Protocol Testing

It Lights Up, So It's Fine — The Gap That Sinks a Whole Batch

The sentence every charger buyer dreads: "But it lights up — why does the customer say it doesn't work?" Nine times out of ten, it lit up but never made a deal with the customer's device.

I get asked this a lot. The answer is usually simple, and almost always missed: it lit up, but it never negotiated with the customer's gear.

Here's the mechanism in plain words. A charger's output isn't fixed. Plug in an iPhone, it should negotiate 9V or 20V over PD. An older Android, maybe QC. A laptop, 20V. That negotiation — the talk — is the protocol handshake. Negotiate well, you fast-charge. Negotiate badly, you fall back to 5V slow, or connect-drop-connect.

Why a light check misses it: the light test uses a dummy load. The dummy never talks. So the handshake — the exact thing the customer's phone does — was never verified. The factory tested "it outputs." The customer tests "it outputs after a conversation." Different test, different result.

Three ways it goes wrong on a real line

  1. Never negotiates. Label says 65W. Customer's phone sits at 5V/1A. Back it goes.
  2. Negotiates but wobbles. Charges a bit, drops, charges again. Customer blames the cable. It's the handshake.
  3. Ports fight. One device fine, two devices and they claw at each other, total power sags.

The real cost isn't one unit. It's the batch. Five thousand pieces, return shipping plus re-test plus a customer who now doubts you — that bill is bigger than the tester that would have caught it on day one.

How the better lines actually prevent it

We put this into a single bench unit: full protocol handshake, ripple measurement, results to MES. If your line still hand-tests with a phone per unit, you're burning time and still missing the wobblers.

If your chargers leave the building on a power-on check alone, it's worth a look before the next batch ships.
The bench unit for this
See the GTI6031H tester we built for this →
← Back to all articles