Robert G. Brown rgb at
Fri Jul 19 14:53:13 PDT 2002

On Thu, 18 Jul 2002, Bill Northrup wrote:

> I thought RGB would be all over the fact that more humid air would
> conduct heat better. Actually I felt let down that I didn't get a long
> and informative physics lesson, maybe I will be more lucky next week.

Well, I could always TRY to oblige, but actually I don't really know
much about the relative benefits of dry vs humid air in system cooling
(as opposed to human cooling, where a high RH retards evaporative
cooling of sweat and makes for oppressive heat on a warm day.:-)
Especially given that the relative humidity of the warm air inside the
systems case (right up against the CPUs) is likely very low regardless
of cold-side ambient RH, since RH is a function of temperature.;-)

Actually, our AC also regulates humidity (I believe), but this time of
year I'd expect that it squeezes more water out of the air than it puts
into it.  In fact, we have a small condensation leak in one of the
overhead returns, although it is fortunately not near any systems (yet).
This is a problem with just dumping water into the air of a server room
with no regulation via e.g. an OTC humidifier.  A server room has a fair
bit of VERY cold metal, and humidity will readily condense out on it.
The AC air vents are in the 50F's or even cooler.  The air in the cool
part of the room is in the high 50's to low 60's (F).  Behind the cases
it is 10-15F warmer.  Air that is 30% RH at 70F is more like 60% RH at
50F and has a dew point (100% RH) around 40F.  Any 40F metal in the room
will condense water readily, and the water will then drip to the floor,
possibly into an expensive machine that hates being wet.

The point about grounding still stands, though -- I've generated sparks
on humid days and not generated them on dry days, depending on things
like what I'm wearing and how I brush it against the seat I'm sitting
on.  The only safe thing to do is to ground yourself on big metal before
touching anything static sensitive.  I generally slip off my shoes to
work inside a case, and although I don't go to the point of wearing one
of those annoying grounding wristbands, I generally lean my forearms
against a grounded case or rack in addition to touching something
grounded before approaching hardware.  

It's been over a decade since I last knowingly fried something.


> Bill
> _______________________________________________
> Beowulf mailing list, Beowulf at
> To change your subscription (digest mode or unsubscribe) visit

Robert G. Brown	             
Duke University Dept. of Physics, Box 90305
Durham, N.C. 27708-0305
Phone: 1-919-660-2567  Fax: 919-660-2525     email:rgb at

More information about the Beowulf mailing list