Water leaked all over GPU and motherboard

vergdm

n00b
Joined
Feb 23, 2017
Messages
11
Hi all, first post here. I was leak testing and found two of my EK rotary fittings were leaking badly, one on the GPU and one on the CPU block. I immediately turned off the pump and put more paper towels down but it was a decent size puddle of water, there was water in the ram slots and basically the top half of the motherboard was soaked. My PC was completely powered off, only had the pump connected via molex.

I inspected everything and there appear to be stains from the water on the video card PCB. I tried cleaning the water stains with a microfiber cloth and it didn't really do much. Motherboard seems to be free of water spots.

I took apart everything and allowed it to dry in front of a fan for four days. Am I totally screwed with those stains on the PCB like that? Don't have photos of the damage at the moment, I can upload them later this evening. Thanks in advance for any advice.

My setup:
SLI 1080 Ti FE
ASUS Maximus IX Formula
i7 7700K
64GB G.Skill RAM
EK Surpemacy EVO
EK PETG tubing
EK TItan X waterblocks (purchased before the 1080 Ti block became available)
CaseLabs SMA8
Hardware Labs GTR 480 radiator
 
You did the important thing and dried everything out quickly. Since you've let it COMPLETELY dry now, it's probably time to hook everything back up and fire it up to see where you stand. In my experience, if everything was off when you started and you've done a good job of drying it out, it should be OK. But there's only one way to know for sure...

Some isopropol alcohol mixed with distilled water may help clean up the PCB, but that might be a secondary concern.
 
You did the important thing and dried everything out quickly. Since you've let it COMPLETELY dry now, it's probably time to hook everything back up and fire it up to see where you stand. In my experience, if everything was off when you started and you've done a good job of drying it out, it should be OK. But there's only one way to know for sure...

Some isopropol alcohol mixed with distilled water may help clean up the PCB, but that might be a secondary concern.

I'm also on the fence about using any other liquids since it seems like the stains are in there pretty good. I'll put everything on the test bench tonight and give it a whirl. thank you!
 
Yup I agree. You should be go to go. Just check in between the ram slots to be sure no water droplets are still hiding. Use compressed air to blast out any unwanted, then fire that sucka up.
 
So finally had a chance to get everything powered on and getting a code 55 on the motherboard, getting no picture. tried removing ram and using different slots, also tried different ram altogether and no dice. tried loosening the CPU block a bit, nada. only thing I haven't tried yet is a different CPU, will order another 7700k and see what happens.
 
So finally had a chance to get everything powered on and getting a code 55 on the motherboard, getting no picture. tried removing ram and using different slots, also tried different ram altogether and no dice. tried loosening the CPU block a bit, nada. only thing I haven't tried yet is a different CPU, will order another 7700k and see what happens.

Did you try with only one ram stick?
I think code 55 is "memory not inserted", if I can remember correctly.
 
After several reboots and code 55 every time (even w/ one RAM stick installed), I let it sit for a few days again and tried the RAM trick again and voila! It finally booted into the OS. Tried two RAM sticks in slots 3 and 4 and it still worked. It's when I try to use RAM slots 1 and 2 that the board won't post, hopefully I can RMA it. Thanks everyone for the input, all is working fine otherwise. Seems like I dodged a bullet this time around.
 
Is this a new build? Or was the motherboard working previously? Distilled water is non-conductive, so it shouldn't break memory slots while your computer is without power. Technically speaking you should be able to soak a running computer in distilled water without doing harm, but let's not go there right now.

I wouldn't think twice about RMA'ing it if that is the case.
 
Is this a new build? Or was the motherboard working previously? Distilled water is non-conductive, so it shouldn't break memory slots while your computer is without power. Technically speaking you should be able to soak a running computer in distilled water without doing harm, but let's not go there right now.

I wouldn't think twice about RMA'ing it if that is the case.

Pure distilled water is non-conductive. It's not non-corrosive. It also becomes conductive as soon as you get a bit of dust or dirt in the mix. Using distilled is far, far better than tap water, but it's not perfect.
 
