Exway doesn't care about USB-C conformity | Hacker News Reader

Exway doesn't care about USB-C conformity

ericswpark.com

159 pointsericswpark114 comments

While I'm glad that they took the time to hold support's feet to the fire over this, doing that sort of thing is almost certainly a waste of time.

Seldom will you find a support team who can understand a technical hardware problem like this, and even more seldom will you find a company that responds to this with anything other than "nothing we can do, sorry". You're not going to get the "Wow, that is our bad, we'll retool our entire production line to account for that issue that nobody else complains about and send you one as soon as it's ready, thank you" that you so desire.

I've sometimes been able to escalate to the right people and effect change, even when I have to go through a seemingly uncaring customer support team.

I was inspired by patio11's "Identity Theft, Credit Reports, and You."[0] It changed the way I raise disputes with companies as a consumer, and it's gotten me good results.

The tl; dr is that large organizations have things they're afraid of, and they typically have processes in place to prevent them from happening. If you figure out what the company is afraid of, tie your grievance to that fear, and it will pressure the company to resolve your issue.

With credit reporting agencies, they're afraid of regulatory incidents. If you give signals that you're gathering evidence for a complaint to regulators, they'll work hard to resolve your issue. Other companies are afraid of a complaint to a distributor or the potential for a lawsuit. They're usually afraid of something, and if you can figure out what it is, you can get the attention of people with the power to resolve the issue.

[0] https://www.kalzumeus.com/2017/09/09/identity-theft-credit-r...

So; in this specific case, if they have a USB-IF trademarked logo or icon on their device or packaging, then maybe the USB-IF could come along to put some pressure to get a better outcome?

I've seen cheap devices with USB ports --- but no logo. Probably for this very reason.

The USB logo is pretty well pointless. All you're paying for is a license for the logo.

Despite USB-IF's efforts to associate that logo with "good" or "safe" or even "compliant" USB devices, I doubt that many consumers know or care.

Part of the reason is that their licensing is pretty pricey unless you're already at the mass manufacturing stage. You have to pay an annual fee for the logo, an annual fee for the VID, and you have to pay to get your widget certified. All so you can use a logo that consumers don't know or care about.

On the other hand, you can implement a USB device without paying a dime to USB-IF. It's super annoying without your own VID, but that's often not even necessary.

So we get cheap Chinese devices with no certification because why would they even bother?

Worked in support for a long time. Very true.

The real trick is trying to get the support guy to tell you what to do to get the right attention ... IF they know.

I used to do that all the time when I did big time networking gear support. "I, the support guy, can't just send you a new router (price like $200k+), that's just the policy, I gotta do X, Y, Z. That will take a bit of time and here is how that works ____ . But if you can get _____ to tell me to do it, I'll do it."

Now as a support guy you have to know the lay of the land before you deliver that line... but I was lucky enough to know and that org had good policies and so on.

What is AMAZING is some customers didn't realize that I just gave them the route they needed to go to get exactly what they wanted and they'd complain and whine. Like guys ... come on. I'd also tell them what to do to get closer to that goal faster without pulling strings, but that meant they had to do extra work, lotta folks didn't want to do that either.

Granted consumer support guy, probably has no tools / doesn't know the magic words / people and so on. Also probably afraid to tell you. Support personnel are most often "valued' for a short time but in reality are seen as a cost in most orgs and treated like garbage / scummy pawns. At one company I worked at the engineers would invite me over to their building when they had food catered. They knew we got jack squat (support almost never got food catered in), we had management who only knew how to prove their worth by penny pinching, and the engineers liked some of us / knew we saved them a lot of time.

Side story: I don't know if Amazon ever had human support but right now I've got something that shipped 3+ months ago and it is "On the way but running late" ... for 3+ months. In the past Amazon would just give me a refund outright... Now Amazon just sends me between two different bots that can't help me at all... The seller has a bunch of posts and feedback all the same, people not getting the product. Amazon bot don't care tells me to talk to the other bot. My review (almost the same as the other reviews now warning people about not getting product) was rejected.

