USB 3.0 Promoter Group Announces USB Type-C Connector Ready for Production [pdf] | Hacker News Reader

USB 3.0 Promoter Group Announces USB Type-C Connector Ready for Production [pdf]

usb.org

70 pointsDeinos82 comments

I'm not an apple fanboy - but I do really like their magsafe & iPad mini connectors (whatever those are called officially). One thing I really like is the lack of "sticky-out-bit" in the middle of the connector, which all USB sockets seem to have, and which always feels kind of unstable or weak to me.

Is there a reason why all USB sockets - including these new diagrams - have a feeble looking centre part rather than it being part of the outer ring?

Came to ask the same thing. Those little pins always break for me.

   > Is there a reason why all USB sockets - including these 
   > new diagrams - have a feeble looking centre part rather
   > than it being part of the outer ring?

The outer part is the RF shield. That increases reliability by protecting the low voltage (but high frequency) USB data signals from coupling to outside interference (50/60 hz stray emf from power lines, switching power supply noise, etc). The magsafe connectors are just pulling power so don't really need that level of shielding.

But what about the Lightning connectors on iPads and iPhones? They don't have that outside shield but they carry high speed signals.

The lightning connector is pretty cool isn't it? Apple figured out you could build the RF shield around the signal lines in a single planar unit. If you look at the original 30 pin connector they did, they used a very similar scheme to USB the shield around the outside of the connector, the signal lines on the inside in a separate piece.

I actually think it was the lightning connector that got the USB group to go back and look at their connector designs. The interesting bit about the shield on the lightning is the fairly tricky way they create an island in the middle of the shield and thread the signal lines through it. Makes for a more expensive cable I expect.

The lightning connector is pretty cool isn't it?

No. It's not. I haven't plugged my phone in since I originally bought it and I don't understand thinking that a connector is cool.

For charging, every major mobile phone company except Apple supports Qi wireless charging. Source: http://www.qinside.biz/en/support/qi-enabled-phones

For connectivity, every major mobile phone company supports wireless connectivity via Bluetooth or WiFi.

I had an iPhone for 6 years (3g, 4, 4s) and saw all the competition fly ahead of Apple in terms of real features and screen size. My next phone may be an iPhone 6, but really Apple has a ton of catching up to do.

My phone has wireless charging; I used to use it, but I eventually switched back to wired charging.

Why? Two reasons, which I think are the biggest ones that explain why wireless hasn't completely taken over wired: efficiency and reliability. Especially the latter.

I used the wireless charging on my Nokia Lumia for two years, every night. Loved it. Problem is, it's dog slow compared to how fast my iPhone 5s charges. And when I tried to charge my Nexus 4 on the same charger as the Nokia (they use the same technology, why not?) it would go haywire and never actually charge.

Apple may be behind their competitors in some way, but they have a way of not pushing out technology until they're sure it's going to work the way the mass market expects it to (which is how the general public lets them get away with calling everything "magic". To a non-technical audience, it might as well be). My Nokia is full of cool features that don't work. If the battery dies, it takes half an hour before you can turn the phone on again. My Nexus 4 has a big screen but the camera is shit and it can't hold a charge through the day. My iPhone doesn't have a single "wow" feature to me, but then again there are no features that make me think "this would be so cool if only it worked". The last non-iPhone I didn't truly hate was my Palm Pre+.

Yet every phone still has a power/data cable... cool or not.

It seems like an RF shield on the male end is unnecessary. Apple's current iPhone connector does without it. I have no idea how they could see the lightning connector, then go and design a next gen standard that is across the board inferior. The Apple connector is sturdier, and has much more latitude for insertion angle. Surely Apple doesn't have a patent on a housing-less male plug.

My guess is the housing serves more as a mount point providing the friction and spring, rather than being necessary for shielding.

Are you so sure that it is "across the board inferior"? I don't really know about the technical details, but the lightning connector is designed for pretty specific uses. Can it carry 10Gbps and 100W like this connector can? Perhaps the housing contributes to that ability.

Maybe not, but it can withstand multiple connections daily for 2+ years without failing, unlike the MicroUSB connectors I've had on my phones. Also, the connector is very sturdy, so it doesn't break off, even on cheaper cables. Additionally, I never plug it in upside-down, because there is no such thing.

