Sign in with
Sign up | Sign in
Your question

A Beginner's Guide For WaterCooling Your PC - Page 3

Tags:
  • Homebuilt
  • Water Cooling
  • Systems
  • Product
Last response: in Systems
Share
March 30, 2007 9:01:41 PM

dont get me wrong cleeve. I think your review is great. I can see a great amount of time was spent in writting it.

You did cover the basics greatly. However i think a few big nono's should of been added on your guide.

Tube diameter, is a major component in watercooling.

Mixing of metals: ALU n Copper Also very very bad

Non Conductive coolants: Did u know good old distilled h2o is non conductive?? and its also dirt cheap at 1.49/gal at almost any local super market.


Simple things that plays a hugh role should of been mentioned, thats all.


But yes it was greatly written, and the pictures were nice. However, im still at errk, about the product. But i guess you need to follow your sponsors.


You should ask gabe over at swiftech for a demo unit of there APEX220 unit. Im sure he will sport it to you guys if you say its for toms hardware as a review.

Just looking at how many koolance fires you started, im sure he wouldnt have any problem. And no watercooling person would have an issue with a swiftech apex kit.
March 30, 2007 11:19:32 PM

Further to discussions regarding using water cooling for the purposes of a silent-running PC, I had previously noted that there are silent air cooling options as well. Also, people have complained that the water cooling was compared to stock cooling only, so I'm going to offer up an example using a better cooler.

My setup: Pentium D 820 @ stock settings for this test
TT Tsunami Dream case (2x 120mm fans @ 1300 rpm, 1x92mm fan, and the PSU)
Arctic Silver 5 TIM.

a) Intel stock cooling - one week of TIM break-in
b) TT Sonic Tower - fresh TIM (mounted today)

I load the CPU up with two instances of folding@home for 100% CPU ustilisation. Temperatures are measured with SpeedFan.

In case a), temperature after 15 minutes (it takes that long to settle - first order response cure) is at least 63C. I didn't run the test longer than that because I don't like my CPU that hot.

In case b), temperature after 15 minutes levelled off to 53C. That's a 10C improvement with no fan on the CPU. I have run the heatsink with a fan before, and that brought temps down to 44C with 20% overclock. The system was still very quiet.

Anyway, that might bring some more perspective to those who are looking for it. My point isn't that this information should be in a beginner's guide, but people might be interested since one of the motivations for using water cooling is to reduce noise.

A drawback of the air cooler is that it's too tall to fit in most HTPC cases. For that case, I'm tempted by the SilverStone water cooling rig. It's pretty.

Cheers
March 31, 2007 7:09:02 PM

Has anyone had any experience with the "fanless" water-cooling solutions, such as those offered by Zalman? I'd imagine they wouldn't be good for overclocking,but it's be interesting to have a completely fanless PC; even with good-quality components.

(Example: Zalman - Looks cool)
Related resources
March 31, 2007 9:51:40 PM

I had a friend once buy an innovatek setup when they came out with the ULTRA version of their passive radiator. He was curious to see how well a fanless radiator would work.

He was using parts exclusively from innovatek - blocks, reservoir, pumps, tubing, fittings, etc. The radiator was mounted along the side of his tower case and covered the entire side door.

Well as the end results showed, it was not suitable for any OC tasks. In fact, the best air coolers gave him lower temps at stock speeds.

Innovatek uses 1/4" connections on all their equipment, the flow rate is very low and the pump must fight backpressure inherent when trying to cycle water through thin tubing and multiple blocks.

As for the radiator, yes, it was silent, but the heat that was stuck there along the entire surface and fins, front and back, had nowhere to go. The fins at the bottom radiated their heat over the ones above, so the top 1/3 of the radiator was always warmer than the bottom.

In any event, the noise of the lousy pump fighting the 1/4" flow killed any positive benefits the silent radiator had achieved.

Needless to say, after just three months, he dumped his bigass passive radiator system and bought some sensible gear from Swiftech.

If quiet cooling is what you are looking for, a good triple radiator mounted at the top of your case with three 11dBa 120mm fans, silicone silencers and RPM control will do the job very nicely.

