Is it possible for desktop drives, even if not these ones in particular, to be powered by USB Type C ports as well as to transfer data through the same port at the same time?
USB type C ports are supposedly capable of providing up to 100w power transfer. So - since most full size desktop drives require about 1.5a during normal operation, but up to 3-4 amps during spinup. So - 50w on startup, well within the 100w max.
But the startup inrush current is for a very short duration, could you not resolve this with an ultracap, charge it over say 3 seconds, then spin up, never drawing more than about 10W.
Wouldn't it be more correct to say that "USB type C ports can NEGOTIATE up to 100W power"? As far as I can tell from looking around, all that is guaranteed available is the same 4.5W (900mA at 5V) that USB-3 promises. Beyond that you have to negotiate and many devices (eg your smartphone) are presumably not going to give you anything more.
Also 100W seems high. The highest I can see is the ability to supply 3A over 5V, which would give 15W. USB Power delivery promises 100W, but that's yet another spec that isn't (as far as I know) an essential part of the USB-C spec, and which I imagine will rarely be supported by non-power bricks. Even something like an iMac can hardly commit to supporting 100W on each of 4 USB ports; it just doesn't come with a 400W (plus enough to run the iMac) power supply.
USB Type-C can support UP TO 100W for power delivery. I was given to understand that the Porsche Design Desktop introduced at CES can do up to 15W only.
When can we expect Thunderbolt 3 external SSDs? PCIe Gen 3x4 Samsung M.2 950 Pro 1TB in an external case with Thunderbolt 3 Type-C port that can do >2Gbps.
The current bridge chips (M.2 to USB3.1 Gen 2) only uses the B-key for SATAIII mode, not the full PCIe Gen 3 x4 bandwidth (M-key). There would be no need for 2x mSATA in RAID0 or 2x M.2 in RAID0 to saturate the USB3.1 Gen 2 10Gbps bus, 1x M.2 M-Key 2x lane mode should be able to do that. Since Thunderbolt 3 (40Gbps) uses PCIe natively, it would be the best choice for external M.2 storage. USB3.1 Gen 2 could be the backup option on such a drive (albeit slower), as both uses Type-C connector. Current solutions use 1x thunderbolt + 1x USB3.0, new ones only need 1x Type-C. Where are the bridge chips?
Computers with Type-C USB 3 or 3.1 only guarantees 4.5 watts, and this is the USB 900ma @5V spec.
Computers with Type-C Thunderbolt 3 implementations guarantee 15W to either USB 3.1 or Thunderbolt connected devices.
Looks like the Porsche Desktop drives likely only guarantee limited power to a connected computer. Goal probably was to be able to connect to Apple's Macbook or the new Alpine Ridge/Skylake Windows laptops for charging while drive adds disk storage at the same time.
Perhaps what needs clarification is how much power the PD pass thru provides.
Note well, the PD power spec is 5v@3A for 15W, and 20v@3A for 60W. To get to 100W with the PD spec requires hipower Type-C cable that handles 20v@5A.
Dig into the USB 3.1 PD spec to understand why, and why it's going to be messy for awhile as the industry and end-ysers adapt.
Also, don't forget that those Type-C passive cables have current AND RF requirements to handle USB3.1 and Thunderbolt 3 reliably at 10Gbps and 20Gbps respectively.
I wish they would a build a raid 1 enclosure. Say 2 2tb drives. And if 1 fails, we can replace that hdd. If the enclosure fails, we can put the hdd in another one. Seems like the safest route. Is that possible/feasible?
We’ve updated our terms. By continuing to use the site and/or by logging into your account, you agree to the Site’s updated Terms of Use and Privacy Policy.
Facebook
Twitter
RSS
12 Comments
Back to Article
adityarjun - Tuesday, January 5, 2016 - link
Is it possible for desktop drives, even if not these ones in particular, to be powered by USB Type C ports as well as to transfer data through the same port at the same time?bill.rookard - Tuesday, January 5, 2016 - link
USB type C ports are supposedly capable of providing up to 100w power transfer. So - since most full size desktop drives require about 1.5a during normal operation, but up to 3-4 amps during spinup. So - 50w on startup, well within the 100w max.rtho782 - Tuesday, January 5, 2016 - link
But the startup inrush current is for a very short duration, could you not resolve this with an ultracap, charge it over say 3 seconds, then spin up, never drawing more than about 10W.name99 - Tuesday, January 5, 2016 - link
Wouldn't it be more correct to say that "USB type C ports can NEGOTIATE up to 100W power"?As far as I can tell from looking around, all that is guaranteed available is the same 4.5W (900mA at 5V) that USB-3 promises. Beyond that you have to negotiate and many devices (eg your smartphone) are presumably not going to give you anything more.
