![]() | ||
|
|
General Liquid/Water Cooling Discussion For discussion about Full Cooling System kits, or general cooling topics. Keep specific cooling items like pumps, radiators, etc... in their specific forums. |
View Poll Results: What WC disasters have you had? | |||
*Never* Had a problem |
![]() ![]() ![]() ![]() |
85 | 31.02% |
Pump leak or total pump failure |
![]() ![]() ![]() ![]() |
63 | 22.99% |
Block SLOW leak (drip, drip, drip) |
![]() ![]() ![]() ![]() |
48 | 17.52% |
Block total failure (FLOOD!) |
![]() ![]() ![]() ![]() |
12 | 4.38% |
Hose / barb SLOW leak (drip, drip, drip) |
![]() ![]() ![]() ![]() |
72 | 26.28% |
Hose / barb BLOW OFF or Hose rupture (FLOOD!) |
![]() ![]() ![]() ![]() |
14 | 5.11% |
Radiator leak or blockage |
![]() ![]() ![]() ![]() |
10 | 3.65% |
CPU / Block separation |
![]() ![]() ![]() ![]() |
11 | 4.01% |
Radiator Fan failure |
![]() ![]() ![]() ![]() |
13 | 4.74% |
Other problem not described |
![]() ![]() ![]() ![]() |
55 | 20.07% |
Multiple Choice Poll. Voters: 274. You may not vote on this poll |
![]() |
|
Thread Tools |
![]() |
#51 |
Cooling Savant
Join Date: Sep 2002
Location: Santiago, Chile
Posts: 403
|
![]()
I voted for block slow drip.
That was on my first block, that was a heatsink on which I bonded a fiberglass cage. The adhesive used was quick epoxy, and it didn't hold the long term water exposure, becoming very soft. One of the barbs loosened and a slow drip developed over the back of the gpu. It was so slow that every drop always evaporated completely after the next. That caused intermitent annoying graphics behaivor only when wet. This was more a manufacturer mistake than an user error, even if in this case both were the same ![]() My lesson was not to trust on a product marketed properties; every material deserves a proper test on the specific aplication before being accepted. |
![]() |
![]() |
![]() |
#52 | |
Cooling Savant
Join Date: Jun 2003
Location: North Billerica, MA, USA
Posts: 451
|
![]() Quote:
However your failure does illustrate an important point, namely the need to ensure that any adhesives or sealants are suitable for the application. It isn't enough to just say 'Water Resistant' or even 'Water Proof', the product needs to either specifically mention the application or at least say 'suitable for constant immersion'. There are lots of 'Water Proof' glues that are intended only for intermittent exposure, and will fail if left to soak for extended periods. Gooserider
__________________
Designing system, will have Tyan S2468UGN Dual Athlon MOBO, SCSI HDDS, other goodies. Will run LINUX only. Want to have silent running, minimal fans, and water cooled. Probably not OC'c |
|
![]() |
![]() |
![]() |
#53 |
Cooling Neophyte
Join Date: Sep 2003
Location: Honolulu, Hawai'i
Posts: 25
|
![]()
Hi Guys:
These are not strictly on topic but I think you will get a laugh out of them. Between 1965 and 1980 I worked as a stagehand both on Broadway and on the road, I also worked in the various shops building the shows so I got to witness and or hear about many funny “disasters”. While moving a show from The Kennedy Center in Washington, DC to Toronto, miscommunication among the Carpenter on the deck, the Flyman on the Flyfloor and the Loading Bridge resulted in the wrong arbor being unloaded. When the load was removed from the pipe that was supposed to have had its counter weights removed, the pipe sped to the grid at the top of the stage house and snapped of some fire sprinkler heads. We had a two hour rain storm until someone could find the shut off valve. While setting up a show in Los Angeles an error by a contractor who was installing a tennis court on the roof of the stage house (he pushed the gravel into mounds that blocked the rain drains) caused a lake to form on the roof. When the water rose high enough, it came cascading down through the emergency smoke vents and rolled into the orchestra pit. It was a mad scramble to remove the musical instruments, the microphones and the wiring before they were submerged. The