Submersible borehole pumps — hydraulics, motor and rising main
Borehole Pump Runs But Delivers No Water — Diagnosis and Repair
- Applies to
- Submersible borehole and well pumps, single- and three-phase, on direct-on-line, soft-start or variable-speed control
- Difficulty
- advanced
- Competence required
- qualified electrician
- Diagnosis complexity
- Moderate. Most of the diagnosis is done at the surface; pulling the pump is expensive and should be the last step, not the first.
- Electrical system
- Single-phase 240 V or three-phase 415 V 50 Hz nominal; pump rating per nameplate
- Safety classification
- multiple hazard
- Author
- EmersonEIMS Engineering
- Technical review
- Mr. Kararaho
- Last reviewed
- 2026-07-27
Scope — read this before relying on the guide
This guide sets out diagnostic method, ranked causes and safety requirements. It deliberately does not state equipment-specific numbers — test voltages, resistance and component values, torque figures, acceptance windows and pin assignments are referred to the manufacturer's documentation for your exact model rather than given here, because a plausible-looking figure for the wrong machine is more dangerous than no figure at all.
Direct technical answer
Establish the water level before you blame the pump, because the most common cause of a running pump delivering nothing is that there is no longer water at the intake. Boreholes draw down while pumping, and a yield that was adequate at commissioning can fall with the season, with neighbouring abstraction, or simply as the borehole ages, so the pump runs, cools poorly, and delivers air. Measure the standing level and, more importantly, the pumping level under running conditions rather than the level after a night of recovery. If water is genuinely at the intake, work through the delivery path in order: a stuck or failed non-return valve, a leak or corroded joint in the rising main returning water down the hole, an air lock, or worn hydraulics. Worn impellers and wear rings are a real and common cause in sandy boreholes, and they produce a pump that draws near-normal current while delivering little, which is why current alone will not tell you. Above all, do not leave a pump running while you investigate: submersible motors are cooled by the water flowing past them, and a pump running dry destroys itself in minutes.
01Symptom description
Controller / display
- Pressure switch not satisfying, so the pump runs continuously without reaching cut-out
- Dry-run or low-flow protection tripping, where fitted
- Overload relay tripped, or drive reporting overcurrent
- Level probe or float indicating low water
Indicators
- Pump contactor pulled in with no delivery at the outlet
- Pressure gauge reading low or zero while the pump runs
- Flow meter showing no or reduced flow
Sounds
- Motor running normally at the wellhead but no water arriving
- Gurgling or spluttering at the outlet, indicating air being drawn
- Cavitation noise — a rattling as though pumping gravel
- Rising main knocking on start, which suggests a failed non-return valve allowing backflow
- Silence at the wellhead, which is a different fault — the motor is not running at all
Smells
- Burnt smell at the starter or control panel
- Any burnt smell means isolate and investigate before another start
Behaviour
- Delivers briefly on start then stops, which is the classic signature of drawdown to the intake
- Delivered normally until the dry season, or after a new borehole was drilled nearby
- Flow has fallen gradually over months, suggesting wear or partial blockage rather than a sudden failure
- Delivers when left off overnight then fails after a short run — recovery masking an inadequate yield
- Runs continuously and never satisfies the pressure switch
- Repeated pump failures on the same borehole, which points at the borehole or the installation, not the pumps
- Water arrives sandy or discoloured, which indicates the pump is drawing near the bottom or the borehole is deteriorating
Visible
- Standing water level and pumping level, measured with a dip meter
- Non-return valve condition at the wellhead where accessible
- Rising main, wellhead seal and pipework for leaks and corrosion
- Pressure gauge and switch settings against the system design
- Cable condition and splice at the wellhead
- Sand or sediment in the discharge, filter or tank
- Pump nameplate and installation depth records
02What the fault means
In plain language
The motor is turning but water is not arriving. That usually means one of three things: there is no water at the pump because the borehole level has dropped, the water is being pumped but leaking back down the hole, or the pump itself is worn and no longer moving water. Finding out which — from the surface — matters, because pulling a pump out of a borehole is expensive.