This is a good move for USB, and I personally can't wait until these become the new standard :)

I have my phone for 3 years, I charge it at least once daily and many times I use the cable to bring it to me, thus the phone ends up hanging from the micro USB connector. It still works perfectly, so I guess YMMV.

You have this backwards, the shield is there to prevent RF noise radiating _from_ the USB signal pairs, not to prevent outside noise coupling onto the signal pairs.

Hmm, while Lightning connector is rather nice, I don't like MagSafe 2 at all... it keeps falling out from the connection, especially if I use the laptop on a bed or similar surface.

The original MagSafe and the connector on my previous Dell M1330 were significantly more reliable while still providing enough protection.

MagSafe 2 works without issues for me, and is plenty strong. There should be no way that it is "falling" out!

The issue I've had with MagSafe2 is that it's very sensitive to vertical force: If there is any tension at all on the cable and I rotate my rMBP more than 15 degrees or so around the axis between my head and the screen (say, while using it on the couch), the connector disengages. Not so if it's rotated around the vertical axis. It's not clear to me that there's any safety or durability justification for this difference, only that they wanted a connector that was much thinner along one dimension than the other.

As an IT guy, I see hundreds of these connectors on a regular basis, so I've noticed something...if it's falling out too easily, odds are that the port has picked up a small piece of debris somehow, and the magnet isn't getting good contact. Take a close look at the plug and the port and brush them out a bit, and see if that helps.

>it keeps falling out from the connection, especially if I use the laptop on a bed or similar surface.

That's the whole idea. It better fall out, than put strain on the port or cable, like it would normally do. Plus, if you yank the cable while it's on a table, the laptop wont fall down with it.

Not sure who's downvoting people for saying they have trouble with the much weaker MagSafe 2 connector, but I'm with you. Same problems, and reviewers noted it when it debuted too.

>iPad mini connector (whatever those are called officially)

It is called the "Lightning" connector.

http://i.imgur.com/GSYJTI1.jpg

The Lightning connector just seems so fragile to me. I haven't broken mine yet, but I have certainly seen a lot of broken ones on displays in stores. I can't say I've ever broken a USB connector. That said, there is something to be said for a reversible connector that doesn't require 3 rotations to plug in :)

Fragile? It's a solid piece of metal. You really have to try, hard, to break it. It's about as durable as it's possible for something of that size and complexity to be.

By contrast, for instance, the design for the connectors for most USB "SuperSpeed" drives right now is incredibly awful. Fragile (both the port and connector designs), only plugs in one way, and it's quite hard to see what way that is without giving it a close look.

If you work in tech support you'll see USB ports mangled beyond recognition, people force-plugging them in backwards or worse, and a horrifying number of cables that have ends that have been pinched or bent.

The Lightning connector by comparison is a solid piece and would have to be smashed in half to fail.

I don't believe you can force-plug a USB connector backwards unless you use a hammer and a clamp.

Depends upon the design of the receptacle, but I've seen badly designed ones where the USB cord (on the USB A side to a PC/laptop/whatever) almost fits in the wrong way, and can easily imagine someone attempting to force it in from there.

You have no idea: http://www.reddit.com/r/techsupportgore

Fragile contrasted with USB cables, perhaps. I have used several counterfeit lightning cables from Shenzhen, some of which just failed after about a month to half a year, the reason of which I guess was mechanical stress. The genuine Apple ones are still pretty solid and work fine, though. There may be some pretty tough requirements in the process of production or the materials.

Comparing usb 3 to the lightning connector is very unfair, since the lightning connector doesn't go any faster than usb 2.0.

A much fairer comparison is to the thunderbolt connector. Which is non-reversible and also has a connector in the centre of the female.

The USB Power Delivery spec is pretty interesting too - up to 100W @ 20V over USB (presumably with modified cabling, too :P)

http://www.extremetech.com/computing/172113-usb-power-delive...

(a link from a link on this comments page)

Powering an external monitor from a laptop is appealing. One cord to the wall.

Even cooler, would be driving an external monitor from a smartphone.

> Even cooler, would be driving an external monitor from a smartphone.