Amazon deeply hides the human chat features. They’re also placed differently in diff countries. Go look for it under the general support page, not starting from the product order status page which rarely exposes helpful support functions

TY, I will try that.

even when amazon tells you on the product page that something cannot be returned, if you can find the chat link elsewhere and ask about the product there, you will find that not only can it be returned, you can also get the option to be refunded without having to return the item (at least mostly the case with food items in my experience)

On the Amazon thing, get the tracking number and track it with the carrier. I’ve found Amazon will say something has shipped when only the label has been created. You can see the truth if you track it through the actual carrier.

There seem to be some sellers on Amazon that just don’t ship things until the customer complains.

It cuts both ways. Some carriers tracking (Glares at USPS) is just bad. I had had stuff delivered that the web still claims is "label created, item not yet received".

The China post tracking number doesn't track, although I suspect there really is no way to make that work. I half suspect someone figured out if it didn't track they could avoid Amazon giving refunds.

> Now Amazon just sends me between two different bots that can't help me at all...

I ran into this issue some months back. The unofficial-official policy seems to be to make you file a dispute with your CC stating that merchandise was not delivered. I was promptly refunded and haven’t had any issues conducting further business with Amazon.

Great comment… It makes me appreciate that last year I bought a part for my car that didn’t work great but was ok. They said exactly that- they would gear up to manufacture a better one and get back to me. About 8 months later I got two redesigned parts free in the mail and got to keep the initial one.

If this is True, then definitely name the company! This is excellent customer service!

Support can’t do anything, but they are often your only chance to get what you want: the message passed to someone who might be able to in the future.

That’s what the rep said they would do here.

And sadly that’s the most you can ask for.

Then you don't bother with the company and report it straight to the regulators. Losing that sweet CE sticker will make them move.

Not implementing USB-PD could cause that to be yanked in a product.l?

Maybe these aren’t sold in Europe though.

I think the issue is worse than not implementing USB-PD, it won't charge at all when plugged into a usb-c port, thus not complying with the usb-c spec while having a usb-c port. I'm no lawyer but the author says it's illegal in some jurisdictions.

In your declaration of conformity you have to list which standards you conform to, and IEC 62680-1-3 would be one of them. I suppose if you found to be lying on your declaration there would be consequences.

> doing that sort of thing is almost certainly a waste of time.

A waste of time for both parties, because the author is wrong.

As Wikipedia helpfully points out: "The designation C refers only to the connector's physical configuration or form factor and should not be confused with the connector's specific capabilities"

From the write up it looks like he was able to get it working with the minimum 3A required by the cable spec. You don't actually have to support C-to-C cables if you choose to not to communicate with the e-marker.

Hi, author here. The quote you mentioned from Wikipedia indeed says that devices don't have to implement all the capabilities of the connector. However, further down, it says the following:

> USB-C devices may optionally provide or consume bus power currents of 1.5 A and 3.0 A (at 5 V) in addition to baseline bus power provision; power sources can either advertise increased USB current through the configuration channel, or they can implement the full USB Power Delivery specification using both BMC-coded configuration line and legacy BFSK-coded VBUS line.

And further down:

> However, to connect a USB 2.0/1.1 device to a USB-C host, use of Rd[57] on the CC pins is required, as the source (host) will not supply VBUS until a connection is detected through the CC pins.

See: https://en.wikipedia.org/wiki/USB-C

As the device doesn't implement the full USB PD specs, it must advertise that it wants power through the CC (configuration channel) pins. This is part of the specifications and it must be properly implemented.

> > However, to connect a USB 2.0/1.1 device to a USB-C host, use of Rd[57] on the CC pins is required, as the source (host) will not supply VBUS until a connection is detected through the CC pins.

Yup. You are referring to the USB 2.0 spec there, not the USB-C spec. If you follow the citation it in turn seems to be citing (not explicitly) the USB-PD spec.

It's a remote that charges via the USB-C port. If it doesn't follow USB 2.0 spec, or the USB-PD spec, or any spec associated with the type-C port, how is it supposed to get power from the host device/charger?

> Seldom will you find a support team who can understand a technical hardware problem like this, and even more seldom will you find a company that responds to this with anything other than "nothing we can do, sorry".

I've had reasonable luck, honestly. Certainly not the majority of the time, but often enough to have some hope.

I've found that if I treat the support person decently -- as in, don't show my irritation and treat them as I'd prefer to be treated if I had their job -- and I demonstrate (not assert) that I've done my homework, know what I'm talking about, and am reasonable, I can get things escalated to someone who has the knowledge and power to address my concern.

