[mythtv-users] Proposed future power saving networked configuration (0.22 in mind)

Michael T. Dean mtdean at thirdcontact.com
Tue Mar 10 18:24:53 UTC 2009


On 03/10/2009 01:32 PM, David Brodbeck wrote:
>> On Feb 18, 2009, at 7:42 PM, Michael T. Dean wrote:
>>> Actually, taking it to a nice pretty sine wave would be a travesty.
>>> Computers run on DC power, so if you attach your 
>>> solar/wind/home-nuclear
>>> generator to batteries, then attach an inverter to the batteries to 
>>> make
>>> AC, then feed that AC to the computer PSU, it will take the AC and turn
>>> it back into DC...  So, you incur an efficiency loss going from DC->AC,
>>> then compound the loss with a second efficiency loss going from AC->DC.
>
> But even 12VDC-powered computer power supplies have to have an AC 
> stage in them, to get the 5V, -5V, and -12V outputs.  Usually this is 
> some kind of chopper circuit to make a high-frequency square wave.  
> This kind of circuit is often referred to as a "DC-DC converter" but 
> internally it's still making AC.
>
> In theory you could reduce 12V to 5V with a linear regulator, but it 
> turns out you waste more power this way than with the other approach.

But it seems to me there would there be more of an efficiency loss going 
from DC->AC (incurring a loss) then using a standard PSU (say an 84% 
efficient 80 PLUS) compared to using a "high-efficiency" DC->DC PSU, 
such as http://www.opussolutions.com/index.php?p=product&id=56 , which 
lists "Power supplies[sic] overall efficiency: >91%" (on a 12V system) .

I'd be more likely to use 
http://www.opussolutions.net/catalog/product_info.php?products_id=48 , 
but, of course, it doesn't list its efficiency.

Mike


More information about the mythtv-users mailing list