I pose a question to those most seasoned hydro-pimps, like Putwig and lcpiper...

it seems apparent, that although it is more expensive, that water cooling provides far superior cooling than fans can. But here's my question: wouldn't a combination of liquid and the fans be even better? My knowledge of thermodynamics would suggest yes, but since practicality beats theory, it would seem better to ask. Would the addition of fans to a liquid system be effective enough to warrant putting in fans? (heatsinks, too)

I don't quite understand your question here Leandra. Most, nearly all H2O cooling systems have radiators with fans blowing over them.

Can't forgo fans. Video cards, MOSFETs, northbridge all need cooling. Besides, it lowers temps even further...

Most high end heatsinks can match the majority of water cooling setups in performance.

And fans are already needed in watercooling setups to cool the water, usually via some form of radiator. I'm as confused by your question as Feral.

I think what she's getting at is making a waterblock that also has a built in heatsink with a fan on it. Which wouldn't be a bad idea...would help to keep it cool if the pump failed. Might give your board enough time to power down.

yes, independent cooling systems so that if one should fail, the others would keep it going until power-down, but in tandem they would be even more powerful....

I suspect that the waters superiority to air cooling would mean that little benefit (likely no benefit) would be gained by using a HSF/water cooling combo block during normal operations.

In the even of pump failure, there would be a thick water layer seperating the core side of the block from the fan side. The core would cook long before the fan was accomplishing any real cooling.

Also, in the even of pump failure, temps rise fairly slowly. Your bios temp failsafes should have ample time to work, specially if they are set low (45-55 degrees). I once booted without my pump plugged in, I was fully loaded and checking email before the temp hit 55 and the comp shut down ...... I was like "WTF" ........ then I looked at the compunurse and saw 71 C.

Anyhow, I don't see it being a very effective alternative.

Too much playing with the direct die idea, methinks.

Formatted properly, the request was, if you run watercooling, do you need fans? Answer-yes.

Could you add a second type of cooling system? Probably should be a pelt/chiller if anything. putwig's right on the block. Can't really mess with that.

Least she's trying to think outside the box, wish more women did that...

I've always wondered what the results would be of taking, say, a SLK-800 and putting a "lid" on it so that the water ran through the fins, heres a bad pic;
FAN
| | | | | | | | | | | | | | | | | |
| | | | | | | | | | | | | | | | | |
| | | | | | | | | | | | | | | | | |
| | | | | | | | | | | | | | | | | |
| | | | | | | | | | | | | | | | | |
===============
| | | | | | | | | | | | | | | | | | <- Water runs through here
| | | | | | | | | | | | | | | | | |
===============
CPU

so the fins of the HS go through h2o as the no. 1 way of removing heat but the heat can also go up the fins and be cooled by a fan

evaporation would kill it dead.

ive thought about this as well,but never brought it to fruition.
what i believe they are saying is something along the lines of a "maze" block with,say, a slk900 for a lid.
heh,heres a quick paint of the idea.....

I have seen entire copper blocks encased in acrylic then water was pushed through the fins, they work pretty good but are BIG. As far as capping a block and using the fins and fan to cool, what Putwig said.

The only problem with adding a fan on top would be the hoses and the fact that no waterblock I've ever seen has had the top and bottom combined in such a way that the two parts would conduct heat as well as if they were a single piece. That would have to be done custom and the inlet/outlets would have to be from the side, smilar to the Swiftech GPU blocks. This in itself wouldn't make it worthwhile because having water jetting directly in over the die is superior to water flowing over the area from a side inlet. I think it would just cost alot for the same, or worse performance than just buying a top-end WB or purchasing a bigger, better radiator.

I don't recall the name but...A couple of years back there was a company that made exactly what Leandra is talking about. They have since disappeared. But they had a Hybrid water and air block that performed reasonably well on just air or just water. Enough to keep the cpu from total melt down. It was made of solid gold anodized aluminum with built in cooling fins above the water block. Overall the performance was no where near as good as the dedicated water blocks of the time. IE Innovatek, Swiftech.

Well, here's an idea. have a heatpipe connect the base of the waterblock and some cooling fins. this way, the waterblock would still cool properly, and the pipe would move heat to the fins.

Quote: Originally posted by BigAkita I have seen entire copper blocks encased in acrylic then water was pushed through the fins, they work pretty good but are BIG. As far as capping a block and using the fins and fan to cool, what Putwig said.

Here's a linky BA.

The Shuttle I recently tore up had a heatpipe arrangement in it. Not extremely effective. It had the heatpipes out the back of the block leading up to the back of the case, where the 'radiator' was. It also had fins all over the block.

I added a 60mm evercool fan to the block itself, in addition to the heatpipes, maybe gained a couple degrees, not significant enough to show a clear benefit, although it traditionally had high case temps already, so a fan wasn't going to help much even in the best case scenario...

Quote: Originally posted by Cheshiredemon Well, here's an idea. have a heatpipe connect the base of the waterblock and some cooling fins. this way, the waterblock would still cool properly, and the pipe would move heat to the fins.

In terms of thermodynamics, that would be an excellent idea, especially with good airflow over the fins, however, the design of such a thing would be more difficult and would require an extremely careful and thorough soldering job to make sure none of teh water escaped, because water +metal+electricity = VERY VERY BAD. I think.. what does everyone else think?

I think it would probably perform very poorly. In order for a heatpipe to work, one end has to be hot the other cool. By water cooling the hot end the dynamics of the heat pipe would cease to function efficiently. If at all. Heat pipes generally have a very narrow temperature range to work within. A solid copper rod would work much better in this instance.

90% of effective cooling of a cpu takes place within 1 cm of the core. Your choices to take advantage of that are: A. Use a very conductive material such as copper to carry the heat away from the core. Like an Innovatek rev. 3. Then use the expanded surface area to dissipate the heat. B. As close as possible to the core aim very turbulent water directly over it at a reasonable flow rate. Like a White Water block with it's thin base and impinged inlet.

BTW here's a link to the block I spoke of. The Infini Pro. I can't believe the reviewer actually reccomended this kit.

Even a year and a half ago many air coolers were out performing this block. 13 degrees over ambient at idle with a pali and having to listen to a 5000 r.p.m. fan at the same time is nothing to brag about. And it was not a budget kit. It cost more than the Innovatek and Swiftech kits of the time. You can see why the idea didn't go far.

Attached Images

infiniblock.bmp (58.0 KB, 45 views)

Why don't we all just go submersion and be done with it?

/end/sarcasm, I just wonder how far we can get on current technologies...