Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

Airflow simulation for pc cases ;)

735 views
Skip to first unread message

Skybuck Flying

unread,
Oct 3, 2009, 9:03:39 AM10/3/09
to
Hello,

So far gpu's seem to be very good at "particle simulations".

One interesting application could be "airflow simulations for pc cases" ;)
:)

So next time I am interested in buying a pc case... I simply download a 3D
model of it...

Active some fans in it... <- would be cool...

And then I could see the airflow in action in 3D preferably realtime... to
know if it's a good case or not ! ;)

And hopefully the simulation will be realistic ! ;) :)

Doesn't seem to though to do ?

Must know theory about air flow equations or so ?

Any theory out there ? ;) :)

Otherwise wing it ! ;) =D

Also if heat equations could be added even better ! ;)

Bye,
Skybuck =D


Timothy Daniels

unread,
Oct 8, 2009, 12:44:15 AM10/8/09
to
IMHO (i.e. a guess based on a physics degree), the surfaces and
drag inside a PC case are too complex to even model, much less
calculate detailed airflow. You'd be better served by experimentation
than by software. Just remember that turbulence increases the
effectiveness of airflow's cubic-feet-per-minute, and flow perpendicular
to sharp edges promotes turbulence. Since bottom-to-top airflow
is helped by convection, and you'd probably want to keep the
hard drives cooler than the power supply, lower-front-to-upper-back
airflow is called for with the hard drives just inside the air intakes.

*TimDaniels*

Charles E Hardwidge

unread,
Oct 8, 2009, 3:58:43 AM10/8/09
to
SF is a troll who babbles and crossposts. Please ignore him or trim your
replies so the OpenGL newsgroup can stay clear of rubbish.

FU's trimmed to alt.comp.hardware.pc-homebuilt

--
Charles E Hardwidge

Rui Maciel

unread,
Oct 8, 2009, 8:02:18 AM10/8/09
to
Skybuck Flying wrote:

> Must know theory about air flow equations or so ?

If you want to model something then you must know how it behaves. So, yes.

> Any theory out there ? ;) :)

You could take a look at fluid dynamics implementations of the finite element method.


Rui Maciel

0 new messages