digg

Facebook Connect Join Digg About

What a watercooling loop looks like after 16 months without coolant change.

forums.pcapex.com — Make sure you aren't eating when you check it out. :) Some nice algae growing in the res as well as goop all in the channels of the CPU water block. The moral? Change the coolant in your loops regularly instead of waiting over a year. :)

Who dugg this? Made popular Dec 25, 2006

Digg DudeWhat is Digg?

Digg is a place for people to discover and share content from anywhere on the web. Digg surfaces the best stuff as submitted and voted on by the community. Take the Tour to learn more.

106 Comments

show profanity settings
expand all
  • SubWolf, on 10/12/2007, -1/+46I haven't changed my coolant in over a year, because there are chemicals in it to prevent algae growth.
  • deanlowe, on 10/12/2007, -2/+22He probably used tap water.
  • inactive, on 10/12/2007, -5/+24Don't be lazy and change it often, and you won't have a problem. Simple. If you wan't something worry-free then I still think air cooling is the way to go.
  • sparkysko, on 10/12/2007, -0/+14maninblac:

    That's all fine and good, your expensive $60/gallon DDI water, however the second that water touches *ANYTHING* in his cooling system, it's gone from 60$/gallon DDI water to 1$/gallon IONIZED distilled water. At the very least, just by pouring it in, oxygen will dissolve into the water, which will aid in oxidizing the copper/whatever else in his cooling system.
  • sockpuppets, on 10/12/2007, -3/+16Uh, I haven't changed mine since installation over a year ago. Oopsie. Guess I'll be posting my own photos for you all in the coming days. :(
  • maninblac1, on 10/12/2007, -1/+14Okay, i'm not a master of thermodynamics, but i am an engineer, and we specialize in systems at equalibrium. And i don't agree with you two, not entirely anyway.

    Thermally, the system does reach equilibrium, additionally, it is always about the tempeture delta as vuke said. Newton's law of cooling states that a hot mass in contact with 2 different cold masses will dissipate its heat more quickly to the colder of the two masses. Additionally this cooling will model exponential decay and the end result (Tout) depends on delta T (the change in time). That being said, a high flow rate gives the fluid little time to be in actual contact with a colder surface. Since we must assume the system is at equilibrium and the steady state temperature of the radiator fins does not change with fluid velocity, which ideally it shouldn't.

    Additionally, if we assume that the flow is turbulent, which it likely is since there are very very very few things that are laminar flow. Higher velocity increases the amount of turbulence in the tube and increases fluid friction, thus generating further heat and decreases thermal conductivity because individual particals do not spend sufficient time against the exterior of the radiator.

    Now, in general we can ignore the time spend on the edge of the tube due to turbulence because the water can not escape the tube, however speed is a factor. And there is a point Q that optomizes thermal efficiency, of which it is impossible for me to tell whether that speed is faster or slower than the pumps sold on the market, also that being said it would depend on the system in question and the environment in question.

    So the generalization that faster is better is meager at best, if not false.

    And lastly to point out, if you logically follow my statements it would seem to conclude that you would want little fluid motion, this is true, but only the case of the radiator, in the case of the CPU, a medium to high flow rate is ideal, so it really goes both ways. If you have a fast flow the radiator can't do it's job, if you have a slow flow your CPU will be hotter than it could be.

    Wow, i can't believe i remembered all that from fluids and thermo.
  • neoian, on 10/12/2007, -2/+13http://duggmirror.com/
  • trampish, on 10/12/2007, -0/+10aahhh yes sexy time
  • maninblac1, on 10/12/2007, -4/+12I'm lucky, i've got chemical/labratory grade DI water (DDI), which for those who don't know is not only dionized, but distilled as well, if there was any living thing in that water to begin with i'd be surprised, i mean, they use this quality of water to cool nuclear reactors because it has zero residue when it boils.

    You can buy it online for about $40-60 a gallon. But i have some friends who let me use water off their system, it's pretty sweet.

    I'll probalby change mine ever 6 months as recommended, but i doubt i'll need to.
  • vuke69, on 10/12/2007, -0/+7Honestly, it's not THAT bad.

