Am I reading this right that by shorting the up and down D+ (and D-) line as one is supposed to by the spec, one essentially creates a HF stub?
Or doesn't the stub matter for signal integrity / EMI because it's only a couple of mm and fully enclosed by the shield of the plug and the receptacle?
I think it just means that there's only actually a single differential pair for USB 2 so connecting the two "D+"s from each side of the connector, and the same for the "D-"s, is recommended.
This doesn't work for the super-speed differential pairs because they're actually different lanes.
I feel like USB-C is cursed enough on its own. The sheer number of times I’ve had to explain to family members why just because the two USB-C ports on a laptop look the same, doesn’t mean they work the same.
This is true of electrical power outlets (15A/20A), HDMI, schrader valves, RJ45 (this one especially!), and many other things in life, and I don't think it's very fair to pick on USB-C in this manner when it does so many other things much better than previous alternatives.
But USB was supposed to be the unifying connector.
Before: "Here in the back of the computer we have PS/2 ports (1 for keyboard and 1 for mouse), a parallel port, serial, 25-pin serial, game port".
Introducing USB: "Everything is USB now! Plug anything anywhere!"
USB version whatever: "Well the sockets look all the same, but you can only plug your display in here, to charge it you have to connect it here", etc, etc.
But the point is that USB was introduced more or less on the promise of not being like that. When USB 1.1 came in the late 90s, it was pretty nice. Lots of different ports and cables went away, and all were replaced by standardized USB ports and cables.
You could geek out and learn the difference of upstream (A, host) and downstream (B, device) ports, the various types of B port sizes, and that was it. Later USB 2.0 was introduced and just (bam!) made it all 40 times faster, but it still logically worked pretty much the same.
Then USB 3.x came and just blew all that to h-ll, which is frustrating for both old bearded geeks, and* new/non-technical users.
For consumers, wasn't USB 3.0 just USB 2.0 but a few times faster? USB-C is what is confusing to some people. Replacing micro USB was straightforward enough, but replacing USB-A is a rough process.
Except that my desk has three USB cords, each with a color-coded label so that I can tell what each one is for. One is seemingly the only one that will charge my e-reader, but only does low speed data transmission so it never really gets used for anything but my e-reader. Another is power only and therefore useful for charging devices without my computer attempting to talk to them. And a third is a proper, modern (ish, I suppose), well-behaved USB 2.0 cable.
I am reminded of Einsteins quote about making things as simple as possible but no simpler. I feel that USB-C is an attempt to make things just a bit simpler than possible.
With USB-A and -B, there is a physical indication of the direction of flow (data and/or current). Just "connecting" two things is sometimes unambiguous, but sometimes it's not.
From what I remember reading about the spec, USB-C feels like an attempt to make things simpler for users (one plug for everything!) and manufacturers (here's all the things you can do, but they're all optional!) simultaneously, which of course just makes things more complicated for everyone.
Where I live (which uses the NBR 14136 standard), the 20A plugs and sockets look identical to the 10A plugs and sockets, the pins and corresponding holes are only slightly larger (so that a 10A plug fits on a 20A socket, but not the reverse).
> This is true of electrical power outlets (15A/20A)
You mean that one can hold 15A and the other - 20A? If so, they do work the same, they just have different capacities (like a tap, some of them will provide more water, others less but they work the same (especially if they look the same))
There are actually different connectors for 15 and 20 amp devices in the US under NEMA.
You do see NEMA 5-15R receptacles on 20A circuits in residential applications, but that's probably because most appliances don't require 20 amps. A 5-15 connector has two vertical prongs, a 5-20 has a horizontal and a vertical, with the special case of receptacles with a T slot that fits both 5-15 and 5-20 plugs.
It goes a bit deeper than that though: A NEMA 6-20 also has a horizontal and vertical prong, but they're reversed from a 5-20 so you can't plug them in in an incompatible (and likely hazardous) way.
The whole standard is really well thought out. I went down the rabbit hole when I got an old table saw with a 6-20 plug and had to figure out what the hell to plug it in to (dryer outlet, via a "custom" extension cord).
I feel like I haven't seen too many uses of Schrader valves that are likely to cause havoc. Most of the use cases for them besides tires are pretty obviously special and distinct from one another. I mean, I'd hope nobody's ever hooked a bike pump up to their air conditioner or a can of R134a to a tire...
> You do see NEMA 5-15R receptacles on 20A circuits in residential applications
This is, of course, what I was talking about. I do appreciate the nuances of the 15/20 T-socket and the 6-20/5-20 symmetry; NEMA is super cool (aside from the fact that bladed plugs that aren't twist-lock are sort of lame/less safe in general, at least compared to the UK standard bladed plug). It's a fun rabbit hole to venture down: my occasion was having to adapt straight-blade wall box receptacles that the electrician installed to the twist-lock of the PDUs we had bought.
There's also a fun and confusing mess of competing DC connector standards. I haven't decided what 12/48VDC connectors I'm going to put into my new house build yet (beyond the obviously mandatory 5V USB).
That's never bothered me, but I guess I see your point. I always figured that the point of having a 20A circuit with 15A receptacles was so that you can plug in and run two 10A devices without popping the breaker.
Stupid question: are there any common household appliances that draw 20A? I've only ever seen the T receptacles in non-household uses.
Window air conditioners, as a sibling dead comment pointed out. You'll commonly see such an outlet below a windowsill in some
apartment buildings for this reason.
safety nit: By using a dryer circuit (30A), you've got the wrong overcurrent protection on the saw. Which means that some failure modes could overload the saw, but not trip the breaker. The proper ways to do this is to run a new 20A 240V circuit, or to incorporate a subpanel with a 20A breaker into your extension cord. I don't think the risk is severe [0], but you should at least know your setup goes against the electrical code. And that the problem would be more pronounced if you did the same thing with a 40A/50A stove outlet.
FWIW if the nameplate of the motor specifies current draw at 120V, you can likely reconfigure the motor to use 120V instead of 240V.
[0] take a look at the gauge of a dryer's internal wiring, which is regulated by UL/CE and not NEC.
You raise a couple points I point I hadn't considered.
Rewiring the motor to 115v is probably a no-go, since it's a 3hp motor (continuous), which I believe is more amps at 120v than could be drawn from a normal circuit.
