1.0MHz Tx/Rx using ASK modulation

Discussion in 'The Projects Forum' started by BrentM, Oct 26, 2012.

  1. BrentM

    Thread Starter Member

    Oct 26, 2012
    43
    0
    I have had an Idea for a project for a while, and havent had a whole lot of time to work on it. Last month I finally began piecing together a simple design. I plan to build a wireless transmitter and receiver system. This system will be as simple as possible using easy to find components. The DSP portion of this will be handled by a microcontroller via software (probably written in C). The goal of this project is not to create a cutting edge technology that will be used in the next iPad, but to create a proof of concept to increase my knowledge of digital processing and wireless transmission.

    The modulation I have chosen to use is ASK, Amplitude Shift Keying. This is perhaps the simplest method of digital modulation and could be accomplished using a few components. Demodulation of an ASK signal is also easily accomplished with minimal components.

    I will be using an RF carrier frequency of 1.0MHz. The main reason I will be using this frequency is largely due to the fact that the only oscillator I have us 1.0MHz and I am too lazy/cheap to buy another one. This frequency is also rather convenient as it will be easier to find discrete component values to build filters, amplifiers....etc.

    I plan to process the received digital signal using a microcontroller. I have yet to decide which MCU I will use, but will probably use an arduino or PIC, because those are the only ones I own and (as usual) I am too lazy/cheap to buy anything else. The bulk of the software will be written in C with a possibility of assembly being used at some point.

    The final design will include two MCU units. One unit will utilize the transmitter to send a 4 bit binary value. The second unit will utilize the receiver to demodulate/process the received signal and display the 4 bit value to a small LCD display.

    Stay tuned for more to come.

    Brent
     
  2. BrentM

    Thread Starter Member

    Oct 26, 2012
    43
    0
    I have come up with a flow chart to illustrate the working of the transmitter. You don't even have to tell me how professional/awesome the flow chart is. I'm sure engineers all over the world utilize the many advanced features of Microsoft Paint in their schematics. So lets get started:

    Since the output of the oscillator is a 5Vpp square wave with a duty cycle of 50%, I will need to filter this to produce a decent sine wave. A square wave, in theory, contains infinite frequencies, and since we want to transmit on a single frequency, a filter will be needed. I have decided to implement a Multiple Feedback Bandpass Filter using an LF-351 op amp. This will give me a decent two pole filter with a single op-amp and 4 other components. It will also slightly amplify the signal. The MFBP filter will be designed to have a center frequency of 1.0MHz and a quality factor, Q, of 6.

    The voltage amplifier will be implemented with an inverting amplifier using the LF-351. depending on the amplitude of the signal going into the voltage amp, the gain will need to be adjusted. I know the input signal will be relatively small, so I will assume a gain of 22 will be sufficient. Since the rail voltages that I will be using for the design will be +5v/-5v, a maximum output of 10Vpp will be possible.
     
  3. BrentM

    Thread Starter Member

    Oct 26, 2012
    43
    0
    I have completed theoretical calculations for the transmitter portion of the design. I will be building and testing the design sometime this weekend. Take a look at the schematics and let me know what you think.

    Brent
     
  4. BrentM

    Thread Starter Member

    Oct 26, 2012
    43
    0
    Thanks for the input. I checked the output in the OScope today, and it's operating just as I had hoped. Im getting an 7.5Vpp ASK modulated signal. The filter didn't completely remove the higher frequencies from the oscillator, so the RF signal is closer to a triangle wave (still very sin-ish). I will be taking more measurements this weekend and will compare the results with my theoretical work. I have attached a crappy snapshot of the OScope signal. I will be uploading better pics soon.
     
  5. BrentM

    Thread Starter Member

    Oct 26, 2012
    43
    0
    I have a question though: How do I go about increasing the range? Do I use a voltage amplifier (as I have done) or should I use some other kind of amplifier?
     
  6. Papabravo

    Expert

    Feb 24, 2006
    10,142
    1,790
    Were you planning on sending this signal through a cable or over the air?
    If over the air you can use a power amplifier or build a resonant antenna. At 1 MHz., it will be a pretty large antenna.
    If over a cable your range will be limited by cable losses so you want to build a bi-directional repeater.