I've even been put directly in touch with devs this way on occasion.

What I've learned is that most support people actually do want to get your issue resolved and don't want to spend all day with you. What you need to do is give them an acceptable (to their supervisors) reason to kick your issue to someone who can be more helpful. And don't be a dick. Nobody's going to go out of their way to help a dick.

So, you have proof their devices are defective, and chosen to be so. Hold on those emails, they are valuable.

Also, this reminds me a lot of the urban legend regarding Van Halen and Brown M&Ms with regard to their contract.[0] If they ignore the easy stuff that is hard for most to know, what other corners might they be cutting?

[0] https://en.wikipedia.org/wiki/Van_Halen#Contract_riders

I wonder how effective is this strategy, and is there a name for it? Reminds me of people planting 20 dollar bills in their auto seats before a oil change/car repair. If the money is missing, what does that actually say about quality of the car repair? Also doesn't trust go both ways and you risk damaging the relationship with the vrndor? I'd be very reluctant to repair someone's car again or who plant judgmental traps.

There is an academic literature in game theory dealing with this topic, the key text is called "Repeated Games and Reputation." The difference from your example and the OP is that OP's experience isn't a repeated purchase (to my understanding). For Van Halen and brown M&Ms, there are elements of leverage in negotiation and other elements of reputation.

I have this problem constantly. I have to work with a number of hardware devices for my job (non-consumer hardware) that proudly claim USB-C.

But they also are exactly like the old USB-micro devices with a different port and don’t charge on real USB-C. So to charge them I use a USB-C to USB-A port adapter so I can plug in a USB-A to USB-C cable.

It’s horrible.

I was kind of resigned for everything to be awful forever. It felt like 60% of the devices I bought had this problem. Makes things like charging these Tribit Surf speakers I otherwise love quite a pain. But I've purchased around a dozen pretty cheap devices with USB-C in the past year & much to my surprise each one has properly resistor terminated CC pins & just works.

I was kind of begging for a specific C-C cable that built in the resistors, specifically for these bad devices. It'd be incredibly easy to make, but what a dumb purpose in life. A 3 inch male-to-female adapter would be ideal, for all these jerk-wad devices.

I'm a bit perturbed but the very excellent power-monitoring AVHzY CT3 device I got recently automatically negotiates 5V, so at least when I go to plug in any of the various problematic devices, they work now. Alas it requires a second usb-c cable to work, plus the device, so it's cumbersome: that male-to-female usb-c 5v adapter would still be appreciated.

In the end, it feels like the real pressure the world needs is better reviewing. It'd be lovely to have a meta-site, that tells reviewers things they need to check for on each product. Slip ups like this should be a notable ding on everyone's name, but there's so many reviewers and so few actually know all the various things to look for. Some progress in solving the review meta-problem - enumerating all the concerns any device-type might have- would be greatly appreciated.

It’s so weird. I have what I would categorize as “cheap crap” that works great. Like an IR thermometer for surfaces or an electric ‘compressed air’ duster (powerful fan with a small nozzle and a battery).

Both work perfectly on proper USB-C and charge. They were both $15 or less. The kind of device where every cent on the BoM may matter.

The devices in my original comment? Hundreds of dollars each. They have to go through certifications to prove they do what they say. The price of two resistors (they don’t need more power) is nothing compared to the MSRP which I already suspect has a lot of profit baked in.

There is no way to guess if a device charges right or not without finding the logo (which I guess could be a lie from an unscrupulous vendor) or finding reviews.

It’s not about the price of two resistors, it’s just incompetence when updating the design from micro-usb to usb-c.

I agree. I think they took an old design for the USB part of things and just swapped the connector too.

Yes, I think it's a case of the engineer finding a connector that doesn't expose the CC pins (where the resistors go), and not even realizing that something is missing when they update a design from microusb to usb-c.

> non-consumer hardware

Presumably not cheap too and they do… that.

Hundreds of dollars, and I wouldn’t be surprised if there were very healthy margins too.

But two resistors? Nah.

I have a feeling this will go over poorly with their customers as USB-A stops being so common and suddenly the devices “won’t charge” when plugged into computers or common power bricks.

(throwaway account for fear of retribution)

Eh, I don't know...

The author seems to assume the USB-IF is a good thing.

