There's been something that's been bugging the hell out of me lately. Why is it that some devices are equipped with USB C, but refuse to work unless the other end of the cable is USB A?

I have at least 4 devices like this at the moment, including a handheld emulator. It's so freaking annoying when I'm trying to standardize around one connector type. Especially because I have no way of knowing that a device will have this quirk when I'm purchasing it.

Slightly related rant: I can only assume that this is the explanation for why so few USB chargers refuse to relinquish the USB A ports. I get that they want to appeal to as many people as possible but it's been crazy to see how few chargers ONLY have C ports. I think I've found like...2.

Ok, I did some more research and I'm extremely unsurprised that the answer seems to be "laziness and cost cutting." The only solutions are the most janky imaginable: permanently affix a USB A to C adapter and cable with USB female plug, so the proper negotiation can occur. It's so dumb.

I wonder if anyone makes a specialized adapter to address this issue... I'm going to look around.

I'm pretty sure that the reason would be a cheap cable. Mayyyybe the device itself is problematic, but most of the time when I have a problem, I get a better cable and the problem is solved.

Have you tested each of those four devices with several well regarded/rated USB-C chargers and similarly qualified USB-C to USB-C cables?

Some (nominally) USB-C cables and some USB-C chargers are not capable of _all_ the various output modes and end-to-end protocols that the USB-C world includes.

Even some of the earlier model USB-C power adapters from Apple were limited spec. Apple did list all the modes they actually implemented in each charger model if one were to go look.

It would not surprise me to learn that various adapters and cables currently in the market are only implementing portions of the USB-C spec.

It would be interesting to learn whether the four devices you mention will work with _any_ USB-A to USB-C cable or only some.

Edit: Also, in some cases the USB-C connector must be oriented one way up. If connected with the other side up, the USB-A internal connection does not work. Sometimes I encounter this and have to ‘flip’ one end of the USB-C plug over, plug it back in, see if it works. Sometimes I have had to do this on the other end or even swap the USB-C cable end for end.

USB-C (and ThunderBolt) continues to impress with the amount of potential and actual user confusion. One connector to confuse us all!

Yes I've definitely tested. All the devices fail to work with every grade of USB C cable, but work just fine with any grade of A to C cable. Apparently this is a common issue where manufacturers simply ignore the spec and leave out the proper resistor to save a tiny bit. It's obnoxious.

See the response here:

https://www.reddit.com/r/UsbCHardware/co...h_usba_to_usbc/

I saw a post somewhere that said, "What just happened? We went from lots of different ports, all labelled, to one port but lots of different cables, all unlabelled..."

Peter

Indeed. USB-C is just a hot mess.

I dunno. Sounds like a mess, but in practice here I'm pretty pleased with how it all just works. Mind you, I have been somewhat careful to only buy/keep cables that support 5gb/s or faster, avoiding the USB2 ones that sport USB-C connectors.

And I'm loving Thunderbolt-4 / USB4 connections over USB-C cabling.

Cheers

I find myself buying ThunderBolt 4 rated cables when I only ‘need’ USB-C.

Especially with power delivery and the higher speed ratings, I suppose I am trying to minimize the number of ‘have limitations’ cables I have in use and laying around. ThunderBolt 4 tends to higher cost and thicker cabling, so I do have some ‘plain old USB-C cables’ in the mix.

I color code them. White = not ThunderBolt 4, and black/grey cable means ThunderBolt.

Ditto. My (10gb/sec, 35W) USB-C cables are all white; my Thunderbolt4/USB4 ones (40gb/s, 130W+) are currently all black.

I use NVMe drives in external USB4 and USB-3.2 enclosures for quick backups of my main NVMe-equipped laptop. Either is quick, but the USB4 (40gb/s) ones finish in less than 1/3 of the time of the USB-3.2 (10gb/s) ones.

I'm nearly positive that the cakes and chargers are not the issue. I like ugreen's products so most of the cables are theirs, but I have a variety of others I'm able to test with. I just have some devices that refuse to work with any c to c cables whatsoever.

Hm. Serious USB-C cables have some electronics in them which are supposed to communicate their power capabilities to the devices. Wikipedia says this:





My guess is that all of the A-to-C cables you've been using are the "minimal" cables mentioned in that quote, and your C-to-C cables have the E-Marker chips. The minimal cables require no special interpretation by the devices and just work (at 3A or less), but then the devices balk when confronted with the E-Marker chip, possibly disliking what it tells them (or maybe just rejecting any cable with an E-Marker at all). Alternatively, maybe the cables are communicating good information, but the specific troublesome devices are misinterpreting that information due to some kind of bug.

Related: I recently saw this video which describes one interesting special case where the E-marker chip might cause a problem. The spec doesn't allow for multiple daisy-chained cables to communicate their capabilities downstream, so if you use a USB-C extension cable, the device that's drawing the power might, in some situations, draw more current than the second-in-line cable can handle.

My understanding is that the USB-C specification has no allowance for any form of USB-C to USB-C ‘extension’ cable.

USB-C adapters to/from USB-A are possible, but there are implementation requirements.

What about a 1-port hub at the end of a cable?

USB-C hubs have apparently been something of a weird thing.

The early days it seemed common to encounter complaints about USB-C hub Implementation issues and not-quite-what-was-expected functionality. I have not been tracking the current state of the art.

I have managed to avoid buying any pure USB-C hubs to date. I do have several ThunderBolt (wall powered) hubs which include USB-C ports (along with an array of other port types). Thunderbolt hubs also went through some early rendition issues.

And yet I've got one. Pretty sure I've run video over it in the past as well...

The market will always attempt to satisfy consumer demands

The problem is that the extension cable can differ in attributes (and capabilities) from the primary USB-C cable it is extending. Per the above referenced video, this can create edge cases where the USB-C end devices can make incorrect assumptions about the overall capacity of the paired cable set.

Fair. I bought the cable and the extender from the same manufacturer at the same time, so that's -- probably -- not a problem for my specific case. But, yeah...

I, too, have successfully used USB-C extension cables, despite the fact that the USB-C spec doesn't allow for them. In my case, I have used them to extend the reach of the PSVR2 headset. Since that's relatively low power, compared to, say, charging a laptop off the cable, I'm probably fine. As that video explains, though, problems may occur if you attach a device which needs more power.

That's all irrelevant to Dignan's problem, of course, since he wasn't using an extension cable, but it's certainly an interesting side note, pointing out how the E-marker chip in the cables just makes life harder and more complicated.

We should just go back to serial ports