mixing console for the musical “The Wiz” was placed in a spot out of audience traffic flow but directly under a fire sprinkler head. The sprinkler gave way and filled the chassis of the console with rusty water. Back at the shop (after swapping in a replacement console) we revived the console by washing the chassis and all the boards in distilled water, drying it slowly and spraying it lightly with WD40 to drive out the last of the moisture. One series of consoles exhibited an intermittent low level noise on random outputs. The shop, the manufacture and the various operators using these consoles were stumped until an examination under magnification revealed small growths of fungus on the boards. That was when we made the connection between the humidity in the venues where the board was used or transported and the noise problem. The solution was to wash all the PCBs in a trisodium phosphate solution and the rinse in distilled water and long slow drying. I hope these have brought some smiles to your faces Grayson |
![]() |
![]() |
![]() |
#54 |
Cooling Neophyte
Join Date: Jun 2003
Location: Belgium\Ieper
Posts: 65
|
![]()
1. my res is a jam jar; it's connected with 18mm tubing to my pump and it is NOT with hose clamps (only using them on my hose adaptors). as I moved my res up and down when i was filling the loop the tubing must have gotten off the pump's barb but not completely. it ran fine, I went away. when I got back after several hours the pump was very LOUD, air everywhere in my loop and the res more or less empty.
quite weird, as it got loose but because the tubing was very stiff it was still pressed agains the pump's barb. so I lost half a liter of water in my pump's wooden case. 2. leaking CPU block. prolly a leakin the taps for the allen bolts. my machine often hung after a while, when I was about to get my new mb/cpu I removed the block... ![]() a small leak, but water got on the CPU and evaporated. the CPU was rusty and there was lime/dirt everyhwere (evaporated water!). there was even dirt IN the socket, one pin broke off because of that. sold the MB/CPU in an entire machine to a computer shop for waay to much $$$. and the MB was completely modded/trashed too: glue/white paste on the mosfets, it got vmodded and I just cut off the wires, a huge scratch on the bottom of the northbridge, black goo because it's been in a prommie,... ![]() ![]() ![]() |
![]() |
![]() |
![]() |
#55 |
Cooling Savant
Join Date: Nov 2002
Location: portugal
Posts: 635
|
![]()
My johnson pump from surplus center burned.
Fortunately it was the only thing. Now i need to find a seller of mcp600 at portugal.... and it isnt being easy. |
![]() |
![]() |
![]() |
#56 |
Cooling Savant
Join Date: Apr 2001
Location: new jersey
Posts: 142
|
![]()
just had a water pump leak and take my mb with it . Although i just instaled my rotor block and my temps droped 8c over my old maze 2 block
|
![]() |
![]() |
![]() |
#57 |
Cooling Savant
Join Date: May 2002
Location: Canada
Posts: 247
|
![]()
About a week and a 1/2 ago I suffered a pump failure of my little ViaAqua. The computer was off at the time, but I didn't know the pump had died and I turned it on. Windows booted, I logged in, started up IE, and started surfing around. After about a minute of time in windows MBM popped up saying my CPU was at ~60 C so I closed IE and shut down windows - it did finish shutting down and everything before it locked up. There was no damage at all to any hardware (not even a dirty windows shutdown). The CPU is a barton core running at 2300Mhz at I think 1.8V and had nothing more than a Maze-2 full of water but no flow at all to dump heat into for the entire time it was running.
On the bright side, it gave me a very good excuse to upgrade ![]() |
![]() |
![]() |
![]() |
#58 |
Cooling Neophyte
Join Date: Dec 2001
Location: Stockholm, Sweden
Posts: 3
|
![]()
A couple of months ago, I would've voted "No failure despite 2 years of water cooling". But when I designed my new cooling system, bad luck struck more than once.
