Motherboard USB 3.1 motherboard header

[jeshikat]

Jan 3 edit: News article posted here: https://smallformfactor.net/news/usb-3-1-motherboard-connector-revealed

And it has a pic of the cable connector.

Original: https://benchlife.info/msi-z270-motherboard-leak-before-announce-12212016/

Here's a good look at the USB 3.1 motherboard header from a MSI Z270 board.

 

[The Dude]

It's smaller than the old pin headers, which is nice, but it doesn't look like it has enough pins for two ports.

 

[jeshikat]

Assuming the connector is symmetric there should be 24 pins total (12 on each side). So it should be possible to run two USB 3.1 ports off this header.

 

[Ceros_X]

Don't really see a locking latch/mechanism?

Also, I may be a sucer but I like the all black boards with metal accents and the M.2 metal covers. Names are meh, but I like the looks. Couple with a Turemetal PSU and you're looking good

 

[jeshikat]

There are slots in the metal piece on the short edges of the connector so maybe there are locking tabs on the connector? I really hope so because there have been several machines we've sold at work where the USB 3.0 header wiggles loose and the front ports quit working.

 

[Phuncz]

Finally. It only took them a year or so after type C became available. But who cares about PCs anymore ? ¯\_(ツ)_/¯
Good find ! It seems like a decent quality and very tiny connector, although the plug is still a mystery. Yes front panel USB 3.0, I'm looking at you if that wasn't obvious.

 

[BirdofPrey]

Assuming the connector is symmetric there should be 24 pins total (12 on each side). So it should be possible to run two USB 3.1 ports off this header.

That would depend on how it's wired and what the controller would do. That's advertised as a USB-C connector and USB-C has 24 pin. The 4 TRX pairs can be remapped at will, but the D+ and D- pairs are all for one port to allow for a dumb controller to work with a cable plugged in both ways, so for 2 USB 3.1 A ports to work off one connector, the controller would have to be able to split the D+/- lines into two ports, and for 2 USB 3.1 C ports to work, the front panel itself would also have to be able to multiplex the 4 signal paths down to two each.

It's certainly possible, but it may not be worth the cost.
On the other hand if it's a 1 to 1 connection, that opens up the possibility of supporting thunderbolt or display-port on front-panel connectors. Now while DP on the front panel is kinda meh unless it's able to use the GPU (if they can then You might start seeing USB C DP VR headsets), but thunderbolt could be interesting.


It is nice to see a proper connector now, though. USB 3.0 pin headers are annoying by being fat yet still having tiny, fragile pins and no locking mechanism.

 

[jeshikat]

That's a good point, the boards we've seen so far that have this new connector seem to also have a USB 3.0 header. So they may just stick with a single 3.1 Type-C.

 

[The Dude]

Assuming the connector is symmetric there should be 24 pins total (12 on each side). So it should be possible to run two USB 3.1 ports off this header.

That was my first thought, but it looks kinda asymmetric in this picture.
And the other thing, judging by this only one out of 15 MSI boards have it.

 

[jeshikat]

Yeah, I haven't seen it on any of the ITX or mATX boards we've seen pictures of so far.

 

[iFreilicht]

Man that's an interesting connector to say the least. I searched around for pluggable connectors with 24 contacts on digikey, but nothing similar to this turned up, so it seems like it's a newly developed one for USB Type-C that can't be sourced yet. Pretty exciting.

I would assume that it's electrically identical to the normal Type-C, so one of these is probably only fuelling one external Type-C. Fine by me, makes the cable assemblies easier to deal with.

 

[EdZ]

I wonder why they designed an entirely new connector, rather than just using a vertical Type-C socket? It appears to be physically larger, and should be operating with fewer insertion cycles than an outside-facing Type C socket so does not need to be any more robust, so I can't imagine any reason for it to exist other than "it's marginally cheaper".

 

[iFreilicht]

My best assumption would be that this larger connector will be more stable when mounted vertically. There are exactly two vertically mounted Type-C sockets, and they are relatively long compared to their footprint. Add to that the length of a typical Type-C cable connector and you've got a lot of leverage on the PCB if the cable is being accidentally pulled during assembly.

 

[Phuncz]

The same connector huh ? That sounds too easy. I personally don't mind it only being one to one connector. 24 pins is a whole lot more than the 5 pins we had with USB 2.0, especially considering the charging spec. I'm hoping the connector is suitable for that or else I'll probably double, maybe triple face-palm.

 

[iFreilicht]

It looks to have a higher contact surface area, so I'd be very surprised if the supported maximum current is lower. That might actually be part of the reason why they chose a bigger connector, too.

 

[|||]

From the Guru3D.com article, it is just USB 3.1 Gen2.

Looking at the front plate, I'm wondering if it is USB-C + HDMI electrically (separate):

 

[iFreilicht]

Ah, that picture also shows another benefit over the previous front panel connector: Continuous shielding. And that's probably required for the additional speed Gen2 gives you.

 

[jeshikat]

News article posted here: https://smallformfactor.net/news/usb-3-1-motherboard-connector-revealed

And it has a pic of the cable connector.

You'll also get a kick out of the post @IntoxicatedPuma

 

[IntoxicatedPuma]

I read through and article is interesting, but maybe i missed the "kick" part >.<

 

[Ceros_X]

News article posted here: https://smallformfactor.net/news/usb-3-1-motherboard-connector-revealed

And it has a pic of the cable connector.

You'll also get a kick out of the post @IntoxicatedPuma

haha