Hacker News Reader

Comment by chx | Hacker News Reader

> If you have a device that wants 20W as 9V/2.22A , your 30W charger may not support that specific combination and will charge much slower than a 20W charger that does.

Such a charger would be not specifications compliant. Don't blame the specs for what the manufacturers did with it.

Oh and don't get me wrong: there are a lot of problems with USB C mostly with the complete impossibility of knowing what the heck a USB C port is actually capable of. Is it Thunderbolt capable? if not, DisplayPort alternate mode capable? Can you charge the device over it? If yes, what's the maximum wattage.

This will ease finally because Microsoft will require PCIe tunneling on all C ports for the device to be Windows 11 certified so there won't really be different USB 4 ports -- except for charging wattage... https://docs.microsoft.com/en-us/windows-hardware/test/hlk/t...

View on HN

A charger does not have to support any particular voltage/current combination. If a device requests a specific config (9V/2A) but the charger only provides 12V/1.5A with 5V/1.5A fallback, then the best the two can negotiate is 5V at 1.5A. Even if the charger provides 9V/1.5A the device might be programmed to only accept 9V/2A with 5V fallback. This would be USB-compliant, but still frustrating to the user.

First of all, I have answered to "If you have a device that wants 20W as 9V/2.22A , your 30W charger may not support that specific combination" saying that's not specs compliant. https://i.imgur.com/rkIJuMr.png I highlighted the relevant part. These are the rules: https://i.imgur.com/b8q96tH.png So, to summarize: since a 30W charger must provide everything a lower wattage charger like a 20W charger does and 20W falls between 15 and 27 then yes it must support 9V 2.22A as well. If it doesn't then it's not compliant. Note the caption in the second image "an example of an adapter with a rating at 50W. The adapter is required to support 20V at 2.5A, 15V at 3A, 9V at 3A and 5V at 3A."

> If a device requests a specific config (9V/2A) but the charger only provides 12V/1.5A with 5V/1.5A fallback

Then it's not compliant. 15-27W is to be delivered via 9V. 12V is optional.

You're almost correct here - if a device advertises a particular power rating, they need to provide the configurations you linked. But they are allowed to provide any arbitrary configuration beyond their marketable power rating. So consider a charger that has a smallish inductor in its DC/DC supply, and a DC/DC chip with an integrated MOSFET limited to say 2.5A switching current. For thermal reasons, that charger can't safely provide more than say 1.7A regardless of voltage. Because it can't supply more than 1.7A at 9V it can't be listed as having a PDP of more than 15W. However, that charger is allowed to provide that same 1.7A at 20V if it wants to, and advertise that configuration to the device - as long as it doesn't list that as its rated power. This sort of charger would be functionally providing >30W, but would only be allowed to put 15W on its packaging/advertising materials. From a user perspective, they see a charger happily providing 30W to one device, plug in another device, and get a lower output power. The lower power is exactly what it says on the packaging (which the user threw out ages ago and never looked at) but the user is still frustrated.

> However, that charger is allowed to provide that same 1.7A at 20V if it wants to,

Is it? Allowed by the standard where?

Looks like I was wrong on this - at least in the latest version of the PD2 standard it's not allowed to advertise any combination that exceeds the PDP. It's only in PD3 that devices are allowed to limit power in particular configurations due to thermal constraints even if they can technically deliver the same power in another configuration, but even that is limited only to >100W configs. So you're right about this - my example is definitely incorrect.

Reply on news.ycombinator.com