Failure #1: Mr. Murphy decided to pay a visit. When reattaching my Maze2, I accidently put the springs on the wrong side of the block, pushing it away from the core instead of towards it. Result: One extra crispy T-bird 1000 and total motherboard failure (probably due to extreme overloading of the voltage regulators when the CPU failed). It was probably time to upgrade anyway. ![]() Failure #2: The reservoir (a plastic box of the type you would build electronics into) had a manufacturing flaw, causing a slow dripping leak. The drips reached my new NF7-S motherboard causing random lockups. After letting everything dry out, it worked perfectly. |
![]() |
![]() |
![]() |
#59 |
Cooling Savant
Join Date: Feb 2003
Location: Phoenix, AZ
Posts: 193
|
![]()
NEVER HAD A PROBLEM. BEEN A YEAR NOW.
KEEPING MY FINGERS CROSSED. ![]()
__________________
NF7-S Bios D1.8 218FSB X 11.5 1700+ DLT3C 0310 @ 2511Mhz TwinMos 2x256 PC3200 BH-5 Albatron TI4800se 310/643 3dmark 15,110 3DMark2001SE Link |
![]() |
![]() |
![]() |
#60 |
Cooling Neophyte
Join Date: Jun 2003
Location: London, UK
Posts: 70
|
![]()
Thought I'd share my late night dilemma with you all!
![]() After tinkering with my watercooled dual Xeon system I turned the power button on only to have the fans spin up, no video signal display and the cold, hard knot in the pit of my stomach form as the words 'uh-oh' ran through my mind. Thing would not turn off, so I had to yank the cord and after several minutes inspecting the machine I could see a few droplets at the bottom of the case. Looking at the bayres I noticed the CD drives were a little wet, no biggie I figured, it's happened once or twice before, unplug them and try the PC again. Nothing. An hour later of desperately trying to coax my expensive toy back to life by intermittently pressing the unresponsive power button I decided to get off my ass and reinspect the PC. To my horror it dawned upon me that if the bayres had dripped water onto the CD drive beneath it and this had dripped down onto the drive beneath that, it didn't take much for the water to drip down the IDE cable to its logical conclusion. My brand new workstation mainboard! No hairdryer to hand (sister's nicked the thing and taken it with her to uni ![]() ![]() ![]() This thing blows hard enough to tear a case fan from its housing, and employed on its lowest setting dried my machine out quickly. (is it my coolant mixture or do additives make the coolant a complete pain to dry out since the stuff is almost oily in its consistency and evaporates incredibly slowly?) After spraying the mainboard liberally with ispropyl, and then drying it out again I pressed the power button.... YEA-ESS! the ****er came on!!!! with a 'too much to hope' grin I checked the bios, the hard disk post, and then after praying for XP to load the system device manager and event viewer. All's good, so far. Fantastic eh? THE RELIEF! ![]() It's 3am and I would have been lonelier had this damned thing left me high and dry. Not quite out of the woods yet however, since the mainboard components around the wet area developed a whitish deposit fairly quick. I washed as much off as I could with isopropyl, but it was a pain rubbing it off all the SM components. Any ideas as to what this residue might be, some corrosion or water residue? *ahem* to stay on topic, yes, I have had a watercooling disaster, the problem was my own clumsiness in spilling water out the reservoir. |
![]() |
![]() |
![]() |
#61 |
Cooling Neophyte
Join Date: Aug 2003
Location: Olsztyn/Poland
Posts: 17
|
![]()
I had one leak. Water was running from CPU block-hose connection. It costs life of my gf3 (I gave it to warranty (one week before its ends) and they gave me Leadtek A310TD) - one mofset burned. I saw water just before going to school. All riva was wet, but it's still working.
