3
\$\begingroup\$

Further to this question: 434MHz PCB Trace Antenna Design.

If I select a suitable 434MHz PCB antenna design and make two of them, what is the minimum extra circuitry I would require in order to de able to transmit a simple square wave from one to the other over a short distance (e.g. 30cm) ?

(10Hz Square wave) -> (Circuit A) -> (PCB Antenna 1) -> (PCB Antenna 2) -> (Circuit B) -> (10Hz Square wave out)

What could I use for circuits A and B? In this application, I would value simplicity, size and cost over reliability.

Could I literally use a crystal oscillator with enable input for A, and something like an AM radio for B?

\$\endgroup\$
  • \$\begingroup\$ It's not the same question, quite, but that sounds solidly like you are re-asking this previous question of yours from March 25th. People put a lot of work into addressing that problem and if you'd answer on list the questions asked and provide information along the lines asked for then you'd get a good answer in no time. The answers you need are, perhaps, along the line I gave you off list at the time but with more detail and I'd be happy to add to them where all could benefit from them. \$\endgroup\$ – Russell McMahon Apr 27 '12 at 22:39
  • 1
    \$\begingroup\$ @RussellMcMahon - Yes, I am re-asking the question, hopefully in a way which keeps the topic focused on RF. My goal there was to try to learn about RF, and I felt that none of the answers there helped me. Instead everyone wanted to solve a hypothetical problem I might have. It was my fault for poorly phrasing the question, and I hope I have done a better job here. \$\endgroup\$ – Rocketmagnet Apr 27 '12 at 22:50
2
\$\begingroup\$

RF comm is a regulated medium. There are specific bands in which unlicensed signals may be sent. Even in these unlicensed bands there are limits on output energy.

Just because the carrier is 434 MHz doesn't mean that all of the energy output will be at that frequency. For example a square wave has infinite bandwidth. When a square wave is multiplied by a carrier, it still has infinite bandwidth centred at the carrier frequency.

Real radios need bandwidth filters to ensure that the energy output complies with a spectral mask. So a square wave will only be parts of a square wave when it goes over the air.

That may be enough in your application. Or you may want to consider one of the usual RF packet protocols.

\$\endgroup\$
  • \$\begingroup\$ Thanks for your answer. I'm not really trying to make a radio transmitter, I'm just interested to learn about RF. How easy would it be to make a clean sin wave oscillator? Or would it be better to use a filter? \$\endgroup\$ – Rocketmagnet Apr 28 '12 at 9:38
  • \$\begingroup\$ one cannot communicate any information with a "clean" sin wave. It has to be modulated (amplitude, frequency, or phase) to communicate information. Communication requires the filter. \$\endgroup\$ – user9224 Apr 28 '12 at 10:20
  • \$\begingroup\$ That said, there may be a way to get what you're after in your application without communicating anything. You're not trying to send a signal necessarily. You just need a "reference" for the micros. So if one of your micros is the master and broadcasts a pure sine wave and all micros (master and slaves) derive their clock - or at least sync their outputs - from this sine, they they won't slip. Such would depend on the pure sine being strong enough to defeat any interference, yet be weak enough to stay below the regulatory masks. Since your micros are 30 cm apart this may be achievable. \$\endgroup\$ – user9224 Apr 28 '12 at 10:27
  • \$\begingroup\$ Thanks. Yes I think it's fairly simple. When I say a clean sin wave, obviously it also needs to be multiplied by the square wave too. \$\endgroup\$ – Rocketmagnet Apr 28 '12 at 11:03
  • \$\begingroup\$ @scorpdaddy, yes, that is AM modulation, albeit a simple form, either the signal is there or not, now calibrate what strength signal at that frequency represents the presence of your transmitter. Sending a square wave could easily be done with FM or even basic phase modulation, this does not generate gigantic sidebands, instead you are going to be at one of two frequencies with their resulting sidebands with very low datarate keeping your sidebands from being too large. \$\endgroup\$ – Kortuk Apr 28 '12 at 23:08
2
\$\begingroup\$

Modulating 434MHz to get a square wave at 10Hz will not require a big spectral footprint. If you AM a 434MHz carrier with a 10Hz square wave, you will get harmonics enough even with a narrow 1kHz BPF that the received wave will be recognizably square.

If you try to turn a 434MHz oscillator on and off at 10Hz, though, you may still 'get outside the band' - oscillators can behave very differently in the brief amount of time when they power up or power down. During the bulk of the 50 msec that the 434MHz oscillator is on, it may produce a very clean line at 434MHz, but at the ON and OFF events, you might see all kinds of other frequencies. A better approach would be to let the oscillator run continuously, and simply gate its output with a transistor.

But that's only the transmit end of things. Receiving the signal is going to be a little more complicated. Someone else may perhaps address that, but I think it's going to be the harder part of your project.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.