    What's really scarier is all those 90deg bends (some say it looks better, but in my eyes all it does is kill coolant flow rate), and none of them have clamps, or even zip ties.
  • inactive, on 10/12/2007, -0/+7I would imagine a lot of people's bedsheets look much worse that this.
  • opensourcemaven, on 10/12/2007, -2/+9damnit why cant i go to sleep. must. open. presents in 5 hours...
  • radu79, on 10/12/2007, -1/+7I doubt the green thing is algae.
    Besides for the fact that the temperature is not really optimal for them, they need light and some nutrients in the water, plus some sort of gas exchange with the outside world.
  • inactive, on 10/12/2007, -3/+9No thanks. I'll stick with PCs and use the money I save by not overspending on being trendy to keep it up to date.
  • woodsja, on 10/12/2007, -0/+5I hate when people say "the fluid doesnt sit in the rad that long to get cooled that much." Heat transfer doesn't work that way. Always: more cold mass in contact with hot mass equals more heat transfer; ;turn up the flow.
  • maninblac1, on 10/12/2007, -0/+5@sparky

    While you're technically correct, water like this acts somewhat like a buffer, it is resistant to changes in ionization and PH. For example, a gallon of this kind of water takes about a 1/4 to 1/2 of a cup of salt (register that as highly ionizing substance) to become mildly electrical transmitible. As for interally corrosive that's the biggest thing to worry about since, the PH level is low enough that it's not condusive to alegea growth since most algea does not do well below PH 7 as DDI water is naturally about 6.3-6.5. But that's what the additives are for, they stop the corrossion. As for their purity, well perhaps that should be called into question, but i'd say having high quality water can't hurt if you want to use a system for a long time.
  • BigKitty, on 10/12/2007, -0/+5When an article like this hits the front page, will the intertubes get clogged up?
    ;-)
  • joost68, on 10/12/2007, -1/+5NSFW link
  • inactive, on 10/12/2007, -3/+7HAHAHA! That's what you get for having that glow-stuff in your water! Stupid modders.
  • agoodm, on 10/12/2007, -0/+4http://savegamehosting.com/pics/water-cool-oops.html

    Full size image mirror. Enjoy!
  • delusion, on 10/12/2007, -1/+5This is dumb; most watercooling guys have already figured out solutions to this. There are numerous additives you can use to prevent growth in your system. I was using Purple Ice in my rig with distilled water. Your mileage may vary.

    I'm guessing that side-ways fan is cooling a heatsink on the motherboard? Weird.

    Also, why do people use antifreeze anyway? Is it actually proven to work better in home-brew watercooling applications versus just plain ol' distilled water or something else?
  • DiamondIce, on 10/12/2007, -0/+4dbr_onix, better some dust and a failed fan that results in some rebooting (nearly every system today has overheating protections built in, and most graphics cards will start locking up long before the heat damages them) than a half gallon of water spraying all over my 2 $700 graphics cards.
  • inactive, on 10/12/2007, -0/+4Dont eat while checking it out??
    Why the overexaggeration in everything on here?
  • angrycat, on 10/12/2007, -1/+4He should have used catfood.
  • Picard102, on 10/12/2007, -0/+3Doesn't seem like alot of build up at all acctualy.
  • cyclonesworld, on 10/12/2007, -0/+3Doesn't really look that bad actually. I was expecting flesh eating algae or something.
  • vuke69, on 10/12/2007, -0/+2I agree. That is one of my pet peeves as well.

    It's all about the temp delta, and the higher the coolant flow rate, the better.
  • sparkysko, on 10/12/2007, -1/+3You need *alot* of light for algae to grow in there. Either direct sunlight or a 20+ watts of fluorescent blasting entirely at that thing to get algae to grow. I'm skeptical that this is algae. Diatoms can grow in water with silicate, and will form a brown crud on the side. This will occur within a month, even with tap water. The gunk stuck on the sides isn't green, *ALL* algae is green. If the gunk ain't green, it ain't algae. (A misnamed 'algae' cyanobacteria also exists which will be a dark green/blue)

    Plausible explanations:

    Copper oxidizes, creating a greenish color. Possible reaction also with the chemicals added to the water. This could also have the side effect of depositing this gunk on the sides.

