[Beowulf] Interesting google server design
Many of your questions may have already been answered in earlier discussions or in the FAQ. The search results page will indicate current discussions as well as past list serves, articles, and papers.
Lux, James P james.p.lux at jpl.nasa.govThu Apr 2 12:01:49 PDT 2009
- Previous message: [Beowulf] Interesting google server design
- Next message: [Beowulf] Interesting google server design
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
> However, I would imagine this is a trivial amount at 12 VDC > vs 5 VDC and the incredibly short distances the electricity > travels on a PCB trace. Actually, it's a non-trivial amount. Consider that a processor that draws 100W at 3.3V is drawing 30Amps. If you're using 2oz copper (typical for a high power application), and you allow a 10C rise, the trace, if inside the board, needs to be 1.7 inches wide. A 3 inch long trace will have a loss of 0.4W (which is small). If you can put the trace on the outside of the board, it can be about half the width, but you'll lose a watt in a 3" long trace. Drop the current to 10A, and you're down to a .4" trace inside the board, and .15" wide outside. You could let it get hotter, but that gets you into junction temperature problems and/or thermal distortion issues. Play around here http://www.circuitcalculator.com/wordpress/2006/01/31/pcb-trace-width-calculator/
- Previous message: [Beowulf] Interesting google server design
- Next message: [Beowulf] Interesting google server design
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
More information about the Beowulf mailing list
