Pump inverter dry run protection is easy to describe and surprisingly easy to commission badly. A contractor enables a protection function, sees no alarm during normal operation, and assumes the pump is protected. The first real water shortage then reveals that the signal was wrong, the delay was too long, or the automatic restart kept repeating the same damaging condition.
Dry running does not always mean a completely empty pipe. It can begin when a well level falls below the pump inlet, a suction tank empties, a valve blocks the inlet, a strainer clogs, the pump loses prime, or air enters the suction line. The motor may still be turning while the water needed for cooling, lubrication and hydraulic load is missing.
That is why commissioning must prove three separate things: the system can detect water shortage, the inverter stops the pump at the right time, and the restart logic waits for a realistic recovery. The seven tests below are intended for contractors working with deep-well pumps, booster sets, irrigation systems and water-treatment pumps.
Table of Contents
- What dry-run protection must detect
- Choose the right detection method
- Test 1: record a healthy operating baseline
- Test 2: prove the field signal
- Test 3: separate a real shortage from a transient
- Test 4: perform a controlled protection test
- Test 5: verify the stop and alarm response
- Test 6: test water recovery and restart logic
- Test 7: hand over a repeatable record
- False-alarm diagnosis table
- W713B and W713 application notes
What Pump Inverter Dry Run Protection Must Detect
Water shortage can damage more than the motor. Grundfos explains that dry running can damage booster-pump bearings and shaft seals, and recommends dry-run protection for booster sets. Its dry-running protection overview identifies three common detection routes: a pressure or level switch, an inlet-pressure transmitter, or a tank-level transmitter.
Other systems infer dry running from reduced motor load. When a pump stops moving water, current or absorbed power may fall. Grundfos describes this approach in its SQ/SQE dry-running explanation, where low power consumption is used as a dry-run signal. A recent Xylem controller manual also shows current threshold, delay and restart mode as separate settings.
These examples prove an important point: dry-run protection is not one universal alarm. The correct method depends on the water source, pump type, available sensors and expected load range.
| Detection method | Best suited to | Main commissioning risk |
|---|---|---|
| Float switch or electrodes | Tank, reservoir, sump or well | Wrong contact logic or poor sensor position |
| Suction pressure switch | Pressurized inlet manifold | Pressure can dip briefly during normal demand |
| Analog level transmitter | Variable tank or well level | Wrong range, scaling or cable interference |
| Analog inlet-pressure transmitter | Booster suction monitoring | Sensor location may not represent pump inlet condition |
| Current or power underload | Stable pump and motor duty | Low-flow operation may resemble dry running |
| Flow switch or meter | Systems where minimum flow is defined | Air, fouling or poor installation can distort the signal |
Test 1: Record a Healthy Operating Baseline
Do not start by creating a fault. First run the pump with a confirmed water supply and record its normal behavior at minimum, typical and high demand.
Record:
- Motor nameplate current
- Actual current and frequency
- Suction pressure or source level
- Discharge pressure
- Flow, when a meter is available
- Time needed to establish pressure after start
- Normal pressure dip when another pump or valve operates
This baseline tells the contractor whether current-based or pressure-based detection can distinguish a real shortage. If normal low-flow current is close to the proposed underload threshold, a current-only method may create false trips. If inlet pressure briefly falls during every start, an immediate low-pressure trip will also be unreliable.
Test 2: Prove the Field Signal Before Testing the Pump
The safest first test is an electrical signal test with the pump stopped. Operate the float switch, water-level contact or simulated transmitter signal and confirm that the correct input changes state on the controller.
Check four items:
- Is the contact normally open or normally closed?
- Does the displayed level or pressure match a known reference?
- Does a broken sensor wire create a fault rather than a false healthy signal?
- Is the signal cable separated from motor and output cables?
The W713/W713B manual includes low-water-level, low-inlet-pressure and sensor-fault responses. Those functions are useful only when the terminal logic, analog range and field wiring are correct. Never assume a connected sensor is working because its value looks plausible at one operating point.
