Phanteks Unveils PSU Power Splitter for Enthoo Mini XL

[btarunr]

Phanteks' highly desirable dual-motherboard case, the Enthoo Mini XL, was displayed with a new PSU power splitter accessory, at Computex. Simply put, this accessory lets you multiply the 24-pin ATX and 8-pin EPS connectors of your PSU, so you can run both the ATX and ITX motherboards with a single PSU, with the added freedom of shutting down and powering up either board without affecting the other. The splitter takes is 'smart' in this way. It takes 24-pin and 8-pin inputs from the real PSU, and to the end user, behaves like two PSUs, so both "systems" (motherboards) can be turned on or off at whim.



View at TechPowerUp Main Site

 

[ZoneDymo]

very cool, though too bad they put it black on black, kinda hard to make stuff out this way.

 

[RazrLeaf]

Now that is something that the case desperately needed. This actually makes the case a valid candidate for my next build. 2 desks, one tower...

 

[ZeDestructor]

If the splitter is a direct pin-split, I can see them being used in the opposite way: get two PSUs and combine them to make single ATX/EPS connections to the PC, and load-balance across boath. Incidentally, also making your PC get some PSU redundancy at low loads...

I've been keeping an eye on such things for a while now, because of my silly watercooled PC plan...

 

[McSteel]

If the splitter is a direct pin-split, I can see them being used in the opposite way: get two PSUs and combine them to make single ATX/EPS connections to the PC, and load-balance across boath. Incidentally, also making your PC get some PSU redundancy at low loads...

I've been keeping an eye on such things for a while now, because of my silly watercooled PC plan...

It doesn't really work that way, though... With the ATX PSU being a voltage source, let's say PSU1 outputs 12.1, 3.31 and 4.95V on the main rails, and PSU2 gives out 11.9, 3.29 and 5.03V. There would be a current flow from PSU1 to PSU2 on the 12V and 3.3V rails since they're on a higher voltage, but PSU2 would sink current into PSU1 on the 5V rail.

The amount of sunk current would be proportional to the difference between the effective voltage at the load end (lets say on the motherboard end of the ATX connector) and the voltage "seen" by the PSUs on the splitter. The current sinking capacity is the same as the current output capacity, so if a PSU can output 30A on a particular rail, that is also how much it can sink into itself.

Put simply, you cannot load-balance between different PSUs by having both connected to the same load. You would need some kind of electronics in place to block the current flow between PSUs.

 

[ZeDestructor]

It doesn't really work that way, though... With the ATX PSU being a voltage source, let's say PSU1 outputs 12.1, 3.31 and 4.95V on the main rails, and PSU2 gives out 11.9, 3.29 and 5.03V. There would be a current flow from PSU1 to PSU2 on the 12V and 3.3V rails since they're on a higher voltage, but PSU2 would sink current into PSU1 on the 5V rail.

The amount of sunk current would be proportional to the difference between the effective voltage at the load end (lets say on the motherboard end of the ATX connector) and the voltage "seen" by the PSUs on the splitter. The current sinking capacity is the same as the current output capacity, so if a PSU can output 30A on a particular rail, that is also how much it can sink into itself.

Put simply, you cannot load-balance between different PSUs by having both connected to the same load. You would need some kind of electronics in place to block the current flow between PSUs.

Couple of diodes should do the trick (which wouldn't really mind in practice). Only ATX would be shared (to reduce the crossloading from non-loaded minor rails). Harddrives and PCIe would be easily shared by just splitting connections across PSUs.

I should take apart a server PSU section at some point to see how it's done...