    My advice is to use distilled or reverse osmosis water. Algae *will* not grow in just this water. There is no need to add chemicals. If you have *REALLY* soft water (like in the pacific northwest), the water is acidic, and can somewhat corrode pipes, add a teaspoon of baking soda to raise the pH. Probably can't go wrong with using car antifreeze either, as long as the tubings good for it.
  • djmarston, on 10/12/2007, -0/+2Love the fact that its christmas morning, and we still have time to check digg!!! and post comments! oh well....

  • OBKenobi, on 10/12/2007, -0/+2Ah, so that's why McDonald's soft drinks sometimes have that signature Chlamydia aftertaste.
  • OBKenobi, on 10/12/2007, -1/+3[quote]than a half gallon of water spraying all over my 2 $700 graphics cards.[/quote]

    You know, you're not really supposed to use regular water in "water cooling" for that very reason. Non-conductive coolant works best. Plus it glows nicely under UV light.
  • DiamondIce, on 10/12/2007, -0/+2Call me old fashioned but to me Liquid (non-conductive or not) + Expensive gadgets = Bad Mojo, no matter how I look at it.
  • mangina, on 10/12/2007, -0/+2that CPU cooler looks like a cup.
  • maninblac1, on 10/12/2007, -0/+2@JWfokker

    Okay, what you describe is a poorly mixed system (a real system), one where T throughout the system is not uniform. Then the higher the flowrate, the better, only because it makes T more uniform, it decreases the temperature at the CPU and increases it at the radiator. The question is at what flow rate do you consider the system well mixed, and by how much will increasing the flowrate improve your performance?

    In my system i have roughly 2 cups of fluid in the system leaving out the reseviour, and a pump that is rated for about 8 cups a minute, a pretty slow system in my opinion right? I'd definately say that this is a poorly mixed system with tempeture extremes, in this case a higher flow rate is beneficial because it significantly cuts down the extremes of the system. So what if my system was more powerful, with 20 or 30 cups a minute, is it well mixed? Well JWfokker your tests say it's better but probably not the best and i'd agree.

    But if i make my system very high pressure/velocity, will i see that much gain, probably not, not because of friction or pump inefficiency, but because that's life, we don't see jet pumps for water cooling systems because it is more expensive than the gain it provides, IMHO.

    As i said engineers specialize in systems that are at steady state, something which our water cooling environment is not, as nothing ever is, unless it is greatly idealized such that T is uniform, which i assumed. At which point the flow rate does not matter, because thermally you can not tell a difference between the system if it's running at 30 cups a minute or if it's standing still, at least not at any specific point in time, dirac(t-x) to be specific.

    The most ideal solution is this, have 2 pumps (because nozzels and expanders (forgot the technical name) have poor coefficients (like elbows etc) and more pressure would be desirable), one at the radiator start and then increase the tube volume/insert a reseviour in the radiator section of the system, then have one at the end. Finally then have a thin tube high pressure/velocity section that passes water through the water block, ironically you see many setups like this because it is thermaly effective, or did you miss that in the real life implementation???

    It's not that i don't understand the "real world", i'm merely pointing out that generalizing the fact that faster is always better isn't true.
  • dbr_onix, on 10/12/2007, -0/+2"If you wan't something worry-free then I still think air cooling is the way to go."
    Air-cooling isn't exactly worry free - Depending on the house, and pets and such, you need to hoover the dust out the machine, and if it's bad enough, swap fans out.
    I've seen air-cooled machines *far* worse looking than a bit of algae, then again they probably haven't been cleaned ever (Two "memorable" ones : One was from a house with a carpet and a dog who sleept near the box, and acctualy stopped working because of the ammount of dust, the other from a small hotel, which was probably 5-10 years old - and was near the ground in the reception-area, but was/is still running fine, if a little loud)

    Not cleaning the watercooling loop for 16 months sounded worse than the pictures made it look. An hour or so every 6-12 months doesn't seem too bad.. Although for now, I'm more than happy with archiac air-cooling
    - Ben
  • Nerys, on 10/12/2007, -1/+3Antifreeze is not for cooling in fact it REDUCES cooling capacity most of the time.