For safety, I'm wondering what kind of faults wouldn't be protected against. The starter has built-in thermal protection (heaters), which should protect the motor from the unlikely use case where I'm pushing the saw too hard, or more likely, a stall on start-up.
The breaker should still protect against shorts. I'm not knowledgeable enough to know what other faults could lead to an over current fault that one or the other of those wouldn't catch.
Would anybody care to educate me?
Also: the most dangerous aspect of a table saw is probably not the potential for electrical faults :-)
3HP is too much for a 120V/20A circuit for sure. Only magic shop vacuums can get >1.5HP out of a standard 120V branch circuit.
I'm having a hard time coming up with a compelling failure mode, given that thermal protector. Perhaps when the rotor locks, the thermal protector has a slower curve than the saw's wiring (which is assumed to be protected by the circuit breaker's curve). I would hope the saw wiring would be a bit oversized such that it too was protected by the motor thermal protection, but without doing an engineering analysis of the saw itself, you don't really know.
Unknown unknowns are my real point though - you're operating outside of the electrical code. Same as if you wired up a workbench of 5-15 receptacles with 10ga and put a 30A breaker on it - realistically it's going to be fine, but we avoid doing this. All those different NEMA plug sizes exist for a reason. 5-15/5-20 is really an anomaly where you can plug the smaller into the bigger, and even there I've got to wonder if eg old table lamps on a nice stiff 20A circuit are a good idea, or if they just haven't been focused on because they're statistically unimportant (see also: strings of Christmas lights with receptacles at the end).
FWIW I'd probably do the same thing you did, but at the cost of making sure I unplugged the saw whenever I was done using it until I made a proper permanent circuit. But that is not professional advice!
> just because the two USB-C ports on a laptop look the same, doesn’t mean they work the same.
This is already a solved problem[1] through color coding. Not sure why they couldn't use that scheme. Maybe combine a few to indicate things like DisplayPort and Thunderbolt capability, could have up to four quadrants for example.
> this is already a solved problem through color coding
Speaking from experience, I don't think most end users are liable to pay enough attention to their ports to notice the color coding. I remember family members being confused by the blue 3.0 ports when those were a thing--now there's also red ones, and different color C ones, and some C ports have the little lightning bolt and some don't, and most users don't know what THAT means.
The value-add of USB C was supposedly you didn't have to buy different cables for different purposes now, you could just slap everything on a C port. But of course that's not truly the case, and we're now in a messy soup of standards and capabilities that even to seasoned tech professionals are hard to disentangle, much less for end users.
Calling this a solved problem 'cuz they slapped some colors on some of the ports is missing the forest for the trees.
My wife's laptop has three USB-C ports and the laptop charges off all 3, but only one has the bolt. I strongly suspect it means Thunderbolt support, i.e. it's the only port that will work well with a dock.
I built a new desktop recently, and the case has a USB-C port on its top panel. I had to buy a $20 PCI-E card to have somewhere legit to plug the port into because my motherboard didn't support it. I had the option of trying to make it USB 3.1 or whatever, thunderbolt, or USB 3.0 or even 2.0 via a series of adapters that probably shouldn't exist but do on ali express.
I doubt I'll ever plug anything but a smart phone into it, but it seemed worth it for the feng shui.
This is already a solved problem[1] through color coding
You would think so, but no. Colors don't help.
Not just because a good chunk of the population has color vision issues. But people for some reason the average person can't latch onto that sort of thinking.
We went through this back in the 90's. Your PS/2 mouse plug was one color, and the PS/2 keyboard plug was another color. But people still tried to stick them in the wrong places.
Same with the sound card. Microphone, speakers, line in, line out, all had their own colors that matched the wires that came in the box. Regular people screwed that up all the time.
Going further back to the 70's and 80's, if you looked at the back of anyone's rack system, there was a 50% chance they had the red/white pairs of RCA plug backwards. Then when video started getting carried over the same plug, but colored yellow, things got worse.
Not when there are many ports out there without the right colours. Or with the wrong colours. Or arbitrary different colours because some manufacturer has a "bright" idea on their own. Or where it doesn't matter (were the keyb/mouse ports generally interchangeable? I certainly had two machines where they were despite the colour coding and I've seen single ports with "key/mouse" iconography since).
Maybe if the next standard has colours to start with, and they are unambiguous, and somehow not even ambiguous if confused with a previous port version, and they are adhered to by all manufactures, consistently, and nothing new arrives that someone adds a colour for before the standards body makes a decision...
Between USB2, USB2-Power, and USB3, I haven't got a laptop that I know what USB port does what in a long time... And I still didn't get any with USB-C.
To make it worse, there's a Thunderbolt port which is indicated by a little lightning-bolt near it, and a charging port which is indicated by a little lightning bolt near it, and neither of those has anything to do with the Lightning connector.
Some USB-C ports have DisplayPort output, so there's a DP near them. Some do Power Delivery, so there's a battery near them. And some are also a power input to the laptop, so there might be other random symbols near them.
It's chaos. I just juggle plugs until I get the expected behavior, and sometimes I never do because it's not implemented. I've given up.
I find it funny how people used to complain about the sheer number of times they had to explain to family members exactly the opposite: "No need to ask. If you can plug it in, then it will work. Computers are fool-proof like that. No, it's not like the jacks on the hi-fi"
As in, someone got a new mouse and didn't know where to plug it into their computer, when in fact there was only one DB-9 connector on the computer where it will physically fit and it would be the correct one.
I'm not even sure this was strictly true at one point. As soon as PS/2 keyboard vs mouse plugs, you had problems. Plus you had (and still have) various audio jacks on the computer anyway. Back to the mouse even prior to PS/2 you often had COM1 and COM2 and if I remember correctly you actually could not replug the mouse on the other port in a plug-and-plug way. You also could have very weird video cabling in the 3DFX 1 and 2 era.
But yes, I agree the USB-C plug mess is probably the worst. It has became worse than what it "wanted" to solve. The only way to make it work as broadly advertised again would be if all plugs supported all protocols (that make sense for a given device) and PD options, etc. on all devices and with common cables, at max bandwidth everywhere, but not only I'm not sure this is technically possible, this would probably be cost prohibitive.
