New Issue with Water flow/air bubble in CPU Block

[animalmom]

Hi, I have had my system stable and O/C for about 2 months. I have had a small (5mm?) Air bubble in the Gpu block. Yesterday I got a CPU temp shutdown, looked at the innards and it looks like this now and there's a bunch of gunk in the CPU Block, the fluid is cloudy (EK Green fluid) and my CPU temps are high/

I turned off the aura for the photos and turned off O/C.

Is this just a normal bleed issue - what should I do. I intend to drain it and then take the CPU block off an clean it.





When I turn the rig on its side the pump starts to suck air and I know that's bad - how do I get rid of the bubble?

 

[WALL-E]

While I can only speculate on the fluid "gunk" I suspect that the most likely culprit is galvanic corrosion. It is almost certainly worth washing out your loop and trying an anti-corrosive agent in the next fluid. I swear by diluted automotive antifreeze for how simple and low-maintenance it is, although the rest of the water-cooling community looks down on this. Air bubbles will greatly speed up the corrosion rate, as the oxygen will pull electrons off the metal forming electrolyte.

On the topic of bleeding a system, I know there are several good guides on this forum. I can offer a personal technique which I have found successful even with very convoluted loops. Of course designing your loop with a high reservoir/tee junction is always the best option, but this isn't always possible.

1: Fill the system as high as it can possibly be filled, all the way up to the top of the fill line.

2: Seal every opening to the loop.

3: (Optional) Briefly and intermittently turn the pump on while performing the next few steps. You can do this with an external power supply or by shorting the GREEN lead on the atx connector to any BLACK lead. This has the potential for damaging or reducing the lifespan of the pump if it runs dry, sucks air, or spins at too high of an RPM.

4: Starting with the air bubble that is closest to the reservoir/tee junction but still "behind" the junction, rotate the computer so that UP is in the direction of the intended fluid flow (relative to the bubble). The bubble will travel along the loop until it reaches the next bend. Keep rotating the system until the bubble reaches the reservoir/tee junction. Refill the system if necessary. Optionally briefly run the pump again.

5: Repeat with the next closest air bubble.

 

[animalmom]

This is a really great and comprehensive response. I will give it a go cleaning and removing the air

 

[trents]

I think you need to ask the question, "Why is there air in the system in the first place"? If you don't find an answer to that question you will be in the same place 2 months after cleaning it.

Concerning the gunk, could this be just frothed coolant?

 

[animalmom]

I think you need to ask the question, "Why is there air in the system in the first place"? If you don't find an answer to that question you will be in the same place 2 months after cleaning it.

Concerning the gunk, could this be just frothed coolant?

Thanks - I am not sure why there was air but I know why its hot now. Yuck.




Sitting in the hot soap bath - I think there was just an air pocket and crap grew because there was air

EKWB says no silver coil or biocide if you use their fluid - what has been everyone’s results with that - I doubt it’s been two months since brand new

 

[Zerileous]

I've had great success with their clear fluid. This could be gunk due to dye or corrosion or junk from parts, it is hard to tell from the pictures. Was this a transparent or an opaque "pastel" fluid? Did you mix a concentrate with distilled water? Did you clean and flush the radiator and other components. Cleaning the radiator, filling and soaking with boiling water, then shaking vigorously for 5-10 minutes once it is cool enough the handle is usually sufficient.

Regarding the air, you're sure those bubbles were not in the blocks when you first started the loop? It's possible that they worked their way out of the rad if you did not change the cases position after filling to help work bubbles out.

 

[animalmom]

Thanks. It was their green transparent fluid. Regardless we can table this for awhile as I took the block apart and now the big gasket won’t fit -it has expanded and it’s 1cm too long now. I froze it and still didn’t work

What’s odd is that the gasket is not shaped like the space it fits In. Just a big ring.

I’ve asked them for more gaskets and for advice on why the block is essentially disintegrating

Yes I cleaned the radiators prior. The whole thing is weird. There must be some contaminant in the loop or I made some other mistake

 

[maxfly]

Thats looks like the coolant/dye started to shear or seperate with the majority gathering in your cpu block. I would show those pictures to ek if your block doesnt come clean.
Your going to have a hell of a time getting all of that out of the rest of your loop.

 

[animalmom]

Thanks. Should I run distilled water through the loop and keep emptying and refilling?