now I'm running on alu cooler waiting for new blocks from www.overcooled.com
__________________
1.Duron 1400@2083(12,5*166) Zalman 7000A AlCu | EPoX 8RDA C1 rev 1.1 | WDC 400JB | Leadtek Winfast A310TD | 256MB Corsair XMS PC4000 | LiteOn CD-RW 48/12/48 | FSP300-60PNA(PF) | 3D Mark 2001 SE | Aquamark 3 2. Celeron 433 | SIS620 (all integrated) | WDC 4,3GB | 64MB SDRAM | CD-ROM x8 |
![]() |
![]() |
![]() |
#62 |
Cooling Neophyte
Join Date: Aug 2003
Location: Ottawa Ontario Canada
Posts: 27
|
![]()
My first watercooling setup consisted of a DIY copper cap waterblock, a 200ghp tetra aquarium pump, and a big rubbermade storage bin.
After hours of using my setup my temps weren't looking to great, so I got the idea of adding huge buckets of ice into the res because I knew that would fix my problem. A copper cap waterblock has alot of surface area, the perfect designed for enhancing the condensaion experience. My new geforce 4 440mx got soaked, also my motherboard agp slot was wet. After drying my parts off with a hair dryer everthing worked fine. This happened almost 2 years ago. |
![]() |
![]() |
![]() |
#63 |
Thermophile
Join Date: Sep 2002
Location: Charlotte, NC
Posts: 1,014
|
![]()
Problem 1- crappy lucite tops crack when barbs screwed on too tight or with too much teflon tape- very VERY slow leak around barb resulting in me trashing the stupid cheapo block.
Problem 2- Bayres leaked very slowly, in fact, so slowly that I only caught it 2 weeks after it started (at least I think 2 weeks judging based on how much it was leaking and how much had leaked already). The water and antifreeze leaked down the IDE cables into the two HDDs, corroding and eventually bridging some of the pins. I was wondering why one of my hdd's stopped working, then the other one stopped a day later. Took the plugs off, scrapped the crappy bay res, made a nice PVC res, cleaned the pins and cables with baking soda and water, 'puter putters along flawlessly. ![]()
__________________
I have a nice computer. |
![]() |
![]() |
![]() |
#64 |
Cooling Neophyte
Join Date: May 2003
Location: Colorado
Posts: 27
|
![]()
The tube on the pump outlet had been slowly moving off over a matter of days. I hear a slight rumble and then the tube got pushed off. THe pump points straight up, so a column of water about a foot high or so from a 3/4 in outlet erupted underneath my desk. Fortunately, no damage except a soaked carpet, thanks to an external case.
A more dangerous incident involving a leaking tube to cpu block was caught before the water trickled onto the motherboard. THat is because my 100% diy blocks use copper pipe, no barbs.
__________________
MSI K7D Master, 2x1800XP |
![]() |
![]() |
![]() |
#65 |
Cooling Neophyte
Join Date: Apr 2003
Location: Kingston, ON, Canada
Posts: 31
|
![]()
While I was on campus studying for one of my exams, the outlet hose on my pump came off. Blew water all over my desk, A/V receiver, and of course, my case. Luckily, distilled water saved the day, and not a damn thing was dead after drying it out. Needless to say, all my hose clamps have been replaced with what I hope are better ones...
