Want to move to all 208V for server racks

Owen DeLong owen at delong.com
Sat Dec 4 06:28:55 UTC 2010


On Dec 3, 2010, at 9:17 PM, Adrian Chadd wrote:

> On Fri, Dec 03, 2010, Joel Jaeggli wrote:
> 
>>> (OK, so it's not as practical when you have other customers to worry
>>> about... but it might not be so crazy when you're looking at the
>>> efficiency numbers for 100,000 small 1u power supplies vs a set
>>> of much larger ones.)
>> 
>> Ohm's law is a bitch. 10kamp -48v DC plants are bad enough as far as the amount of copper required, running 12v for significant distance is comical, this is the reason small boats airplanes and diesel trucks adopt 24v systems. There's probably some model where top of rack rectifiers makes sense but that's really pretty much what a blade server is. When you look at a motherboard in a server a big chunk of of real-estate is devoted to taking 12v and switching it down to 1.2-1.8 for distribution to the CPU/memory, a 4 socket server might have to carry 400amp around in a space of around 300cm^2 on a layer of the pcb. 
>> 
>> The justification for running 208 or 480 all the way to a cabinet is all about smaller conductors.
> 
> Isn't this one area where Google have already (re-)pioneered recently?
> 
> Besides, there's a reason why AC won over DC for carrying 0 < x < few hundred 
> (or thousand? Amps) over a reasonable distance. IANA-PowerEngineer, but
> ISTR the behaviour/efficiency of voltage/current over distance for both
> AC and DC is well understood. (And no, ISTR it isn't "AC wins." :-)
> 
> If you're at all serious about discussing this, I bet spending 15 minutes
> doing some research and then an hour or so crafting some simultaneous equations
> to solve/graph would be very very eye-opening.
> 
> Come on guys/girls, you're a bright bunch, post some models and discuss
> those rather than un-substantiated datapoints! :-)
> 
> 2c,
> 
> 
> Adrian
> 
This isn't rocket science and doesn't require much math...

1.	For long distances, you need higher voltages to overcome line loss.
2.	For larger loads, you want to use higher voltages to have lower amperages so
	that you can use reasonable wire sizes.
3.	It's a whole lot easier to change AC voltages than DC.

The system that won is a system of very very very high voltages for the core
distribution with transformers converting that to intermediate distribution
voltages which are then further transformed down to even lower voltages
for service delivery. This is easily done with AC and would be quite complex
and inefficient (especially with the technology available at the time this
decision was made) with DC.

It would probably be more efficient to run the entire country on 200,000 VDC,
but, the dangers of exposing the general public to that kind of voltage are, well,
probably just one of the reasons we use 110 VAC instead.

Owen





More information about the NANOG mailing list