Telit GM862-GPS Power supply and UART

Working with the GM862 GPS turned out to be a little tricky. It can be somewhat picky about a stable voltage during and works at CMOS logic levels.




The first issue was transitioning the 3.3V of my Bus Pirate to the 2.8V levels the GM862 required.
On the TX line this can be done very simply with a resistor divider. I used the values R1=100k, R2=18K
The Bus Pirate was able to use the CMOS voltage levels on it's RX line without issue, but I put a 1K resistor inline to prevent it from being overdriven.

The second issue I ran into was the device 'browning out'. This manifested itself as the device never registering to the cell network and an unusual status led pattern. It would blink rapidly 3 times, pause, then blink rapidly again. During the normal course of operations it blinks rapidly while it is searching for a signal, and blink slowly once it is registered. I observed a mixture of these, which was fairly confusing.

Borrowing an oscilloscope from school I was able to observe a voltage drop on my 3.4V line that occurred during the pause in the blinking pattern. The GM862 has a peak current of 2A which my bench supply was unable to respond to these peaks without voltage drops.
The solution to this, as described in the user guide, is to use a decoupling capacitor. I used two DigiKey 478-1899-ND which is more than nine times the capacitance suggested. YMMV.

The next issue in my power supply was that my ancient bench unit can only provide 1A. The solution to this was to design out my circuit more completely. In the end I'm going to be powering the device from a 11V RC battery, so I needed a DC-DC converter. I oped to use a APTS006A0X-SRZ from Sparkfun. For the trim resistor I used a 1.758K trim pot giving a Vout of 3.8V
The unit generates about 108F at 11.5Vin in approximately a half hour. The unit's operating rage is listed to 185F, so hopefully this won't be an issue. I will be carefully monitoring the temperature in the future.

My last problem with the power supply was that the 3.4V line was still dropping too far during association. The tantalum capacitors helped, but there was still a .4V drop occurring. The simple solution to this was to bring my operating voltage up to 3.8V. With the .4V drop I'll still be within the operating range.