I worry somewhat about the smartphone's battery in this context. I wonder if any have good 'passthrough' for when the battery is full and a wall socket is connected - even then I doubt it. :/

> Even cooler, would be driving an external monitor from a smartphone.

That will be expensive though. You'll have to pay the medical bills for getting your back fixed after having to carry the suitcase with the batteries :).

Current cell-phones have batteries around 2 Amp-hours. At 5 volts, that's 10 Watt-hours. The least-power hungry display according to [1] is a 15 inch LED; it clocks in at 15 Watts. This means you'd be able to use your external 15 inch display for 40 minutes before completely draining your battery. That is not counting anything the phone is actually doing, just powering the display.

[1] http://energyusecalculator.com/electricity_lcdleddisplay.htm

So with this new tech, it looks like you could charge the smartphone from the monitor while sending video from phone -> monitor? Sounds awesome.

It'd be cool with a laptop, but it'll require an extra 20 watts (typical for a ~22" screen) to be sized into the laptop's power brick. Since laptops have pretty predictable power characteristics, the manufacturers don't necessarily give you that much extra.

Would we upsize everybody's power cord on the off chance that 1% of people would use it? Or does "bigger power brick" become another configurable option, and users have to decide in advance if they'll want display powering capabilities? Neither one's a good option, in my opinion.

More likely we'd end up with a separate class of monitor, something limited in size and with a lower maximum power draw. At which point I'd rather just deal with the power cord and get a real screen for my desk. But frequent travelers who want an accessory screen would love it.

I think what will be more likely is the other direction. The monitor powers the laptop. Most laptops are well under a hundred watts and could easily be charged and powered that way.

Which is a great new other feature of USB Power Delivery: peripherals can provide bus power. This way you might have an external USB monitor (USB A/V you were announced 2011: where the _hell_ are you?!) that also provides power.

For a while it seemed like DisplayPort or DockPort might be the one cable to rule them all... USB PD seriously changes the footing of the game. Alas, USB A/V (encoded video to an output device) being completely absent is seriously damning USB's chance to be that one cable.

Experience dictates that about 95% of manufacturers will be too cheap to implement that bus power on their devices, because it'd cost them 50 cents more. So it won't matter much.

You've got it backwards... powering your laptop from the monitor is the killer app, meaning an end to proprietary laptop power supplies.

Well, you've got one out of two already. :) Apple laptops can be powered by Apple external monitors.

It'd be nice to have a more "universal" standard -- where I'm sitting at work right now I see one coworker with a ThinkPad plugged into an Apple monitor, and another with a MacBook Air plugged into a Samsung monitor, and that means both of them need one more outlet than I do. But that's a meeting of the brands I wouldn't hold my breath for.

> Well, you've got one out of two already. :) Apple laptops can be powered by Apple external monitors.

Well. Kinda. You still have to plug in two cables. This would allow to go down to one single cable powering the laptop and sending signals to the screen.

Though, honestly, I don't trust USB to be a good protocol for transmitting a high bandwidth and very lie latency signal with it's inherently polling architecture.

Isn't that the whole point of USB's isochronous transfer mode, where a peripheral schedules part of the total USB bandwidth budget for a regular transfer of data? USB 3.1's got similar bandwidth to Thunderbolt and HDMI 1.3, so that could work for many video applications, although not for uncompressed 4K/60Hz, although VESA Display Stream Compression could allow it to work.

No cord is the future. At least Intel want to get rid of all cables with skylake platform.

http://www.cnet.com/news/intel-aims-to-eliminate-all-pc-cabl...

100W @ 20V (5 Amps Max)- There are a few uses I would be interested in if it supported higher amperage, robotics is one.

> Reversible plug orientation and cable direction

That's all I needed to read.

Now you only have to plug it in one time instead of three!

http://cdn.thedroidguy.com/wp-content/uploads/2014/04/USB-pl...

Source article actually relevant to topic: http://thedroidguy.com/2014/04/next-generation-usb-type-c-ca...

Sounds great! For anyone not sure of the context, right now Type-A is the "host" end and type-B is the "device" end. However since OTG was introduced that correlation isn't concreted and can reverse during the connection. I might have misunderstood but it sounds like Type-C will be a symmetrical affair (the cable will be reversible). I guess this opens the door for totally new use cases for USB? With extensions to the OTG system we could be connecting two hosts for e.g. transferring files between laptops. Again I could be wrong about that.

