The methods described below apply to valves designed to isolate equipment, such as chillers, boilers, and cooling towers, from the rest of the plant when they are not in use. Generally, tests on isolation valves are performed to find large leaks.
This article covers the Automatic Isolation Valves Test Procedures. Refer to Valve Leak-By Test Procedures for full testing guidance.
1. Visual Inspection Method
1.1 Condenser Bundle Isolation Test - Command all chillers OFF, all condenser bundle isolation valves closed, all cooling tower isolation valves open, and condenser water pump(s) ON. With all of the condenser bundle isolation valves closed and the condenser water pumps operating, this in effect creates a “dead head” condition on the pump and should not be operated for more than 5 minutes at this condition. Visually inspect for any water entering the cooling towers. If water is flowing, then the test fails. The source of the problem must be fixed before the system can be retested.
1.2 Cooling Tower Isolation Test - Continuing from above with all chillers OFF and condenser water pump(s) ON, command all condenser bundle isolation valves open and all cooling tower isolation valves closed. With all of the cooling tower isolation valves closed and the condenser water pumps operating, this in effect creates a “dead head” condition on the pump and should not be operated for more than 5 minutes at this condition. Visually inspect for any water entering the cooling towers. If water is flowing, then the test fails. The source of the problem must be fixed before the system can be retested.
1.3 Return System to Normal - Once all tests are complete, return all control parameters back to original set points and conditions per the design sequence of operations.
2. Balancing Valve Method
Most chillers and boilers designed to operate in parallel with other units will have automatic isolation valves to prevent water from flowing through the unit when not in operation. The distribution pumps serving these units may be configured as constant or variable primary flow. Each pump will be fitted with either a triple-duty valve or a dedicated balancing valve at the discharge of the pump, which can be used to help identify valve leakage. The following procedures are quick to perform and should identify a small leakage rate through the valve(s) being tested.
2.1 Test Setup - Carefully mark each balancing valve’s setting before making any adjustments. For linear gauge types, mark the gauge. For screw type adjustments, mark the stem as well as count the turns as the balancing valve is fully closed. Adjust each balancing valve serving the isolation valve being tested to 10% open to ensure a small valve leak can be detected. Command all chillers and boilers OFF to ensure they will not try to operate without any water flow. Command the isolation valve on the unit being tested closed, and close the manual isolation valve on the units not being tested. Command one pump ON.
2.2 Valve Closure Performance - Measure the pressure drop across the balancing valve. If there are no leaks in the system, the differential pressure across the balancing valve should be zero. A pressure drop of more than two psi indicates leak-by. If results are inconclusive, close the manual isolation valves on the tested unit and see if the differential pressure across the balancing valve is lower that the first reading. If it is, then it confirms the valve is leaking. When taking pressure measurements, be sure to keep the transducer at the same elevation during each measurement of a given port so that atmospheric pressure doesn’t skew the results. Repeat this procedure with all other units to be tested.
2.3 Return system to normal - Relocate the balancing valve precisely to the pre-test setting and return controlled equipment to auto or pre-test conditions.
3. Installed Flow Meter Method
Many larger boiler and chiller systems contain permanently installed flow meters on their primary or secondary loops, and occasionally on the primary leg to each chiller or boiler. In such cases, using the installed flow meter offers an easy way to detect isolation valve leakage if the device can measure accurately at very low flow rates. The procedures for using installed flow meters are described below.
3.1 Test Setup - Command all chillers and boilers OFF to ensure they will not try to operate without any water flow. Command the isolation valve on the unit being tested closed, and close the manual isolation valve on the units not being tested. If the flow meter is in the primary loop, proceed directly to the Test and Evaluation step. If the flow meter is in a secondary decoupled loop, manually valve off the bypass leg and command the secondary pumps OFF.
3.2 Test and Evaluation - Turn ON the primary pump. There is leak-by if the flow meter consistently reads a flow rate. Repeat this procedure with all other units to be tested.
3.3 Return all systems to normal - Return all manual valves to pre-test positions. Return all automatic devices to auto or pre-test conditions.
4. Other Methods for Isolation Valves
There are other viable methods for detecting leak-by on chiller and boiler isolation valves, but they are generally more time consuming or less conclusive. They are briefly mentioned here for reference.
4.1 Ultrasonic Flow Meter - The test procedures are similar to the Installed Flow Meter method described above. This test may be more time consuming due to the potential necessity to remove and replace at least two feet of pipe insulation in order to measure water flow. In addition, the accuracy of the ultrasonic meter at reduced flow may make it difficult to identify small leaks.
4.2 Pressure Differential Across Chiller or Boiler - This method measures the change in differential pressure across the evaporator chiller bundle or boiler heat exchanger. The method is fast, but can only detect larger leakage rates, since a small leak represents a very small pressure drop across the unit. The procedures are to turn the chiller or boiler OFF; turn ON the pump and close manual isolation valves to ensure no flow. Measure the differential pressure across the device (which should be zero). Then, command the automatic isolation valve closed, open the manual isolation valves, and measure the differential pressure again. If the readings with the automatic isolation valve commanded closed show a consistent differential pressure over 2 psi, then leak-by is likely.
4.3 Mixed Temperature - The method works for equipment that is piped in parallel and the output from each unit is mixed together before it is delivered to the loads. The intent is to measure the mixed water temperature to determine if there is leakage past a unit that is not operating. This is best illustrated through the following example: Chiller 1 is ON and producing 42°F water. Chiller 2 is OFF with its automatic isolation valve closed. With no leakage through the Chiller 2 isolation valve, the mixed water temperature will be 42°F. If there is leakage, the mixed water temperature would be higher due to blending of warm return water with the 42°F water. The test procedure follows this example – command one unit ON and the other(s) OFF with the isolation valves closed. If the mixed temperature is higher than discharge water temperature from the unit that is operating, there is leakage. The procedures would be repeated until all units have been tested. This method may be inconclusive due to the difficulty in measuring the mixed water temperature.
4.4 Pump Differential Pressure - This method is similar to the balancing valve method except the balancing valve is 100% closed rather than 10% open, and the differential pressure is measured across the pump rather than balancing valve. A change in the differential pressure across the pump with the automatic isolation valve closed indicates leakage. However, many pumps have a very flat impeller curve, making a small flow rate due to valve leakage difficult to detect.