Pure distilled water is non-conductive. It's not non-corrosive. It also becomes conductive as soon as you get a bit of dust or dirt in the mix. Using distilled is far, far better than tap water, but it's not perfect.
Corrosion is not really an issue when the board is littered with goldplated connectors. The odds of contaminated distilled water damaging unpowered electronics is also highly unlikely. If you left your motherboard soaking in water for weeks then we could start thinking about damage from corrosion on traces or something, but not from accidental spray from a pressure test. We could start down a trail of all sorts of theories of course, what was in the loop that contaminated the water and could cause damage? Could be metal particles in the radiator for all we know. Could be cosmic rays... Sun flare that hit at an inconvenient time? ;)

But that doesn't change the fact that if this is a new motherboard that had never worked prior to the accident I'd still RMA it in a heartbeat.
 
Corrosion is not really an issue when the board is littered with goldplated connectors. The odds of contaminated distilled water damaging unpowered electronics is also highly unlikely. If you left your motherboard soaking in water for weeks then we could start thinking about damage from corrosion on traces or something, but not from accidental spray from a pressure test. We could start down a trail of all sorts of theories of course, what was in the loop that contaminated the water and could cause damage? Could be metal particles in the radiator for all we know. Could be cosmic rays... Sun flare that hit at an inconvenient time? ;)

But that doesn't change the fact that if this is a new motherboard that had never worked prior to the accident I'd still RMA it in a heartbeat.

I'm not going to argue the RMA point. That's really an ethical matter, not a technical one. Is it morally right to request an RMA when you know the MB had water spilled on it without telling the vendor?

On the technical side though, motherboards are complex beasts with lots of very tight clearances between pins. Gold plating is used on the CPU and memory pin contacts, but where those attach to the board are simple tin-solder and copper. The boards aren't sealed in most cases, so a little water can even soak into the PCB material itself, and then sit right against a copper layer and cause corrosion and swelling, even if the board appears dry. It wouldn't be inconceivable at all to have some dust or debris washed under a memory slot, or a PCIe slot, or a surface-mount controller. It doesn't take much to throw things off, it doesn't even need to be a direct short, just enough to give you some capacitor effect.

I did my EE Master's thesis on making digital logic resistant to radiation (cosmic rays), so I can say with some certainty that's not a likely cause here.
 
Motherboard was brand new out of the box when the leak happened, so I can't say if the ram slot issue was present before the leak. The fittings, radiator, pump and tubing were all new as well, the only thing that was previously used (from my last build) were the GPU and CPU blocks (and those blocks were only a few weeks old).

The replacement Formula mobo arrives today. Now I'm really wondering if this board had the ram slot issue before the leak - I didn't actually test anything prior to building the loop around it, bummer.

On a side note, I purchased three motherboards because I couldn't decide which one I wanted: Maximus IX Hero, Formula and Extreme. The Hero had issues posting at first, but eventually worked after 3 or 4 reboots. Formula has the RAM slot issue, and also took a few reboots for it to finally post. The Extreme mobo is the only one I didn't have issues with out of the box.
 
do a test boot before you put it all together.

Not sure what you mean? The motherboard that had the leak (Formula) is up and running now, completed a few RealBench hour tests since. No issues other than the two ram slots not allowing the system to post.
 
test it out of the case before putting it all together. that way you can check its ok, confirm no ram issues, so you don't have to tear it apart if there is.
 
Oh gotcha. I always do any testing on a bench before putting into the case. And I haven't unboxed the replacement Formula motherboard yet, will do some testing this evening.

Have also been reading more into this ram slot issue on various ASUS motherboards. Folks without any water involved were having to RMA their motherboards too due to one or two ram slots preventing post.
 
Motherboard was brand new out of the box when the leak happened, so I can't say if the ram slot issue was present before the leak. The fittings, radiator, pump and tubing were all new as well, the only thing that was previously used (from my last build) were the GPU and CPU blocks (and those blocks were only a few weeks old).

The replacement Formula mobo arrives today. Now I'm really wondering if this board had the ram slot issue before the leak - I didn't actually test anything prior to building the loop around it, bummer.

Using an untested motherboard in a fresh water loop build, you are much braver than I.
 
Not sure what you mean? The motherboard that had the leak (Formula) is up and running now, completed a few RealBench hour tests since. No issues other than the two ram slots not allowing the system to post.

Yep. If water becomes a puddle during a leak test, it really seems that you should have used more paper towels from the start. I also like to use a few ziploc bags in the mix so that dripping will flow over those instead of inwards onto the board.

But I agree, I'd RMA the board anyways noting that you don't know if the slots were broken by the water. For a loop test it seems unlikely, unless the water has caused further damage to the internals of the board.
 
Back
Top