How important are ambient temperatures in WC'ing? (ie., high amb.temps are how bad?)

[Xylo]

I'm still working on piecing together my new water-cooled system, and am at a crossroads in a couple areas with choosing cases and fans...

So my question is, high ambient temps impact the effectiveness of a WC'ing solution just *how* much really? A lot? A little?

If the inside of my case is fairly warm, how bad is that going to be? I mean, lots of people (most?) pull air from the outside through their radiator inside, and that at the very least is going to be slightly warmer air.

But then in terms of fans, is there a bare minimum that should be done? I currently am planning on one 120mm in the front, one 120mm in the back, a PSU with one 120mm, and a radiator with two 120mm's (location TBD)...that seems like plenty to me?...and yet I still see tons of cases with a whole lot more fans, mounted in side panels, blow-holes, etc.

In addition to how much does the ambient temp effect the WC'ing, what about just the general case itself (ie., non WC'd components)? My last computers will not be running anywhere even remotely near as hot as this one will, so I'm a little behind the times in getting a good feel for what will work.

Thanks!
Xylo

 

[wwparrish]

Think of it this way, the rad cant get any cooler than the air that surrounds it.

So in a chilly room in winter at 65 F, you get X temp at full load.

Same setup in summer in a 85 F room you will get X + 20F temp at full load.

This is not exact for a lot of reasons but in general it applies. Thats why big comptuer rooms are air conditioned (plus the heat buildup from the many machines wouls soon make it like a oven).

 

[Socket7]

Just in case you didn't know already.

Liquid cooling will not nessicarily run cooler then an air cooled system. The advantage to water is thermal mass. Because you have more stuff to heat up, you are less likely to generate more heat then your cooling system can get rid of. Water cooled systems usualy only have a tempature difference of 2 or 3 degrees between idle and load because of this, and this stability under load is what makes water better then air.

 

[Erasmus354]

Just in case you didn't know already.

Liquid cooling will not nessicarily run cooler then an air cooled system. The advantage to water is thermal mass. Because you have more stuff to heat up, you are less likely to generate more heat then your cooling system can get rid of. Water cooled systems usualy only have a tempature difference of 2 or 3 degrees between idle and load because of this, and this stability under load is what makes water better then air.

Wrong, watercooling is much more efficient at moving the heat away from the processor than a regular air cooled solution, that is what makes it better. In a watercooling system the mass of the water doesn't matter, if you leave your computer on for longer than 10 minutes the water will heat up to an equilibrium temperature anyways. At that point you are dissipating the same heat load as an air heatsink. There is no difference in the amount of heat being dissipated, just that watercooling dissipates the heat away from the processor.

Also, the temperature difference between idle and load is only 2-3 degrees in a well designed system that has a more than adequate radiator. Many systems have closer to 5-6 degree difference, and poorly designed ones (underpowered radiators) have much higher differences.

 

[Xylo]

Think of it this way, the rad cant get any cooler than the air that surrounds it.

So in a chilly room in winter at 65 F, you get X temp at full load.

Same setup in summer in a 85 F room you will get X + 20F temp at full load.

This is not exact for a lot of reasons but in general it applies. Thats why big comptuer rooms are air conditioned (plus the heat buildup from the many machines wouls soon make it like a oven).

Right...I'm well aware of all this, but I'm wondering in practice just how helpful it is to have a nice chilly ambient temp. inside the case, vs. one that's a few degrees higher?...end result of effects on temp. of waterblocks, tubing, etc...
And if the radiator is mounted outside the case?...

 

[Xylo]

Just in case you didn't know already.

Liquid cooling will not nessicarily run cooler then an air cooled system. The advantage to water is thermal mass. Because you have more stuff to heat up, you are less likely to generate more heat then your cooling system can get rid of. Water cooled systems usualy only have a tempature difference of 2 or 3 degrees between idle and load because of this, and this stability under load is what makes water better then air.

Well, 1) I partially disagree...but, 2) even if I did agree, what's your point?

 

[Russ]

I don't think he's talking about overall room temps. He's talking about will it be ok if he pulls air through his rad into his case.

 

[r00k]

In my wc rig, i have the radiator mounted just outside below the psu, in a push / pull configuration. In my initial design, logic told me that I should pull the cooler air from outside to the inside, as the 2 parts that heat up the most will be cooled by the rad / water, and not to worry about the rest of the case.

However, I monitored my internal temps, and realized that if i push the internal air out, i can keep the inside cool enough that there would be little difference in internal and external temperature. This in mind, i turned my 120s around, and now exhale the internal air over my rad. In doing so, i watched my internal temps drop several degrees, and my cpu temps seem to have dropped by about a degree.

Also, the temperature difference between idle and load is only 2-3 degrees in a well designed system that has a more than adequate radiator

I'll definitely have to take this as a compliment. My Idle temperature is 39-40C and load temperature is 41C, and quiet.