Solar Power Pack — Farmers IoT Toolkit

Why you would want this

Run sensors anywhere on the farm — no power outlet needed
Free to run once it is built
One pack can carry several sleeping nodes at once
Sensors that sleep are almost free: ~5.5 Wh/day against a 50–60 Wh/day panel

How it goes together

Wiring diagram: a 20 W solar panel feeds a CN3722 MPPT charger module, which charges six 18650 lithium cells arranged 3S2P through a 3S 40 A BMS. A screw terminal block takes 12 V off the BMS to power the other modules. Parallel wiring is purple, series wiring yellow, BMS wiring red. — Panel → MPPT → BMS → pack, and a terminal block where the other three modules tap 12 V. The purple links are the parallel pairs, the yellow are the series joins, and the red are the BMS balance taps — get those three colours right and the pack is right.

What you will need

Bill of materials for Solar Power Pack
Picture	Part	What it does	NZ$	Link
	20 W solar panel, Vmp 18.6 V Vmp matters more than watts. Subtract ~11% for heat and the result must still sit ≳2 V above your pack's full-charge voltage, or the pack never finishes charging.	Charges the pack	$49.99	Buy 20 W solar panel, Vmp 18.6 V ↗
	CN3722 MPPT solar charger	Gets the most out of the panel and charges the pack safely	$15.24	Buy CN3722 MPPT solar charger ↗
	18650 lithium cells ×6 ×6 Ours are Samsung INR18650-32E (3200 mAh). Use the real figure off the wrapper — treat unbranded or salvaged cells as 60–70% of whatever they claim.	Store the day's sunlight — three in series, two in parallel	$120Six cells at about $20 each.	—
	3S 40 A BMS protection board Every load taps pack+ / P−, never B−. Tapping B− silently bypasses the protection.	Guards against overcharge, over-discharge and short circuit	$5	Buy 3S 40 A BMS protection board ↗
	Battery holder or spot-welded pack	Holds the six cells in 3S2P	~$8.50	—
	5 V / 3 A synchronous buck converter For a phone, do not use a tiny Mini360 — it is rated 3 A and misbehaves above ~1.5 A in a hot box. Set 5.15–5.2 V and short USB D+ to D−.	Steps the pack down to 5 V USB for the base-station phone	~$5	—
	Fuse holders + fuses	Cuts power if a leg shorts	~$8.50	—
	Screw terminal block	Where the other modules tap their 12 V	~$1.70	—
	Weatherproof enclosure	Keeps rain, dust and sun off	~$17	—
Approximate total			$230.93

Prices are NZD, read off the linked listing on 2026-08-08. A figure marked ~ is an estimate — we have no listing for that part yet, so treat it as a placeholder rather than a quote. Prices move, and shipping to a small island is its own line item. Links are not affiliate links and we get nothing if you use them.

Wiring

Everything hangs off one chain: panel → MPPT → BMS → pack → terminal block. The order is not negotiable, and neither is where the loads tap — pack+ / P−, on the protected side of the BMS, never B−. Tapping B− routes discharge around the protection FETs and you lose over-discharge and short-circuit protection silently.

Set the charger's CV pot to 12.6 V and its MPPT pot to the panel's hot Vmp (≈16.5 V hot), not the 18.6 V @ 25 °C on the label. The pack sits at 12.6 V full and ~9.0 V near empty, with an ESR of 53 mΩ — so voltage sag under load is telling you the state of charge, not that something is dragging the rail down.

The energy budget

This is the number that decides whether the system runs forever or dies in a week. Compare two figures in watt-hours per day: what the panel collects, and what your devices spend. Not what the battery holds — the battery is a bucket, not a tap.

What each thing costs you per day, against a panel that collects 50–60 Wh/day clear and 6–15 Wh/day under heavy cloud
Device	Power awake	Awake how often	Wh/day
Sleeping soil node	0.6 W awake	~5%	~1.5
Water probe + node	0.5 W	~5%	~0.6
Solenoid valve	~7 W	20 min/day	~2.5
Field gear, total	—	sleeping properly	~5.5
Android phone, plugged in	0.7–2.0 W	100% — cannot Doze while charging	20–57

Our ~70 Wh pack gives 56 Wh drained to 80% depth, but that is an emergency depth. Cycling that deep daily in tropical heat wears the cells out in a year or two, so plan against ~38 Wh everyday. Charging is rated 0–45 °C only; discharging −20 to 60 °C.

Build it, step by step

Test every cell before it goes in

Multimeter each cell: it should read between 3.0 V and 4.2 V. Discard anything below 2.5 V or that looks damaged, swollen or smells wrong.

