Question Constant air bubbles in loop

[Ronnie30]

Hi,

I have a custom waterloop that I've been running for about 1.5 years now. It was last drained and refilled about 6 or so months ago. I run a res/pump combo to a cpu block, to the rad and back to the pump. As usual, over the first few days of a new fill, some air bubbles show up, and I simply top up the reservoir. I also rotate the PC all around to eliminate as many of the air pockets as I can.

I've previously asked on these forums about air bubbles showing up much later, as I would sometimes hear and see a couple of bubbles here and there, even weeks and months afterwards, but was told that this is normal and isn't an issue, if temps remained ok. From today, I started contributing to the fold@home project which is pushing my CPU and GPU to 100% utilization. Currently, temps are still fine, but I had several minutes where I started getting a massive rush of air bubbles travelling to the res, which added nearly 1cm of additional air in the res (see image here- ignore the level of dust). This is also the first time that I can actually see the water line at the top of the res being disturbed; I'm guessing it's due to the pump working harder than normal, even though I didn't witness this back when I was stress testing. It seems to have stopped producing air bubbles now, but the sheer amount of air that came through has me worried. Could this all have really been in my rad? Or is it possible I have an opening in the top of the rad that is adding air to my loop over time, and is only visible now due to the pump working harder? Basically, should I be concerned that there's an issue or did I just do a poor job of bleeding it?

 

[grimfox]

There are really just a couple options. 1) you did have a bubble stuck in there, and the added vibrations from high flow and fan shook the bubble loose. 2) you have a leak somewhere. 3) The prolonged high temps are causing increased evaporation.

Did you run your stress tests since your last refill?

 

[Ronnie30]

I ran a short stress test, but the folding had my CPU running at 100% for 90-120 mins. The temps were fine.

I'm not even sure where a leak could exist. There's no openings on top and there's no liquid leaking internally. I may take the system apart in a couple of months and give it a full clean and rebuild.

 

[rubix_1011]

Reservoir should have fluid level higher than what you show. It doesn't have to be 100%, but having more liquid volume allows for cavitated air bubbles to dissipate up and collect rather than being pulled back into the coolant flow.

The tiny bubbles you see are more common early in a cooling loop and should dissipate out over the course of several days or a few weeks.

The rush of air and the drop of the reservoir level tells me that you likely have pockets of air trapped somewhere in your loop, likely a radiator or GPU block. The coolant level really should not be 'dropping' like that - might be a good idea to vent the cap of the reservoir and allow the loop to run for several hours to see if this normalizes at all, and topping off once it does.

The idea is that you want to allow the pressure to vent so that air can more easily be pushed out and remain out with only minimal headspace of air at the top of the reservoir.

 

[Ronnie30]

Thanks Rubix. Since no more air travelled through, I assume it's mostly or completely bled at this point. As mentioned, I'm going to be draining it again in a month or two, so I'm just going to leave it as is until then. My primary concern was that there was a leak that I wasn't seeing, but that doesn't seem to be the case.