resistor and would be much more stable with temperature.  A carefully designed circuit could allow a multi-turn pot to adjust the radio transmitter frequency over a narrow range. In fact, it might be possible to use fixed resistor values to select the desired frequency.  A chart could be made which showed the relationship between the resistor value and the transmitter frequency. Resistors with a 1% accuracy are pretty common.
The conventional way to produce an Amplitude modulated RF signal is shown below.  A RF oscillatorís Amplitude is varied by a DC voltage, corresponding to the audio information.  This is the classic ďamplitude modulationĒ method.  The signal emerging from the circuit is usually connected to a LC tank circuit, to filter out the harmonics and provide a better impedance match to an antenna.  If I want to keep the range of an AM radio transmitter short, I should try to restrict the size of the antenna.  A pure magnetic field transmitter should work, using a coil of wire as the antenna.
   
A variation of the classic oscillator modulation method is shown below.  It varies the Amplitude of a square wave RF signal generator.  Its output would then drive a series resonant LC network with a medium Q value.  The square wave drive would be turned into a nice sine wave transmitter signal.  The inductor of the LC network would send magnetic information to the carís radio.  In addition, a short piece of wire would provide the electric field connection.  I like this idea but again, it would take some experimentation to see what kind of coil would work best.
   
Another way to produce the Amplitude modulated signal would be to use a pulse width modulation (PWM) scheme as shown below.  This method is often used in ďclass DĒ audio power Amps and allows for a very efficient driver.  It might be an interesting experiment to try the PWM technique.  Instead of varying the drive voltage with audio information, the pulse width of the RF drive signal would be modulated.  The PWM signal would be set for a duty cycle of about 25%.  The audio signal would then increase and decrease the pulse width of the RF.  The Amplitude would be maximized at a 50% duty cycle and would be much less, as the duty cycle (pulse width) became shorter. Again, a series resonant LC tank circuit would smooth out the signal into a clean sine wave whose Amplitude would vary with the audio signal.  Although this type of circuit would be more complex, it would be more efficient than the classic AM method.
To send the information from the transmitter box to the carís radio, some kind of antenna would be needed.  I would like to try a pure magnetic method to bridge that distance.  Inside nearly every AM radio receiver is a ferrite ďloopstickĒ antenna as shown below.  This type of device might make a nice antenna for the transmitter as well.  It is a bit directional and would need to be oriented properly to the carís own antenna to work but it would keep the system compact.  If necessary, a better antenna might use a larger ferrite rod.  These are available in lengths out to about 10 inches.  Another option would be a large coil loop as shown below.  This would certainly work but would be fairly large.
AM Radio Loop Stick Antenna AM Radio Coil Antenna  Ferrite Rod
I think the first order of business would be to connect my signal generator, set at 530KHz, to a push-pull type driver circuit similar to the circuit below.  I can chop the RF at a 50% duty cycle at say 400Hz, to produce an AM signal.
Push-Pull Driver
I would then measure the inductance of an off-the-shelf loop stick antenna coil and find a suitable capacitor, so the LC network will resonant at 530KHz.  I will then see if that circuit can transmit enough RF signal to the carís AM radio.  I can vary the driverís supply voltage, to control the power to the antenna.  This will give me an idea if a small loop stick coil will work.  It those results are disappointing, I can move up to a larger ferrite rod or maybe even an open air coil. I will let you know what I discover.

LTC1799 Data Sheet

Please send comments to me


October 2010     Issue 12

Page 1 Circuit
Diagnosis
Experimenter's
Corner
Good Idea
gone Badly
New Products What the World
needs Now
Wily Widget


HOME Schematics Index Hobby Corner Dave's Circuits Electronic Resources Contact Info
Imagineering Ezine    Discover Solar Energy Dave Johnson & Associates Faraday Touch Switches


 About Us   |  Advertise on DiscoverCircuits.com   |   Report Broken Links  |    Link to DiscoverCircuits.com  |    Privacy Policy

Copyright  2002 - 2015 David A. Johnson & Associates.  All Rights reserved. 
 Linking is ALLOWED but COPYING any content or graphics to your web site is EXPRESSLY PROHIBITED.