76 comments

[ 3.9 ms ] story [ 131 ms ] thread
"A third benefit of Type-C is that it can deliver greater amounts of power -- up to 100 watts. That means you should be able to charge something as big as a laptop via a USB cable, as well as the phones and other small devices you can charge today."

How awesome would it be to not need a dedicated power port on your laptop? In my experience with repairing laptops, that's one of the first things to break if you're keeping the thing plugged in constantly. It gets old disassembling the entire machine just to J-B Weld or solder the female end back in place. Then there's the idea that you may eventually be able to charge your other internet devices with the same cord you charge your laptop with. You get a cord with every device, so that means you'll have backups ready if your main charge cord breaks or disappears. This just seems like an all-around Good Thing.

I'm a little worried that if these get made to the same standard that most usb cables are that one of these is going to kill someone.
That's 20V over 5amps that won't kill anything... Lenovo and some other companies already charge their laptops via USB (Normal A Type connector) on a dedicated port that connects to a 65W power adapter...
20V over 5A could definitely kill someone.
in order to pump 5A with only 20A you would first need to liquefy someones flesh
You should stop posting false information. A simple wound is enough. 5A @ 20V is 4 ohm. A wound is lower than that.

And you don't need 20A to kill anyway. 20V with sweaty salty skin is right at the edge of dangerous territory.

go ahead and stab yourself with a multimeter, your gonna read somewhere around 300-1000ohms
http://darwinawards.com/darwin/darwin1999-50.html

> (1999) A US Navy safety publication describes injuries incurred while doing don't's. One page described the fate of a sailor playing with a multimeter in an unauthorized manner. He was curious about the resistance level of the human body. He had a Simpson 260 multimeter, a small unit powered by a 9-volt battery. That may not seem powerful enough to be dangerous… but it can be deadly in the wrong hands.

> The sailor took a probe in each hand to measure his bodily resistance from thumb to thumb. But the probes had sharp tips, and in his excitement he pressed his thumbs hard enough against the probes to break the skin. Once the salty conducting fluid known as blood was available, the current from the multimeter travelled right across the sailor's heart, disrupting the electrical regulation of his heartbeat. He died before he could record his Ohms.

"It's the volts that jolt but the mills that kill" - voltage gives you the electric shock feeling but current across the heart will kill you.

It's a Darwin Award because you have to go out of your way to achieve it. When was the last time you plugged in a USB device while bleeding profusely from both hands and/or standing barefoot in a pool of your own blood?

Unless you are actively, visibly bleeding, it's going take well over 50V to cause any sort of lasting harm. Not to mention that it'll almost certainly default to 5V unless both ends are plugged in to devices supporting higher voltage.

He was not bleeding profusely from both hands nor standing in a pool of blood.

He stuck the probes into his thumbs.

This sub thread is not so much about whether it is probable that a person will die but whether it is possible that a person could die. And, with some convolutions, you can work out a situation where a person could indeed die.

Importantly rasz_pl is claiming that 50 mA cannot kill you. The rule of thumb is 50 mA at 50 V across the heart is enough to kill you. This voltage level is dropped if the skin is broken - and you don't need a big wound.

If you're saying it's fantastically unlikely to actually happen: I agree.

so many clueless people downvoting :/

>50 mA at 50 V across the heart is enough to kill you.

1 Voltage doesnt matter!

2 AC 60hz, not DC. I dont know the state of safety education in US, but in Europe it was the first thing on the first day of EE lab. Quick google:

"The threshold for ventricular fibrillation for direct current shocks longer than 2 seconds is 150 mA as compared with 50 mA for 60-Hz shocks; for shocks shorter than 0.2 seconds, the threshold is the same as that for 60-HZ shocks, that is, approximately 500 mA"

so >2 seconds at 45 volts if you jam two live wires at 45V into your open wounds across the heart

and here are some funny, but educational yt clips

https://www.youtube.com/watch?v=hp97GjuULX8

https://www.youtube.com/watch?v=XDf2nhfxVzg

The downvotes are because you are wrong.

