I think the answer would depend on whether or not your losing a lot of water through evaporation….

@Chris Wilcox Hard to say. There’s a make-up water supply but it’s not metered independently. I would guess there isn’t a ton of evaporation happening, but definitely a non-zero amount.

Reason I ask is, based on my limited understanding of chilled water systems, I expect to see positive pressure readings on pump suction side and discharge, maybe around 40 psi each. In this case, there’s negative pressure (~20 in.Hg) on suction and around 20 psi on discharge. This splits off into three chillers, each of which has a more extreme pressure drop as they get further from the pump. By the time the chilled water reaches the load, pressure is at 15 psi inlet and 12 psi out. So we’re getting water back at the pump, but not enough pressure to keep it from going into vacuum. I suspect the water holding tank being unsealed is causing a loss of pressure that would otherwise be recirculated.

Any thoughts on this?

I would suspect a partial blockage somewhere in the system. Try back flushing the system with the outlet set to discharge into some sort of collection vat so the fluid can be inspected for particulate. 

I would start looking at the area with the greatest pressure loss. The open tank may affect slightly, but it is a holding tank not a pressure vessel, correct? The pump should still do its job. I would suspect a restriction or blockage (like Justin said) somewhere in the chiller area. If the pressure is rated for 40 psi, I would check the pump.

@Mike If you have -20 in Hg at the suction of the pumps, it is cavitating badly, been long time?, replace.

When was the last time the loop was flushed and cleaned?

What type of water did you fill it with, tap, softened, decimalized? Do some checking of the water, pH, conductivity, colour. Maybe send it for metals.

Are you putting in any inhibitors? 

There are companies that will do flushing and refilling. Seek assistance soon, as this summer you are not going to have enough cooling.

Sounds like you have a plugged line somewhere or a pump on the verge of failure. Do you have any in-line filtration? 

We use chilled water and the amount of chemicals our city uses was causing corrosion issues which was creating scale and blockages. We installed an in-line canister strainer with 10 micron mesh and a second canister fitted with a wool “sock”. We monitor suction and discharge pressure and when we see 15-20 psi differential, we know it’s time to replace the sock and clean the basket strainer.