Technical explanation
A submersible borehole pump is a multi-stage centrifugal machine coupled to a submersible motor, delivering through a rising main with a non-return valve to prevent backflow. Delivery therefore depends on suction availability at the intake, the mechanical condition of the impellers and wear rings, and the integrity of the delivery path. Water level is the dominant variable: a borehole draws down while pumping, and the pumping level rather than the standing level determines whether the intake remains submerged. Because the submersible motor is cooled by water flowing past it up the borehole annulus, loss of submergence causes rapid overheating in addition to loss of delivery, which is why dry running destroys pumps quickly and why dry-run protection is not optional on a borehole that has ever drawn down. Hydraulic wear behaves differently from a blockage: as impellers and wear rings erode, internal recirculation increases, so the pump develops progressively less head while its power draw falls rather than rises, which is why a pump can be badly worn while its running current looks unremarkable, and why current alone is a poor diagnostic here. Abrasive sand accelerates this wear dramatically, which is the normal condition in many Kenyan boreholes. A failed non-return valve allows the rising main to empty back down the hole between runs, giving strong initial flow that fades, and repeated water-hammer that damages pipework. Electrically, the cable, splice and motor are in series down the hole, so an insulation test at the wellhead measures all three together and cannot by itself distinguish a damaged cable from a failed motor — that separation requires testing at the splice after pulling.
03Common causes, ranked
These are ordered by likelihood. Presenting every possible cause as equally probable is a failure of diagnosis, not thoroughness.
Most likely
start here- Water level drawn down below the pump intake — seasonal, abstraction-related or a declining borehole
- Non-return valve stuck open, allowing the rising main to drain back between runs
- Worn impellers and wear rings, commonly from abrasive sand
- Leak or corroded joint in the rising main returning water down the hole
Possible
check next- Air lock in the rising main or delivery pipework
- Blocked pump intake or strainer, from sediment or biofilm
- Pressure switch or control set incorrectly, so the pump never satisfies
- Closed or partly closed valve in the delivery line
- Supply voltage low, so the pump runs slowly and delivers little
Less common
after the above- Broken pump shaft or coupling — the motor runs and the pump does not
- Motor running in reverse rotation on a three-phase installation, which gives reduced flow
- Borehole collapse or siltation reducing yield or burying the intake
- Rising main separated at a joint, dropping the pump or discharging into the borehole
- Motor failed while the control still indicates running
Model specific
verify per unit- Duty point, head and flow come from the pump curve and nameplate for that specific pump — do not assume performance figures
- Minimum submergence and required cooling flow velocity are manufacturer specifications and differ between motors
- Some motors require a flow sleeve where the borehole diameter is large relative to the motor, and its absence causes overheating even with water present
- Direction of rotation checking differs between single- and three-phase installations
- Dry-run and low-flow protection options differ by controller
Environmental
site conditions- Seasonal water table variation, which is substantial in much of Kenya
- Abstraction by neighbouring boreholes lowering the local level
- Abrasive sand content, which is the dominant wear mechanism
- Aggressive or mineral-rich water corroding the rising main and pump
- Biofouling and encrustation reducing intake area over time
Installation related
built in- Pump set too shallow for the borehole drawdown behaviour
- No dry-run or low-water protection fitted
- Rising main material unsuited to the water chemistry or the depth
- Pump selected on flow alone without matching the borehole yield
- No provision for measuring water level, which makes proper diagnosis impossible
- Undersized cable causing voltage drop over the drop length
Maintenance related
deferred work- Water level never measured or recorded, so decline is invisible until failure
- Flow and pressure never recorded at commissioning, so there is no baseline to compare against
- Borehole never test-pumped to establish sustainable yield
- Sand content never assessed despite visible sediment
- Insulation resistance of the pump and cable never trended
Component level
electronics- Impellers and wear rings eroded
- Non-return valve failed
- Pump shaft or coupling broken
- Motor failed
- Drop cable damaged or splice failed
- Intake strainer blocked
04Safety requirements
Isolation
- Isolate the pump at its starter or drive, lock off and prove dead before any work at the wellhead
- Confirm automatic control — pressure switch, level probe or building management system — cannot restart the pump
- Where a variable-speed drive is fitted, its DC bus capacitors remain charged after isolation; verify discharge
- Treat the wellhead as electrically live until proven otherwise, because water and supply cabling share the space
Lockout and tagout
- Lock and tag the pump starter and the control that can call it to run
- Tag the pressure switch or level control specifically — an automatic restart during work at the wellhead is the hazard here
- Keep the only key with the person doing the work
PPE
- Insulated tools rated for the system voltage
- Eye protection and gloves
- Safety footwear and hand protection for pipework handling
- Head protection where lifting equipment is in use
- Harness and fall protection where working over an open borehole
Stored energy
- The rising main holds a substantial column of water under pressure — releasing a joint without relieving it can cause injury
- Pressure vessels and accumulators stay pressurised after the pump stops and must be relieved before work
- Variable-speed drive capacitors hold a lethal charge after isolation
- A suspended pump and rising main represent a large stored mechanical load on the lifting equipment
Specific hazards
- NEVER leave a pump running to "see if it picks up". A submersible motor is cooled by water flowing past it, and running dry destroys it within minutes. If there is no delivery, stop it.