    Are you aware of the rules on interfering with other users of the RF spectrum?
     
    Last edited: Oct 26, 2012
  7. BrentM

    Thread Starter Member

    Oct 26, 2012
    43
    0
    This is designed to be a wireless system. I will be transmitting ASK modulated binary data short distances (probably between two arduino's or something).

    Yes, there are rules and regulations regarding the us of the AM spectrum. As long as the range and output power of this circuit does not exceed certain guidelines, I should be all set. This is a small low-power transmitter, and will probably not transmit more than 20 or 30 ft.
     
  8. BrentM

    Thread Starter Member

    Oct 26, 2012
    43
    0
    What is the basic design behind a power amplifier? Does it amplify current or voltage? It wouldn't make sense that the current would be amplified, because the antenna would have no path to ground and therefore there would be no current flow into the antenna. If I remember correctly, a rising and falling voltage on a wire creates a electromagnetic field proportional to the amplitude of the voltage applied(I think?).

    Also, how does one measure the power output of a transmitter? I just saw another post on here that said to measure the output signal with an Oscilloscope and presented the equation:

    (Vpp*Vpp) / (8*RL), where RL is the input impedance of the scope probes.
     
    Last edited: Oct 26, 2012
  9. Papabravo

    Expert

    Feb 24, 2006
    10,142
    1,790
    Given the range of 20-30 feet a "power amplifier" seems like overkill. An ordinary transistor will produce a suitable power gain at that frequency of 1 MHz. A couple of milliwatts should be sufficient for you purposes.

    To answer your original question a "power amplifier" amplifies both voltage and current. The antenna with "no DC path" to ground nevertheless has AC voltage and current and can give you a nasty burn that will take a long time to heal.

    Power can be measured with an RF Wattmeter. The impedance can be measured with an SWR Bridge. When you minimize the SWR you make the transmitter-feedline-antenna system more efficient at radiating power. Given the long wavelength of a 1 MHz signal you will be in the near field which has different characteristics from the far field.

    In short -- without spending a boatload of money on RF gear -- just build it and try it. A word of warning. If your neighbors complain and you get a letter from the FCC -- do anything you want except don't ignore the letter. Contrition and apology go a long way toward avoiding jail. Don't believe me? Google Jack Gerritsen KG6IRO for an apochryphal tale.
     
  10. BrentM

    Thread Starter Member

    Oct 26, 2012
    43
    0
    Interesting, the AC current flow makes perfect sense. I'm learning something new everyday.

    Do you think a simple CE amplifier with a standard NPN transistor will work?

    The voltage amplifier(using the op-amp) I have already implemented, puts out a maximum of around 25mA with a max voltage swing of 10Vpp. Would this cut it? or is there simply not enough current output?
     
  11. THE_RB

    AAC Fanatic!

    Feb 11, 2008
    5,435
    1,305
    This sounds very similar to a really old project I saw called "RF PIC" or something like that. From memory he used a PIC and one transistor to make coded RF, the freq was from the PIC xtal.

    And I think at the receiver it used a PIC and one transistor as the detector. It's worth a google if you want to continue with this.

    Otherwise, you should google for cheap RF modules, you can get a 315/433MHz transmitter AND receiver pair from ebay for about $3. Much better range, an RF band that should not get you in trouble and it is already tuned and working.
     
  12. BrentM

    Thread Starter Member

    Oct 26, 2012
    43
    0
    Simply buying RF modules would defeat the entire purpose of doing this project. I'll take a look at the RF PIC.
     
  13. KCHARROIS

    Member

    Jun 29, 2012
    292
    1
    Keep up the good work I'm really interested in this. I'm currently trying to build an AM receiver to further my knowledge in RF seeing this helps alot.

    Thanks
     
  14. BrentM

    Thread Starter Member

    Oct 26, 2012
    43
    0
    No problem. This is a learning experience for me too. This is the first time I am able to develop something RF with the help of oscilloscopes and function generators. I will be working shortly on the receiver portion of this design, and I will post every detail as I go. It will basically be using a "tuned" LC resonator with a few amplifiers and a simple Envelope Detector. I will try to be as descriptive as possible when doing my calculations/explanations to make everything as clear as possible for you.