Test 3: Separate a Real Shortage From a Transient
A dry-run delay should be longer than harmless start-up behavior but shorter than the time that could damage the pump. There is no responsible universal delay for every pump.
Measure how long the pump normally takes to prime, build discharge pressure and stabilize current. Then review the pump manufacturer’s limit for operation without adequate water. Deep-well pumps, self-priming pumps and booster pumps can require different logic.
Avoid two extremes:
- A delay so short that normal pressure dips cause nuisance trips
- A delay so long that the protection only reacts after the pump has already run dry
Write down the reason for the selected delay. A parameter copied from another project is not an engineering justification.
Test 4: Perform a Controlled Protection Test
Only qualified personnel should perform a water-shortage test. Use the least harmful method available and stay within the pump manufacturer’s instructions. For a tank system, lowering or simulating the level signal is usually safer than physically running the pump empty. For an inlet-pressure system, use a controlled signal or approved valve procedure.
During the test, record:
| Test point | Required observation |
|---|---|
| Trigger condition | Level, pressure, current or flow at the trip point |
| Detection time | Seconds from shortage signal to alarm |
| Motor response | Controlled stop without repeated start attempts |
| Fault display | Clear code or message for the service team |
| Output state | Pump and auxiliary outputs return to the designed safe state |
Stop the test immediately if there is abnormal noise, heat, vibration or seal distress. The purpose is to prove the protection logic, not to demonstrate how long the pump can survive without water.
Test 5: Verify the Stop and Alarm Response
After the trip, confirm what actually stopped the pump. A thermal overload or overcurrent fault is not proof that dry-run protection worked. The intended water-shortage input or underload logic should be the first protective action.
Also confirm whether the alarm is retained, sent to a relay output, or available to a control system. For unattended sites, a local fault code that nobody sees has limited value. The handover document should tell the operator whether to check the well, tank, suction valve, strainer or sensor before resetting.
Test 6: Test Water Recovery and Restart Logic
Automatic restart can restore supply, but it can also repeat dry running if the source has not recovered. The restart delay should reflect the source.
A booster tank may refill in minutes. A low-yield well may need much longer. An irrigation canal can remain unavailable until an upstream gate changes. Test at least these states:
- Water returns before the restart delay expires
- Water has not returned at the first restart attempt
- The fault occurs repeatedly
- Power fails while the system is waiting to restart
Define the maximum number of automatic attempts and when manual inspection becomes mandatory. The goal is controlled recovery, not endless restarting.
Test 7: Hand Over a Repeatable Record
The final protection record should include the sensor type, wiring terminal, normal readings, trip threshold, delay, restart interval, maximum attempts and test result. Save the final inverter parameters and attach a simple action list for the operator.
| Alarm observation | First checks before changing parameters |
|---|---|
| Low level but tank contains water | Float position, contact logic, cable continuity |
| Low inlet pressure during peak demand | Suction pipe, strainer, valve, tank refill rate |
| Underload trip only at low flow | Normal current range, sleep logic, pump curve |
| Pressure remains low with normal current | Wrong rotation, air, blocked pipe, worn pump |
| Repeated restart without water | Recovery delay, well yield, source-level signal |
| Sensor fault appears intermittently | Shielding, grounding, terminals and transmitter supply |
W713B and W713 Application Notes
BEDFORD W713B and BEDFORD W713 can be used in pump systems requiring pressure or level-related protection, automatic operation and fault handling. The final dry-run method still has to match the field sensor and application.
For a deep-well project, include well level, recovery time and motor-cable length in the technical inquiry. For a booster system, include suction arrangement, tank level and minimum inlet pressure. For irrigation or treatment systems, define whether the source can change suddenly and whether automatic restart is safe.
Pump inverter dry run protection should be accepted only after the full chain has been tested: field condition, sensor, inverter logic, motor stop, alarm and recovery. When contractors record that chain, the protection becomes a verified project function instead of a checked box on a parameter list.