• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

OMG-Sprung a Leak

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.

RDWest2005

Member
Joined
Sep 1, 2005
Location
South Carolina
Guys - Its really my fault
The plastic barbs on the apogee wasn't tight enough
I thought I bared down and was going to strip them
but I didn't have the bottom flat part pushing the o ring all way in and man - i thought it was dead

the water had dripped down onto northbridge block lines
then was dripping onto top 7800

would power on and cut right off in 1 second

I took it all out and wiped it all down
I had an extra pad for the gpu block which was soaking wet

and even took a knife and pushed tissue into pcie slot to see if it was wet and nothing showed up on tissue

still would just cut on and then right off after i done put it all back
i been at this 4 hours i guess

after tearing out all the water cooling about 3 times

i took a blow dryer and blowed into pcie slot about 5 minutes and she booted up

have you guys sprung any leaks like this?
and does it normally fry **** and am i just lucky or what?

i just now got it back running

~RD
 
I am glad you didnt destroy anything. I personally would have given it an alcohol bath and let it sit for a few days before trying it. Esp if I invested the money for a 7800GT

JT
 
Kinda makes me want to switch to the non conductive liquid now

it should be ok now - i PRAY ^
i'm not over clocking for a few days to make sure its good and dried out

~RD
 
I've had leaks like that before. A few times. I let it dry overnite, next day worked fine. However my last leak (barb on my Danger Den chipset Z leaked like a mofo) cooked my mobo and my cpu :(
 
Sir Golitech said:
WOW :eek: dude, I would say you were lucky. Personaly I don't have a water cooling system, but I was giving it some serious thought on getting one. My one fear of all of it, is getting a leak. Now you got me thinking twice.

You should always run the loop outside of your system for some time to see if it leaks and like RDWest2005 said, he didn't. If you want to be safe you have to take the propper percautions.
 
thorilan said:
0 leaks for me and i build a LOT of systems
That's cause you're supa sweet like that.

How many times did you try to turn it on and have it short? I hope if I ever get a leak I'm lucky enough to not break anything.
 
This isn't a very big problem if you use distilled water, as it doesn't conduct. Frankly it seems like using anything BUT distilled water is a mistake in any WC situation.
 
RD, what were you using for coolant? Have you checked it with an ohmmeter? My guess is that it's more physics than luck. :)
 
This isn't a very big problem if you use distilled water, as it doesn't conduct.

It does... moreso than "inert" coolants such as MCT-5 and FluidXP+... hence they exist... but at the end of the day no coolant in use in these systems is "non-conductive" in a literal sense.

Distillate quality may vary depending on the quality of the water source. The pH and conductivity of distilled water changes when exposed to the atmosphere. This is a result of carbon dioxide absorption that is temperature dependent, and is not an indication of a change in overall purity.

We've been using bottled distilled water for about 10 years for both bioassay and micro. Conductivity can range from as little as 1.0 µS to as high as 8.5 depending if the 5 gallon bottle sits around for awhile. We take the conductivity from each bottle that we open up.

Distilled water is purified using a distillation process, which is an excellent way to remove contaminants. The quality of distilled water found in grocery stores can vary, however, so it is a good idea to try a different brand if results are questionable.

Just a few quick quotes from a google result or two

Long story short - you cannot guarantee conductivity of distilled fluid once you take the top off the bottle. Better to assume NO coolant in use in your system is non-conductive... assume all are conductive - final result - you take more precaution during assembly to avoid leakage.

My recommended procedure...

Whenever you get a leak, just power it off and leave it off. Don't even try to turn it back on. The potential shorts created by the coolant could cause more damage when u try turning it back on then any already done.

Locate drowned components, remove, mop up excess with loo roll or similar, and place somewhere warm (airing cupboard) and dry and leave there for 3 days or so.

After 3 days, rebuild, all should be peachy...

The "inert" coolants tend to evaporate-off slightly, so dry out quicker, but other than that can see no tangible benefits to them other than one includes a UV exciter, and neither require end-user blending with a separate additive.
 
Last edited:
First off, if you have a leak, turn the thing off and unplug it! Trying to power it on without making sure the water is gone is not a good idea. Repair shops bathe computer parts in distilled water sometimes if they get really nasty. It's not the water that's the problem, it's the water conducting electricity and shorting stuff out.

Bathe whatever got wet in rubbing alcohol and use a hairdryer/fan to dry it thoroughly overnight. Rubbing alcohol will take water with it when it evaporates. This really is a good idea to do especially when you have slots that get wet as it can be a problem to get all of the water out of the inside of the slot.

I leaktest with everything in place, not with it outside of the case, usually overnight. I use a spare PSU to power just the pump and use only distilled water while leaktesting. I haven't had a leak on my systems, but if I were to have one while doing this, it wouldn't do any damage and could be easily fixed.

I watercooled two computers for a friend of mine, and on one, I used a non-barbed copper elbow to make a sharp bend. Even with clamps, after about 3 weeks it worked loose (of course it didn't help that he forgot to turn his air on and had it over 100 degrees in his shop during the summer one day). The entire bottom of the thing got bathed in water/coolant. He was working on it when he heard the leak and cut it off right away. After calming him down given that he was freaking out, I took out the MB and literally bathed it in rubbing alcohol and set it in front of a fan overnight. The next morning I plugged it up and it fired right up just fine. Of course I learned never to use unbarbed fittings again.
 
Last edited:
thorilan said:
0 leaks for me and i build a LOT of systems

same here but i still test every loop outside the tower before i put it in.

sorry to hear about the scare there though RD i just won an apogee and ill keep this story in mind when im putting the barbs in.
 
5|*42 said:
That's cause you're supa sweet like that.

How many times did you try to turn it on and have it short? I hope if I ever get a leak I'm lucky enough to not break anything.

sheeewww - i'm lucky dude - i hit power over and over - i didn't know it was leaking - after removing the top card about twice and the top loops, and hand drying with towel, it still would just turn on and blink off...

then i pulled the top card out of slot - still hooked up and in loop but power unplugged to it -

it fired up
so i knew it had to be shorting in the pcie slot
so i used a blow dryer into slot and then it fired right up

RD, what were you using for coolant? Have you checked it with an ohmmeter? My guess is that it's more physics than luck.

i'm using that GM DEX COOL Prestone orange coolant for the late model cars with aluminum blocks and pastic intakes and stuff.

about 1/3 coolant and 2/3 of just tap water

my 840 dual core never got over 32c while running 3dmark05 at 3840ghz

~RD

PS - another question coming in a sec
 
another guy at the eoc forums had a similar problem with the barbs on the apogee

I tighten the barbs down about like you do on the BIX rad - you only barely tighten to o ring or it will push it out

so thats how i did with plastic barbs but a wee bit tighter

but now i really bared down and almost have the round flat part above threads flush with block

meaning the o ring is forced down into its groove

the apogee dropped my cpu temp from the thermaltake block about 10 degrees lower - so the apogee is a real good block

~RD
 
This is why I always use metal worm clamps instead of the plastic clamps. No chance of breaking or getting loose. Never had a leak.
 
i use all fuel line worm clamps

i only have the plastic clamps on pump

but i just PM a guy about how to clean up the mess now - i think its left a slick feeling film over it all

i don't know whether to soak the whole board under alchahol and card then blow them off with air compressor and let dry or what

~RD
 
I'd just wipe it down with alchohol, paying special attention to the card edge contacts and sockets. But I've never soaked my mobo with antifreeze, so don't take my word for it.
 
Back