[Beowulf] Re: Beowulf of bare motherboards

[Andrew Piskorski]

On Mon, Sep 27, 2004 at 11:16:36PM -0400, Glen Gardner wrote:

> I'd like to sequence the turnon of the nodes such that the inrush 
> currents at turnon are lower, allowing the 12V supply to come up to a 
> stable 12VDC a bit faster.

How would you do that, if you were using standard ATX power supplies?
Is there some sort of simple circuit you could place inline which will
delay the ATX "power good" signal by different amounts to different
boards?  I'd discarded this idea as too complicated, but you made me
think of it again.

The Finnish guys powered 3 motherboards per power supply by soldering
on extra 20 pin connectors.  I'm also going to try running multiple
nodes off one supply, but without any soldering, by using a
CB-Z-Y2020-EXT "ZIPPY Power Cable Splitter: ATX 20 pin to Two ATX 20
pin for ATX Power Supplies".  At $55 shipped for an MGE "600 W" Active
PFC supply (from Newegg) vs. about $8.50 for each 20 pin ATX Y-cable,
the choice seems obvious.

  http://www.ateck.com/viewItem.asp?idProduct=453056250
  http://www.newegg.com/app/viewProductDesc.asp?description=17-167-010

I've read that maximum current is usually drawn at boot up, so at
least in theory, it could be better, and might allow running more
nodes per supply, if the nodes on each supply DIDN'T all try to boot
at exactly the same time.

-- 
Andrew Piskorski <atp at piskorski.com>
http://www.piskorski.com/