Not built yet. This is my plan and parts list. When it's done I'll add what actually happened.
What to expect
This is a demonstration, not a power source. It's worth being clear about that before you start.
- Voltage is easy, current isn't. One copper and zinc cell gives about 0.9 V with nothing connected. The current it can push is tiny, roughly a milliamp or less, and it drops as soon as you draw on it.
- An LED needs several cells. A red LED needs a bit under 2 V before it lights at all, so you need three or four zinc cells in a chain, or two magnesium ones. Even then it glows dimly. A dark room helps.
- Nothing is free. The zinc or magnesium slowly dissolves. That metal is the fuel. When it's gone, so is the electricity. The earth or the lemon only carries the current.
- It fades. Fruit dries and goes off in days. Soil dries out. The plates corrode. Expect to clean and rebuild it.
Why this way
Two metals that are cheap, safe to handle and far enough apart in how readily they give up electrons to make a useful voltage. It's the classic pair.
Magnesium gives up electrons even more readily than zinc, so a magnesium and copper cell gives about 1.5 V. Fewer cells to light the LED, but the magnesium gets used up faster.
A galvanised nail is steel with a zinc coat. It's the easy way to get zinc into a lemon, and it shows what happens when the coating wears through.
More metal touching the soil or juice means more current. Strips cut from sheet have far more surface than a wire.
Red LEDs light at a lower voltage than most other colours, and they glow visibly on a very small current. That makes them the best thing to light from a weak source.
A needle meter measures volts using the cell's own power. It needs no battery of its own to do it, which suits a page about battery-free things.
How it works
Put two different metals in something damp that conducts, like lemon juice or wet soil, and join them with a wire. The zinc slowly dissolves, and each zinc atom that does leaves two electrons behind in the metal. Those electrons flow through the wire to the copper. At the copper they're used up in a reaction with the acid in the juice, or with water and oxygen in the soil. The copper itself isn't used up; it just gives the reaction somewhere to happen.
The voltage depends on which two metals you use, not on how big they are. Size changes how much current you can get.
One cell isn't enough for an LED, so you chain them in series: the copper of one cell to the zinc of the next. The voltages add up. Each cell must be in its own pot or its own lemon. If they all share one patch of soil, the current takes a short cut through the soil between them and the voltages don't add.
| Metal pair | Roughly, per cell | Notes |
|---|---|---|
| Zinc sheet + copper | ≈0.9–1.0 V | The classic pair. Steady and cheap. |
| Galvanised nail + copper | ≈0.9 V | Works while the zinc coating lasts. Falls away once the bare steel shows. |
| Magnesium + copper | ≈1.5 V | Highest voltage here. The magnesium fizzes and wears away quickly in acid. |
| Copper + copper | ≈0 V | Same metal both sides, so nothing to drive the current. A useful check. |
These are open-circuit figures, measured with nothing else connected. They're approximate and vary with the fruit or soil, the temperature and how clean the metal is.
What to buy
The lemons or soil you have already. Listings change, so check each one before you order.
These are affiliate links. If you buy through them I make a small commission, at no extra cost to you. It helps cover the cost of running this site. This is a hobby, not a shop.
Metals
Load and measuring
You'll also need: tin snips, a file or sandpaper for the cut edges, four lemons (or four small pots of garden soil and some water), a marker, and a dark room.
Build it
Measure one cell before you build a chain. It tells you whether everything else will work.
-
Cut the plates
Cut copper and zinc into strips about 20 mm wide and 70 mm long, four of each. Snip a corner off each copper strip so you can tell them apart in a hurry. File the edges smooth, then rub both faces with sandpaper until they're bright.
Why: a clean surface reacts properly. Tarnish and oil from your fingers cut the output. Sharp edges from snips cut skin.
-
Make one cell
Roll a lemon firmly on the table to break up the inside. Push in one copper strip and one zinc strip about 3 cm apart. They must not touch, inside or outside. For soil, fill a small pot with garden soil, wet it until it's damp all through but not swimming, and push the strips in the same way.
Why: rolling the lemon frees the juice, and the juice is what carries the current between the plates. If the plates touch, the cell shorts itself and you get nothing at the wires.