    Brent
     
  15. BrentM

    Thread Starter Member

    Oct 26, 2012
    43
    0
    I have a question: What are the unlicensed RF bands? Are the amateur bands unlicensed below a certain output power?
     
  16. Papabravo

    Expert

    Feb 24, 2006
    10,142
    1,790
    Well...
    10 V P-P --> 3.535 VRMS
    Now if the 25 mA is also a peak current then you have 17.675 mA RMS

    Power = 3.535 * .017675 = .0625 watts

    Yeah.. I think 62.5 milliwatts should get you the distance you want. I suggest that you try a ferrite loopstick antenna from an old AM radio. A halfwave dipole at 1.0 MHz will be 150 meters long which is about five times as far as you want the signal to go.

    Is the op-amp ground referenced or is the 10 V P-P from ground to +10V? If so you will want to AC couple to the antenna.
     
  17. BrentM

    Thread Starter Member

    Oct 26, 2012
    43
    0
    The op-amp is ground referenced. Each op-amp has a positive and negative rail.

    How is the RMS of 10 Vp-p = 3.535 V? I thought you multiply 10(.707) = 7.07Vrms? I'll look around for an old loopstick antenna. I should have one somewhere.

    Brent
     
  18. BrentM

    Thread Starter Member

    Oct 26, 2012
    43
    0
    Nevermind that last question, I figured it out. You divided the Vrms by 2. Also, I looked at the Tx circuit with the OScope, and the Peak-Peak was around 7.5 Vp-p. Since I was using a 50ohm probe, would the current to the load be something like this:

    7.5 Vp-p *(.707) = 5.305 Vrms / 2 = 2.65 V

    Now to find the current through the load,

    I = 2.65 V/ 50ohm = 0.053 A through the load

    and then to find power,

    P = I*V = (0.053 A)*(2.65 V) = 0.1405 W = 141mW

    Does that look right?
     
  19. Papabravo

    Expert

    Feb 24, 2006
    10,142
    1,790
    The division by two is applied to VP-P. For example 10 V P-P swings over the range [-5V to +5V]. The factor of 0.707 or 1 over √2 is applied to the peak voltage of the sine wave.

    so +5V Peak over √2 gives the RMS voltage which is used to compute AC power. For this to be valid I think you may also need an impedance match from transmitter to antenna, otherwise the power going out will be reflected back and nothing will radiate. Of course as we've already explained this may not matter for nearfield work. My experience is with far field.
     
  20. vk6zgo

    Active Member

    Jul 21, 2012
    677
    85
    I can't help but feel that "You're cruisin' for a bruisin'" with this monstrosity.

    A square wave,has frequency components out to at least,5 times the fundamental frequency of,in your case 50kHz,so this implies a highest frequency component for your square wave data signal of around 250kHz.

    A bit of revision on Amplitude Modulation:-

    In the Frequency domain,you may regard the useful output of an Amplitude Modulator,for the simple case of modulation with a single tone, fm,as three discrete signals:-

    The carrier,-----------------------------fc

    The lower sideband (LSB),----------- fc-fm

    And the upper sideband (USB),------ fc+fm

    This is in fact,what you will see with a Spectrum Analyser for this simple case.

    Complex signals,such as square waves are a bit different,as the sidebands are no longer discrete frequencies,but,as the name implies,bands of frequencies.
    In your case,bands with a maximum frequency of 250kHz.

    Now,we have fm=250kHz,fc =1000kHz

    so the LSB reaches down to 1000- 250kHz=750kHz

    & the USB up to 1000+250kHz=1250kHz

    So we have a data transmission 500kHz wide in a MF Broadcast allocation of around 1.2Mz width.

    The low power dispensation for playing around on the AM Broadcast band was probably conceived for relatively narrowband "pretend radio stations" for people wanting to play music,so the Authorities may be a little peeved.

    They may not be your main problem,however,as the lady next door may come after you with a shotgun!

    This is all of course,providing you don't cut carrier,(which you easily could),extending crud at 1MHz intervals across the HF bands as well.

    By the way,any tuned circuit worth the name in your receiver will cut off the higher frequencies,so the received data may not be usable.
     
Loading...