Why Solar Water Pumps Fail: Troubleshooting and Maintenance

The call almost always describes the same thing: the solar pump is not producing the water it used to, and somebody wants to know whether it needs a new pump.

Usually it does not. A solar pumping system has five places a fault can live, and the pump is neither the most likely nor the most expensive to be wrong. Replacing it without diagnosis is a common and costly reflex, and the replacement frequently fails the same way within months because whatever killed the first one is still there.

This article works through the diagnosis by symptom, because that is how the problem actually presents, and then sets out the maintenance routine that prevents most of these faults from occurring. It builds on our guide to solar water pumping systems and the borehole-side issues covered in solar borehole pumping.

The Five Places a Fault Can Live

Before any symptom, it helps to hold the system in mind as five stages, because a structured check beats guessing every time.

The solar resource is what is actually arriving: irradiance, shading, season and weather. The array converts it: modules, wiring, connectors, combiner and mounting. The controller conditions it: settings, protection logic, voltage window and switching. The pump and motor do the work. The well and hydraulic system absorb it: water level, rising main, valves, tank and distribution.

A fault in any one of these produces reduced water at the tap, which is why the symptom on its own never identifies the cause. The diagnostic value comes from the pattern: when it happens, how it changed, and what else is true at the time.

Symptom: No Water At All

Start with whether the pump is running. This sounds obvious and it is regularly skipped, and the answer splits the problem in half.

If the pump is not running, check the controller display or indicators first, because most controllers report the reason. Common causes are a tank-full signal stuck active, a dry-run trip that has not reset, an under-voltage condition from an array fault, and a blown fuse or tripped protective device. A float switch that has failed in the full position is a frequent and undramatic culprit.

If there is no controller indication at all, the array is not delivering. Measure open circuit voltage at the controller input in good sun. Zero or very low voltage points at a broken connection, a failed isolator, water ingress in a junction box, or corroded connectors. In dusty and hot climates, connector degradation is a leading cause and it is often visible once the enclosure is opened.

If the pump is running but no water arrives, the pump is either not primed, running dry, running backwards, or the delivery path is blocked. Reverse rotation on a three-phase installation after any electrical work is a classic, and it is fixed by swapping two phases. A closed valve somewhere in the delivery line accounts for more of these calls than anyone likes to record.

Symptom: Output Has Fallen Gradually

Gradual decline is the most informative symptom, because gradual means wear or accumulation rather than a discrete failure.

Array soiling is the first thing to check and the cheapest to fix. Dust accumulation in arid conditions reduces output substantially between cleanings, and in the environments where most of this equipment operates the loss between rain events can be severe. Clean the array and re-measure before investigating anything else. If output returns, the fault was soiling and the finding is that the cleaning interval is too long.

Shading that was not there at commissioning is worth ruling out. Vegetation grows, buildings appear, and a new structure that shades one corner of the array for two hours a day removes considerably more than that fraction of the output, because a partially shaded module drags down its whole string.

Pump wear produces exactly this pattern. Abrasive particles progressively open the running clearances, the pump loses the ability to generate head, and the daily volume drops. If the array is clean, unshaded and producing rated current, and the pump is drawing normal power but delivering less water, the pump is worn. Sand in the tank or at the outlet confirms it.

Screen encrustation in the borehole reduces the well's yield over time, particularly in waters with high iron or carbonate. The symptom is increased drawdown at the same pumping rate. If nobody is measuring water level, this is invisible until it is severe.

Module degradation is real but slow, and it is rarely the answer to a decline noticed over months. If a string is measurably down against its neighbours, look for a specific failed module or a wiring fault rather than assuming general ageing.

The five stages to check in order. A fault in any one produces the same symptom at the tap, which is why guessing is expensive. Source: MIMAH engineering practice.

Symptom: Works in the Morning, Stops by Midday

This is the signature of over-abstraction, and it is the most misdiagnosed fault in solar pumping.

What happens is straightforward. As irradiance rises through the morning, the pump speeds up and the pumping rate increases. If that rate exceeds what the well can sustain, the water level in the borehole falls progressively. Around the middle of the day, when the pump is running hardest, the level reaches the intake or the dry-run probe, and the system stops. It recovers overnight and does the same thing again tomorrow.

The give-away is the timing: it fails when conditions are best, which is the opposite of what an array or controller fault would do. A system that struggles on cloudy days and works in full sun has an energy problem. A system that works on cloudy days and fails in full sun has a water problem.

The fix is not a bigger pump. It is matching the pump's maximum output to the well's safe yield so the system is physically incapable of drawing the level down that far, which the Rural Water Supply Network has consistently identified as the right design default for unattended installations. Where the pump is already installed, capping the maximum operating frequency at the controller achieves the same result without new hardware.

Symptom: Motor Keeps Failing

A submersible motor that fails repeatedly in the same installation is telling you about the installation, not about the motor.

Cooling is the leading cause. Submersible motors rely on water flowing past them at a minimum velocity, and a pump set in a large-diameter borehole or an open source without a flow sleeve sits in effectively still water and overheats. Solar makes this worse, because the pump spends much of each day at reduced speed and therefore reduced flow. The cooling check has to be done at the minimum operating speed, not at rated flow.

