Solar Borehole Pumping: Well Yield, Drawdown and Submersible Pump Selection

When a solar borehole scheme fails, the panels are almost never the reason. The pump is rarely the reason either. In the great majority of cases the borehole was asked to produce more water than it could sustainably give, and everything downstream was designed around a number nobody tested.

This is the specific risk that solar introduces. A diesel pump runs when somebody starts it and stops when somebody stops it, and that person usually notices when the water goes off. A solar pump runs whenever the sun is up, unattended, every day, and if it has been sized above the well's safe yield it will pump the water level down to the pump intake day after day until something burns out or the well is damaged.

This article covers the borehole half of solar pumping: what a pumping test tells you, how to establish a yield you can design against, where to set the pump, how to keep the motor cool, and the protection that has to be in place before the system is left alone. The whole-system design method sits in our article on designing a solar water pumping system, and pump selection across the power range in the solar pump sizing guide.

The Borehole Is the Constraint

A borehole is not a tank of water. It is a hole intersecting an aquifer, and the rate at which water flows into it is governed by the properties of the rock, the construction of the well, and how much you are prepared to draw the level down.

Three numbers describe this, and a design needs all three.

Static water level is the level at rest, with no pumping and enough recovery time to be genuinely at rest. It varies seasonally, often by several metres, and the figure that matters for design is the level at the end of the dry season rather than whatever was measured on the day of drilling.

Pumping water level is the level once pumping has been running long enough to stabilise at a given rate. It is always lower than the static level, and it is the level from which the pump actually has to lift.

Drawdown is the difference between the two at a stated rate. It is the well's response to being pumped, and it increases with the pumping rate. Drawdown is not a fixed property; it is a relationship, which is why it has to be measured at more than one rate.

Dividing yield by drawdown gives specific capacity, in cubic metres per hour per metre of drawdown, and this is the single most useful number for comparing wells and for predicting what will happen at a rate you have not yet tested.

The Pumping Test

There is no substitute for this and no responsible way to skip it. A design built on an assumed drawdown is a guess.

A step drawdown test pumps the well at several increasing rates, holding each long enough for the level to approach stability and recording level against time throughout. It reveals how drawdown grows as the rate increases, and it identifies the point at which the relationship stops being proportional, which indicates the well is being pushed past its efficient range.

A constant rate test then pumps at a single chosen rate for an extended period, commonly many hours, watching whether the level stabilises or continues to fall. A level that stabilises indicates the aquifer is supplying at that rate. A level still falling at the end of the test indicates it is not, regardless of how much water came out.

Recovery monitoring after the pump stops is as informative as the drawdown itself. How fast the level returns towards static tells you about the aquifer's ability to replenish, and a well that recovers slowly will not tolerate daily deep drawdown even if it survived a single test.

The output that a designer needs from all this is a rate the well will sustain indefinitely, with a margin, at a drawdown that leaves adequate submergence over the pump. That is the safe yield, and it is usually well below the maximum rate the test achieved.

The three levels that describe a borehole, and the number derived from them. Drawdown is a relationship with pumping rate, not a fixed property. Source: MIMAH engineering practice; Rural Water Supply Network guidance.

Safe Yield Is Not Test Yield

This distinction causes more failed schemes than any other single factor.

A pumping test might demonstrate that a well produces 12 cubic metres per hour. That is a tested rate over a bounded period, under the conditions of that day, at whatever the water table was doing that season. Designing a solar system to abstract 12 cubic metres per hour every sunny day for twenty years is a different proposition entirely.

Safe yield applies margins for seasonal variation in the water table, for long-term decline where an aquifer is under regional pressure, for the reality that the pump will run every day rather than for a test period, and for the fact that other abstraction may increase nearby. In practice safe yield frequently sits substantially below tested yield, and the gap is a judgement informed by hydrogeology rather than a fixed percentage.

There is a design elegance available here that is worth taking. If the pump is sized so that its maximum output at full irradiance does not exceed the well's safe yield, the system becomes physically incapable of over-abstracting. It cannot damage the aquifer or itself, whatever the weather does and whoever is or is not watching. The Rural Water Supply Network has made this point repeatedly as solar pumping has scaled up across rural water supply, and it is the correct default for any unattended installation.

Setting the Pump

Pump setting depth is decided by four constraints at once, and getting it wrong produces failures that look like equipment defects.

Below the lowest pumping water level, with margin. The intake must remain submerged at the maximum drawdown the system will ever produce, in the driest season, with the water table at its lowest. Margin here is not optional. A pump that breaks suction intermittently will fail.

Above the well screen where possible. Setting a pump within or below the screen draws water horizontally across the screen at high velocity, which encourages sand entry and can damage the formation behind it. Above the screen, water flows up past the motor before entering the intake, which is what the design intends.

Deep enough for adequate submergence, shallow enough to limit head. Every additional metre of setting depth that ends up as lift is array capacity spent for nothing. There is a genuine optimisation here and it is worth doing rather than defaulting to the bottom.

