Area vs Spot Cooling: Mix & Match?

Felger Carbon · Post by **Felger Carbon** » Tue Jan 16, 2007 1:55 am

This is Off Topic because I want to make it clear that I'm musing.

Big fans are absolutely ideal area coolers. They're simply outstanding. Every case needs area cooling.

Certain small, hot components don't benefit the way the rest of the case does from area cooling. Since I don't do video cards, I'm talking HDDs and the CPU - this is the arena of spot cooling.

A good fan for spot cooling should be as large as the object being cooled, but no larger (or at least very little larger). The garden-variety 3.5" HDD is 101.6mm wide; a Scythe 100mm fan makes an ideal spot cooler for it - and using push-pull (as I do) one fan can cool two HDDs. That's what I've been doing for a while now, and it works very well.

For a long while, I've used passive Ninjas to cool my CPUs. The area-cooling fan provides the gentle breeze needed by the Ninja. At this time, the mid-tower ATX cases that come with 220mm side fans cannot mount a Ninja. So until these cases get an inch wider, a 126mm HS or HSF is about the tallest that can be used. Hello, Freezer 64 Pro! But these smaller heat sinks come with closely spaced fins (Scythe decided not to put their dwarf Ninja into production, darn it!). A gentle breeze won't push much air between tightly packed fins.

So the 104mm-wide 64 Pro is a candidate for spot cooling. I've already tried, temporarily, a 100mm Scythe fan and it worked well. And now I come to the point: I'm considering putting a fan back on my 64 Pro, and leaving it there this time.

With a CPU fan, I can turn the big fan down close to the starting voltage and get more than adequate area cooling - I'm talking good area cooling. And I can concentrate on providing no more spot cooling than is necessary, where "necessary" is an individual judgement about CPU temperatures.

In other words, I think adding a fan will result in a quieter computer because the big fan can be run more slowly.
------------

Hmm. Upon reviewing this, I started with this sentence: A good fan for spot cooling should be as large as the object being cooled, but no larger (or at least very little larger).

Come to think of it, the big fan is a spot cooler for the case! AKA area cooler, where the case innards are the area. And the big fan darn near is as big as the case!

Bluefront · Post by **Bluefront** » Tue Jan 16, 2007 3:11 am

Nothing wrong with spot cooling, provided the fan you use is quiet. Without any direct connection to the outside, an internal fan at low rpms can be almost inaudible. The idea is to mount it as close as possible to the components that need more airflow. You can use ducts for this purpose, but mounting the fan properly can avoid ducting.

Why don't you try mounting the big fan externally? All you need to do is construct a shallow housing that attaches over the hole in the panel. But I really don't think a Ninja is suited for airflow from the side. Unless you constructed a set of deflectors, most of the airflow would be obstructed by the top fin of the Ninja. I'll bet an XP-120 would be ideal for these big fans.

I'm waiting for one with an rpm sensor, maybe with a standard fan frame (square). Would simplify installation and control considerably......

Felger Carbon · Post by **Felger Carbon** » Tue Jan 16, 2007 7:03 am

Bluefront wrote:I really don't think a Ninja is suited for airflow from the side. Unless you constructed a set of deflectors, most of the airflow would be obstructed by the top fin of the Ninja.

Actually, I proved that's not a problem with my first big fan, the one I mounted on the outside of the left panel. I used a paper-tape duct (mode of 3M Micropore tape), and it worked, and guided the output airflow, just great. The closeness of the fan to the Ninja top proved not to be an issue.

That shouldn't surprise you because of your advocacy of positive pressure. All these cases with big fans on the left side panel are, in fact, positive pressure cases. The airflow is guided by the placement (and ducting) of the outlet vents and has nothing whatever to do with the input fan placement.

It may dismay you that our interests have converged...

jaganath · Post by **jaganath** » Tue Jan 16, 2007 8:15 am

I wonder if big fans could be used in a negative pressure arrangement? In theory if the inlets were well-defined and ducted and undesired intake holes are taped up, it shouldn't matter if the "polarity" of the airflow is reversed.

Felger Carbon · Post by **Felger Carbon** » Tue Jan 16, 2007 10:53 am

The only cooling problem with using the side panel as an exhaust would be the UPS heat and the CPU heat would be dragged into the case instead of immediately expelled. This would obviously raise the internal temperature of the case. It would also require reversing any hot video card coolers that previously expelled air out the PCI ports.

There is also a mechanical issue with turning the big fan around. Unlike the common 120mm fan, the big fans are not front-back symmetrical. This is not just something to fix with a screwdriver.

For those two reasons, I would not even attempt the project unless somebody presented proof that the gain was worth the pain.

Bluefront · Post by **Bluefront** » Thu Jan 18, 2007 3:44 am

FC.....I have no doubt a Ninja can be cooled well by blowing a lot of air directly at the top fin, like what a case with this 250mm fan will do. But this flies in the face of all the airflow mods I have used in my quiet computer projects.

I have advocated making the maximum use of all the airflow, rather than a "shot-gun" approach. This has proven to be the quietest way to cool a computer.....to me anyway. The big fan cases waste much of the airflow with a Ninja-type heatsink.......in all of the current designs anyway. I suspect with proper ducting, the design would be more efficient, but as it stands, it cannot work as quietly as a case making better use of airflow.

One other thing.....positive pressure in a case, does not equal airflow. You could have great pressure (imagine a pressurized tank of air) without any cooling at all. The air must flow over the hot spots to have any cooling effect. More air passing over a component equals better cooling....not the pressure in the case.

I'd like to try out this big-fan solution, but the cases I've seen so far using this fan, have been terrible. Maybe I'll just buy the fan and use it in a better case......

Felger Carbon · Post by **Felger Carbon** » Thu Jan 18, 2007 2:05 pm

Bluefront wrote:I'd like to try out this big-fan solution, but the cases I've seen so far using this fan, have been terrible. Maybe I'll just buy the fan and use it in a better case......

I definitely recommend giving it a shot. Let me point out that the Xclio 3060 is just $35 at Newegg, and includes the big fan and its accessories (the fan-bezel etc). Once you have the case+fan, it would make sense to give the case a try before discarding it for just the fan. Sort of a cheap 2 fer 1 deal!

Remember, the 3060 plastic bezel is screwed in with 6ea screws. Do not try to remove it without first removing the screws!