__________________
http://www.devhardware.com |
![]() |
![]() |
![]() |
#66 | |
Cooling Savant
Join Date: Jun 2003
Location: North Billerica, MA, USA
Posts: 451
|
![]() Quote:
1. Stainless worm clamps - big, a bit clumsy, but will hold better than anything else. About the only failure modes are either not tightening them enough to hold, or over tightening them to the point where they cut the hose. This works well on either barbs or smooth pipe ends. The following I would use ONLY on fittings with barbs or at least a bulge on the end of the fitting to catch the hose on. IMHO they aren't as foolproof as the worm clamps, but they are a bit neater. 2. Stainless steel safety wire - put 2-3 wraps around the hose, pull tight, and twist the ends together (works best with safety-wire pliers!) This looks neat, and is actually FAA approved for use on aircraft hose applications, which says how good it is... 3. A heavy duty nylon zip tie pulled really tight. I don't trust those nylon 'pinch together' type clamps at all. Gooserider
__________________
Designing system, will have Tyan S2468UGN Dual Athlon MOBO, SCSI HDDS, other goodies. Will run LINUX only. Want to have silent running, minimal fans, and water cooled. Probably not OC'c |
|
![]() |
![]() |
![]() |
#67 |
Cooling Neophyte
Join Date: Apr 2003
Location: Surrey BC
Posts: 42
|
![]()
On my first install of a MCW5000A I neglected to tighten one of the worm clamps enough, it didn't leak untill I put it into my system. I turned on the pump, neglected to check for leaks, turned on the computer, and for two weeks my computer was down.
![]() By the grace of a certain Diety, (God) after two weeks of trying to start the computer (three times a day, all ending in failure) the computer started fine, and still works now. ![]() The situation could have been used as the base for a geek sitcom. The geek breaks his computer, his real life friends see this and he says the problem is the CPU failed, not wanting the truth to be known. He suspects the problem is the MB, but thinks that the warrenty has expired as he didn't read the small print. (he thinks it is a one month warrenty when it is a one year warrenty) The CPU goes out to be RMAed, but comes back with a note saying "no problem found." geek tries to start his computer, but it fails. His real life friends come over and say, "Start the computer, we want to do some computing and CPUing." The geek proceeds to make a big show of inserting the CPU and installing the cooler, presses the power button, and the computer starts normally. ------ But I learned from my mistake, and I plan on reinstalling the WC sometime within a week.
__________________
I'm just like a superhero without powers or motivation. |
![]() |
![]() |
![]() |
#68 |
Cooling Neophyte
Join Date: Apr 2003
Location: Kingston, ON, Canada
Posts: 31
|
![]()
I had crappy "pinch type" ones, and am now using zip ties. I might try your "wire" trick though, when I get back to school. If it's good enough for aircraft, it's damn well good enough for a simple water cooling system!
__________________
http://www.devhardware.com |
![]() |
![]() |
![]() |
#69 |
Cooling Savant
Join Date: Jun 2003
Location: Florida
Posts: 266
|
![]()
hm. Accidently knocking on a GPU hose would be the worst I think for me. The pump was on and the water/dye went in my case a little... but went on my white carpet
![]() |
![]() |
![]() |
![]() |
#70 |
Cooling Savant
Join Date: Feb 2003
Location: USA - Boston area
Posts: 798
|
![]()
Several years of building water cooled PCs. All with Swiftech blocks. The Swiftech blocks with John Guest Speedfit fittings had push-in-to-barb adapters. All with thick wall silicone tubing. All with various Eheim pumps (1046, 1048, HPPS). All but two with Innovatek agb-o-matic push-on reservoirs (and those with simple 'T' lines). Some Black Ice radiators (pros - for low noise), the rest heater cores. All Panaflo fans (mostly L1A) except for my current desktop machine which has a motorized impeller in it. All distilled water + ~5% Zerex Racing Coolant.
No failures - still pretty cautious, though. |
![]() |
![]() |
![]() |
#71 |
Cooling Savant
Join Date: Jul 2003
Location: Michigan, USA
Posts: 456
|
![]()
I had the hose barbs on my black ice micro II radiator both loosened and wiggled free. Luckly, the system was turned off right before it gushed. Nothing damaged, but all was very wet.
__________________
Thou art God. |
![]() |
![]() |
![]() |
#72 |
Cooling Neophyte
Join Date: Aug 2004
Location: Denmark
Posts: 4
|
![]()
The only problem ive had was with a ready-to-go watercooling package i bought when i first tried out watercooling. It was filled with water and connected prior to delivery. Unfortunatly there was a leak where the tube connected to the CPU block, causing water to drip onto my gfx card and down onto the CPU socket.