Watch out
Check the number on the wrapper before you trust it. Cheap and salvaged 18650s are routinely relabelled — if yours are unbranded, or claim something like “5000 mAh” (no such 18650 exists), assume 60–70% of the printed figure. A pack built on a fictional capacity will surprise you on the first cloudy week.
Arrange the cells 3S2P and fit the BMS

Three cells in series makes ~11 V nominal (12.6 V full); two of those chains in parallel doubles the capacity. In the picture above the purple links are the parallel pairs and the yellow are the series joins.
Follow the BMS markings for B+, B− and the balance taps — the taps read ascending: 0 → 4.2 → 8.4 → 12.6 V.

This will bite you
Every load taps pack+ / P−, never B−. Tapping B− routes discharge around the BMS protection FETs and you lose over-discharge and short-circuit protection silently.

This will bite you
Never short the cells, never puncture or crush one, always fit the BMS, always fit a fuse, and keep the pack dry. If a cell looks swollen or gets very hot, stop and dispose of it safely.
Connect the MPPT charger and panel

Panel to the controller's PV input, pack to its battery terminals. Set the CV pot to 12.6 V and the MPPT pot to the panel's hot Vmp (~16.5 V), not the 18.6 V on the label.

Watch out
Power-on order is not arbitrary — a panel has no off switch. Cover the panel, connect the battery through the BMS to the charger output, then uncover. Teardown is the reverse: cover and disconnect PV first, then the battery.

This will bite you
Charging is rated 0–45 °C; discharging is −20 to 60 °C. That gap is the important part. A sealed box in tropical sun runs 20–30 °C above ambient, so at midday it is easily above 45 °C inside — exactly when the panel is trying hardest to charge. Powering your sensors in that heat is fine. Charging in it damages the cells and is how a pack starts to swell. Shade and vent the enclosure.
Fuse each leg out of the terminal block

The terminal block on the right of the diagram is where Modules 2, 3 and 4 tap their power. Put a fuse between the pack and each load rather than one big fuse on the pack: 3 A is a sensible start for a sensor node, and the phone charger gets its own 2 A so a chafed USB cable cannot take your sensors down with it.
If you are charging the base-station phone from this pack, feed the 5 V/3 A buck from the protected side of the BMS and set it to 5.15–5.2 V with a meter on it.

Watch out
Short USB D+ to D− on any socket that charges a phone — otherwise the phone decides it is plugged into a computer, limits itself to 500 mA, and the failure looks exactly like a solar problem.

Worth knowing
Module 2 and Module 3 each do their own conversion from what this block gives them — Module 2 bucks 12 V down to 5 V and boosts a second rail up to 18 V; Module 3 runs on its own small pack entirely. This block only has to be a clean 12 V.
Do the energy budget before you close the box

This is the most important step in this module, and the one people skip. Batteries are a bucket, not a tap. The panel fills the bucket; your devices empty it. Adding more batteries makes the bucket bigger — it does not make the tap run faster. If your devices use more each day than the panel collects, more batteries will not save you; you will just take longer to run flat.
A sunny-day surplus tells you almost nothing. Any solar system looks fine in the sun. What kills them is a run of overcast days, and in the tropics that means the wet season — not the dry months when you probably built it.

Worth knowing
Ranked by value for money when the budget doesn't close: (1) fix the panel you have — point it north in the southern hemisphere, tilt it, clean it, un-shade it, often worth 2× for free; (2) add a second 20 W panel, the only fix that addresses a cloudy run; (3) duty-cycle harder; (4) more batteries, last and only for riding out cloud.
Charge the phone in bursts, not continuously

If you're running Module 4 off this pack, do not leave the phone plugged in. A plugged-in phone can never enter Android's Doze (it only engages when unplugged), so it runs at its awake floor forever — 20–57 Wh/day out of a panel that collects 50–60 Wh on a good day.
The trick: the phone's own battery is a second battery pack. A 5100 mAh handset holds ~19.4 Wh against the pack's ~38 Wh everyday-usable. Charge it from daytime surplus and let it coast overnight, exactly as it would in your pocket. A $3 pack-voltage relay does the whole job with no code: ON at ~12.3 V, OFF at ~11.6 V.

This will bite you
Charging must be the DEFAULT state. If the phone is the thing that decides when to charge and the phone goes flat, nothing is left to turn the charger back on — your base station is dead until someone walks out to it with a power bank. “Stop charging” must be a command the phone actively sends, and it must expire.

This will bite you
A lithium cell held at 100% and high temperature degrades fast, and swollen phone batteries catch fire. A phone sealed in a box in tropical sun, plugged in permanently, sits at both extremes at once. Shade and ventilate, don't hold it at 100%, inspect every few months, and never mount it against anything flammable.