 

[WALL-E]

Gross! The green stuff is malachite and the black stuff is copper oxide. Mild acid will work for both of these! Since that is (I assume) an electroplated piece of copper, and because the block probably has tight tolerances for reassembly, any abrasive is out. I'll bet you could have good luck with a vinegar soak and then a spray with pressurized water (finger over the garden hose?) Just make sure you rinse it well since trace vinegar will destroy your rubber seal. Since you've already disassembled the loop, you can probably just wash each piece individually. Clean. Pick a correct fluid (listen to Zerileous). Bleed properly. Back in business.

 

[thenrz]

Having air in the system at all is an issue. I'm not going to go into anything else as it's been covered in detail above, but you certainly didn't purge the system correctly. Once you get everything cleaned and reassembled, it will be necessary to fill the system much more carefully this time. Physically rotating the case with the pump running into several orientations will augment the relative water level at different locations and displace the air with fluid.

What additives are you using? Galvanic corrosion inhibitors, fungicides/bacteriacides, and a slight lubricant (usually glycerol) will be in most retail additive products. It's important to utilize these to keep your loop running properly.

 

[animalmom]

Gross! The green stuff is malachite and the black stuff is copper oxide. Mild acid will work for both of these! Since that is (I assume) an electroplated piece of copper, and because the block probably has tight tolerances for reassembly, any abrasive is out. I'll bet you could have good luck with a vinegar soak and then a spray with pressurized water (finger over the garden hose?) Just make sure you rinse it well since trace vinegar will destroy your rubber seal. Since you've already disassembled the loop, you can probably just wash each piece individually. Clean. Pick a correct fluid (listen to Zerileous). Bleed properly. Back in business.

Thanks, they are nickel plated copper per the website - si I can soak it in vinegar to clean? soapy water did almost nothing.

I have asked for new rubber gaskets so I have a week or so to work on it.

- - - Auto-Merged Double Post - - -

Having air in the system at all is an issue. I'm not going to go into anything else as it's been covered in detail above, but you certainly didn't purge the system correctly. Once you get everything cleaned and reassembled, it will be necessary to fill the system much more carefully this time. Physically rotating the case with the pump running into several orientations will augment the relative water level at different locations and displace the air with fluid.

What additives are you using? Galvanic corrosion inhibitors, fungicides/bacteriacides, and a slight lubricant (usually glycerol) will be in most retail additive products. It's important to utilize these to keep your loop running properly.

I am using only EKWB coolant, no additives. They said that’s what you should do. No biocide, no silver coil - their fluid has biocide. I ran it with distilled water, drained it and then added the coolant. I cleaned the radiators.

I will certainly do as you suggest and report back. In the meantime I’ll run it with distilled water to see if I can clean out the system more.

 

[Zerileous]

We would only recommend using the clear fluid. However since this is their ecosystem of products they should hopefully make things right. I have seen 3d white toothpaste recommended

 

[maxfly]

Ive been using clr to clean really dirty blocks and rads for years. Fill the block up with it and let it sit for 5 or 10 minutes. Rinse it out really well with tap water then distilled and thats it. Most if not all of the crud should be gone. Use it the same way with your rads. Just be sure to flush them until no more bubbles come out. It wont damage nickel or copper.
Clr is the only way ive found that will truly clean crud out of rads and blocks.

 

[animalmom]

Ive been using clr to clean really dirty blocks and rads for years. Fill the block up with it and let it sit for 5 or 10 minutes. Rinse it out really well with tap water then distilled and thats it. Most if not all of the crud should be gone. Use it the same way with your rads. Just be sure to flush them until no more bubbles come out. It wont damage nickel or copper.
Clr is the only way ive found that will truly clean crud out of rads and blocks.

Thanks,

What is CLR?

Since my blocks are nickel plated copper EKWB says dont use vinegar - but I on't care if the nickel gets removed anymore, they will always be ugly now so I just want them cleaned.

If I use vinegar and the nickel goes away and there is just copper will the blocks still be ok to use? I will replace all the gaskets.

- - - Auto-Merged Double Post - - -

We would only recommend using the clear fluid. However since this is their ecosystem of products they should hopefully make things right. I have seen 3d white toothpaste recommended

Thank you - will do that in the future

 

[WALL-E]

I've gotta walk back my advice to try vinegar. I didn't think about it removing the nickel plating. While it will react slowly with nickel due to its relatively high standard reduction potential of -0.25v, the plating is thin enough that apparently it will strip right off. I wish I could give you good advice, but Ill search around and see if I can find something.

Edit: Found several folks online have tried dilute vinegar on a toothbrush and not stripped through the nickel. Some people have even had luck with toothpaste!