And so what is the point? The net number of ports decreases on laptops, and there is no need on most gadget devices (beyond data+PD). The only point is for manufacturers to economize 50 cts and make the new inferior laptops pass as better than the old ones because they use the shinny new USB-C plugs. I would be happy if the price was actually 1/2 of the old models.
I think this was at its peak when basically every computer had only USB 2.0 ports. You pretty much could plug in anything to anything, so long as it fit. Not to say there were no issues -- USB 1.1 hubs, cable length limits -- but they were so incredibly minor compared to today.
USB 3 introduced blue plugs, complicating things slightly but still everything basically worked, and if something was operating slower than expected it was at least obvious by the color.
Then 3.1 and USB-C came out, and apparently USB-IF lots its collective mind. Now you have you pay attention to see if your ports, cables and devices are "USB 3.0" or "USB 3.1 gen 1" or "USB 3.2 Gen 1" or "Superspeed" or "Superspeed 5Gbps".. (Btw, trick question: all of those are the same thing).
I don't understand why the color coding plugs wasn't brought back for C. The 3.0 vs 2.0 thing was nice, and made it easy to tell what I needed to plug into. Now I have a phone with a 2.0, a laptop with a 3.1, and a macbook with a thunderbold, and my USB to HDMI port only works with one of the three.
Monochrome adapters (Hercules) indeed had a DB-9, but it was a different gender than serial (COM) ports. Monitor output had a female DB-9 and COM ports had a male DB-9.
DE-9 was used for MDA, CGA, EGA, and Hercules. Also some early VGA hardware used it, before DE-15 became sort of a de facto standard. (EGA had separate wires for red, green, and blue, they just carried digital signals. VGA used analog signals for red, green, and blue, so it was sensible to just map them to the same pins. Some monitors could accept either signal format and adapt to the timing of almost anything, the NEC MultiSync and Samsung Syncmaster for two.)
The Tektronics TDS-4xx oscilloscopes produced VGA-signals-on-EGA-pinout, for one example I've encountered in the wild.
DE9 connectors were also used for RS232C serial ports, which included mice and modems and all manner of other things speaking RS232. Wand-style barcode scanners were quite frequently on a DE9 connector, speaking some obscure protocol which may be at RS232C-compatible voltages but probably had positively Martian numbers of data/stop bits, etc. Or maybe just a raw photodiode signal so plugging into a serial port wouldn't do anything useful at all. Heaven help you.
Beyond that they were also used on Token Ring network cards (the hermaphroditic connector being to large to fit on a slot bracket), Fibre Channel in its copper variant (using the same pinout as Token ring, thank goodness they wouldn't blow each other up if mistaken), and pretty much anything else that only needed a few signals. It's a durable, ubiquitous, easy-to-work-with connector.
And of course in industry, the DE9 connector is still used for a zillion more things, like CAN. Pretty much every CAN test tool you'll find uses pins 2 and 7 for CANL and CANH respectively, with 3 and 5 as ground.
Pin 9 has quietly emerged as a sort of de facto phantom power supply, and if you stick 9-to-12-ish volts on it with a polyfuse, a lot of devices will run from that and not need batteries. Some RS232 GPS receivers would accept that, and the Toughbooks with serial ports still produce it.
Here's a funny: RS232 on DE9 is still produced today by the millions, on the backs of fancy TVs, because it's still the most sensible way to interface with high-end home automation systems. (USB never arrived at a good way to do peer-to-peer, and requires entirely too much software.) It's a 60-year-old standard and still has legs.
I feel like this is a temporary state of affairs that falls into the bucket of "regrettable but unavoidable". Over time I think the standard march of mass manufacture, refinement, miniaturization, and so on will result in every single USB-C plug on a general purpose device (like a notebook) doing the max of the standard. They'll all be Thunderbolt 4/USB 4 or whatever, and thus able to do everything. Cables production will also scale up and become cheaper, and at some point every single cable one would typically find will be capable of handling whatever. There aren't any fundamental scaling blockers, like some dependency on rare metals or something.
But it's still costly at the start of the refinement cycle, and it's hard to bootstrap without allowing manufacturers to do cheaper cut down feature versions to get the ball rolling. It needs to provide some marginal value in the mean time and real promise, even if it means some higher overhead in the establishment phase. So for someone buying a new device in 2025 or 2030, USB-C may still exist but I think issues with what ports do what will be lessened. We already see devices like Macbook Pros where all the USB-C ports are functional, and while those are on the higher end of pricing they aren't stratospheric enterprise territory either. I suspect it won't take that many hardware generations for that sort of thing to spread downward. The question is if that's worth the immediate trouble.
And at least for USB-C, I personally think it'll be worth it. It's a good connector, and the improved capabilities are valuable. We've had USB-A for a long, long time, and it doesn't seem unreasonable to guess that USB-C will stick around similarly. There was demand for something more compact and reversible with other QoL benefits along for the ride, and I think it's worth the upfront cost to get the whole industry onboard something that can converge rather then ending up with various manufacturers going their own routes.
There will always be some inconsistencies that consumers will need to be aware of. For example, Thundetbolt cables are limited in length, to about 0.5 m for passive cables and 1 m for active cables (and those are very expensive). Meanwhile, a bog-standard charging cable can easily be several metres long. I doubt every cable will become an active Thunderbolt cable; that's simply not realistic or economical.
I've had more issues with USB-C connections wearing out than other formats, even micro&mini-usb.
My phone barely charges at this point, which is frustrating because it otherwise performs fine. I'm going to have to buy a new phone because of the damn charging port!
I was considering how many times I've actually plugged it in. Worst case for me would be 3x/day. At 3 years, that's only ~3300 times. Surely they stress tested the connection more than that before production? Of course, these days I'm plugging it like 20x/day because it falls out or just doesn't work.
my phone had an issue with charging cables falling out. Turns out the port had compacted lint jammed in it. Turned the phone off and used a needle to clean it all out, phone retains cables now like the day it was new.
This. My phone reports "moisture detected" or oscillates between charging and not charging when there's crud inside the socket. I mechanically scrape it with the tip of a sharp pair of tweezers when it's lint, or I use the tweezers to swab with little pieces of paper towel soaked in isopropyl alcohol when it's something more nefarious.
I don't even bother turning my phone off, but it's rated as water resistant and so I am pretty confident that there aren't any voltages present when disconnected. Even on non-waterproof phones, though, any voltage persistently applied to pins on an external connector will result in corrosion over time, so it shouldn't be a problem to just mechanically scrape gently with a dry metal needle.