Having gone through the despicable $4,000 shakedown that is required to get a vendor id from the USB-IF, and implemented multiple devices against the outrages that are the specs, I dream about the remote possibility of living in a post-USB world someday.

Yes, I was around in the bad old days before USB...

Author here; I don't think my blog post is in support of the USB-IF (unless I missed a sentence I wrote somewhere that insinuates otherwise...) I just think selling defective products that don't charge with certain cables shouldn't be allowed.

Thanks for clarifying.

In support of your point, this sentence, "I mean, the USB-C spec is really long, and probably very complicated thanks to the USB-IF committee" makes me think we probably agree more than disagree about the USB-IF.

I was probably reacting more to my own bitter experience with USB-C/3.xyz-it-even-has-electrolytes version(s).

It was relatively trivial to stand up a USB 2.x system.

USB 3.x with Type C connectors? Orders of magnitude more challenging throughout the stack. Even 99% of the Type-C connector hardware is bullshit.

I guess I'm empathizing with the opposite view that your title presents.

I truly don't care about _USB-C conformity_.

That said, I'd be embarrassed if my product that had a USB-C connector did not (at least slow) charge with a standard Type-C cable.

come on, we just got rid of crossover cables for ethernet, do you really want rs-232 again? :)

Yah, there are lots of good reasons to be grateful for USB (especially before the fiasco that is type C).

I'm just bitter and resentful about being forced to pay thousands of dollars for an artificially scarce number.

do you really need to get a vendor number? I've assumed there's an unregulated section somewhere.

I think most cheap devices I've tangled with use the vendor number of whoever made the chipset. (or a random one, because chineseium)

Yes, you do.

You can beg for one from a chip vendor, but they aren't really permitted to dole them out, and they will help once, but not for production distributions.

You can ignore it, make one up, or just try to take someone else's, but there be dragons that way too.

If you plan to distribute an actual electronic product under a real business name, you must first pay these trolls thousands of dollars.

It's probably a rounding error for companies who make billions of things, but for small market goods?

USB-IF says, "Get rekt nerd! Gimme yer lunch money!

I upgraded all my old microusb and non-conforming usb-c devices to working usb-c. I think in every case of a non-conforming usb-c device, the connector itself only had V, ground, d+ and d- (and sometimes not even the last two). So it's kind of understandable how a junior engineer, when asked to update a design from microusb to usb-c finds a connector without CC pins and puts it in the design without thinking twice.

This also means the easiest DIY fix is the same as upgrading a device from microusb. Buy a connector with a small breakout board that has the correct resistors like this one[1]. Desolder the old connector, and solder the new one onto any convenient ground pads. Using the linked connector, it needs to be soldered upside-down but you can find a few different styles on aliexpress depending on what will fit in your device. You may need to insulate or scrape away some traces to avoid short circuits. Then solder small wires in place for V, ground, d+ and d-. There's always a nearby capacitor to solder to V, and nearby resistors to solder to d+ and d-.

[1]https://www.tindie.com/products/casualcoders/c-usb-type-c-br...

Meh, this problem is super, super common. I have many random electric devices that charge using USB-C that don't have the pulldown on CC.

Maybe I should get/make a shim for this.. a USB-C female-to-male thing that just passes through everything but has a pulldown on CC.

Wouldn’t it be dangerous to plug this into a device that does not expect to be powered?

No moreso than using a USB-A to USB-C cable, unless I fundamentally misunderstand something

Doesn’t USB-C deliver more volts and amps?

Not if you’re using the resistors to demand power. It will only provide 5v, the same as USB-A. It’s sort of a “backwards-compatibility” mode.

Right. You need USB Power Delivery, which requires the chip mentioned in the article, to get real power.

My USB-C to 240v dryer plug will work great at doing something.

I have a small kit I use when traveling. It includes a short c-c cable, an anker 45W charger (which is good enough for an m1 MacBook), a c-lightning adapter, a c-a adapter and an a-c adapter. Using both the c-a and a-c adapters in series works.

For the curious, there's a hierarchy defined by the USB-C cable spec.

Table 4-17 Precedence of power source usage https://www.usb.org/sites/default/files/USB%20Type-C%20Spec%...

- Baseline: it behaves like a USB cable from 1996. Sync gets 100mA@5V (USB2) or 150mA@5V (USB3), and then as part of USB enumeration you can get up to 500mA@5V (USB2) or 900/1500mA@5V (USB3 single/dual lane) depending on what happens during USB enumeration.

