Custom loop with D5 Vario that may be airlocked
Custom loop with D5 Vario that may be airlocked
Moatt clarified the situation, noting the reservoir plays a role. He observed the air bubble behavior and its stopping point, mentioning it maintains pressure without fully pushing through. He discussed possible solutions like adding another radiator or a second post, but was unsure about feasibility without a reservoir. He confirmed his current pump is functional but lacks sufficient strength for the entire loop, questioning if the issue lies in the system setup or equipment condition.
That's what I considered. I'm curious if there might be an obstruction in the actual port on the pump or reservoir. Water exits without issue, but no flow comes back in. At this stage, it seems like the only option is to retrieve the reservoir and inspect both the reservoir and the pump.
Moatt explained his thoughts. He questioned whether something might be obstructing the actual port on the pump or reservoir. Water should flow out without issues, yet nothing seems to come back in. He believes it’s time to inspect everything thoroughly, right?
Ok, so not the pump. Bought a brand new D5 Vario kit and swapped it out. Still doing the same thing. I also found out that if the pump is on, I can disconnect the short hose from the graphics card to the cpu (disconnected at the cpu block) and suction will hold the water in place...not even a drop comes out. I filled it up with pure distilled water directly into that short hose and burped all air bubbles completely, and there still doesn't seem to be any movement.
Picture
The water between the cpu and graphics card is starting to change colors a little but it's still not mixed like the water in the rest of the loop. And my flow indicator spinning wheel doesn't move at all. Shouldn't there be movement? I get that we're looking at laminar flow here, but seems like the key word here is "flow". Thoughts?
Yes, it's hard to understand what's happening there. I plan to simplify the loop to see how it behaves. Could there be more than one "in" port on the reservoir or pump unit? Is one side of the loop facing the "out"?
And as usual, I'm making a mistake. The top outlet on the reservoir is labeled "out." I thought that was an odd design, but I connected it from the "out" side of the reservoir to the "in" port on the CPU block. Eventually, I chose to ignore the marked out port and switched it so the flow comes in through the reservoir's in port when the CPU block is active. The entire system operates like a high-speed jet ski. The flow indicator wheel, which had been completely still, was now spinning rapidly.
The most puzzling part was that I had set it up this way for nearly a year without noticing any issues. There was never been any overheating. It wasn't until I added the flow indicator to test the fluid that I realized something was wrong.
Vellinious :
Examining the images here, it seems the "out" marked on the reservoir should connect to the outlet that leads into the CPU block.
https://static1.squarespace.com/static/5...ayres2.pdf
This was my original setup, but it appears the outlet is incorrectly marked or the inlet on my raystorm CPU block is off. If I connect a hose from that outlet to the CPU block, the pump operates and nothing flows out through the other end of the CPU block. After swapping them so the reservoir inlet connects to the CPU block inlet, everything works perfectly.
I left the pump running overnight, and everything remained fine. The reservoir stayed cool now, whereas before, if I ran the pump for just 5 minutes the reservoir would get warm (likely because no water was exiting, so the pump kept spinning).
It's strange to me that the outlet port is positioned at the top of the reservoir. From a fluid dynamics perspective, it seems more logical to place the outlet at the bottom, ensuring the pump can still draw water if the level drops below the top port. With the outlet at the top, the pump would be completely dry if the level fell below that point. Or at least that's my assumption.