- An open borehole is a fall hazard and must be physically covered or guarded whenever it is not being worked on.
- Pulling a pump is a lifting operation with real crush and drop risk. The rising main full of water is far heavier than the pump alone, and improvised lifting kills people.
- Never work at a wellhead with the control in automatic — a pressure switch can start the pump without warning
- Water and electricity share the wellhead; a damaged drop cable can make the rising main and casing live
Stop and call a qualified professional if
- The pump must be pulled and you do not have proper lifting equipment and a trained crew
- There is any indication that the casing, rising main or wellhead is electrically live
- Borehole collapse or siltation is suspected
- Repeated pump failures on the same borehole, which needs a hydrogeological assessment rather than another pump
- Abstraction changes are contemplated, which may require WRA authorisation
05Tools required
| Tool | Why it is needed |
|---|---|
| Dip meter (water level meter) | Standing AND pumping water level — the single most informative measurement, and the one most often skipped |
| True-RMS clamp meter | Running current on each phase against nameplate; note that a WORN pump draws less, not more |
| True-RMS multimeter | Supply voltage at the starter under load, and continuity of the drop cable |
| Insulation resistance tester | Motor and drop cable insulation from the wellhead, but note it measures cable, splice and motor together |
| Pressure gauge | Delivery pressure against the pump curve duty point |
| Flow meter or a timed volumetric measurement | Actual delivered flow, which is what the complaint is really about |
| Pump curve and nameplate for the installed pump | Performance can only be judged against the specific machine, never against a general expectation |
| Borehole log and installation records | Depth, intake setting and commissioning yield — without these you are guessing about drawdown |
06Diagnostic decision tree
1. Is the pump currently running with no delivery?
Yes → STOP it now. Running dry destroys the motor within minutes.
No → Continue
2. Is the motor actually running at all?
Yes → Continue — this is a hydraulic or delivery-path fault
No → This is an electrical fault, not a pump fault — diagnose supply, control and motor
3. What is the PUMPING water level, measured while running?
Yes → Water proven above the intake — continue
No → Drawdown below intake is the cause. This is a borehole yield problem, not a pump fault.
4. Is every valve in the delivery line open?
Yes → Continue
No → A closed or partly closed valve explains it and costs nothing to check
5. Does flow start strongly then fade away?
Yes → Suspect a failed non-return valve or a rising main leak draining back
No → Continue
6. On three-phase: is the direction of rotation correct?
Yes → Continue
No → Reverse rotation gives reduced flow and is easily missed after electrical work
7. Is running current well BELOW nameplate with poor delivery?
Yes → Consistent with hydraulic wear — impellers and wear rings
No → Continue
8. Has the ROOT CAUSE been established from the surface?
Yes → Plan the pull with proper lifting equipment
No → Exhaust surface diagnosis first — pulling is expensive and often unnecessary
07Step-by-step diagnosis
Step 1Stop the pump and secure the installation
- Inspect
- Whether the pump is running without delivery, and whether automatic control can restart it
- Where
- At the starter and the control
- Instrument
- Observation
- Expected result
- Pump stopped, control locked out, borehole guarded
- If the result is abnormal
- A pump running without delivery is destroying itself. Stopping it first is not a formality — it is frequently the difference between a repair and a replacement.
- Next
- Step 2
Safety: Never leave it running to see whether it picks up.
Step 2Gather the records before measuring
- Inspect
- Borehole log, pump nameplate, installation depth, commissioning yield, flow and pressure baselines
- Where
- Site records and the pump nameplate
- Instrument
- Documentation
- Expected result
- Known intake depth and a commissioning baseline to compare against
- If the result is abnormal
- Without the intake depth you cannot interpret a water level, and without a baseline you cannot judge whether flow has declined. Where records are absent, say so rather than assuming.