I believe to properly conform to the USB specification, the device must be tried one way, tried the other way, then tried the first way again before it plugs in properly.
I knew this was not going to be ideal in the late '90's when USB 1 came out.
Radio Shack was still going strong but this was the first PC connector that you could not purchase plugs or jacks for to solder your own. Also no suitable 4-conductor cable.
Even though the USB 1 approach was for a 92 foot maximum cable length (since it was only intended for the low-bandwidth webcams at the time) this was rapdily nonpublicized since USB 2 was already in planning, for much faster speed and a 25 foot max cable length, eventually shortened to 16 feet.
Well once you got suitable cable, you still had to cut pigtails off of USB cords so you could get the cable ends.
Though, in all seriousness, it would make an interesting security through obscurity feature. Everything works fine with the device, but only allows/shows certain features/files after Satan's rotation pattern. More secret squirrel than "security" in reality.
Now I want a USB-C thumb drive that has different data on different sides. Like the old 5 1/4" diskette drives where you flipped the disk for double capacity. Using it for retro-gaming would be fun. When the game says "Flip the Disk" you flip the USB drive.
It was to reduce BOM cost. That is such a driver that there were (/are still?) cables without any data lines at all and only power, just to save money.
I used them for plugging into public power supplies, but they were never sold to consumers labeled for that purpose.
Firstly, reversible plugs are harder to make structurally resilient, they only become feasible cost-wise as stronger materials become cheaper and manufacturing grows more efficient.
Secondly, reversible plugs are a lot more complex. Compare the 4(ish) pins of USB-A to the 24 pins on USB-C. Not all of them are strictly necessary for parity with USB-A, but you could cut that number in half quite easily if the plug were not reversible! Sure, this complexity impacts manufacturing, but, less obviously, it also pushes that increased complexity onto the USB controller hardware/firmware.
Neither of these issues were necessarily dealbreakers, but... for a long time they certainly did conspire to make reversible plugs into a pipedream. I have to imagine that were always bigger problems to solve and fewer resources to work with! We're just now seeing reversible plugs really take off because we've finally crossed that magic threshold where it makes sense to try surmounting those barriers.
Assuming the structural issues are solved, isn't it literally as easy as doubling each pin in a 180° fashion and shorting each pair? What additional controller complexity would this lead to (assuming termination issues are solved)?
I suppose there's no reason you couldn't do that. I imagine they didn't want to do that because cables (as a platonic ideal) should just carry signals from point A to point B without fudging anything. When you start to screw with that dynamic, you end up with lots of non-standard issues that people blame on the standard (I'm reminded of contractors jamming two 100mb/s lines through a single Cat5e cable instead of just switching on a single 1000mb/s line)
Not for high speed data (USB3.1 ++). The electrical length of the unused pins causes reflections in the signal. It's a consequence of the finite speed of light.
A standard USB 3.1 type A connector has 9 pins. The reversible type C has 3 extra pins (SBU, CC, VBUS), not counting the doubling caused by making it reversible. So while the reversibility doubled the number of pins, the increased complexity of the standard itself tripled it (from 4 to 12).
Good catch. I should have either compared the 9 pins of 3.1 or been more specific in saying 4 pins for USB 2.
Fun fact: USB 2 Mini/Micro connectors actually have 5 pins (hence my original 4-ish statement). The 5th pin doesn't actually go over the wire, it's just used as a signalling pin to the controller that tells it if the plugged in device is compatible with USB OTG.
> A standard USB 3.1 type A connector has 9 pins. The reversible type C has 3 extra pins (SBU, CC, VBUS)
Not precisely. The type C connector has four distinct USBSS pairs; the type A connector only has two. This allows the type C connector to be used to carry DisplayPort or other high-speed data alongside USB.
Not everything. The USBFS pair (D+/D-, at the center of the connector) is mirrored; the two D+ lines are connected together. This is not true of the four USBSS pairs -- TX1+/- and TX2+/- are separate, and can both be used at the same time for different purposes. The CC1/2 lines (which aren't mirrored either) are used as part of the process to negotiate how these pairs are used.
I meant it the other way around. The non-reversible type A does not have this extra pair regardless of standard. But the type C is reversible and short of having pins automatically reconfigure based on the detected position, the only straightforward way to achieve this reversibility is to double the number of pins. So one side of the type C connector perfectly mirrors the other (but flipped for reversibility) mechanically and electrically.
I'm fairly certain they are using double the number of USBSS because they're available due to the mirroring. It was the easiest way. Unlike the control channel or side band, you can actually benefit from using all the pairs for data transmission since they're already there.
This! Even just comparing USB-A 1.0 to what it replaced, it's hard to imagine a reversible plug like USB-C having been feasible/cost-effective to produce in the mid 90s. The PS/2 mouse and keyboard ports had pins of a similar size and spacing, but it was hard to orient the circular connectors. You could get the connector lined up and twist gently until the pins found their holes, but eventually you would bend a pin. I remember being very impressed at the way USB solved the issue of bent pins, but wondering why they didn't just key the connector like a Zip disk (which had beveled edges along the top of the disk, and wouldn't go into the drive upside-down).
Meanwhile we were connecting printers, scanners, and other peripherals using a 25-pin parallel (or sometimes SCSI) plug. Granted, the DB-25 was already pretty old at that point, but I have to imagine a 24-pin reversible USB plug would have been close to that size if it had been developed in the 90s.
This thread makes a convincing case for why the original USB connector couldn’t be reversible, but there’s a parallel question I’ve never seen addressed: Why couldn’t it have been asymmetrical?
The physical design of FireWire was one of its many advantages, in that you could instantly tell, visually and by touch, which side was up by the beveled corners on one side of the plug. I’m guessing the answer here was again cost, that manufacturing an interconnect with two 90-degree angles and four 45-degree angles was more expensive than one with four 90-degree angles, but I’ve never seen any confirmation of this.
> This thread makes a convincing case for why the original USB connector couldn’t be reversible
The sole reason as stated by Ajay Bhatt, the project lead at Intel, was cost. They wanted to encourage mass adoption of USB and every cent saved helped in that initiative.
The compromise for UX was to make the connector rectangular rather than circular, so that there was a 50% chance of getting it correctly orientated on first try.