Fortunately both survived(though the gfx card had to dry off for a week ![]() After that i made my own watercooling system, and too avoid problems like these, i made a piece of tube that stuck up into the air, so that the water could expand and contract without causing any pressure changes in the system. After that ive never had any problems at all. |
![]() |
![]() |
![]() |
#73 |
Cooling Savant
Join Date: Sep 2004
Location: Los Angeles, Ca
Posts: 310
|
![]()
the only problem ive had is a total system failure when i hooked up my water cooling, but the irony of it all is that the water coling had absolutely no failure and is in pristine leak free working condition. But the system doesnt turn on...
__________________
Excuse me, I believe you have my stapler. |
![]() |
![]() |
![]() |
#74 |
Cooling Neophyte
Join Date: Jul 2004
Location: EL, MI - USA
Posts: 6
|
![]()
I happen to just have an almost disaster. I turned my WC system on a week or so ago in the morning and went to take a shower. When i came back my computer was off. I looked inside and saw water dripping from my cpu block barb. Apparently the clamps came loose over time, i swear i tightened them before i came back to college here.
Anyway, my Radeon 9800 pro was shot and my mobo was covered in water. I skipped all my classes that day and found that the agp slot and a shitload of water underneath it, i blew that out with compressed air. I let that dry for a day with a fan along with my radeon. I wiped everythign off with alcohol and cleaned off the white hard stuff from the drying of the water. Then when it was all put together it was not working....and i was pissed. but yesturday when i was about to return the radeon (found out that was broke) i tryed it again and it worked. After some tests i found its perfectly fine. so apparently it needed to dry out more. so i am happy ![]() ...now i just gotta get a replacement for my crappy amd athlon mobile 2600+ that cant get past 2ghz for some reason. |
![]() |
![]() |
![]() |
#75 |
Cooling Neophyte
Join Date: Mar 2005
Location: Washington
Posts: 93
|
![]()
Well, i am one of those awesome people who lost a system for good in his very first attempt at WC. The real problem is that I was new to it and I was rushing to complete the system. Before I put it all togeather, I stupidly didnt use any hoseclamps (sysfx said you didnt have to on thier block, but damnit, just use hoseclaps no matter what) on the waterblock. Naturally while installing the mobo one of the tubes came off and spilled water all over my mobo and processor. So **** I say, and let it dry overnight. Well, after getting it all togeather, I turned it on after a leaktest. The system came on, but there was some wierdness in the display, then after a while it happened - system shut off. No matter what amount of drying it never worked again. later on i confirmed the cpu was toast in another board, and the mobo was fried as well from using known good components to test it. Needless to say, that wasnt a very good week. A good lesson in not getting too much over your head in these things.
The next thing is a near disaster. I basically scrounged a system togeather a few months after my first disaster, and eventually I got the itch to try liquid again. Amazingly i got it running without a problem, using most of my origional gear. But one fateful day, I had my bay res slid a little ways out because I was filling it a little more earlier, but didnt bolt it back down yet. I went out with my friends thinking "that will be fine for now". Well, I stepped into the room and what do I see - the bayres was hanging out of the bay with the inlet and outlet completly cut off, and a BSOD on the screen. First thing that comes to my mind is OH ****, I shut it down without a moment to spare. I notice that the tubes surrounding the processor (just a cheap duron 1.3) were hot. So i put the res back in its position and fire it up, everything cools down immediatly and begins to boot. Whew. Unfortunatly the biggest penalty was the pump, as it was sitting there pumping nothing for an hour the impeller shaft was somewhat damaged and had a rattle ever since, looking at it you can see where it is scuffed up. Lessons learned, this stuff is deffinatly not for the faint of heart. |
![]() |
![]() |
![]() |
Currently Active Users Viewing This Thread: 1 (0 members and 1 guests) | |
|
|