USB-PD PSU almost thrown away because it would not charge a simple bicycle light

[daveyosborn@infosec.pub]

I bought a Lenovo laptop charger with USB-C connector. There was no mention of USB PD but the label listed voltages (5v, 9v, 12v, 15v, 20v), which encourages us to guess that it may possibly be USB-PD. But since there is no incentive for Lenovo to disclose whether it’s USB-PD compliant, it chooses to minimise its legal liability and write nothing about USB-PD (as most PSUs do, even aftermarket ones).

Since I bought the PSU from a dodgy street market where people sell dumpster-dive products, I needed to quickly test it. I attached it to a simple USB-C bicycle light which has an LED to indicate charging. It did not charge. So it appeared the Lenovo PSU was dead. I tossed it into a box of e-waste to decide later whether to repair, harvest parts from, or dispose of.

Some manufacturers apparently believe when the USB-PD standard says “5v is the default”, it means 5v should be delivered in the absence of a handshake. Other engineers have interpretted “default” to mean both sides of the negotiation should conform to 5v /if/ the handshake negotiation yields nothing higher -- but that a handshake is strictly required nonetheless. No handshake → no power.

Couple that discrepency of USB-PD interpretation with the fact that manufacturers resist claiming to conform to USB-PD anyway (as the law does not demand it and consumers don’t either), we have a recipe for disaster. I am sure a lot of functional USB-PD devices will be thrown away because of this kind of shit that I encountered. Perhaps the Lenovo PSU I have was already in a dumpster and fished out by whoever sold it to me.

I almost tossed the Lenovo PSU but decided to try powering a 9v device with it, and it worked. I almost didn’t do the experiment because I originally thought if it cannot power a 5v bicycle light it’s probably dead.

Worth mentioning this thread which shows how consumers are blindly overrating USB-PD unaware of the pitfalls.