On Wed, 13 Jul 2005 20:20:58 -0400, "Frank"
<Frank@hotmail.com> wrote:

>This is very interesting information. Kinda like it that two point of views
>sort of debate over this as
>even though both of you are sometimes on or off target, you both provide me
>with very very useful
>info and un-thought of difficulties I might encounter other than just
>finding "the" perfect fan for my
>needs. Just don't slash at each other as you both have points I agree with
>and are what I was also thinking of.
>
>My 99.99% goal if I could say that way is to have a 24/7 silent pc. The only
>things I can think of in
>the pc that will make a sound are my 2 HD and the 2 auto-controlled fans in
>my PSU which run at
>900RPM. No other fans are to be added except the obvious radiator fan. That
>alone is what I call
>a silent pc. I'm not going to use a fan controller as I only have 2 bays
>left and they are used as front
>exhaust for the rad. I could mange to go down to one bay only but would have
>too much cutting and
>adjusting to do again plus it might possibly hurt the air flow. I stick with
>a manually controlled fan or
>a switch type.
>

You still seem to be overlooking the obvious, that you can't
simply "choose" to use lower RPM fans, you also have to move
the heat out of the system. Merely thinking "it's idle
therefore I need almost no airflow" is typically incorrect.
Remember that water cooling with a radiator, tank, lines
inside the system, does nothing to remove any heat from the
chassis. Chassis heat accumulation is inevitable if you
don't move enough air through it. However, again I do not
have specifics on your radiator, if it is mounted external
to the chassis that will help some.


>What would you think of this fan? I Might just be satisfied with it at low
>speed all the time as I
>doubt 17 db is very noticeable.
>http://www.silverstonetek.com/products-fmc3xw.htm
>Whatever static pressure is, would it be good at performing at low speed?

No, it is a junk fan in a novel color. Choose high quality
fans and go from there, as most fans can be throttled to the
appropriate RPM. The particular linked fan is also only
25mm thick, sub-optimal if you can fit a 38mm wide fan.


>
>As for my reservoir, it is indeed and rad/reservoir combo, however, I don't
>expect cooling miracles

It isn't a radiator. It might be marketed as such but that
doesn't begain to make it a significant factor in cooling.

The distinction of a radiator is a very large metal surface
area to coolant ratio. The storage tank is just the
opposite. What little benefit it will have is not enough to
alter any other cooling plans by much.

Even if it did have a more significant impact than it will,
mounting it inside a case does nothing to remove heat from
the case. You have not described in detail exactly how you
will implement this, as any cooling system of this nature
has some flexibility due to changing hoses or choosing
different tank, radiator, pump, etc as desired in addtion to
mounting location.


>from it. However, logically (I'm no expert) the fact that I placed the
>reservoir behind the PSU fan
>outside the case will make the PSU fan blow on the reservoir. The air coming
>out the PSU might be
>30 some degrees

Huh? You're making up numbers. a 900 RPM PSU exhaust fan
will have higher than 30 degree temp.


>which would be cooler than the water itself

No evidence of that yet. In fact, the more you reduce the
airflow the less likely that's true.


>and should allow
>the reservoir to have
>some kind of air flow and and being the first thing after the "waterblocks"
>should dissipate heat more.

What's your point?
Do not plan your cooling around the reservoir placement. In
fact, too much impedance from PSU exhaust will simply cause
even more chassis heating.


>Hopefully it won't store heat.

At this point I respectfully suggest that you have no idea
what you're talking about, and have been listening to the
wrong sources of info. It WILL heat up. It will stay warm,
that is heat storage in a loose way of looking at it. That
is impossible to avoid and is not a problem, issue, etc,
etc. In fact, based upon your plan to try to cool it, it is
manditory that it "store heat" in order to get any benefit
out of that airflow, as stored heat is necessary for the
thermal rise above exhaust temp (which is still a conjecture
not in evidence) which would allow airflow to have even a
trivial benefit in cooling it.


>
>(Again I'm no expert here) but I've learned that a hot liquid will drop in
>degrees faster when
>the difference between it and surrounding is greater. Hard to explain with
>my limited experience
>but basically, I think I might gain a degree here and there which could make
>a difference in running the
>system with fans off or at low speed.

I'm just going to be blunt. You don't yet have enough
experience to run a system without fans turned on and will
end up damaging components- if not immediate they'll just
have a shorter lifespan. Your goal at this point should
only be to reduce fan noise some and then proceed from there
seeing if further reduction is possible after measuring
component temperatures. That is not a concession, it still
allows eventually throttling down even turning off one or
more fans if that seems possible.

I'll put it another way. You seem to keep assuming there is
some drastic heat difference between idle and full load (for
example gaming). While the difference can be a large
percentage, it is not as significant as you anticipate.
Primarily, the very parts you'll be water cooling are the
ones that have a heat reduction, so other parts not
water-cooled continue to have the need for fan-forced
airflow. To put it in perspective, suppose your system
produces 130W idle and 250W under full load. You want to go
from almost no airflow at 130W to almost full airflow
potential at 250W. That's not a realistic presumption to
make about airflow need. Realistically the need will be
closer to 130/250 of the full load flow rate IF the
guesstimations for idle and full load wattage were in the
ballpark.