Guru
Joined: 07/09/2016
Location: United KingdomPosts: 2171
| Posted: 01:05pm 29 May 2025 |
Afternoon forum, hope you are all well.
I have built a data logger that has a small battery of 4xAA NiMh cells (so nominally 4.8V but actually seems to sit around 5.2V here)
From the scrap box, I found a solar panel about A6 size that gives 6.1V in bright sun with a short circuit current of about 120mA
The panel currently charges the battery through a single schottkey blocking diode and it all seems quite happy - it does what I expect. I have some quite severe energy saving designs in the hardware for the sensors and radio transmitter (MOSFET hi-side switches) and a *lot* of CPU SLEEPs in the code. I cache the data overnight and only transmit during daylight.
Battery voltage drops to about 5.2V overnight but as the sun comes up, in today's blustery weather with bright bits, VBAT comes up as usual to about 5.4V with VSOL occasionally hitting 5.8V. I sense both solar and battery voltages through potential dividers with D1 and D3 clamping anything nasty to the supply rails and then onto AIN pins, applying a multiplier in the prog to give "real" voltages. All works nicely, but we are coming up to maximum daylight and I am wondering about six months from now when we might be lucky to get 8 hours of heavy overcast.
I am conscious of the voltage drop across D2 and sure enough the trend shows the expected 0.3v-ish difference between VSOL and VBAT (of course VSOL will be dragged down towards VBAT as it tends to short circuit). Clearly this is very blunt and I don't have space or $$$ for some MPPT controller (hence NiMh cells which are quite forgiving).
So now I start thinking about an "ideal diode" circuit (4 components) in an effort to eliminate the "lost" 0.3V and bench tests suggest I might get it down to about 50mV, but does pose some questions for me as I have never done this stuff before.
Ideal Diode schematic and discussion
1. I expect to get a lot more usable energy into the battery - do you see any problems?
2. How do I measure the energy going into the battery? With no voltage drop, I expect ~voltage parity on both sides
3. Should I put a resistor in series? I am thinking 10Ω just to be a bit nicer to the panel and allow the voltages to seperate a bit.
4. what else am I missing?
all input gratefully recieved - thanks in advance
h
Edited 2025-05-29 23:30 by CaptainBoing