FireWire 400 was a great connector! I resisted using FW for the longest time mostly because the cables were more expensive and less common than USB. Once most of my computers had it it started to make more sense for faster file transfers and external drives. They were so much easier to plug in without looking than USB, but the FW800 and mini connector went to a more rectangular shape and somewhat lost that advantage.
I never ended up getting a camera or anything else that supported FW, but USB always seemed to work well enough for my (relatively light) use case.
You imply that the USB-C could have only 12 cables.
Only the USB2.0 pins are duplicated, there is 4GND and 4PWR cables for power need, and the rest of the connector are not duplicated.
If it was not reversible, USB-C would be 22 connectors, not 12.
Plus USB-A is the only one with 4 cables, what was used on most phones was USB-B micro which has 5 cables.
What are you trying to argue here? All pins on the connector are rotationally symmetric. They may not be useless, because they still increase overall bandwidth, but they are indeed doubled. A non-reversible connector would not use this design and there would be fewer pins as a result (as much as half).
Also: USB-B micro connectors might have 5 pins, but the cable itself has 4 wires. The OTG pin is grounded in the connector. This is actually why I said "4(ish)" in my original post.
To me it's more baffling that a one-way plug would have a (roughly) symmetrical profile. Or even worse, the round connectors (like old PS/2 mouse/keyboard connectors).
A one-way plug would be fine if the profile were sharply asymmetrical, to the point where one could find the right orientation by touch alone. Maybe something like an even more pronounced trapezoid than hdmi.
You need to do a certain sequence of flips in order to unlock the secret partition. Without that, it presents a honeypot partition. Any attempt to request a file in the "SECRET DOCUMENTS" folder discharges 220V of capacitors across the data pins.
same for my Motorola. It will only charge one one side. The best part is that if plugged in on the "wrong side". It is great when it runs out of batteries and you have to gamble that you plugged it in correctly.
I often have to flip old Lightning cables if there is corrosion on one side of the plug contacts. Also, Lightning has a problem with dirt and lint accumulation, so the end of a large paperclip is useful to clean-out the port on the device occasionally.
I've tried that. Toothpicks break off and don't clean the gunk. Round cylinders are fine because the contacts are flat. There are even keychain metal lightning cleaners that are basically just large paperclip-like ends, and that's what many Apple store employees use.
Edit: as long as you're not using a needle-like piece of metal to dig at or bend the contacts, it should be fine.
I've actually used a device that used USB-C but still required you to plug it in a certain way around. It was a Bluetooth protocol analyser. There was an LED next to the socket that only lit up in one orientation.
USB, HDMI, Cable ethernet, DisplayPort, PCIe, ... they all use differential signals to transmit data, just different protocols. We could really and truly have just one type of plug to connect everything to everything a long time ago and without proprietary licenses. As much as I like what USB-C offers as much I hate it and other "stupid" standards that duplicate same function just with differnt plugs - transfering data.
Usb already offers couple of different data rates and power levels, so wonder why they use just one usb-c connector for all of it? ;)
Ok, sarcasm aside. Technicaly you dont need separate plugs for different connection levels and power transmission up to certain powers. Initial data exchange at some base speed and minimum power requirements can provide necesarry information for devices to establish or refuse connection
The variety of what is "supported" is also why it's taken years for USB-C to see any modicum of use outside mobile phone chargers that required the higher power and data rates. It's still not commonplace.
But it offers exactly things like allocating some lanes to DisplayPort.
The unfortunate thing is that Thunderbolt was created, and instead of a simple way to make PCIe external, we have this Intel-created mess that tunnels PCIe over some cursed "MPLS-like network" or whatever it is. And it was completely Intel-proprietary until recently. At least now the spec is open.
1) Long distance communications require much more powerful transmitters/receivers, which also makes it hard to design for high data rate. 10 Gbps ethernet is only just getting popular for consumer electronics, while 10 Gbps USB and video have been around for a very long time. It's not just market forces, it's inherent in the physics of the interface IC design. It also impacts power consumption; a 10 Gbps ethernet PHY takes a lot more power than a USB3.1 hub (by like, 10x per port).
2) Ethernet is galvanically isolated. This takes up a huge amount of space, both volume and PCB area, and makes it much more difficult to provide power. Galvanic isolation is important for robustness across variable electrical environments, but isn't that important for devices in the same room, and especially not on the same AC circuit.
3) The connector is optimized for cost and easy installation in the field, but it's not particularly durable or electrically robust (locking tab is delicate, can't be overmolded easily, not electrically shielded most of the time, huge transmission line stubs, etc.).
I can definitely envision a world where everything used maybe two standards, perhaps Ethernet for long distance and Thunderbolt for short, but there are lots of valid reasons for differentiation. There will always be tradeoffs. In the imaginary two-standard world, you wouldn't want to pay for a 20 Gbps Thunderbolt controller in your mouse.
Can confirm. Suzy Q cable. SparkFun sells one. Had to diagnose a Pixelbook that got bricked on an update that Google was refusing to warranty. Thought the cable was bad before someone in Discord suggested that I flip the plug. Still not sure how I feel about a reversible plug having a specific orientation.
We have a bunch of cursed USB-C cables that exhibit this behaviour. So the orientation matters, but only with some cables and only with some devices (that use 2.0)… good luck explaining that to office staff. Sigh.
When I worked help desk, I once found a USB mouse that instantly killed any computer it was plugged into. Instant power off.
I also found a laptop power supply that would cause the CPU of whatever machine it was plugged into to run at 100MHz. That was a fun one to troubleshoot.
I have a 7-in-1 USB-C dongle with this behavior. Some of its ports won't work in its natural orientation - logo on top, rubber feet on bottom - but will if I flip it. I remember being amazed and disgusted by this when I discovered it.
I have this problem in practice plugging my Pixel 5 phone into my car for Android Auto. The cable only works in one orientation. Android Auto is notorious for being very finicky about connections; no one's really figured it out but it seems to work better with shorter cables and of course higher quality cables. I've assumed my particular problem was some quirk of the connector fit, but now I'm wondering if it could be related to the asymmetry in the data pin design.
142 comments
[ 5.4 ms ] story [ 194 ms ] threadOr doesn't the stub matter for signal integrity / EMI because it's only a couple of mm and fully enclosed by the shield of the plug and the receptacle?