Also 100W power is great - The power delivery capability of current USB is really limiting.

And reversible cable orientation? We are entering a glorious new era.

Where are the photos of the connector?

I think it's this http://9to5mac.files.wordpress.com/2014/04/usb-type-c-01.png

No, that's a mock-up from earlier in the year, made by Foxconn. (It's still fairly accurate, mind you - but the link above gives you a better idea of the socket/connector layout.)

A few diagrams here: http://www.extremetech.com/computing/187882-reversible-usb-t...

Some different connector configurations at the Ars article:

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

Um.. wouldn't the small piece in the computer's end break easily since it is so thin.. and small.

And how does it delivery 100W of power with cable that thin?

I'm guessing it does 100w only at 20v, which is 5A, which, with proper engineering, could be carried over 22-24 AWG wire according to https://www.eol.ucar.edu/rtf/facilities/isff/LOCAL_access_on...

For an example of how thick 24AWG wire is (22 is thicker), your standard CAT5E wire is 8 strands of 24AWG, so with the current-standard USB wiring of VCC (5v DC pos), Data (neg), Data (pos), GND (5v DC neg), it'd be ~half as thick as CatV (but would probably have different shielding). However, according to http://pinoutsguide.com/Slots/usb_3_0_connector_pinout.shtml , the powered USB 3.0 connector has 11 pins, which would be ~lamp cord thickness, or probably comparable to those ~10ft USB printer cables.

Consider how thick overhead power lines are and how much power they carry in watts, and how thick + long your vacuum cleaner cord is with how much that carries (a Dyson DC33 takes 1190 watts): http://www.cockeyed.com/science/power_use_database/dyson_vac...

Here is a good primer on wire capacity and how it relates to volts, amps, watts, and wire diameter: http://www.rowand.net/Shop/Tech/WireCapacityChart.htm

TL;DR totally possible with 20vdc to use a wire thinner than a BIC pen to deliver 100w ~6ft in addition to high-speed data.

They also said it supports the USB PD to up to 100W (@ 20V) which is enough current (5A) to support stuff like a Macbook Air.

Its enough to support any laptop I know of. Very vew exceed 100W.

Pretty much the only way to exceed 100W is to be running a high end gaming laptop, while actively playing games.

A desktop equivalent(ish) quad core hyperthreaded i7-4700MQ that turbos up to 3.4ghz only has a 47W TDP, which is a worst case number, so even heavy duty number crunching isn't enough.

Even exceeding 100W, you'll still be draining the battery much more slowly than when unplugged. At 125W the battery will last 5x longer.

It's hard to imagine a real world situation where 100W would be insufficient, where the user wouldn't be very much better off using a desktop machine for gaming.

I'm surprised they're not calling it ~"USB Smart" like Bluetooth: http://www.bluetooth.com/Pages/Bluetooth-Smart-Devices.aspx

The logo for "USB 3.1 SuperSpeed +" is very similar to the current "SuperSpeed USB" logo: http://low-powerdesign.com/donovansbrain/wp-content/uploads/...

EDIT: Here's the USB 3.1 logo (USB 3.0 logo above for comparison): http://upload.wikimedia.org/wikipedia/commons/4/4c/Usb-3.1-l...

Oh joy, they still use that damn diving board design on the device side. I've broken 3 devices when that diving board breaks.

Really? How aggressively are you plugging in your USB cables?

I have the same problem. One faulty or aggressive plug angle, and you break some pins or the inner paddle cracks.

It happens in any environment with a lot of vibration.

Seriously, I've only ever damaged a USB plug when forcing it to go in the wrong way, which this will prevent.

One step closer to that glorious day when the back of my desktop has like 20 USB sockets and nothing else.

This will be interesting, and its needed to keep up with the evolving use of USB. The downside is that once this is everywhere you will be keeping a drawer full of USB adapters (kind of like those USB/PS2 keyboard adapters that are everywhere, or DVI/VGA)

Just as Randal publishes this...http://xkcd.com/1406/

Doesn't look at all like lightning connector :)

Reply on news.ycombinator.com