How do I size a centrifugal pump for a closed-loop cooling system?

[Jeffrey23]

I’m trying to size a centrifugal pump for a closed-loop cooling system, but I keep getting stuck on the system curve calculation. My theoretical head requirement seems way too high compared to the actual pressure drop I measured in my test loop, even after accounting for all the fittings and pipe length. I’m wondering if my assumption about the fluid’s specific gravity is throwing things off, or if I’m misapplying the Darcy-Weisbach equation somewhere.

[JohnJ]

I have done this and ran into a density trap. Our loop sits warmer than room temperature so the SG you plug in for water isn't accurate anymore. I pulled SG from the fluid property tables at the actual operating temperature and reworked the head and the curve looked closer to reality. Also I double checked that hydrostatic head from elevation was included in the right direction

[Harper.W]

Darcy Weisbach got me tangled once. I kept using the wrong friction factor because of pipe roughness and Reynolds number. I switched to Swamee Jain with the correct roughness and gave the f value the right units, then rechecked the losses. It didn't fix everything but the mismatch shrank a bit

[Olivia_J]

Air in the loop bites you more than you expect. I tested with a nearly full loop and the measured drop felt too high until I bled a bit and watched the pressure stabilize. After that the system curve didn't look phantom anymore. Could be an air trap somewhere you're not seeing

[AnthonyH]

Maybe the real issue isn't the math at all but the actual flow you are delivering If the pump operating point sits on its curve somewhere you might be chasing the wrong intersection Are you sure the test rig is delivering the GPM you think it is