This doesn't work for the super-speed differential pairs because they're actually different lanes.
Before: "Here in the back of the computer we have PS/2 ports (1 for keyboard and 1 for mouse), a parallel port, serial, 25-pin serial, game port".
Introducing USB: "Everything is USB now! Plug anything anywhere!"
USB version whatever: "Well the sockets look all the same, but you can only plug your display in here, to charge it you have to connect it here", etc, etc.
You could geek out and learn the difference of upstream (A, host) and downstream (B, device) ports, the various types of B port sizes, and that was it. Later USB 2.0 was introduced and just (bam!) made it all 40 times faster, but it still logically worked pretty much the same.
Then USB 3.x came and just blew all that to h-ll, which is frustrating for both old bearded geeks, and* new/non-technical users.
Except that my desk has three USB cords, each with a color-coded label so that I can tell what each one is for. One is seemingly the only one that will charge my e-reader, but only does low speed data transmission so it never really gets used for anything but my e-reader. Another is power only and therefore useful for charging devices without my computer attempting to talk to them. And a third is a proper, modern (ish, I suppose), well-behaved USB 2.0 cable.
With USB-A and -B, there is a physical indication of the direction of flow (data and/or current). Just "connecting" two things is sometimes unambiguous, but sometimes it's not.
You mean that one can hold 15A and the other - 20A? If so, they do work the same, they just have different capacities (like a tap, some of them will provide more water, others less but they work the same (especially if they look the same))
You do see NEMA 5-15R receptacles on 20A circuits in residential applications, but that's probably because most appliances don't require 20 amps. A 5-15 connector has two vertical prongs, a 5-20 has a horizontal and a vertical, with the special case of receptacles with a T slot that fits both 5-15 and 5-20 plugs.
It goes a bit deeper than that though: A NEMA 6-20 also has a horizontal and vertical prong, but they're reversed from a 5-20 so you can't plug them in in an incompatible (and likely hazardous) way.
The whole standard is really well thought out. I went down the rabbit hole when I got an old table saw with a 6-20 plug and had to figure out what the hell to plug it in to (dryer outlet, via a "custom" extension cord).
I feel like I haven't seen too many uses of Schrader valves that are likely to cause havoc. Most of the use cases for them besides tires are pretty obviously special and distinct from one another. I mean, I'd hope nobody's ever hooked a bike pump up to their air conditioner or a can of R134a to a tire...
This is, of course, what I was talking about. I do appreciate the nuances of the 15/20 T-socket and the 6-20/5-20 symmetry; NEMA is super cool (aside from the fact that bladed plugs that aren't twist-lock are sort of lame/less safe in general, at least compared to the UK standard bladed plug). It's a fun rabbit hole to venture down: my occasion was having to adapt straight-blade wall box receptacles that the electrician installed to the twist-lock of the PDUs we had bought.
There's also a fun and confusing mess of competing DC connector standards. I haven't decided what 12/48VDC connectors I'm going to put into my new house build yet (beyond the obviously mandatory 5V USB).
Stupid question: are there any common household appliances that draw 20A? I've only ever seen the T receptacles in non-household uses.
FWIW if the nameplate of the motor specifies current draw at 120V, you can likely reconfigure the motor to use 120V instead of 240V.
[0] take a look at the gauge of a dryer's internal wiring, which is regulated by UL/CE and not NEC.
Rewiring the motor to 115v is probably a no-go, since it's a 3hp motor (continuous), which I believe is more amps at 120v than could be drawn from a normal circuit.
For safety, I'm wondering what kind of faults wouldn't be protected against. The starter has built-in thermal protection (heaters), which should protect the motor from the unlikely use case where I'm pushing the saw too hard, or more likely, a stall on start-up.
The breaker should still protect against shorts. I'm not knowledgeable enough to know what other faults could lead to an over current fault that one or the other of those wouldn't catch.
Would anybody care to educate me?
Also: the most dangerous aspect of a table saw is probably not the potential for electrical faults :-)
I'm having a hard time coming up with a compelling failure mode, given that thermal protector. Perhaps when the rotor locks, the thermal protector has a slower curve than the saw's wiring (which is assumed to be protected by the circuit breaker's curve). I would hope the saw wiring would be a bit oversized such that it too was protected by the motor thermal protection, but without doing an engineering analysis of the saw itself, you don't really know.
Unknown unknowns are my real point though - you're operating outside of the electrical code. Same as if you wired up a workbench of 5-15 receptacles with 10ga and put a 30A breaker on it - realistically it's going to be fine, but we avoid doing this. All those different NEMA plug sizes exist for a reason. 5-15/5-20 is really an anomaly where you can plug the smaller into the bigger, and even there I've got to wonder if eg old table lamps on a nice stiff 20A circuit are a good idea, or if they just haven't been focused on because they're statistically unimportant (see also: strings of Christmas lights with receptacles at the end).
FWIW I'd probably do the same thing you did, but at the cost of making sure I unplugged the saw whenever I was done using it until I made a proper permanent circuit. But that is not professional advice!
This is already a solved problem[1] through color coding. Not sure why they couldn't use that scheme. Maybe combine a few to indicate things like DisplayPort and Thunderbolt capability, could have up to four quadrants for example.
[1]: https://en.wikipedia.org/wiki/PC_System_Design_Guide#Color-c...
Speaking from experience, I don't think most end users are liable to pay enough attention to their ports to notice the color coding. I remember family members being confused by the blue 3.0 ports when those were a thing--now there's also red ones, and different color C ones, and some C ports have the little lightning bolt and some don't, and most users don't know what THAT means.
The value-add of USB C was supposedly you didn't have to buy different cables for different purposes now, you could just slap everything on a C port. But of course that's not truly the case, and we're now in a messy soup of standards and capabilities that even to seasoned tech professionals are hard to disentangle, much less for end users.
Calling this a solved problem 'cuz they slapped some colors on some of the ports is missing the forest for the trees.
I doubt I'll ever plug anything but a smart phone into it, but it seemed worth it for the feng shui.
You would think so, but no. Colors don't help.
Not just because a good chunk of the population has color vision issues. But people for some reason the average person can't latch onto that sort of thinking.
We went through this back in the 90's. Your PS/2 mouse plug was one color, and the PS/2 keyboard plug was another color. But people still tried to stick them in the wrong places.
