1
\$\begingroup\$

I am designing a wireless sensor node and want to implement a solar power charging circuit. From my research, I'm aware you shouldn't use a LiPo to power your system load while it's being charged. This is okay when charging via USB because there is enough power from the USB to power both the LiPo and system load.

However, when it comes to solar I don't believe (from research) there's enough power to charge the LiPo and power the circuit due to my solar panel being too small (space constraints) to output enough power for both load and charging.

How do wireless sensor nodes charge themselves? Do they shut down while being charged to prevdnt damaging the LiPo?

The only feasible idea I can think of is having two LiPos that switch between charging and the other providing power to the system.

I have read some other posts, however as stated in their answer:

You cannot simultaneously charge and discharge a battery. I_battery = +Icharge - Iload. If I battery above is +ve the battery is charging. If Ibattery above is negative the battery is discharging. This is not a problem, just needs to be understood.

Which clarifies that you can't simultaneously charge and discharge, however, doesn't provide a solution. There must be a way deployed nodes in industry are recharged via solar and continue to transmit.

\$\endgroup\$
17
  • 1
    \$\begingroup\$ when it comes to Solar I don't believe (from research) there's enough power to charge the LiPo and power the circuit - think again; all the solar has to produce is more power than the load needs and the excess can be used to charge the battery. It seems that your whole question is based around this falsehood. \$\endgroup\$
    – Andy aka
    Sep 26, 2021 at 11:58
  • \$\begingroup\$ Alright, but if my wireless sensor node is constrained by space and my panel cannot be large enough to output enough power to sustain both the load and the charging, then what? \$\endgroup\$
    – Explorex
    Sep 26, 2021 at 12:00
  • 2
    \$\begingroup\$ Then you have a problem. \$\endgroup\$
    – Andy aka
    Sep 26, 2021 at 12:01
  • 1
    \$\begingroup\$ I have a couple of gadgets in my yard that do exactly what you describe. Wake up once an hour, read data from a bunch of sensors over Bluetooth, send to server over WiFi, sleep again. They use solar panels about the area of typical smart phone screen and have a 10000 mAh lipo cell. They run 24/7, and have done so since March (call it six months.) The solar cells keep the batteries charged. They charge continuously in sunlight, even when the ESP32 is collecting and reporting data. \$\endgroup\$
    – JRE
    Sep 26, 2021 at 12:49
  • 1
    \$\begingroup\$ The only problems have been software bugs that sometimes left an ESP32 running for as much as 24 hours straight without sleeping at all. That kind of thing did run the battery down, but a couple of days of sunlight (while still operating normally) got the batteries charged again. \$\endgroup\$
    – JRE
    Sep 26, 2021 at 13:19

3 Answers 3

1
\$\begingroup\$

If you can't get enough power from your solar cells to charge while operating than obviously your system must power down sometimes. You either reduce the average energy consumption of the system or increase the power provided by the solar cells, it's as simple as that.

If your solar panels provide more power than the IoT device needs, then the excess power can be used to charge the battery while the IoT device is operating.

\$\endgroup\$
0
\$\begingroup\$

The answer is in the text you quoted. It’s just a matter of which is higher.

If your load is higher than what the solar panel is providing, then part of the power will come from the panel and the rest from the cell. You won’t be charging it, for sure, you’ll just be drawing less.

If your load is lower than what the solar panel is providing, then it can be fully powered by it, and the remainder can be used to charge the battery.

But you should probably not try to reinvent the wheel. There are ICs which are specifically designed to handle all that for you, especially in cases of very low currents: connect the solar panel, the cell, and your load, and it’ll handle all the interactions between them. You may want to check Fujitsu E-Peas energy harvesting ICs like the AEM10491 for instance (disclaimer: I have never used it, so I can only hope it works as described). There are also ICs from TI and probably a few others.

Note that you mention space constraints. This makes me think those are indoor devices. Remember that light available indoors is just a tiny fraction of what you get outdoors, and you probably need dedicated panels which are actually suited for that kind of light (very different spectrum and intensity). Even then you’ll probably only get a very small amount of power out of it, which makes it very difficult to compete with a regular non-rechargeable battery which will probably be cheaper and smaller (but choose carefully, pay attention to self-discharge).

Of course depending on your actual constraints (size, indoors or outdoors, placement, average and peak current, lifetime…) things may be quite different.

\$\endgroup\$
0
\$\begingroup\$

You seem to be under the incorrect assumption that you have to charge the Li-ion cell at a particular current. That is not so.

Li-Ion cells have a nice advantage over NiMh and lead-acid chemistries in that their leakage currents are extremely low so even a very small current will charge the cell. Slowly yes, but it will charge.

As others have said there is no issue with charging the cell and simultaneously powering the load with the cell, excess power will go into charging.

You do need to ensure that in the event that there is significant excess power from the solar cell that it will not charge the cell to the point that it exceeds its fully charged voltage (probably 4.1-4.2V depending upon the particular cell). You may want to set that voltage lower to increase the life of the cell.

You should also put a low voltage limit to prevent over-discharge. If there is a lack of sunlight for a significant period and the voltage gets to that low point disconnect the load just keeping the voltage monitor powered so that the load can be reconnected when the sun has recharged the cell to a higher voltage. If there is a very low power processor in the system that could do the management with only a few tens of microamps of consumption.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.