- Next
- Step 3
Step 3Measure the standing level, then the PUMPING level
- Inspect
- Water level at rest, then again while the pump runs, and how far it draws down
- Where
- Down the borehole, using the dip tube or annulus
- Instrument
- Dip meter
- Expected result
- Pumping level remaining comfortably above the pump intake
- If the result is abnormal
- This is the measurement that resolves most of these jobs. A standing level measured after overnight recovery looks healthy and tells you almost nothing — the pumping level is what determines whether the intake stays submerged.
- Next
- Step 4
Safety: Run the pump only briefly for this measurement, and stop it if delivery is absent.
Step 4Check the delivery path from the outlet backwards
- Inspect
- Every valve, the pressure switch setting, filters and any strainer
- Where
- Along the delivery line
- Instrument
- Visual inspection and pressure gauge
- Expected result
- All valves open, controls set correctly, no blockage
- If the result is abnormal
- A partly closed valve or a wrongly set pressure switch costs nothing to check and explains a proportion of these complaints outright.
- Next
- Step 5
Step 5Observe the flow behaviour on start
- Inspect
- Whether flow is absent entirely, or starts strongly and fades
- Where
- At the outlet
- Instrument
- Observation, pressure gauge, flow meter
- Expected result
- Flow establishing and holding steady
- If the result is abnormal
- Strong flow that fades points at the rising main emptying back — a failed non-return valve or a leak in the main. No flow at all from the outset points at level, blockage or hydraulics.
- Next
- Step 6
Step 6Measure supply voltage and running current on all phases
- Inspect
- Voltage under load and current per phase against nameplate
- Where
- At the starter
- Instrument
- True-RMS multimeter and clamp meter
- Expected result
- Voltage balanced and adequate; current consistent with the duty
- If the result is abnormal
- Interpret current carefully here. A WORN pump draws LESS than nameplate while delivering poorly, because internal recirculation reduces the work done, so low current with poor delivery supports hydraulic wear rather than ruling out a problem. High current suggests a mechanical bind or sand.
- Next
- Step 7
Step 7Verify direction of rotation on three-phase installations
- Inspect
- Phase rotation and, where practical, delivery comparison in both directions
- Where
- At the starter
- Instrument
- Phase rotation tester
- Expected result
- Correct rotation for the pump
- If the result is abnormal
- A reversed pump still runs and still delivers, but noticeably less. It is easily introduced by upstream electrical work and just as easily overlooked.
- Next
- Step 8
Step 8Test insulation from the wellhead, and understand what it includes
- Inspect
- Insulation resistance of the motor and drop cable together, from the surface
- Where
- At the wellhead terminals, fully isolated
- Instrument
- Insulation resistance tester
- Expected result
- A value acceptable for the installation, compared against its own history
- If the result is abnormal
- A low reading proves a fault somewhere in cable, splice or motor but cannot say which — they are in series down the hole. Separating them requires testing at the splice after pulling, so do not condemn the motor on a wellhead reading alone.
- Next
- Step 9
Safety: Discharge the winding and cable after testing.
Step 9Only then plan the pull
- Inspect
- Whether surface diagnosis has produced a specific cause
- Where
- Review of all findings
- Instrument
- All prior measurements
- Expected result
- A specific reason to pull, and proper lifting equipment arranged
- If the result is abnormal
- Pulling a pump is expensive and hazardous. Where the finding is drawdown below intake, the answer may be lowering the pump or reassessing the yield rather than replacing anything.
- Next
- Pull with proper lifting equipment and a trained crew
Safety: The rising main full of water is far heavier than the pump. Improvised lifting is how people are killed on borehole work.
08Repair procedure
Level, control and protection — usually the real fix
configuration- Fit dry-run or low-water protection where none exists; on a borehole that draws down it is essential, not optional
- Correct pressure switch and control settings to the system design
- Where drawdown is the cause, consider lowering the pump only if the borehole depth and yield genuinely allow it
- Restrict the duty to match the sustainable yield rather than the pump capacity
Where the borehole cannot sustain the demand, no pump will fix it. That is a yield problem and needs a hydrogeological answer.