Same with the sound card. Microphone, speakers, line in, line out, all had their own colors that matched the wires that came in the box. Regular people screwed that up all the time.
Going further back to the 70's and 80's, if you looked at the back of anyone's rack system, there was a 50% chance they had the red/white pairs of RCA plug backwards. Then when video started getting carried over the same plug, but colored yellow, things got worse.
Maybe if the next standard has colours to start with, and they are unambiguous, and somehow not even ambiguous if confused with a previous port version, and they are adhered to by all manufactures, consistently, and nothing new arrives that someone adds a colour for before the standards body makes a decision...
Some USB-C ports have DisplayPort output, so there's a DP near them. Some do Power Delivery, so there's a battery near them. And some are also a power input to the laptop, so there might be other random symbols near them.
It's chaos. I just juggle plugs until I get the expected behavior, and sometimes I never do because it's not implemented. I've given up.
As in, someone got a new mouse and didn't know where to plug it into their computer, when in fact there was only one DB-9 connector on the computer where it will physically fit and it would be the correct one.
But yes, I agree the USB-C plug mess is probably the worst. It has became worse than what it "wanted" to solve. The only way to make it work as broadly advertised again would be if all plugs supported all protocols (that make sense for a given device) and PD options, etc. on all devices and with common cables, at max bandwidth everywhere, but not only I'm not sure this is technically possible, this would probably be cost prohibitive.
And so what is the point? The net number of ports decreases on laptops, and there is no need on most gadget devices (beyond data+PD). The only point is for manufacturers to economize 50 cts and make the new inferior laptops pass as better than the old ones because they use the shinny new USB-C plugs. I would be happy if the price was actually 1/2 of the old models.
USB 3 introduced blue plugs, complicating things slightly but still everything basically worked, and if something was operating slower than expected it was at least obvious by the color.
Then 3.1 and USB-C came out, and apparently USB-IF lots its collective mind. Now you have you pay attention to see if your ports, cables and devices are "USB 3.0" or "USB 3.1 gen 1" or "USB 3.2 Gen 1" or "Superspeed" or "Superspeed 5Gbps".. (Btw, trick question: all of those are the same thing).
The Tektronics TDS-4xx oscilloscopes produced VGA-signals-on-EGA-pinout, for one example I've encountered in the wild.
DE9 connectors were also used for RS232C serial ports, which included mice and modems and all manner of other things speaking RS232. Wand-style barcode scanners were quite frequently on a DE9 connector, speaking some obscure protocol which may be at RS232C-compatible voltages but probably had positively Martian numbers of data/stop bits, etc. Or maybe just a raw photodiode signal so plugging into a serial port wouldn't do anything useful at all. Heaven help you.
Beyond that they were also used on Token Ring network cards (the hermaphroditic connector being to large to fit on a slot bracket), Fibre Channel in its copper variant (using the same pinout as Token ring, thank goodness they wouldn't blow each other up if mistaken), and pretty much anything else that only needed a few signals. It's a durable, ubiquitous, easy-to-work-with connector.
And of course in industry, the DE9 connector is still used for a zillion more things, like CAN. Pretty much every CAN test tool you'll find uses pins 2 and 7 for CANL and CANH respectively, with 3 and 5 as ground.
Pin 9 has quietly emerged as a sort of de facto phantom power supply, and if you stick 9-to-12-ish volts on it with a polyfuse, a lot of devices will run from that and not need batteries. Some RS232 GPS receivers would accept that, and the Toughbooks with serial ports still produce it.
Here's a funny: RS232 on DE9 is still produced today by the millions, on the backs of fancy TVs, because it's still the most sensible way to interface with high-end home automation systems. (USB never arrived at a good way to do peer-to-peer, and requires entirely too much software.) It's a 60-year-old standard and still has legs.
But it's still costly at the start of the refinement cycle, and it's hard to bootstrap without allowing manufacturers to do cheaper cut down feature versions to get the ball rolling. It needs to provide some marginal value in the mean time and real promise, even if it means some higher overhead in the establishment phase. So for someone buying a new device in 2025 or 2030, USB-C may still exist but I think issues with what ports do what will be lessened. We already see devices like Macbook Pros where all the USB-C ports are functional, and while those are on the higher end of pricing they aren't stratospheric enterprise territory either. I suspect it won't take that many hardware generations for that sort of thing to spread downward. The question is if that's worth the immediate trouble.
And at least for USB-C, I personally think it'll be worth it. It's a good connector, and the improved capabilities are valuable. We've had USB-A for a long, long time, and it doesn't seem unreasonable to guess that USB-C will stick around similarly. There was demand for something more compact and reversible with other QoL benefits along for the ride, and I think it's worth the upfront cost to get the whole industry onboard something that can converge rather then ending up with various manufacturers going their own routes.
My phone barely charges at this point, which is frustrating because it otherwise performs fine. I'm going to have to buy a new phone because of the damn charging port!
I was considering how many times I've actually plugged it in. Worst case for me would be 3x/day. At 3 years, that's only ~3300 times. Surely they stress tested the connection more than that before production? Of course, these days I'm plugging it like 20x/day because it falls out or just doesn't work.
I don't even bother turning my phone off, but it's rated as water resistant and so I am pretty confident that there aren't any voltages present when disconnected. Even on non-waterproof phones, though, any voltage persistently applied to pins on an external connector will result in corrosion over time, so it shouldn't be a problem to just mechanically scrape gently with a dry metal needle.
You and everyone else.
I knew this was not going to be ideal in the late '90's when USB 1 came out.
Radio Shack was still going strong but this was the first PC connector that you could not purchase plugs or jacks for to solder your own. Also no suitable 4-conductor cable.
Even though the USB 1 approach was for a 92 foot maximum cable length (since it was only intended for the low-bandwidth webcams at the time) this was rapdily nonpublicized since USB 2 was already in planning, for much faster speed and a 25 foot max cable length, eventually shortened to 16 feet.
Well once you got suitable cable, you still had to cut pigtails off of USB cords so you could get the cable ends.
Though, in all seriousness, it would make an interesting security through obscurity feature. Everything works fine with the device, but only allows/shows certain features/files after Satan's rotation pattern. More secret squirrel than "security" in reality.