    ANtifreeze surve 2 purposes (none of which should be needed in a computer) one it lowers the FREEZING point and I believe also raises the BOILING point. Important in a car when the weather drops below freezing :-)

    It also I believe has lubricants in it that inhibit rusting in the water pump and radiator. (not totally sure on this aspect)

    Otherwise pure water is better for cooling. I would just add some bleach to the water - don't see how anything can grow in that :-)
  • jmke, on 10/12/2007, -0/+2Preventing Microbial Growth in Water Cooling Systems
    http://www.madshrimps.be/vbulletin/showthread.php?t=29607
  • ScottMaximus1, on 10/12/2007, -2/+4Only communists don't celebrate Christmas
  • japroach, on 10/12/2007, -0/+2I agree that isnt too bad a state.

    If you have a decent percentage of antifreeze (or similar) in the water, and use something filtered/distilled, it should stay good for a while.
  • Derrekito, on 10/12/2007, -0/+1I'd be afraid of adding sodium anything ... wont it rust?
  • elsagacious, on 10/12/2007, -0/+1The glow-stuff is flouresceine, which is in some brands of antifreeze so that mechanics can find radiator leaks with a UV light source (black light).
  • thealliedhacker, on 10/12/2007, -0/+1I use this crazy/evil stuff that came with my ThermalTake system. It's got agents in it for "water quality leveling, anti-freezing, anti-rusting, and water scale reducing." It also mentions something about enhancing thermal efficiency.

    I'm not really sure if all of this crap is useful or necessary, but I do know one thing: this crap looks dangerous. It's NEON GREEN and it glows under black-light (and I don't think they added coloring or anything like that).

    Anyone else using this know how resistant it is to microbial growth?
  • galore, on 10/12/2007, -0/+1Ha! You should see the cooling loop of a soda chiller, like used in every commercial soft drink dispenser, and the aluminum heat exchanger for the carbonated water and the syrups and the dispenser nozzles after a few months use. Just hope that the restaurant uses chlorine bleach on a regular basis...
  • zodieman, on 10/12/2007, -0/+1You need to use distilled water dude in you're going to use a glycol mixture... I myself prefer Evans NPG+ waterless coolant. I use it in my aircraft and it works like a champ. Never have any boiling issues and never needs to be changed.
  • dagamer34, on 10/12/2007, -0/+1Unless the guy used tap water, that ain't algae. No light, no food source, no life. Most likely it's probably some type of rust that's turned green after a year.
  • elsagacious, on 10/12/2007, -0/+1Agreed, Misogyny. Now I bet if you could get your hands on some freon, that would work really well. But you'd need to really seal the joints well so it wouldn't leak out be off-gassing.
  • Hegemony, on 10/12/2007, -0/+1About 1 gram of sodium azide will keep anything from growing in it.
  • ogre2112, on 10/12/2007, -0/+1The green is from a UV additive, I believe. Someone can probably back me up on that.

    I think the algae was the chunky stuff, although it didn't look all that bad to me. Couldn't hurt to clean it though.
  • inactive, on 10/12/2007, -0/+1Because it will clog up and eventually kill the pump.
  • elsagacious, on 10/12/2007, -1/+2Antifreeze is ethylene glycol or propylene glycol ("environmentally safe" antifreeze), not ethanol.
  • Fetching more discussions...
    Show 51 - 100 of 105 discussions

  • Add a Comment

    Login or register to comment!
    Or, sign-in super fast with your Facebook account ...

© Digg Inc. 2009 — Content created and posted by Digg users is dedicated to the public domain | Terms of Use Updated | Privacy Policy
DIGG, DIGG IT, DUGG, DIGG THIS, Digg graphics, logos, designs, page headers, button icons, scripts, and other service names are the trademarks of Digg Inc.