You start this post saying "Voltage doesn't matter!" but end it with "so >2 seconds at 45 volts if you jam two live wires at 45V into your open wounds across the heart" - so which is it?

https://van.physics.illinois.edu/qa/listing.php?id=1335

https://www.physics.ohio-state.edu/~p616/safety/fatal_curren...

Fatal current is between 100 mA and 200 mA. Internal resistance of human body (and this doesn't require "open wounds") is between 300 ohms and 1k ohm.

Ohms law. Voltage doesnt matter because its a redundant information, only current matters.
You can actually kill with 20V - just stab the copper wire into the thumb of each hand.

20V is fine through skin - too much resistance, but if you push it under the skin the resistance is much lower and it can kill.

You might be able to do it with sweaty skin and squeeze the wire hard. Not certain.

Not likely to happen by accident though.

Also it's enough to cause a burn if you put it in your mouth.

no you cant, thats gonna be ~60mA of DC across your body provided you have open flesh wounds on both hands :)

painful yes, but not deadly

60 mA is a big deal, as it's getting into the current levels that are more likely to cause fibrillation (at least for AC). But the only way 20V will push that much current through your body is through open wounds. In a worst case scenario, yes, 20V could electrocute a person.

More likely, though, a crappy USB cable carrying 5A will kill people by burning their house down.

Some interesting points on "low" voltage and wounds here. I wonder about child proofing (and, as you say, fireproofing!). Will most people regard these as "100 watt dangerous" or "peripheral/signal-cable safe"?

Frankly, I'm worried enough about the possibility of fire due to power over ethernet (eg: hotel wifi).

All that said, modern, useful, standardised dc cabling does sound convenient.

You are wrong. 20v below the skin will kill.

And if this story is to be believed 9v has in fact killed: http://darwinawards.com/darwin/darwin1999-50.html

9v is not that far away from the usual 5v of a USB charger and some tablets charge with 2A.
They are very safe. Toyota uses them for cabling around compressed hydrogen fuel tanks.
That's very cool. Do you happen to have a picture of it?
It will be featured it the advertising photos. Frayed USB cables sparking next to compressed hydrogen. It will work a little like the subliminal death messages in the Marlborro cigarette ads, which were also very cool.
That doesn't mean it's safe for a person. For hydrogen you want to avoid sparks, for a person you want to avoid current (which is delivered by having enough voltage).
This would also be a great way to standardize DC power delivery throughout buildings:

http://www.economist.com/news/international/21588104-humble-...

This would allow for more efficient use of solar power and batteries. Instead of converting DC from the panel/battery to AC and then back to DC again (losing a lot of power in the process), you could run a lot of appliances directly off a DC power network with USB connectors.

Plugging your phone or laptop into a random usb slot into a wall to charge it isn't safe from a security standpoint. Since data flows with power, it is very easy to load up some malware while charging or steal data from the device. Unless I am misunderstood about what the article is proposing. It would be a nightmare if the only way I could charge my phone was to plug it into rouge usb ports. It is a security nightmare.

it isn't like this is some sort of far off concept.

http://www.nbcnews.com/id/44038933/ns/technology_and_science...

I haven't seen the specs of the Type C, but all of today's generation of USB cables can be modified to be 'power only' - Pins 1 and 4 of almost all USB cables are +5V and Gnd, respectively.

It wouldn't be a stretch to get a cable with pins 2 & 3 disabled or snipped if you're that security concious. :)

Right, you can buy them now. I just mean if it becomes required/acceptable to plug in all your devices whenever you see a usb port, we WILL start seeing malware transmitted that way. These things should be thought of ahead of time.
FWIW, Apple's universal connector doesn't have this security lockout feature either. The data could potentially be stolen via their solution as well.

Honestly, I agree- there needs to be a serious discussion about how physical security could be implemented with all connections of this type.

(comment deleted)
Isn't that already occurring with Type A connectors? I've seen them cropping up all over the place, including in public areas.
So you end up needing a special cable or a special port which just kills the notional of a universal port/cable.
Well, two cables - one for 'just power' and one for power+data.

I'd be fine carrying two small USB cables around.