I am not saying passive cooling is a completely dead solution, but stay away from 1/4" gear, get the heat away from your radiator, and use a pump that is quiet and efficient.
April 2, 2007 8:01:21 PM

I think Koolance makes a very good product. Looks great, easy to use and does a decent job of cooling. The problem with Koolance is it is EXPENSIVE!!! You pay a whole lot of money for no performance increase, literally twice as much as a system that will do as well or better.
May 20, 2007 1:23:17 PM

Quote:
Since someone has deemed to call Koolance cooling systems and parts junk and toys you should in all fairness read their Water Cooling 101. http://www.koolance.com/technical/cooling101/001.html 8)


I've read their Watercooling 101, and in all fairness, it's full of misinformation and outright fabrication.

They say aluminum, as is used in their radiators, is a superior conductor of heat, better than copper, which is why they use it. They further justify this position by saying "Look at automobile manufacturers.....they use aluminum radiators in their autos instead of copper, so it must be great!" And they also go on to mention gold plating is also one way they enhance the heat transfer properties in their crap.


Unfortunately, none of that is true or accurate. Aluminum and gold both suffer in comparison to copper in thermal conductivity.

Aluminum's TC is 250W/m k at 25C, gold's TC is 310W/m K at 25C, while copper has a TC of 401W/m K at 25C.....and that margin continues as temperatures rise.....even to 225C.

So, you may ask, why are heatsinks made of a lot of aluminum and not all copper? Well, weight is one thing. Copper is denser than aluminum and hence heavier. Also, a pound of aluminum can fashion a much larger heatsink than a pound of copper, giving a larger effective surface area. But in no way is aluminum or gold a better conductor of heat.....both are worse, and aluminum is much worse, than copper. Ever wonder why all those aluminum heatsinks have copper cores or bases?

As far as autos and aluminum radiators.......well, it's true they have them, but not for better thermal reasons. Auto makers moved to aluminum radiators for a couple of reasons.....aluminum is cheaper and aluminum is lighter.....so better gas mileage. Of course, auto makers make up for the known short-comings of aluminum by using high pressure, high flow systems.....something not seen in computer watercooling at all. In the real world, computer watercooling is decidedly a low-flow, low-pressure system, no matter the pump used.

And in that end, copper water blocks and radiators provide the best thermal conductivity available......including aluminum and gold.

And Koolance's assertion that the copper radiator assembly is somehow inferior to aluminum radiator construction......copper is "glued" and terrible flux is used, etc., just shows how biased Koolance is. Coper radiators with brass fins are NOT glued anywhere, and the flux used in soldering the radiator assembly is most excellent causing no leak problems at all. If anything, aluminum radiators have a historical problem with galvanic corrosion, something copper systems do not have.

Koolance, like a lot of manufacturers, produces a decent product......very high priced for what you get, but good for someone who has done no research into alternative products.

Of course, we could bring up Koolance's insistence on using 1/4" and 3/8" ID tubing and fittings in their cooling lineup of products.....of which both diameters are well known to provide horrendous back pressure problems and very reduced flow rates compared to 1/2" ID tubing.
May 20, 2007 1:41:08 PM

And we could go on about the galvanic corrosion problems with aluminum......

But forget the tech info for a second. Here's the reality:

Any company will push their product. They're in business to make money, not to give it to their competition. Koolance will say their product is superior (or at least effective). So will Swiftech. So will Toyota, McDonald's, etc. Why would McD's say their food will make you fat? It's fact, yet they don't show it in their literature. Why not? Because you won't buy their products and they'll lose money. It's not that hard to figure out.

Koolance makes good products. But how good is "good"? Are they better than a similarly-priced Swiftech? No. How do we know this? Read the independent reviews. Read the other data. Read all the real-world experiences that hundreds and hundreds of people have posted around these forums. Read the experts' opinions over at XtremeSystems. And then you will see the "real" story. Is getting a Koolance bad? No. It's very good for those who don't want to build their own setup, it's convenient, it's effective for the average user, and it's clean and easy to set up. But you have to pay for that convenience, and when you then balance that cost-to-performance ratio, Koolance is not very good. You can build a much better setup for far cheaper.....but that goes with most things around here.