-
Measure it
Set the meter to the 2.5 V DC range, or the nearest range above 1 V on your meter. Red lead to copper, black to zinc. You should see roughly 0.9 V. If the needle tries to go backwards, swap the leads.
Then switch to the lowest DC current range and clip the meter straight across the cell. That shows how much current the cell can push into a short circuit. Expect very little.
Why: the volts show the chemistry is working. The current shows why one cell can't do much. Never do the current test on a real battery: it's only safe here because these cells are so weak.
-
Try the other metals
Swap the zinc for a galvanised nail, then for a folded bundle of magnesium ribbon. Measure each. Then try two copper strips together and see the needle stay near zero.
Why: it shows the voltage comes from the difference between the two metals, not from the lemon. The copper and copper test is the proof.
-
Build the chain
Set up four cells, each in its own lemon or pot. With croc leads, join the zinc of cell 1 to the copper of cell 2, the zinc of cell 2 to the copper of cell 3, and so on, like Fig 2. Leave the copper of cell 1 and the zinc of cell 4 free. Measure across those two ends. With zinc cells you should see roughly 3.5 V.
Why: in series the voltages add. Separate pots matter: cells sharing one patch of soil leak current between each other and the total won't build up.
-
Light the LED
Draw the curtains. Clip the LED's long leg to the copper end of the chain and its short leg to the zinc end. Look closely: it should glow faintly. If nothing happens, turn the LED round and try again.
Why: an LED only conducts one way. The long leg is the positive side, and copper is the positive end of the chain.
-
Try it in pulses
Unclip one LED lead and touch it back on now and then, rather than leaving it connected. Each touch after a short rest should flash a little brighter than the steady glow.
Why: when current flows, the plates get coated with a thin film from the reaction, mostly hydrogen bubbles on the copper, and that chokes the cell. A short rest lets some of it clear, so the first moment after reconnecting is the strongest.
Looking after it
- Keep soil damp. Dry soil barely conducts. A splash of water brings a fading soil cell back.
- Clean the plates. When the output drops, pull the strips out and sand them bright again. The zinc will look pitted. That's the fuel being used.
- Replace the fruit. Lemons dry out and go mouldy within days. Compost them. Don't eat fruit that has had metal in it.
- Watch the zinc wear. Over weeks you'll see the zinc thin and the nails' coating go dull and patchy. That's the honest answer to "where does the power come from".
- Store dry. Rinse and dry the strips after use. Keep magnesium ribbon sealed and dry.
If it's not working
| What you see | Likely cause | Fix |
|---|---|---|
| Needle doesn't move | Plates touching, wrong meter range, or a bad clip | Separate the plates. Check the DC volts range. Clean the clip jaws and the plate tops. |
| Needle moves backwards | Leads swapped | Red lead to copper, black to zinc. |
| Chain reads less than the cells added up | Cells sharing soil, a reversed cell, or a bad link | Use separate pots. Check every link goes zinc to copper. Measure each cell on its own. |
| Good voltage, LED dark | LED backwards, or not enough cells | Turn the LED round. Add a cell, or use magnesium cells. Try a darker room. |
| Output fades over minutes | Plates coated by the reaction | Disconnect for a while, or pull the plates out and clean them. |
| Soil cell weak from the start | Soil too dry or too sandy | Water it. Try a heavier garden soil, or add a pinch of salt to the water. |
Safety
- Magnesium burns. If ribbon catches fire it burns with a blinding white flame that water won't put out. Keep it well away from flames and heat, don't try to light it, and store it sealed and dry.
- Sharp edges. Cut sheet metal is sharp. File every edge and handle offcuts with care.
- Don't eat the fruit. Metal dissolves into it. Wash your hands after handling the plates and the lemons.
- Children. Small LEDs and offcuts are a swallowing risk. Supervise younger children and put the magnesium away yourself.
Notes
This is the same chemistry as the first batteries of around 1800, which stacked zinc and copper discs with salty cloth between them. People later buried big plates in the ground to make "earth batteries", and in the 1840s Alexander Bain used one to run electric clocks. They worked, but always on the same terms: the zinc wears away.
The voltages here are well-known approximate figures. Real readings depend on the fruit or soil and the state of the metal. I'll add my own measurements when I've done it.
Related guides: needle meters, compasses and sundials (how the meter works without a battery), the crystal radio, and Stirling engines.