Protected and unprotected cells

Every lithium cell you can buy is one of two things: a bare cell, or a bare cell with a small board on the end of it. You cannot tell from the outside of the cell, because the cell is identical in both.

A protected 18650 is a bare 18650 with the board tucked under the wrap at the negative end, which makes it two to three millimetres longer — 68 to 70 mm against 65. That is why a protected cell sometimes will not seat in a holder that was made for an unprotected one, and it is the quickest way to tell them apart with the cell in your hand.

Hobby LiPo pouches sold with a JST plug almost always have the board potted at the wire end already. Two wires out of a small PCB, not three. Cells pulled out of old devices almost never do, because the protection was on the device's own board.

What the board is watching

The usual pair is a DW01A and an FS8205A. The DW01A is the brain, six pins, a few cents; the FS8205A is two N-channel MOSFETs in one package, wired back to back so that one of them blocks each direction of current.

The DW01A opens one or both of them for four reasons:

• Over-charge, above 4.3 V ±50 mV.
• Over-discharge, below 2.4 V ±100 mV.
• Over-current, after about 10 to 20 ms.
• Short circuit, in tens of microseconds.

Notice what those thresholds are. 4.3 V, not 4.2. 2.4 V, not 3.0. The board is not managing the cell — it is standing at the edge of the cliff, well outside the range you should be operating in. Everything between 3.0 V and 4.2 V is somebody else's job.

The measurement everybody misreads

Press play. When the cell reaches 2.4 V the discharge MOSFET opens and the pack terminals read 0.00 V — while the cell behind them is still sitting at 2.4 V.

This is the single most common lithium question a beginner asks. The battery is not dead. The protection board has done exactly what it was built to do, and the pack will come straight back.

It comes back because current can still get in. The discharge MOSFET is open, but its body diode conducts into the cell. Put a charger on the terminals, the cell climbs back above the release threshold, the DW01A closes the switch, and everything works again. Modules that "will not charge a flat battery" usually need nothing more than a moment at a slightly higher voltage to start.

What it costs, and what it does not do

The two MOSFETs are in series with the cell, permanently, adding a few tens of milliohms. On a 2 A discharge that is a few tens of millivolts and a little heat — the same arithmetic as everywhere else in this track.

What it does not do is longer than what it does:

• It does not charge. There is no constant-current stage and no 4.2 V target in it. It is a cut-out at 4.3 V and nothing else.
• It does not balance anything. One board, one cell. Multi-cell packs need a real BMS with a balancing circuit, which is a different part entirely.
• It does not gauge. It has no idea how much charge is left.
• It does not protect against charging when cold. Cheap boards have no thermistor. The invisible damage from the previous page is still yours to avoid.