Also 100W seems high. The highest I can see is the ability to supply 3A over 5V, which would give 15W.
USB Power delivery promises 100W, but that's yet another spec that isn't (as far as I know) an essential part of the USB-C spec, and which I imagine will rarely be supported by non-power bricks. Even something like an iMac can hardly commit to supporting 100W on each of 4 USB ports; it just doesn't come with a 400W (plus enough to run the iMac) power supply.
ganeshts - Tuesday, January 5, 2016 - link
USB Type-C can support UP TO 100W for power delivery. I was given to understand that the Porsche Design Desktop introduced at CES can do up to 15W only.Dahak - Wednesday, January 6, 2016 - link
As others have stated, while USB Type C ports can support UP TO 100W for power delivery, there is no way of knowing if the port is designed for that.It is not a required part of the spec as you have to design additional power delivery components to the port
Dahak - Wednesday, January 6, 2016 - link
Really need an edit button..I should clarify, the port on the motherboard/laptop has to be designed to provide the extra power over and above the minimum required for the spec
bill44 - Tuesday, January 5, 2016 - link
When can we expect Thunderbolt 3 external SSDs? PCIe Gen 3x4 Samsung M.2 950 Pro 1TB in an external case with Thunderbolt 3 Type-C port that can do >2Gbps.The current bridge chips (M.2 to USB3.1 Gen 2) only uses the B-key for SATAIII mode, not the full PCIe Gen 3 x4 bandwidth (M-key).
There would be no need for 2x mSATA in RAID0 or 2x M.2 in RAID0 to saturate the USB3.1 Gen 2 10Gbps bus, 1x M.2 M-Key 2x lane mode should be able to do that.
Since Thunderbolt 3 (40Gbps) uses PCIe natively, it would be the best choice for external M.2 storage. USB3.1 Gen 2 could be the backup option on such a drive (albeit slower), as both uses Type-C connector. Current solutions use 1x thunderbolt + 1x USB3.0, new ones only need 1x Type-C.
Where are the bridge chips?
m2inor - Tuesday, January 5, 2016 - link
Computers with Type-C USB 3 or 3.1 only guarantees 4.5 watts, and this is the USB 900ma @5V spec.Computers with Type-C Thunderbolt 3 implementations guarantee 15W to either USB 3.1 or Thunderbolt connected devices.
Looks like the Porsche Desktop drives likely only guarantee limited power to a connected computer. Goal probably was to be able to connect to Apple's Macbook or the new Alpine Ridge/Skylake Windows laptops for charging while drive adds disk storage at the same time.
Perhaps what needs clarification is how much power the PD pass thru provides.
Note well, the PD power spec is 5v@3A for 15W, and 20v@3A for 60W. To get to 100W with the PD spec requires hipower Type-C cable that handles 20v@5A.
Dig into the USB 3.1 PD spec to understand why, and why it's going to be messy for awhile as the industry and end-ysers adapt.
Also, don't forget that those Type-C passive cables have current AND RF requirements to handle USB3.1 and Thunderbolt 3 reliably at 10Gbps and 20Gbps respectively.
m2inor - Tuesday, January 5, 2016 - link
According to info on LaCie website:"Delivers up to 30W of power to charge compatible laptops such as the Apple® MacBook®."
zodiacfml - Tuesday, January 5, 2016 - link
For the price, I should have seen a Wi-Fi module there like 2x2 802.11ac or better, 802.11adadityarjun - Wednesday, January 6, 2016 - link
I wish they would a build a raid 1 enclosure. Say 2 2tb drives. And if 1 fails, we can replace that hdd. If the enclosure fails, we can put the hdd in another one.Seems like the safest route. Is that possible/feasible?