I have a SunControl board with a 6600 mAh battery and two 6V 5.6W solar panels. The load is a Raspberry Pi ZeroW, with HDMI disabled (drawing about 100 mA). The Pi is connected to the SunControl I2C Grove connector.
This setup worked fine for a couple of weeks. I then decided to add a solar tracker circuit (separate battery, separate solar panel, separate voltage booster), and brought my SunControl inside for testing. While I had the opportunity I charged the 6600 mAh battery overnight from an USB socket. It charged up to about 4.1V.
When I brought the setup back outside in the morning the sun was already out, but still behind the trees. After a while my Raspberry PI stopped responding. By then the sun was fully out. Checking it I noticed that the USB Output LED was off. I also noticed that both the orange and the green LED on the Suncontrol were on. When obscuring the solar panel (or unplugging it altogether) the USB output would switch on again.
I think I didn't see that issue before because the battery never reached full charge - just not enough sun in the day at that location.
Do I have to live with this quirk or is there a way to keep everything connected regardless of the battery level?
Small side note: I have a thermistor wired to the board as instructed, and placed next to the battery. Current temperature is around 23F. So I would expect charging to stop, but that should not impact the discharging and the USB output, or?
Hmm. I would guess that your solar panel are supplying too much voltage to the system, when the battery is fully charged and the system is shutting down (It can be damaged by too high of voltage on the solar panels).
Solar Panel voltages will climb up when the battery is charged (you can think of this as the current in the solar panel has no where to go, so the voltage goes up, quite possibly up to VOC (Voltage Open Circuit) a typical solar panel specification. The low temperature will put your system in this spot faster (that's to protect the LiPo batteries.)
The quickest fix for your solar panels (which are pretty hefty!) is to put a 6.2V or6.4V Zener Diode from the output of the solar panels to Ground. Size it for a couple of Amps.
P.S. This was a very well written request and had all the information I needed to help you out. Thank you!
Thanks, I managed to source a 10W Zener, will try both options.
It feels like the Zener should be placed close to the battery in winter, and outside the box in summer. I wish there were some place for a Peltier in this story, like the TES1-4902. Whatever current the battery rejects should go to the Peltier (which could then also be placed this way in winter and that way in summer where it actually might provide cooling to the battery).
I don't think it will make much difference. If you have a tight box, mount it outside (but cover it). During the winter it won't be used much and during the summer a lot (which is why it being outside can make a bit of a difference).
I love your idea about the Peltier. What an interesting idea! Cooling the battery in the summer and heating it in winter. Have to be careful about using them though. They require a heat sink on the hotside. Since it is up to 6.2V, you could put it across the Zener and use the current to just do that. Maybe have two Peltier's. One for summer and one for winter. I would think you would want to control it using a relay or something like that to control the heating and cooling. How interesting!
I have one in my lab box, I should try this.
I got on Ebay and bought two. I just noticed that if you switch polarities, you can change which side heats and which side cools.
I tried the bank of 10 1W Zener diodes - left them in their paper strip packaging and just soldered the wires across. I then connected that contraption to the "Solar Cap" + and - pins (assuming they are directly linked to the panel input - please correct me if that is not the case).
They heated up a little, providing some heating to the battery. But they were not really able to keep the voltage to 6.2V - more like 6.5 to 6.6V Still better than the open voltage that I had seen earlier, but not as good as expected. No idea if these Zener diodes all have the same characteristics or if the yield differences have an impact. I can test that once my single 10W Zener arrives.
I also had to disconnect one of the solar panels during periods of intense sunshine. Together the two panels were showing 6.5W, but when I disconnected one the other produced 4W in the same conditions. The charging circuit seems to have been overwhelmed by the two panels. This is a shame as these moments of intense sunshine are rare, and I specifically oversized the panels to harvest as much energy as possible in "normal" conditions - cloudy or overcast. Wonder if there is a smart way to selectively enable or disable individual panels based on actual conditions (insolation and load)?
On a side note - not impressed with the 6600mAh Li Ion battery. I know cold weather reduces the capacity by 30% or so, but with a miserly 95 mA on the Pi Zero and a circuit efficiency of (supposedly) 96% when not charging I would expect much longer runtime than I am seeing.
Expected from full battery: 3.7V, 6600mAh * 0.7 / 5.17V * 95 mA = 34 hrs Actual = 20 hrs max, often not even enough to last through the night and catch the first rays in the morning
Wonder if a Li-Poly battery fares better. Definitely eyeing a larger battery as I know the panels can serve it.
Last Edit: Dec 11, 2018 10:34:51 GMT -8 by lbendlin
Each solar panel has a diode in it (I don't recall what diode, but something with a low Vf) to keep back current flowing into non-performing solar cells. This might be your problem.
The charging circuit will stop charging at some level (1500mA I think on SunControl and around 1000ma on SunAirPlus) and then the panel voltage will go up. You can see this by looking at what is going with the solar data.
My solar panels are barebones, don't think they have any circuitry. That board looks interesting. I ordered it, will report back if it helps or makes it worse (with the .45V drop etc). How will these diodes interact with the Zener once the voltage goes over 6.2V ?