After the measurement points are corrected and flow is verified independently, differential pressure becomes a useful indicator of chiller water-side condition instead of an isolated number. The installation has 13 water-cooled liquid chillers, including older units with limited engineering data, so establish a repeatable baseline before changing valves, pumps, or controls.
Which delta should be checked?
Separate hydraulic performance from cooling performance. Water-side differential pressure, written as ΔP, is the pressure difference between two points in the same circuit. Temperature difference, ΔT, is the change in water temperature across a heat exchanger. They answer different questions.
| Measurement | Points | What it indicates | What it cannot prove alone |
|---|---|---|---|
Heat-exchanger ΔP
|
Water inlet and outlet of the same exchanger | Hydraulic resistance at the current flow and fluid condition | Actual flow without a valid pressure-drop curve |
Balancing-valve ΔP
|
Valve test ports | Flow when combined with the correct valve position and chart | Heat-exchanger cleanliness by itself |
Water ΔT
|
Temperature inlet and outlet | Thermal change across the exchanger | Cooling duty without flow and fluid-property data |
| Direct flow | Suitable straight pipe section or collection point | Volume or mass flow, depending on the method | Cause of abnormal resistance without pressure measurements |
Follow the water. Begin at the pump discharge, identify each branch, locate valves and strainers, and then identify the inlet and outlet of the exchanger being tested. Both pressure readings must belong to the same water circuit. A condenser-water reading cannot be combined with a chilled-water reading.
Where can a differential-pressure reading go wrong?
Layer one first: inspect the physical measurement path before interpreting the number. A pressure instrument only reports the pressure delivered through its hose, port, impulse passage, or sensor connection. Blocked test ports, trapped gas, liquid loss from hoses, mismatched instruments, and reversed high/low connections can create a plausible but false result.
- Trace both pressure connections to their actual process locations.
- Confirm that both points are on the same side of the refrigerant boundary and across only the intended component.
- Inspect hoses, fittings, test plugs, and ports for leakage, obstruction, or trapped gas.
- Use one differential instrument where practical. If two gauges are used, check their zero and resolution before subtracting readings.
- Record pump state, valve positions, active chiller branches, and water temperature with every reading.
Placement matters. A reading taken across an exchanger plus a strainer and isolation valve is the combined loss of all three. It cannot be treated as exchanger pressure drop. If a shared header changes pressure while branches start or stop, the measured ΔP can change even though the tested exchanger has not fouled.
Which flow-checking approach fits the available equipment?
Four approaches are practical when original chiller data are incomplete. Select the method by access, required confidence, and whether a component-specific calibration is available.
| Approach | Required information | Strength | Recurring limitation |
|---|---|---|---|
| Heat-exchanger pressure-drop calculation | Geometry, fluid properties, condition, and a valid calculation method or performance curve | Relates measured resistance to expected hydraulic behavior | Old, fouled, modified, or undocumented exchangers may not match calculated clean-condition performance |
| Clamp-on ultrasonic flowmeter | Pipe material, dimensions, fluid, suitable mounting location, and correct transducer setup | Measures flow without opening the pipe | Poor pipe condition, bad coupling, incorrect dimensions, or disturbed velocity profiles degrade the result |
| Balancing-valve differential pressure | Exact valve identification, valve setting, measured ΔP, and the matching chart |
Fast branch-flow check where test ports exist | A chart for another valve size, design, or setting produces the wrong flow |
| Container fill test | Safe accessible discharge, known volume, and measured fill time | Direct volumetric reference | Usually impractical on closed, high-flow chiller circuits |
For an operating closed-loop chiller system, use a properly configured clamp-on ultrasonic flowmeter as the primary independent flow check. Use balancing-valve readings as a second method where the exact chart and setting are known. Reserve exchanger calculations for comparison with measured behavior, and use a container test only where the circuit can discharge safely into a known volume.
How should the baseline test be performed?
- Create one record for each of the 13 chillers. Identify the chilled-water and condenser-water circuits separately, along with the exchanger inlet, outlet, nearby valves, strainers, and accessible test points.
