Monday, April 5, 2010

Testing the SRB MX146

I just got an SRB MX146 module! It puts out 0.5 watts and operates on any frequency from 144 to 148 MHz. The frequency can be chosen from among a set of pre-programmed memories, or interfacing directly with the controller using I2C. Changing frequencies on-flight would allow transmitting position messages on the APRS frequency, then switch to a different frequency and send non-APRS data (real-time photos, for example).

This is my current set-up:

The rig

MX146 outputting 1/2W

LDO regulators

On the first picture:

  • An Arduino, the "brain" of the system. I use the internal UART to interface with the GPS and four pins for the radio, one carries an AFSK 1200 baud signal modulated with PWM, another controls the PTT, and two more for I2C.
  • The GPS (the little red board) is a Venus 634FLPx I got from here. It is supposed to work above 18 Km according to their datasheet. Plus, they have a very responsive tech support (I actually confirmed the CoCom limits with them). It fixes the position 5-30 seconds after a cold start, but you can feed it the ephemerides of the next 7 days and it will then fix in less than 5 secs. This is a 3.3 volts device, so it needs some kind of voltage conversion. The easiest and safest way to convert 5v to 3.3v is by using a diode as explained in this nice tutorial. The 3.3 volts don't need conversion back up to 5, since 3.3 volts is enough for the AVR chip to consider it a logical high.
  • On the breadboard, there is also a low-pass filter to get rid of the PWM high-pitch frequencies. PWM is done at the maximum rate allowed by the AVR's timer 2, ie. 62.5 KHz (16 MHz / 256), so this is probably not needed anyway.
  • The MX146 module is connected to the output of the LPF via a coupling capacitor and a voltage divider to bring the 5 volts off of the Arduino down to the 500mV peak-to-peak required by this device. The PTT is internally pulled-up to 3.3v by the MX146, so an open-collector transistor justs shorts it to ground on transmission.
  • Finally, I had to use three different regulators, since all 3 devices take different input voltages. In order to save some power, I used LDO regulators rather than the classic 78xx family. The LDOs usually require around 1-1.5 volts less than the matching 78xx, which means one less battery cell to lift!