Then, in priority order:

- USB PD: If both sides negotiate a USB PD contract, that overrides baseline, and you get up to 5A@20V (or more now with the new EPR stuff)

- USB Type-C current: The source drives a voltage on USB-C CC (pin only in USB-C cables) to say if it can give 1.5A@5V or 3A@5V, the sync pulls CC to say if it wants it (what the author is talking about here). If both sides have the right signaling, the source gives the current requested to the sync.

- USB BC 1.2: Intended for charging bricks; brick shorts the USB2 D+/D- together to signal device it gets up to 1.5A@5V. Or 2.4A@5V if you use Apple's extension (see, every iPad brick back in the day)

So, wonder if the USB-C to USB-A case in the article is just working because it's hooking up to a USB BC brick with that USB-C to USB-A cable, and the remote needs more than 100mA to charge and only supports the USB BC case?

Note only baseline is required to be compliant with the spec; there's no rule that the device has to use any of the other features.

Welcome to USB :)

Also worth noting this is super common on devices designed to have a micro USB-B port, which then are later switched over to USB-C; they just hook up the pins from USB-B and don't bother with the new pins from USB-C.

The spec is specifically designed to allow this, because USB has backwards compatibility as a core tenant (i.e. you can plug in your USB keyboard from 1996 and it will probably still work). Also a USB-C to micro USB-B cable or USB-C to USB-A cable is explicitly allowed in the spec, and how could such a thing possibly work if the spec somehow required using the new CC pins instead of making them optional, since those pins are not in USB-A/USB-B?

Still, not the nicest experience for users :(

Minor nits: _sink_ instead of sync in this sense, and _tenet_ instead of tenant.

In my experience and understanding, those are voltages usb-c must support. Section 4.5 explains how the voltage is zero until a correct resistance is detected, then it can be negotiated up from there. Usb-c to usb-a cables always contain the right resistors to work at up to 5V, 3A. In fact if you go on aliexpress and buy male connectors to make a usb-c cable they will almost always have the resistors needed to make a usb a-c cable.

Chinese products don't have to conform... they just have to be cheap.

They’re not even cheap. Exway boards can go for upwards of $2000.

I have cheap products that work perfectly. I have expensive ones that don’t.

More expensive things are usually right, but it is by no means a rule.

Since it's somewhat related to the topic, does anyone have a recommendation for where to buy USB 3.0 or 3.1 is cords is that are relatively inexpensive? I've gotten burned by Amazon too many times to trust.

If I really really really need it to work I buy an Apple one.

Otherwise go to a real manufacturer like monoprice or a smaller company that specializes in cable. They will usually work well.

You can’t get good cables for aliamazon bargain prices unless you go used/open box or similar.

I’ll second monoprice.

I wanted some thin and various length hdmi cables, Amazon sold me garbage (no surprise these days). Went to monoprice and they were cheaper than Amazon and they worked. Not basement prices like some places... but I don't have a problem paying for reliability.

I'll third them. I don't buy a lot of cables, but when seemingly every USB-micro cable in the house crapped out simultaneously, I put in an order for a fistful and they were all good.

Monoprice is solid and their prices are fair. I think I got half a dozen cables shipped for about what one costs at e.g. CVS.

> a real manufacturer like monoprice

Monoprice does not manufacture anything. Also, a lot of the apple USB-C cables are power only.

They do actually design/spec the stuff they sell though, they're not just some random reseller slapping branding on whatever product is cheapest on AliBaba this week.

That’s the key. You can find perfectly capable cables on alibaba but it’s hard. Let monoprice do the work for you.

Amazon basics might work, but I trust monoprice more.

These are my go to cables

https://www.amazon.com/dp/B093YVRHMB

- USB C/USB A on one end

- USB C/Lightning/micro USB on the other end

- 10 GB/s data

- supports video over USB C (I have a portable external display that gets power and video over USB C)

- 100W charging

- USB 3.1

Okay, but that's Amazon, so they can just commingle inventory with a fake. Hence

> I've gotten burned by Amazon too many times to trust.

I’ve ordered the sand cord on three separate occasions with no issues. It’s easy to verify that they are real.

> It’s easy to verify that they are real.

