Supply 3.3V from a LiPo

[pete_l]

Here's the plan.
I intend to have a PIC processor connected to an NRF24L01 2.4GHz Tx/Rx. This will be powered by a 3.7V LiPo battery that is permanently connected to a TP4056 based charger that gets its supply from a 6V solar panel. It's a requirement that the NRF24L01 and the PIC can be run in sleep-mode for most of the time, with only a short period of active use and power consumption. I except the duty cycle to be less than 1%.

My problem is: how to power the NRF24L01 (max. supply 3.6V @ 15mA) when the supply voltage (from the TP4056) could rise to 4 Volts while it's charging the LiPo?

I am considering 2 options. Both of them assume I can power the NRF24L01 from one of the output pins on the PIC microprocessor by setting its output to "1".
The first plan is simply to place a diode in series with the supply and hope that it will reduce the voltage to within the NRF24L01's acceptable range.
The second idea is similar, but would use a 3.3V zener and a resistor to set an upper limit on the NRF24L01's supply.

I'd value any observations about the suitability of either plan - bearing in mind that the NRF24L01 can consume variable amounts of power, depending on what it's doing. If you can think of a better (and cheap!) way of running this, I'd also like to hear from you.

with thanks
Pete

 

[kam1787]

i'd go with the diode plan

 

[betwixt]

I agree with Kam1787. If you use the Zener it will always consume some current. To actually work as a voltage stabilizer, the Zener has to consume at least as much as the variation in 24L01 so you would have to set the resistor to drop say 0.7V at maximum 24L01 current and let the Zener soak up the rest at lower currents. A diode will only vary the drop as much as it's dynamic resistance over that current range allows whch will be quite a small amount. If you use a diode, it may be worth considering a high value bleed resistor to allow a few uA to flow while the 24L01 is idle otherwise the diode leakage may allow the voltage to rise slightly.

I have a solar powered remote temperature sender here which uses a 3.3V regulator fed from a pin on a PIC processor. It's tricky to keep the regulator stable due to the need to use a capacitor across the input while not overloading the PIC as it charges up but it can be done. In my case it's a 433MHz FM module but the principle is the same, it wakes up once a minute, reads a 1-wire temperature sensor, transmits the data then goes back to sleep. It's worked for several years without any problems.

Brian.

 

[pete_l]

Thanks guys, you've confirmed my thoughts. The bleed resistor is not something I'd thought of but it sounds like a worthwhile addition.