HN Explorer

21 comments

[ 4.5 ms ] story [ 14.6 ms ] thread
megous
> I’ve never done this circuit, but the way I see it, it should be enough to use two FETs, one per CC pin, both put in parallel.

Yeah, don't do it. CC also carries BMC signalling and pull-up/downs are normally dynamic and can change at runtime after plugin.

CH32X035 looks very interesting for playing with USB-PD. :)

MenhirMike
Isn't something like that what made the original Raspberry Pi 4 not actually USB-C compliant, because they combined two signals that shouldn't have been combined?
megous
I don't think so. Rpi shorted CC pins together.
greggsy
Isn’t it just for power, which would make it non compliant anyway if it’s not transferring data? Assuming that would make it non compliant…
stavros
This post gave me the idea of making a USB-C Russian roulette chip, where plugging the USB jack in on one side works fine, but the other side fries your device.
mschuster91
New product idea for USBKill... for the uninitiated, a device that is used to test hardware defenses against malicious input using high voltage [1].

[1] https://usbkill.com/

midoridensha
A more interesting and complicated version would use a random number generator to give you a 1 in 6 chance of frying your device when you plug it in.
stavros
Yeah but then that's not a commentary on how USB-C is supposed to be the same in both directions but isn't.
gsliepen
Oh! How about a USB-C device that detects its orientation this way, and requires you to replug it in the different orientation twice before it starts to work?
gumby
Nice back compatibility thinking!
dist-epoch
This could work in audiophile land.

Start a legend that the DAC sounds "digital", "harsh" and "jittery" if the USB-C plug is not inserted in the "correct orientation".

And of course, your device has a small LED near the plug which is green when plugged with the correct orientation.

LargoLasskhyfv
Green isn't audiophile enough. Needs to shine in royal purple pink, or something like that.
eternityforest
LEDs in general aren't audiophile enough. It should be like, a bare hot loop of nichrome behind some chicken wire for an indicator.
burstmode
Every few sentences there there is a remark like "This circuit might have edge cases,... "...perfect for the majority of use cases,..". In other words: It will, at best, work on the developers table, but nowhere else. Why is somebody publishing examples that won't work in the real world ?
saikia81
hackaday is not about making consumer products. It's for hackers and hobbyist. The real world is vague, but should include people trying different implementations.
m0llusk
USB is a mess. Lots of devices that claim support are actually way out of spec. As a result it is hard to say exactly what will happen when people start plugging stuff in.
MichaelZuo
This is especially true for USB hubs of any kind, so much so that I think even Caldigit sold a hub that wasn't fully compliant to it's advertised capabilities.
bloopernova
For testing your USB-C cables, I highly recommend the caberQU C2C: https://caberqu.com/home/20-c2c-caberqu-746052578813.html

So many cables aren't power delivery, just USB 2 spec: If you're wondering why your phone doesn't charge quickly, the cable tester might be able to help you find the issue.

wlesieutre
You'd think you could plug a cable into a computer and have it tell you the capabilities instead of needing a dedicated device
bloopernova
Yeah but that would be useful. Like being able to switch USB sockets on or off from software. Or lock them to a specific speed/bandwidth.
extraduder_ire
You'd need to plug the other end into something and, unless you know what that device is, you wouldn't know if it's the cable that's lacking or the device you plugged in on the other end.

Better control of USB devices from computers would be swell though.