Here's the patient.
The problem: the sensor doesn't light up and the mouse doesn't react to movement. USB connection works and buttons also do.
It's a two layer board with components on top and three ICs: some MCU on the right, a 3.3V LDO in the upper left and the sensor in the center.
Sensor pinout, clockwise:
- top two pins - a 32MHz oscillator, this appears to be working
- ground
- 3.3V rail
- ground
- internal 1.8V rail apparently, bypassed by C22
- 4 signals that appear to be SPI coming from the MCU: CS, CLK, MISO, MOSI
- a mystery I/O, pulled up to 5V by R22 and going to the MCU
The voltage rails are clean.
Activity on SPI:
CS: pulses low at 1kHz
CLK: bursts of 1MHz clock, 7 × 8 bits each 1ms
MOSI: a single bit each 1ms
MISO: a repetitive bit pattern
The sensor appears to be backfeeding the 3.3V rail and drawing ~150µA through the mystery I/O and its pullup. There is 1.5V drop across the 10kΩ pullup regardless of USB voltage and 3.3V is 11mV below the mystery I/O. Therefore 3.3V rail is 3.4V on 4.9V USB and 3.6V on 5.1V.
The MCU has its internal 3.3V rail sitting at 3.2V. SPI runs at 5V and is clamped by the sensor to 4.2V (120Ω series resistors are included), MISO is 3.4V~3.6V.
I presume the sensors should be drawing several mA at least. Any idea what's going on and what could be wrong here?