Worth knowing
Three wins from duty-cycling the charger: no round-trip loss through the 18650s; the fire risk goes away; and when a cloudy run empties the pack the phone stops charging while the sensors keep logging. You lose your view of the farm before you lose the data. That is the correct order.

Supporting files

Everything in the repository worth having on your own disk before you start — the sketch you flash, the flow you import, the datasheet you check a pinout against. Download saves the file; the path opens it on GitHub if you would rather read it first.

doc hardware/3S2P.md The pack itself: cells, voltages, charge limits and the energy budget every power figure on this site derives from Download hardware/3S2P.md ↓
doc project/bench/03-bench-phone-power.md Runnable bench sheet for measuring what a phone actually draws over 24 h — tests 1, 2 and 6 need no special equipment Download project/bench/03-bench-phone-power.md ↓
image docs/wiring-diagram-module-1-power.jpg The wiring picture at full size, for printing or pinching-to-zoom on a phone Download docs/wiring-diagram-module-1-power.jpg ↓
datasheet hardware/18650-cells-INR18650-32E.pdf Samsung INR18650-32E cell datasheet — the real capacity and temperature limits Download hardware/18650-cells-INR18650-32E.pdf ↓
datasheet hardware/Manual - MPPT Option A.pdf CN3722 MPPT charger manual Download hardware/Manual - MPPT Option A.pdf ↓
datasheet hardware/Manual - MPPT Option B.pdf The alternative charger we costed, kept for comparison Download hardware/Manual - MPPT Option B.pdf ↓

Or take the lot: download the whole repository as a zip, or git clone https://github.com/sunrise-labs/farmers-iot-toolkit.git. Each sketch needs its config.example.h copied to config.h and edited — the real config.h is never committed, because it holds your WiFi password.

When it doesn't work

Ordered roughly by how often it has actually been the answer.

Symptom	Almost always	What to do
No output voltage	Fuse has blown	Check and replace it — then find out why.
Pack charges partway then stops, never reaches full	Panel Vmp too low once hot	Subtract ~11% from the label Vmp and check it is ≳2 V above the pack's full-charge voltage.
Charges fine in the morning, stops by midday	Same cause — the panel heated up and Vmp fell	Higher-Vmp panel, or fewer cells in series.
Panel gives far less current than its rating	Facing the wrong way	In the southern hemisphere the midday sun is in the NORTH. Check Voc and Isc disconnected to confirm the panel itself is healthy.
BMS cuts power suddenly	A cell is faulty or badly connected	Check every connection, test each cell individually.
Phone charges very slowly — looks like the panel is too small	USB data pins not shorted	Short D+ to D− at the socket. Confirm with a USB power meter that it now draws >0.5 A. This failure impersonates a solar problem.
Phone charges but never reaches full on a sunny day	Voltage drop in a long thin USB cable at 2 A	Shorter, thicker cable; set the buck to 5.15–5.2 V.
Enclosure gets very hot inside	No ventilation	Vent holes on the shaded side, meshed to keep bugs out.

How this fits with the others

Why 3 cells in series and not 4

An earlier version of this guide said 4S2P. That was wrong, and it is worth explaining why, because the mistake is easy to repeat and it fails in a way that is hard to diagnose. A charge controller cannot push energy uphill — it needs the panel about 2 V above the pack's full-charge voltage. And a panel's Vmp drops as it gets hot: the 18.6 V on the label is measured at 25 °C, and a panel in tropical sun runs at 55–60 °C and loses ~11%. Here is the nasty part. A 4S pack on this panel does not look broken. It charges on a cool morning, settles around 15 V, and reads like a working battery. What it never does is finish — and a BMS only balances cells at the top of a full charge. So the cells drift apart over months, the weakest starts hitting cutoff early, and the pack's usable capacity quietly collapses long after you stopped suspecting the panel. The 4S failure isn't “it won't charge”. It's “it charges just enough to look fine, and destroys itself over a year.”

Arrangement	Full charge	Panel voltage needed	A hot panel gives 16.5 V
3S — this guide	12.6 V	~14.6 V	✅ works, ~1.9 V to spare
4S — the old, wrong version	16.8 V	~18.8 V	❌ short by 2.3 V

Why you would want this

How it goes together

What you will need

Wiring

The energy budget

Build it, step by step

Test every cell before it goes in

Arrange the cells 3S2P and fit the BMS

Connect the MPPT charger and panel

Fuse each leg out of the terminal block

Do the energy budget before you close the box

Charge the phone in bursts, not continuously

Supporting files

When it doesn't work

How this fits with the others

Why 3 cells in series and not 4

The source