Okay, I'll bite; how?

(In all sincerity - I don't believe you, but if there really is an easy solution, that would make my life better and would be appreciated)

Big minus: Max length is 5ft. That doesn't work for me.

Also, a lot of the reviews are really bad, everything from poor construction to not being flexible enough to the connectors not fitting.

5 feet is actually the red flag here.

It is highly unlikely they are selling a compliant 100W cable that is over 3 feet.

Is there a retail store near you that you trust to not carry junk? I never buy cables from Walmart because I got a junk 3.5mm male-to-male aux cord once, but places like BestBuy or Staples/OfficeMax are reliable.

Monoprice's own website

This will almost always be the case with cheap things with a Usb-C port, even from big brands like Sonicare (toothbrushes and water flosses).

If you don't see USB PD (Power Delivery) on it, chances are it won't work with your usual chargers. Even with PD on it, it can be hit or miss, but at least at an 80% success rate vs 0% without it.

Is there a royalty on the chipset to do it better than resistor pullups? The article implies its a relatively high BoM cost to have the negotiation. I can see its pinouts, board design, test but the actual chip.. surely is still down in the 0.0001c range? or is this one "pay the cartel" expensive?

The chip is very expensive, sadly. If it were just a matter of the cartel, that would be one thing (and would have Shenzhen-made workarounds). No, the USB lunatics made USB-PD one of the most insane specifications you'll ever find, so that no one can implement it correctly period. (Seriously, it's actually the description of the actual behavior of an old TI part, bugs and all, turned into a spec. It's horrid. It's the genuine worst specification ever.)

So you end up needing a 48MHz Cortex-M0 microcontroller just to do the god damned power delivery. At least, I've never seen it done by a less capable part. And processors of that class are, alas, just not that cheap.

As someone who has read the PD spec and agrees with your assessment- I’m curious what that chip is?

I don't, I'm sorry. I also don't think I ever tracked down that part myself (but who knows, it's been years), so it may not actually be true. But it's sure plausible!

The chip lets you draw more than the minimum USB-A standard power. And that’s what, 500mA?

Yes, thats it's role, but it does it "intelligently" where the resistor pullups do it the "dumb" way. His complaint is they cheaped out on even the resistor path.

I just wondered if some hypothesized USB-C consortium decided to "patent" how the signals work, and charge $ for compliance/conformance to the chipset for a logo badging and green tick.

It's more usual higher up the complexity food chain like MP4 decoding. Frauenhoffer wants its IPR respected. Philips made coin on conformance to the audio cassette form factor. Somebody made book with CD-ROM size, encoding. It's normal.

Without such a chip you can use resistors to configure up to 5V, 3A.

Yes it's annoying but still don't get what the fuss is.

You said it's because they don't care about the bigger picture. But the bigger picture is they are doing everything to cut cost and adding extra chips is not worth it. They aren't exactly hailed for their built quality.

The required resistors are literally a tenth of a penny each ($0.0011/ea) on DigiKey[0], probably cheaper elsewhere. You need two of them. Two tenths of a single penny. Exway is charging between $350 and $1650 for their skateboards, from what I can see on their website.

I highly doubt this was done for cost reasons, even including the additional assembly costs, which for surface mount resistors would be very low. This feels like a design error, and I doubt the support emails went anywhere meaningful.

[0]: https://www.digikey.com/en/products/filter/chip-resistor-sur...

materials costs are not the only consideration. they have to arrange for a secure supply, retooling, QC checking etc. to do it properly.

I imagine they got the cheapest engineer to make a functioning product. Changing their whole production line for a $0.001 component is hardly worth their time in this product segment, however debatable the ethics or PR of it is.

Again, they're selling products worth nearly $2000. I'm not talking about changing their production line. I'm talking about why this happened in the first place. You said it was to save money because "adding extra chips is not worth it". I highly doubt it. This appears to be a design error. A design error while saving money on engineering? Maybe, but that isn't what your first comment said.

C'mon, dude, we're talking about SMT resistors, which I'm sure are already used elsewhere in their design. If any electronics company in the world can't "secure" a supply of SMT resistors, we have much bigger problems.

I agree that doing even a small redesign of something existing isn't free, but every company comes out with new designs for new products or to make an existing product more cheaply. At the very least they could acknowledge the issue and put the fix on the queue for the next rev of the board.

