Dual pump loop issues with leaks
Dual pump loop issues with leaks
I've been developing this setup for two years now. After gathering everything I needed, I encountered some problems and am looking for guidance. The leak from the bottom of Res A happened because both pumps were running at full speed, creating back pressure. Adjusting them to 20% (1500 rpm) resolved the issue. However, I've noticed another problem: Res B is slowly draining into Res A, pushing the fluid upward and causing leaks at the top. I've tried increasing the pump speed on Res A to move fluid faster, but it didn't fix the situation. I'm seeking advice without wanting to set up a dual loop, as I'm unsure how the lines would work. Thanks for your help.
Begin with separation. Empty the loop (it's not easy), insert a leak tester into Res B and verify the connections are secure. Once confirmed, seal the termination points of each segment coming from Res B and ensure they’re airtight. Keep adding more components back into the sequence until you hit the first leak. It takes time, but it’s a method that always succeeds. I get it—spotting leaks can be frustrating. I’ve had to replace a radiator POM piece with damaged threads, which required careful inspection since the leak was minimal. The soapy water technique didn’t cut it. Also, I appreciate your chassis—what model is it?
You could have an overflow equalizer at the top if you secured tops for the res that included fittings, helping balance the drain between the two issues.
I just noticed I missed the main issue. Sorry about that—it happens sometimes. You shouldn't be seeing this if both pumps are running at the same fixed speeds, moving fluid in the same direction, and both started with full tanks. Under normal conditions, pump B should be pushing more fluid than pump A is drawing it in. If both reservoirs were empty and there was no cavitation or turbulence, this shouldn<|pad|> to not occur. It’s definitely a challenging situation.
Two pumps sharing a single reservoir in a loop seems risky at best. If they were originally connected in series, switching them to parallel might improve balance, though I’m not sure why you’d need two pumps in the first place. Beyond that… it’s a serious concern. There’s still a significant risk of leakage, but the lack of sufficient pressure makes it hard to overcome the seal. The seal is defective—replace or repair it.
I concur. The main purpose of running two pumps is only if you have separate circuits for CPU and GPU. Beyond the aesthetic appeal, connecting them in series isn't practical and may lead to problems. I see you've already prepared everything properly, but if I were you, I'd remove one of the D5 pump's internal components and use a single pump instead.
Of course, from a fluid dynamics perspective you should divide the loop prior to the pumps, install a backcheck valve on the outputs, and then reconnect them. Even so, this still lets "pump A" push pressure into the "pump B" reservoir throughout the loop without a proper "ballcock" mechanism to counteract it (which might just involve reducing the opposing pump output when a reservoir exceeds a specific level). If you're working with custom pump control, it's worth noting the pumps' intended operating ranges and driving requirements.
Hey there... thanks for the support so far. I've already swapped out all the o-rings in the bottom section. I also ran the EK-Leak Tester on the loop, and it looks good. The loop is fine, but for some odd reason, both the pump and the reservoir seem to leak slowly. It's also curious that pressure builds up when you try to remove the tops—something stuck from the pressure or the running system is at play. I was thinking about adding dual pumps for backup, and I wanted a loop with a double radiator for better cooling. Also, just a heads up—this setup uses a Corsair 900D Super Tower Case.