Prevent voltage to drop under threshold in solar panel

Question

This is a solar charger for my smartphone built with two panels wich I want to keep attached all day long with the phone turned on. When the sun is going away, voltage output starts to go under 5V and my smartphone repeatedly turns on the display to alert of power disconnection wasting battery charge for a long time. Under a certain voltage (< 4.7V) finally smatphone accepts disconnection and stops turning on the display. Is there a way to disconnect the circuit if voltage drops down a certain threshold or better to keep voltage up until it's not possile anymore?

Answer 1

The 1N4733 zener diode is a 5.1 volt type meaning the maximum voltage you get across it is 5.1V +/-5%. The 1N5819 schottky diode drops typically 600mV at 1A and 400mV at 0.1A.

Either way, the voltage you might be seeing at the terminals is significantly less than 5V and I would estimate it to be: -

• 4.7V at 100mA charging current
• 4.5V at 1A charging current

This seems to be at odds with what you are expecting and I'm wondering if you are aware of this?

Assuming you are aware of this limitation, maybe you'd like to offer a new "low-voltage" limit. However, as with most solar cells that I'm aware of, as soon as you disconnect it from a load (via the automatic under-voltage idea) the terminal voltage will undoubtedly rise above the under-voltage threshold and the circuit will re-enable - this might end up in a series of on-offs lasting several minutes and probably isn't what you want?

Answer 2

Understanding Andy's answer is important before attempting an automatic under-voltage circuit. This circuit should provide a basic route to do so. D3 is the solar cell. D1 and D2 are the same as in your circuit. I've added in the under-voltage circuit in-between. The voltage divider made up of R3 and R2 should give a voltage close to the NMOS threshold when the solar cell voltage hits the threshold you're after. When the NMOS no longer has enough voltage to turn on, R1 will not have a path to ground and the PMOS will shut off because the PMOS Vgs will be 0. This will open up the path between the solar cell and the phone. When the PMOS shuts off, the voltage from the solar cell will increase because there won't be a load anymore. C1 voltage will rise then causing the NMOS to turn on. R3 and C1 should be sized very large to create a very long rise time. In that way, you can avoid a constantly oscillating under-voltage cutoff circuit. C1 will at least slow down the oscillations to the point where you can hopefully achieve a single cut off.

^{simulate this circuit – Schematic created using CircuitLab}

Answer 3

Have you considered a 5V buck (step-down) switching circuit? While not highly efficient, something like the LM2574 (ti.com/lit/ds/symlink/lm2574.pdf) would give you better than 75% efficiency with few parts. You would put your two solar cells in series to feed it, but would get a regulated 5V until the cells couldn't put enough current out. This would most likely stay in charge longer. You would have to see how much power the zener is wasting to pick which is best.

You should be able to integrate the slick shutoff show by horta in the other answer. It might be more accurate in cutting of with cells in serial, as the voltage difference when dropping out will be larger.