My guess is that the support person didn't actually forward anything on to any engineering team. At most they sent it to their supervisor, who told them to just make OP go away.

That’s to change it. There would have been no additional cost to do it right at the start other than the resistors.

I don't think anyone is expecting them to release a new revision for just this change. They are just asking that the next revision includes this fix with the other changes that would justify a new revision anyway.

You apparently have no idea how electronics are manufactured.

There is no such thing as "changing the whole production line".

Also, the marginal value the additional components provide exceed their costs by such a wide margin. The ROI is easily in the 10000%.

The whole point of the post is that they didn't even have to add an extra chip -- just a single resistor. It's not even a cost-cutting behavior, it's just a simple mistake.

pedantic: they have to add 2 resistors, one on each CC line. The Raspberry Pi famously screwed this up and shared a resistor for both lines. https://www.theverge.com/2019/7/10/20688655/raspberry-pi-4-u...

I’m not sure it’s a mistake so much as a lack of caring.

It's a fraction of a penny, and super annoying if everything else you have supports USB-C

Totally unrelated to the substance of your post… enable word wrapping in those text boxes if possible.

Yup, I noticed it after publishing. I'll check with the blog theme and see if that is possible. Sorry!

Reminds me of issues I had with the Wahoo Elemnt Bolt bike computer:

For some weird reason it cannot charge if you plug the other end of the USB-C cord to a Apple made charger. It works with all other USB-C wall bricks.

Never understood what that was about

Apple follows the standards. It probably tries to negotiate power before sending it; the other bricks just apply voltage no matter what.

This is my understanding as well.

Pretty sure the apple chargers will only work if the cable supports the charger’s max watt output.

Things like this are the reason the EU mandate for USB C bothers me. No, it doesn’t allow universal charging if the company is still going to be USB C-C charging compatible.

I have this problem with cheap disposable vapes I get from corner stores. Elfbar did seem to add a resistor at some point, though, which is great :)

>“Just charge it with a USB-A to C cable! What’s the big deal here?”

The European Union would like to have a word with you....

This is a classic example of where "Shibboleet" would avoid problems.

We really need a separate certification agency for USB / HDMI / DP devices.

With teeth.

Originally USB-C was supposed to be convenient and easy to use. One port for many different purposes. Instead we have what we have now: a complete mess of a standard that makes it more confusing than USB ever was.

Given how complicated and inconsistent the USB-C spec is, I’m still surprised that Apple chose it even though it’s so terrible.

Apple is one of the key companies involved in the spec and guided it heavily. Why wouldn't they use?

Apple may be one of the few companies that actually follows it, too.

Because it’s terrible.

Apple helped define it!

I don't care either. They'll probably fix it at some point if they themselves get annoyed by this behavior. But TBH, very few people will be inconvenienced by this at the current time. I have exactly one C to C cable in my house, and a lot of A to B or C cables. So what if it violates the USB C spec? The Nintendo Switch violates the C spec too. Perhaps they should have made the spec a little more straightforward?

> I have exactly one C to C cable in my house, and a lot of A to B or C cables. So what if it violates the USB C spec? The Nintendo Switch violates the C spec too. Perhaps they should have made the spec a little more straightforward?

You may or may not be surprised, but other people have different mixes of tech. I'm past the halfway point in A to C, it's more troublesome now to find an open A port or cable than the other way around.

If you excuse bad behavior by companies, they'll gladly take advantage of that. Letting them blame the spec is foolish. It's not _that_ hard to get at least the basics right. Many do it fine.

My Switch charges just fine with USB-C to USB-C cables. These days all I buy are GaN chargers with 3 USB-C / 1 USB-A and they can charge my MacBook, my Lenovo laptop, my work Dell, my Galaxy phone, basically 9/10 of my electronics at their max charging speed with the same USB-C to USB-C cable. Which makes these cheap devices missing pulldown resistors that cost a penny or something extremely annoying.

The problem with Switch is really the charger. If a device pulls down the resistors it puts out 15v. It is dangerous to plug into other devices.

Can you get high wattage charging from A to C cables? I got my first 65 watt charging phone what four or five years ago? So I use only the C to C cables at all my chargers. Plugging random other devices into those chargers and not having them work would be an annoyance to me.

Reply on news.ycombinator.com