Running below the minimum frequency to extract water from poor light is a slow way to destroy a motor, for the same cooling reason plus reduced internal lubrication in some designs. Controllers set to start too early or run too low will shorten motor life while appearing to improve daily yield.

Cable and insulation faults produce repeat failures that get blamed on the motor. A splice that was not properly sealed will admit water, and the resulting insulation failure looks like a motor fault. Test insulation resistance before and after any intervention and record the values.

Voltage quality matters on hybrid systems where a generator or grid supply backs up the array. Unbalanced or out-of-tolerance supply damages motors steadily.

Symptom: Nuisance Trips

Frequent dry-run trips on a system that is not actually running dry usually indicate the probe is set too high, the restart delay is too short, or the controller is inferring dry running from current rather than measuring level directly. Current-based dry-run detection is harder to get right on a variable-speed solar system because the current signature changes with speed throughout the day.

Repeated over-voltage trips in the early morning are a string configuration issue. Open circuit voltage is highest when cells are cold, and a string sized on warm-condition assumptions can exceed the controller's input limit at dawn on a cold morning. This is a design fault that appears seasonally and confuses everybody the first winter.

Rapid on-off cycling near the tank-full point is a hysteresis problem, fixed in the controller settings rather than by replacing anything.

The Maintenance Routine

Most of the faults above are prevented rather than fixed, and the routine is not demanding.

Weekly, in dusty conditions, clean the array. This is the single highest-return maintenance task in solar pumping and it is the one most often deferred. Water and a soft brush in the early morning or evening; avoid cleaning hot modules with cold water. Where labour is scarce, at minimum clean before and after the dusty season and after any dust event.

Weekly, check that the daily volume delivered is in line with expectation for the season and record it. A simple flow meter and a logbook turn a vague sense that output is down into a dated trend, and the trend is what allows a fault to be caught while it is cheap.

Monthly, inspect the array visually for cracked glass, discoloured cells, delamination and hot spots, and inspect wiring for UV damage, rodent damage and loose or corroded connections. Open the combiner and controller enclosures and check for water ingress, insects and dust.

Monthly, check structural fixings. Wind loading works fasteners loose over time, and a module that detaches takes its string with it.

Quarterly, measure and record water level in the borehole, both static and while pumping. This is the measurement that reveals well problems early and almost nobody takes it. Also check the tank for sediment, the float switch for free movement, and the delivery line for leaks.

Annually, measure string voltages and currents against commissioning values, test insulation resistance on the motor cable, inspect the pump discharge for wear indications, service valves, and clean the tank properly.

Keep the commissioning figures somewhere the maintenance team can reach them. A maintenance routine without a baseline can only find gross faults, which is the same problem that undermines industrial condition monitoring and is covered in our article on why solar systems fail early.

The single most useful diagnostic in solar pumping: read the timing, not the symptom. Midday failure in full sun is over-abstraction, and a bigger pump makes it worse. Source: MIMAH engineering analysis.

When It Is Worth Replacing Rather Than Repairing

Pump repair is often economic and often not, and the decision turns on a few factors.

If the pump is worn from abrasives and the well still produces sand, a replacement will wear the same way. Redevelop the well first or accept a shortened service life and plan for it.

If the motor has failed once from a correctable installation fault, fix the fault and replace the motor. If it has failed twice for the same reason, the fault has not actually been corrected.

If the system has been running for many years and multiple components are degraded together, replacing the pump alone frequently disappoints, because the array output and the well condition have also moved. Assess the whole system before committing.

If the original design was wrong, no replacement fixes it. A system that has never met its design output on a clear day at full array capacity was mis-sized, and the answer lies in the design rather than the equipment.

Frequently Asked Questions

Why does my solar pump work in the morning and stop at midday? Almost always over-abstraction. The pump draws the borehole level down faster than the aquifer can replenish, and by midday the level reaches the intake or the dry-run probe. It is a well and sizing problem, not a pump fault, and a bigger pump makes it worse.

How often should solar panels be cleaned in a dusty climate? More often than most schemes plan for. Where dust is heavy, weekly cleaning is justified by the output recovered. The practical test is to measure output immediately before and after a clean; if the gain is significant, the interval is too long.

My pump is running but delivering less water than last year. What is it? Check in this order: array soiling, new shading, borehole water level, then pump wear. The first two are free to rule out and account for a large share of these cases.

Do solar pumps need annual servicing? The array, wiring, controller and hydraulic system benefit from a scheduled annual check with recorded measurements. The submersible pump itself is generally left alone unless performance data indicates a problem, since pulling it is the expensive part.

The routine that prevents most of the faults above. Quarterly water-level measurement is the item almost nobody takes and the one that reveals well problems early. Source: MIMAH engineering practice.

Diagnose the System, Not the Pump

The consistent lesson from solar pumping installations across Sudan, Nigeria and Egypt is that the component that failed is rarely the component that caused the failure. Motors burn out because of cooling. Pumps wear because of well construction. Output falls because nobody cleaned the array. Systems stop at midday because the pump was sized against a yield the borehole was never tested for.

A structured check through the five stages takes an hour and repeatedly finds a cause that costs very little to correct. Replacing the pump takes a rig, a crew and a lead time, and does not address any of the above.

Have a system that is under-delivering? Talk to our engineering team. We will work through the diagnosis with your output records and site data, and tell you what actually needs replacing before you order anything.