You assume some circulation of the oil and that does not seem to be evident.
No, I assume heat transfer even if it didn't circulate, but
on the other hand, it will circulate because that's what
heated liquids do. Put a pot of water on the stove, turn
the burner on and leave it alone... did the water at the top
get hot? Why? Passive circulation is not "as" effective,
but if there is enough margin, it is enough.
The only circulation will occur through the heat rising and radiating and
will be slow, so the oil in the centre near the CPU and GPU will be hotter
than the outside.
Yes, some, but so what? Same goes for air in an air-cooled
system and yet, they manage to stay cool enough.
I don't know what amount of heat the CPU and GPU will be
putting out, but I'm sure someone (probably kony, cos he's clever) will be
able to work out how quickly the oil will heat up and how quickly it would
cool and therefore derive a stable temperature (terminal temperature?).
Anyone know the specific heat capacity of cooking oil?
Tom's already demonstrated that the system ran.What useful
purpose is there in this? Seems more like triva than
anything else but I dont' think all the details would be put
in next year's edition of Trivial Pursuit. I suppose if you
packed as much hardware as physically possible into the
case, it might run too hot but this is why it's an experment
and temps will need be taken. Even knowing the absorbtion
rate of the oil we can't continue to a final conclusion
because there's also the room ambient temp, the conductivity
of the case walls, whether the top stays on the case or
not... lots of variables. In the end, I don't plan to build
such a system so there is no need to try to come up with
some set of charts that would encompass all variables.
On the other hand, you could just email the author of the
article and ask. He'd probably be pleased that someone was
so interested in it.
If this temperature (or the predicted temperature around the CPU and GPU) is
lower than the threshhold for problems in the CPU or GPU, then this solution
looks like it would actually work!!
.... so? Work doesn't mean it has any benefits.
As I"d menitoned in another post, the PSU is still above the
oil and it's fan must still run... that is the larger % of
system noise if the system were otherwise optimally set up.
What it'll do is make a big mess, maybe foul the capacitors,
maybe leak, and smell horrible. If you want to use oil I"d
suggest some other type and a bit more time to seal up the
case than the sloppy job on the one in the THG video.
So it would appear that oil is not conductive then? What about the corrosive
properties - anyone?
In general oil is not particularly corrosive or conductive
but it might depend on the type and the impurities in it, as
well as the materials in the system (components). I doubt
THG used vegetable oil for any reason other than that it was
common and cheap.