Koolance is good, but "good" is subjective.

But certainly don't buy it based on their own literature, that is pure foolishness. Ask around, post in other forums and you'll see the real information. But I'll give it to you already......it's 1 paragraph up ;)  It's good.....for a specific set of criteria and circumstances. But if you're willing to do some of the work yourself and/or buy components, then you'll do ALOT better. Might not be as sexy as Koolance, but it will perform much better and far cheaper. But if you prefer style over substance, or convenience over performance, then definitely go with Koolance and you'll be happy with the look and the ease, but not necessarily the performance or the price ;) 
May 23, 2007 5:16:21 PM

www.hwlabs.com

incase all you koolance junkies still think koolance is a people company.

They LIED, and tried to bad mouth one of the worlds largest radiator company.


SILLY KOOLANCE, TRIX are for KIDS.


Enough of this koolance glorifying. There products are absolute JUNK.

You'll get simular performance on a Ultra120 extreme.


-lates.
July 29, 2007 8:20:30 AM

aigomorla said:
*please dont take me as a newbie. I have been watercooling longer then most people on this fourm. Going on to be my third year this november to be exact. So i know WAY more info then the average h2o cooling person.


Duh! I am running exos over 5 years now.

I don't understand your logic in ridiculing people using koolance but it is a damn good product that saw me through 2 new computers in UTTER QUIETNESS -- cooling 5 components on 6mm ID tubing with a single fan at its lowest setting while i watched my friends fry multitude of components and mobos fiddling with their wc setups that exhibited noise levels not any less than rocket launchpads.

Faulty pump... AMD up in smokes! I don't see you having a logic board on your setups. Kolance has one that shuts the PC down in case of trouble with wc system. What better protection would anyone need?

And i spent not more than few hours for setup and maintenance. I changed coolant twice, tubing once, and took the kit apart and washed the radiator and parts. Only changing the coolant was a necessity, others i did for fun. I didn't spend hours searching for biocides, germicides or miscellaneous chemicals -- just plain Caltex Coolant/Antifreeze... same one i use for my car. Compare this to the days and days they spent on their systems. Some people have all the time in the world... Well.. some just find good products.

All on koolance parts... except for the Innovatech waterblocks for my 15k cheetahs (I didn't like koolance's thermal paste or other absurd hd packs).

And yeah... all on 6mm ID, 8mm OD tubing... easier to work with, less invasive... and it doesn't look like a garden hose is running through my casing... and I don't need to spend 5 hours trying to take my computer apart to install a sound card or memory module.

My friends are shocked by the quietness when i beam stocks on the wall while playing HDTV on my pc (total silence in the room in silent moments of the movie)... And i love the 36Cs i get on my Cheetahs.

5 years and more without a single trouble... I love my exos -- the peace of mind and tranquility i got is priceless. I will change it now with Exos-2 LX... not because something wrong with it, but just because i believe i used it enough. No need to wait for a failure.

And all your insults and belittling attitude towards koolance users... I believe my exos directed them to the appropriate place. You know where it is.

Jimmy
June 26, 2008 9:10:49 AM

PEOPLE UNDER ESTAMAT A ZELMAN, dont use the stock pump. they have a fan for the top of the water tower. inside put a tube on the returning water so it puts the warm water at the top of the tower. dont use a high volume pump that wont give the water time to cool down. 1st you in no way need a mb chipset cooler,or a gpu cooler, buy a top of the line chipset cooler,same temps as water cooled i know iv tryed both! get a arctic cooler for you video, use thermal epoxey for heat syncs/arctic5 for thermal paste.

i have water cooled cpu and 2gb's water cooled 1066 ram,quad core q6600-w/go-2.4ghz
overclkd at 3.8ghz temps are idle 24c cpu/34c mb
high cpu 34c /mb 40c sounds good to me!
also have a cool-it system on a 775 3.2ghz oc'd 3.8 server. [#2a00d4]19 c cpu:) 
      • 1
      • 2
      • 3 / 3
!