Hacker News new | ask | show | jobs
by mike_d 982 days ago
Which is perfectly in spec. Feel free to email the USB Implementers Forum and let them know they are clowns and USB-C was a mistake.

USB-C simply describes a connector. USB-PD is one of many optional power over USB-C specifications you can use, or choose not to use. Heck, USB-PD v1 supports power delivery over USB-A connectors.
3 comments
thekombustor 982 days ago
My understanding is that an in-spec USB-C power supply will supply 5V by default - then, upon negotiation with the client, can go up to a higher voltage.

In the case of the Zimaboard, this is basically a 12V barrel jack with a different shape. It just dumps 12V out on the pins with no negotiation or considerations of the client device.

I suspect if the downstream device was already 12V compliant, it might survive, but I wouldn't count on that nor would I be willing to test it.

link
mike_d 981 days ago
USB-C must deliver a minimum of 3 amps at _at least_ 5V. If it delivers more than 5 amps it must contain a marker chip. The Zimaboard charger could deliver 4 amps at 12 v for 48 watt power delivery and be entirely within spec. (I haven't actually confirmed if this is what they do, so take it with a grain of salt)
link
nrp 982 days ago
This is incorrect. A USB-C power source can’t be in-spec if it doesn’t supply 5V by default. Other voltages or alt modes can be negotiated after that, but everything starts at 5V. This is defined in the base “ Universal Serial Bus Type-C Cable and Connector Specification”, not the optional USB-PD spec.
link
mike_d 981 days ago
A common misconception, Wikipedia spells it out maybe clearer than I did: "A device with a Type-C connector does not necessarily implement any USB transfer protocol, USB Power Delivery, or any given Alternate Mode: the Type-C connector is common to several technologies."

The document you mention is the specification for implementing USB 3.0 on a USB type C connector.

link
nrp 980 days ago
That’s not correct. We’re USB-IF members, and I’ve read through large chunks of USB-IF specs. The spec I am referencing is the one that defines the actual connector. You cannot build a power source that uses a USB-C connector and doesn’t support 5V and call it “USB Type-C”.
link
mike_d 978 days ago
> We’re USB-IF members

I hope so that you can pressure the Promoter Group into not making the same mistakes with USB-D. :)

Have a look at https://web.archive.org/web/20161220102924/http://www.usb.or... slide 45. Note how all the power requirements also list an associated USB spec.

Looking back I will give you that 5v is required for vconn.

> You cannot build a power source that uses a USB-C connector and doesn’t support 5V and call it “USB Type-C”.

Yet this whole conversation thread shows that is not the case. By not joining the IF you can do as you please.

A great example of this is the Ubiquti Theta cable (https://tinyurl.com/5asf8wpp) where they implemented a completely proprietary cable (as far as I can tell) with "USB Type-C" on both ends.

link
RockRobotRock 981 days ago
It's kinda silly that Android phone manufacturers came up with their own protocols. PD was long overdue, it's a godsend.
link