- Choose a stable operating state. Record which pumps and chiller branches are operating. Mark every manual valve position that could alter the tested path.
- Check the pressure instrument zero, connect the high side upstream and low side downstream, clear the measurement lines, and read exchanger
ΔP. - Measure balancing-valve
ΔPwhere applicable. Record the valve identification and mechanical setting; convert pressure to flow only with its matching chart. - Install the clamp-on ultrasonic flowmeter on a suitable pipe section. Enter the measured pipe and fluid information required by the instrument rather than relying on an assumed nominal size.
- Check signal quality and reading stability using the meter's own diagnostics. Reposition the transducers if the signal fails the instrument's acceptance criteria.
- Record flow, exchanger
ΔP, inlet temperature, outlet temperature, pump state, valve positions, and active parallel branches together. - Repeat the measurement without changing the hydraulic state. Investigate unstable or nonrepeatable results before using them as a baseline.
If a safe open discharge is genuinely available, a container test provides volumetric flow from Q = V/t, where V is collected volume and t is fill time. Keep the volume and time units compatible. Do not extrapolate this method to a closed circuit by partially draining a system whose pressure and valve state change during the test.
How should pressure drop and flow be interpreted together?
Pressure drop depends strongly on flow. A higher exchanger ΔP is not automatically evidence of fouling; it may result from increased flow or a changed valve lineup. A lower ΔP may indicate reduced flow rather than a clean exchanger. Compare equipment only at matched hydraulic states.
| Observed result | First checks | Decision |
|---|---|---|
High ΔP, normal verified flow |
Test-point scope, strainer condition, valve position, exchanger restriction | Localize the loss across individual components before planning cleaning |
High ΔP, high verified flow |
Pump operation, bypass position, branch balance | Correct the operating state before judging exchanger condition |
Low ΔP, low verified flow |
Pump state, closed valve, blocked strainer, air, parallel-path bypass | Restore the hydraulic path, then retest |
Expected ΔP, poor cooling result |
Flow verification, temperature sensors, heat-transfer surfaces, operating load | Continue with thermal diagnosis; pressure alone does not establish cooling duty |
Unstable ΔP or flow |
Air, control-valve motion, pump changes, branch staging, sensor connection | Stabilize the system and measurement path before comparison |
When no reliable factory curve exists, do not turn one pressure reading into an assumed flow. Build an empirical baseline from independently measured flow and ΔP at documented operating states. Trend each chiller against itself because exchanger geometry, piping, valves, and instrumentation may differ across the fleet.
How is the result verified without creating a false trend?
Verification requires repeatability and an independent cross-check. Take measurements under the same pump configuration, branch lineup, valve positions, and comparable operating condition. Use the same pressure points and instrument orientation each time. Record units explicitly.
Compare the ultrasonic flow result with the balancing-valve result where both are available. A disagreement calls for checking transducer setup, valve identification, valve position, chart selection, and pressure-port condition. Do not average incompatible results.
After cleaning a strainer, repositioning a valve, repairing a test port, or servicing an exchanger, return the system to the documented baseline state. Repeat both flow and ΔP. A valid correction changes the targeted measurement while repeated readings remain stable and the independent flow check agrees.
FAQ
What happens if chiller differential pressure is high?
Check verified flow first. High ΔP at normal flow points toward resistance in the measured path, but isolate the exchanger, strainer, and valves before assigning the cause.
What happens if differential pressure is low but cooling is poor?
Measure flow independently and confirm the valve lineup and pump state. Low ΔP can be the result of low flow, an open bypass, or measurement-port trouble.
What happens if two pressure gauges do not match?
Check both gauges at the same pressure point and verify zero and resolution. Prefer one differential-pressure instrument to remove subtraction error between separate gauges.
What happens if the ultrasonic flow reading is unstable?
Check pipe dimensions, transducer mounting, coupling, pipe condition, and the meter's signal diagnostics. Move to a more suitable pipe section when the instrument does not accept the signal.
What happens after a valve or strainer is corrected?
Restore the recorded pump and branch configuration, repeat the flow and exchanger ΔP measurements, and accept the result only when repeated readings are stable and the independent flow check agrees.