The average user isn't going to know the difference.
One cable with a small switch would be sufficient.
Perhaps we need to reconsider the trustworthiness of unknown devices connected via USB. A simple "xxx device connected—allow?" prompt could suffice.
https://lockedusb.com/

There are now adapters available for USB which will not allow a data connection. I'm sure it won't be long before these are made for Type C as well.

It appears that the next MacBook Air will feature power charging via a USB Type-C port (the only other port on the laptop is a 3.5mm audio jack).

http://9to5mac.com/2015/01/06/macbook-air-12-inch-redesign/?...

Omigod is anyone else freaked out that on the pictures the power button is where the ESC key should be? My boss has a habit of giving us Apple hardware on which i have a habit of installing Linux or whatever, but that keyboard would require some seriously ugly scripting to avoid reprogramming my muscle memory :/
Power button definitely shouldn't be mixed into the keyboard with the other keys where its easy to hit on accident. On the other hand it probably doesn't do anything unless you hold it down.
If it's like the predecessors, it'll bring up the "Shutdown/Restart/Logout?" dialog on OS X, and fire the ACPI shutdown event on Linux. Of course in Linux this can be remapped (i use it to tell my laptop to suspend), but indeed, top-left is horribly stupid. Top-right i've never banged it by accident, though.
On OS X (recent versions) you need to hold it down to do anything. A quick tap as described does nothing.
I just got a Windows laptop where the power button doesn't do anything unless you hold it. Been throwing me off.
I wouldn't give any credence to anonymous hearsay.

If those mockups end up looking like the real thing, I'll eat my iHat.

If only they end up being more solid than current mini and micro USB female ends. It's ridiculously easy to end up bending the female port sufficiently out of shape to get poor connection with the smaller USB ports - I have several devices that will only charge with specific cables that fits slightly snugger than average. There's also an ongoing problem with cables where the male end is not deep enough / set far enough back, so again many devices won't work with all cables.

Looking at pictures here [1], it looks like the housing may be just as liable to poor/cheap implementations to end up with the same issues as micro-USB.

[1] http://arstechnica.com/gadgets/2014/08/small-reversible-usb-...

I've had the same problem.

Half the things my kids got for Christmas charge off Micro-B cables and almost all the cables and/or ports are damaged already. Yes, they're cheap chinese knockoffs, but even if they were more robost I don't hold out much hope for Mr Busy Travelling Businessman relying on this cable to run his entire laptop system.

I've found Micro USB connectors to be horrible in comparison to Mini connectors. I never once had a Mini cable become useless while I regularly end up tossing Micro cables. And if you are rough on your devices, like my wife, charging become an exercise in frustration where you have to move the cable juuuuust so to get a charge.
Every single microusb cable I ever had ended up like that... what am I doing wrong?
From my experience, if you regularly put any pressure in any direction you damage the cable or, even worse, the port.
I wonder if the increased power capacity is implemented by the device and power supply negotiating a higher voltage. 100w at 5 volts would be 20 amps, which seems a bit high.
About time. The usb physical standards have sucked, and hard. Mini usb tended to break on the device side, usb3 micro usb foot print is obscene etc.
Oddly, I never once had a failure of a Mini cable or receptacle. Micro is a horrid design though.
In connection with this article, according to 9to5mac the next MBA should have a unique USB Type-C connector.

http://9to5mac.com/2015/01/06/macbook-air-12-inch-redesign

The spelling mistake with "connectors" (plural) confused me. After looking at the linked web page, I now get what you wanted to say:

The 12" MacBook Air is rumored to have ONE SINGLE USB PORT and NO other peripheral port (except a headphone jack). 9to5 claims it will be a USB Type-C port. Take with a grain of salt.

The great thing about releasing a computer with 1 port is that when you release the next version, you can trumpet how it now has more than 1 port.

I wonder if something got lost in translation there, "only usb 3" is an interesting thing to do, and "only usb 3c ports" isn't that far off from "only a usb 3c connector".

Another benefit that this doesn't mention is that it can support DisplayPort connections at the same time as USB and charging.
According to Wikipedia, alt modes include DisplayPort, MHL, PCI Express, and Ethernet. I really hope USB doesn't make the terrible mistake of exposing an insecure protocol like PCIe over a "public" port.

