bq25570 Board Debug

In my spare time (which is a rare commodity), I’m in the process of debugging the bq25570-based Solar Li-Poly Charger and 3.3V Buck board. You can see my messy workbench in the photo below. I’m working through the issues with the board and the limitations of the circuit. My current setup uses eight “garden light” solar cells for the power source and a 1S (3.7V nominal) Li-Poly battery rated at 145mAh for energy storage. I initially had the solar cells chained together in series, which was fine indoors. However, when I took the system outside in bright sunlight, the output voltage violated the 4V maximum input voltage for the bq25570. I re-arranged all of the solar cell connections to be in parallel (~2V open-circuit). The final output voltage sits between 0V and 2V depending on light input conditions and the output load.

bq25570 Solar Li-Poly Charger and 3.3V Buck Board Debug Bench Setup

bq25570 Solar Li-Poly Charger and 3.3V Buck Board Debug Bench Setup

The photo below shows my initial setup with a smaller single solar panel from a driveway marker light that puts out ~200uA in indoor fluorescent light. On the right you can see the front of the cuckoo clock that serves as my target microcontroller system (more about that in a future post).

solar cell, PCB, battery and microcontroller system

solar cell, PCB, battery and microcontroller system

One thing I’m learning about using solar power is that the available power sources can be orders of magnitude apart. For example, a typical garden light solar cell can produce tens to hundreds of milliamps when in bright sunlight. In contrast, the same solar cell only produces tens to hundreds of microamps when under bright fluorescent lights indoors. This definitely impacts system design — how many solar cells are needed (or how much solar cell area is necessary) to maintain the system’s power requirements. When using solar energy to power an electronic system, my advice is to take some readings in the location(s) you plan to use the system. These readings can be taken using a solar cell and an appropriate resistor. If you’re planning to use an IC like the bq25570 that can track the solar cell’s maximum power point (a.k.a. “MPPT”), then do something like the following:

  1. Measure the voltage and the current with different resistors (try to keep the voltage between 0.5x and 1x the open-circuit voltage of the cell). For example, using a 2.0V cell, the resistor should be sized to keep the voltage between 1V and 2V.
  2. Multiply the voltage by the current for each resistor and you get the power output curve. The maximum value is the maximum power output for the amount of light available.
  3. Multiply this by roughly 60-70%, and that gives you an estimate for the amount of average continuous power you can expect to get out of the buck converter output from a circuit like the bq25570 one.