[1] Like this: https://imgflip.com/memetemplate/153127632/Cross-fingers
I used them for plugging into public power supplies, but they were never sold to consumers labeled for that purpose.
Firstly, reversible plugs are harder to make structurally resilient, they only become feasible cost-wise as stronger materials become cheaper and manufacturing grows more efficient.
Secondly, reversible plugs are a lot more complex. Compare the 4(ish) pins of USB-A to the 24 pins on USB-C. Not all of them are strictly necessary for parity with USB-A, but you could cut that number in half quite easily if the plug were not reversible! Sure, this complexity impacts manufacturing, but, less obviously, it also pushes that increased complexity onto the USB controller hardware/firmware.
Neither of these issues were necessarily dealbreakers, but... for a long time they certainly did conspire to make reversible plugs into a pipedream. I have to imagine that were always bigger problems to solve and fewer resources to work with! We're just now seeing reversible plugs really take off because we've finally crossed that magic threshold where it makes sense to try surmounting those barriers.
Lots of cool applications using that technique.
Fun fact: USB 2 Mini/Micro connectors actually have 5 pins (hence my original 4-ish statement). The 5th pin doesn't actually go over the wire, it's just used as a signalling pin to the controller that tells it if the plugged in device is compatible with USB OTG.
Not precisely. The type C connector has four distinct USBSS pairs; the type A connector only has two. This allows the type C connector to be used to carry DisplayPort or other high-speed data alongside USB.
[0] https://upload.wikimedia.org/wikipedia/commons/thumb/0/07/US...
I'm fairly certain they are using double the number of USBSS because they're available due to the mirroring. It was the easiest way. Unlike the control channel or side band, you can actually benefit from using all the pairs for data transmission since they're already there.
Meanwhile we were connecting printers, scanners, and other peripherals using a 25-pin parallel (or sometimes SCSI) plug. Granted, the DB-25 was already pretty old at that point, but I have to imagine a 24-pin reversible USB plug would have been close to that size if it had been developed in the 90s.
The physical design of FireWire was one of its many advantages, in that you could instantly tell, visually and by touch, which side was up by the beveled corners on one side of the plug. I’m guessing the answer here was again cost, that manufacturing an interconnect with two 90-degree angles and four 45-degree angles was more expensive than one with four 90-degree angles, but I’ve never seen any confirmation of this.
The sole reason as stated by Ajay Bhatt, the project lead at Intel, was cost. They wanted to encourage mass adoption of USB and every cent saved helped in that initiative.
The compromise for UX was to make the connector rectangular rather than circular, so that there was a 50% chance of getting it correctly orientated on first try.
I never ended up getting a camera or anything else that supported FW, but USB always seemed to work well enough for my (relatively light) use case.
This is misleading.
If you want to only cover the USB-A spec with USB-C, you need only 8 pins.
I have a cheap USB C on my desk that got 8pins because it do only USB and charge.
USB-C can pass HDMI signal on the other pins, which your USB-A can't do.
> Not all of them are strictly necessary for parity with USB-A, but you could cut that number in half quite easily if the plug were not reversible
Is 4 not half of 8?
Also: USB-B micro connectors might have 5 pins, but the cable itself has 4 wires. The OTG pin is grounded in the connector. This is actually why I said "4(ish)" in my original post.
Thats why USB-C have more or less the same speeds than HDMI, while the HDMI have 22 connectors.
Compare what is comparable, you can't use your USB-A to display 4K-60Hz video, but you can with your USB-C, or HDMI.
A one-way plug would be fine if the profile were sharply asymmetrical, to the point where one could find the right orientation by touch alone. Maybe something like an even more pronounced trapezoid than hdmi.
One is Read Only Memory, the other is Write Only Memory (the holy grail)
They only do that in this combination. Not sure if it's consistent though - I should probably mark one side and check.
Edit: as long as you're not using a needle-like piece of metal to dig at or bend the contacts, it should be fine.
Now there are 4 combinations to try rather than two (flipping each end of the cable - 2x2 = 4)
Hm HTTP 451, or JS error?
We should have different connectors for different data rates and power levels. Don't pass the buck to designers, they'll screw up.
Ok, sarcasm aside. Technicaly you dont need separate plugs for different connection levels and power transmission up to certain powers. Initial data exchange at some base speed and minimum power requirements can provide necesarry information for devices to establish or refuse connection
And it's still only useful for low power devices.
The unfortunate thing is that Thunderbolt was created, and instead of a simple way to make PCIe external, we have this Intel-created mess that tunnels PCIe over some cursed "MPLS-like network" or whatever it is. And it was completely Intel-proprietary until recently. At least now the spec is open.
I've never gotten a real answer, one that involves technical reasons. It could be done, but it isn't.
I guess people like having different kinds of cables?
1) Long distance communications require much more powerful transmitters/receivers, which also makes it hard to design for high data rate. 10 Gbps ethernet is only just getting popular for consumer electronics, while 10 Gbps USB and video have been around for a very long time. It's not just market forces, it's inherent in the physics of the interface IC design. It also impacts power consumption; a 10 Gbps ethernet PHY takes a lot more power than a USB3.1 hub (by like, 10x per port).
2) Ethernet is galvanically isolated. This takes up a huge amount of space, both volume and PCB area, and makes it much more difficult to provide power. Galvanic isolation is important for robustness across variable electrical environments, but isn't that important for devices in the same room, and especially not on the same AC circuit.
3) The connector is optimized for cost and easy installation in the field, but it's not particularly durable or electrically robust (locking tab is delicate, can't be overmolded easily, not electrically shielded most of the time, huge transmission line stubs, etc.).
I can definitely envision a world where everything used maybe two standards, perhaps Ethernet for long distance and Thunderbolt for short, but there are lots of valid reasons for differentiation. There will always be tradeoffs. In the imaginary two-standard world, you wouldn't want to pay for a 20 Gbps Thunderbolt controller in your mouse.
https://chromium.googlesource.com/chromiumos/third_party/hdc...
(but that's with a special cable)
I also found a laptop power supply that would cause the CPU of whatever machine it was plugged into to run at 100MHz. That was a fun one to troubleshoot.
I have a Lumia 950 (one of the first USB-C phones) and a short USB-A to USB-C (or apple) cable/adapter.
One orientation does not charge the phone, so hes right it does happen.