With FireWire, I could just destroy the port physically since I didn't use it. But if USB gets the Apple treatment of giving DMA access to anyone that can get a peripheral connected to your machine -- well, that'd suck.

After the recent publishing of how Apple Macbook Pros with Thunderbolt ports can be bootkitted ... Yeah, knowing that USB 3.1/C can provide PCIeX/DMA access to the system is a touch worrying.
(comment deleted)
About time. USB + DisplayPort + Ethernet + power (- PCI Express) is wonderful.
USB + DP + Ethernet + USB-PD?

I haven't seen that mentioned. Do you have a link?

Do you just mean that you could have an attached USB <-> Ethernet adapter while DP alt mode is active?

I've read that DP mode is possible in conjunction with USB-PD and USB 3.1 but have seen nothing about Ethernet in particular.

I'm incredibly excited for the possibilities of this connector. Imagine a laptop that has a bunch of USB Type-C ports—and nothing else. Plugging a (universal!) charger into any of them charges the laptop at a rate of up to 100 watts. Displays are connected via DisplayPort-over-USB, etc. In fact, the display could actually be the charger as well (or, if you have a desktop, the display could be powered from the PC). A single USB Type-C cable could run an all-encompassing, manufacturer-agnostic docking solution.

Phones would also feature the same connector, so a laptop charger could also charge a phone, with no adapter needed.

I'm really looking forward to having a single connector for virtually all of my devices (excluding special situations, such as water-resistant devices) within the next few years.

Unless they changed the protocol, USB will always be undesirable for transporting video and network data. Not that you can't, but thunderbolt is much better suited.
MiniUSB was far better than MicroUSB. I can't count how many MicroUSB devices I've somehow broke (maybe it's just me). Always the tongue that seems to snap.

Thankfully wireless charging is now becoming more and more popular.

USB Type-C still has a tongue whereas Apple's new connector port does not; and Apple's is still reversible. Shame they patented it.

(comment deleted)
I'm with you here. I hate micro and loved Mini. I've bought and tossed more Micro cables than I can count. And if you are rough on the device you widen the port just a fraction and get crappy connections that you have to wiggle to get working.

Also, the Apple connector style is far nicer and it is indeed a shame it's patented and limited to Apple devices.

Having designed some analog electronics, it continues to amaze me that cheap USB cables costing on the order of pennies are capable of carrying GHz-rate signals over a couple conductors, despite not being coaxial. There are so many opportunities for loss and noise, it is incredible it works at all. And it works well. Imagine if we had to use a double(or triple)-shielded coaxial cable with BNC connectors to connect our monitors and high data rate devices. The USB spec is amazing. USB-C finally brings decent usability to the connectors.
I haven't been able to find anything saying how many wires are in the cable, does anyone here know and have a reference?

Is it the full 24 wires mapped end-to-end or are some of the connectors mapped to a single wire? Like, the 4 USB 2.0 connectors, I imagine it makes little sense to have 4 physical wires there, right? Yet they obviously have the SuperSpeed lanes wired separately because they can run DP alt mode on one side of the connector and USB 3.1 on the other.

I'd love to see something discussing this and it could affect the longer term viability of the physical connector spec.

EDIT: Honestly, I'd love it if the spec said they must have 24 wires AND that balanced pairs would be twisted pairs. As long as the connector is durable I'm willing to pay more for a good cable designed to that spec.

(comment deleted)
From the official spec the max wires is 18 + shield.

Also, from all my reading, USB 3.1 + DP Alt run at the same time. USB 3.1 only needs 2 of the 4 lanes.

http://www.usb.org/developers/docs/

Reading the spec a little more, during Alt mode negotiation TX1n/TX1p, TX2n/TX2p, RX1n/RX1p, RX2n/RX2p, and SBU1/SBU2 are all available for reconfiguration. That's 4 balanced pairs + 2 wires.

Additionally, if the connection is a direct-connect, like a docking station, then 3 additional pins become available: B5, B6, and B7.

So, a dedicated docking station could have Power + USB 2.0 + 13 pins (4 balanced pairs + 5 extra) or Power + USB 2.0 + USB 3.1 + 9 pins (2 balanced pairs + 5 extra)