Delivery path
mechanical- Replace a failed non-return valve
- Repair or replace corroded rising main sections and failed joints
- Clear blocked intakes and strainers
- Relieve air locks and fit air release where the pipework profile causes them
Pump and motor
component replacement- Replace worn impellers and wear rings, or the pump end, where wear is confirmed
- Replace the drop cable and splice where insulation testing after pulling implicates them
- Replace the motor where testing at the splice confirms motor failure rather than cable failure
- Fit a flow sleeve where the borehole diameter requires it for motor cooling
Supply integrity
wiring- Correct undersized drop cable causing voltage drop over the depth
- Correct phase rotation on three-phase installations
- Remake the wellhead splice properly — it is a common failure point and a common source of insulation faults
Borehole itself
manufacturer level- Refer suspected borehole collapse, siltation or declining yield for hydrogeological assessment
- Provide the water level records, flow measurements and pump history
Repeated pump failures on one borehole are a borehole problem until proven otherwise.
09Post-repair validation
- Confirm delivery flow and pressure against the pump curve duty point, measured rather than estimated
- Measure and record the pumping water level under sustained running, not just at start
- Confirm the pumping level stabilises above the intake with an adequate margin
- Measure running current on all phases and record it against nameplate
- Confirm dry-run protection operates by testing it, not by assuming it
- Record insulation resistance as a new baseline
- Run for an extended period and confirm flow is sustained rather than fading
- Record water levels, flow, pressure and current so the next visit has a baseline to compare against
10When not to repair
- Where the borehole yield can no longer sustain the demand — that needs a hydrogeological answer, not a bigger pump
- Boreholes with collapse or heavy siltation, where pump replacement will simply repeat the failure
- Pumps heavily eroded by sand where the borehole sand problem remains unaddressed
- Small pumps where a replacement costs less than pulling, stripping and rebuilding
- Where records are so absent that depth and intake setting are unknown and cannot be established safely
11Prevention
- Fit dry-run or low-water protection on every borehole pump — the single most effective measure, because dry running destroys pumps in minutes
- Measure and record standing and pumping water levels at every service visit so decline is visible early
- Record flow, pressure and current at commissioning so later readings mean something
- Provide a dip tube at installation; without one, proper diagnosis is impossible
- Match the duty to the sustainable yield rather than the pump capacity, and use controls to enforce it
- Address sand at source where possible — abrasion is the dominant wear mechanism
- Trend insulation resistance from the wellhead rather than waiting for failure
- Treat repeated failures on one borehole as a borehole investigation, not a pump replacement cycle
12Questions engineers actually ask
The pump runs but no water comes. Is the pump finished?
Not necessarily, and the cheapest checks come first. Measure the PUMPING water level — the most common cause is simply that the borehole has drawn down below the intake, which is a yield problem rather than a pump fault. Then check the delivery path: a stuck non-return valve, a rising main leak, a closed valve or an air lock. Pulling the pump should be the last step, not the first.
Current is below nameplate, so the pump must be fine — right?
The opposite, potentially. As impellers and wear rings erode, internal recirculation increases and the pump does less work, so its current FALLS while delivery falls with it. Low current alongside poor delivery is consistent with hydraulic wear, not evidence of health. High current points instead at a mechanical bind or sand.
It pumps fine first thing in the morning then stops delivering. What is that?
Almost always drawdown. Overnight the borehole recovers, so the first run has water at the intake and delivers normally; once the level draws down below the intake the pump loses suction and delivery stops. That pattern is a yield and duty problem — the fix is matching the demand to what the borehole can sustain, plus dry-run protection to stop the pump destroying itself.
The insulation test at the wellhead is low. Is the motor dead?
It proves a fault in the cable, the splice or the motor, but it cannot say which — all three are in series down the hole and the test sees them together. The splice is a common failure point and far cheaper to fix than a motor. Separate them by testing at the splice once the pump is pulled, rather than condemning the motor on a surface reading.
Standards and references
- ISO 9906 — rotodynamic pumps: hydraulic performance acceptance tests
- IEC 60034-1 — rotating electrical machines: rating and performance
- IEEE 43 — recommended practice for testing insulation resistance of electric machinery
- Kenya Water Act and Water Resources Authority (WRA) requirements, where abstraction rates or borehole works are being changed
- The pump and motor manufacturer's curve, nameplate and installation data for the specific machine — the only valid source for duty point, minimum submergence and cooling flow requirements referred to throughout
This guidance is written from engineering principle and is not a substitute for the manufacturer's model-specific documentation. Where a figure is model-specific, confirm it against the service data for your unit before acting on it.
Need this diagnosed properly?
Send us the make, model and any fault codes shown, and photographs of the controller display if you have them. Our mobile workshop covers all 47 counties.