Clear of the well bottom. Sediment accumulates at the base of a borehole. A pump set close to it will draw that sediment through the impellers, and abrasive wear will end the pump early.

The single most common field error is setting the pump as deep as the cable allows on the theory that deeper is safer. It is not. It increases head, increases sand risk, and complicates every future intervention.

Motor Cooling Is Not Optional

Submersible motors are cooled by water flowing past them. This is a design requirement with a specified minimum velocity, and it is the requirement most often violated.

The classic failure is a pump set in a large-diameter borehole or in an open source, where water enters the intake without any obligation to flow past the motor first. The motor sits in effectively still water, overheats, and fails. The pump is then replaced with an identical unit set the same way, and it fails again, and the fault is attributed to poor quality equipment.

The fix is a flow sleeve, a shroud around the motor and pump that forces all incoming water down past the motor before it reaches the intake. Flow sleeves are inexpensive and they are mandatory in any installation where the annular velocity past the motor would otherwise fall below the manufacturer's minimum. That includes most large-diameter boreholes and every open-water installation.

Solar adds a wrinkle here. A solar pump spends a significant part of every day running at reduced speed and reduced flow as irradiance ramps up and down. Cooling flow at low speed is proportionally lower, and the check needs to be done at the minimum operating speed rather than at rated flow. Manufacturers specify minimum operating frequencies partly for this reason, and running below them to squeeze water out of a dull morning is a way to cook a motor slowly.

Why the tested rate is not the design rate. Size the pump so its full-irradiance output cannot exceed safe yield and the system becomes physically unable to over-abstract. Source: Rural Water Supply Network guidance; MIMAH engineering practice.

Sand, Abrasion and Well Development

Sand is the quiet killer of borehole pumps, and its effect is cumulative rather than sudden.

Abrasive particles pass through the impellers and wear the running clearances. The pump loses efficiency progressively, which on a solar system shows up as reduced daily volume that gets blamed on the panels. Eventually the clearances open enough that the pump can no longer generate head, and it fails.

Sand comes from inadequate well development, from an unsuitable screen slot size or gravel pack, from over-pumping that draws fines through the formation, and from setting the pump in the wrong place. Most of these are construction issues rather than pump issues, which is why the state of the borehole matters before any equipment is specified.

A well that produces sand should be redeveloped before a new pump goes in it. Installing a pump into a sanding well is buying a consumable. Where sand cannot be eliminated, pump selection should favour designs that tolerate abrasives, and the maintenance interval should reflect that they will still wear faster.

Water Quality Belongs in the Design

A borehole scheme for drinking water has an obligation that an irrigation scheme does not, and it should be established before the equipment is ordered rather than after commissioning.

Water chemistry affects both people and pumps. Iron and manganese cause staining and encrustation, the latter of which will progressively block a screen and reduce yield. High salinity affects both potability and material selection. Aggressive water attacks components, and the appropriate response is material specification rather than hoping.

The WHO Guidelines for drinking-water quality are the reference framework for the health-related side of this, and the relevant national standard applies alongside them. The engineering point is narrower: get a full chemical analysis from the pumping test water, not from a grab sample taken at the wellhead after the rig left, and specify materials against it.

Protection for an Unattended System

A solar borehole pump runs without anybody present, which means the protection has to do the job a human operator would otherwise do.

Dry-run protection is the essential item. A probe or level sensor in the borehole, set above the pump intake, stops the pump when the water level approaches it and restarts after a recovery delay. This is what prevents a well that is being over-drawn from destroying the pump. Systems that infer dry running from motor current alone are less reliable than a direct level measurement, and on a variable-speed solar system the current signature is harder to interpret.

Over and under voltage protection guards against the wide input range a solar array presents across a day and across weather.

Dry-run recovery logic matters as much as the trip. A controller that restarts immediately after a dry-run trip will cycle repeatedly, which is hard on the motor. The restart delay needs to reflect the well's actual recovery behaviour, which the pumping test told you.

Tank full switching stops the pump when storage is full, either by float switch or by pressure, and it should be wired so that a failed sensor stops the pump rather than running it.

None of this is expensive relative to the pump, and all of it is cheaper than a borehole intervention. In remote installations the protection is frequently the difference between a scheme that runs for fifteen years and one that needs a rig back on site in year two.

Common Failures and What Caused Them

Pump burnt out within months. Usually motor cooling. Check for a flow sleeve, check the annular velocity at minimum operating speed, and check whether the pump was set in the screen.

Declining daily output with no visible fault. Usually abrasive wear in the pump, or encrustation of the screen reducing well yield. Both are progressive, both are diagnosable by comparing current output against the commissioning figures.

System delivers well in the morning and stops by midday. Classic over-abstraction. The well cannot sustain the rate, the level falls to the intake, and the pump loses suction. The pump is not the problem.

Sand in the tank. Well construction or development, or a pump set too low. Redevelop before replacing the pump.

Works in the wet season, fails in the dry. The design used a static water level measured at the wrong time of year. The seasonal range needed to be in the head calculation from the start.

Four constraints decide setting depth simultaneously. Setting the pump as deep as the cable allows satisfies one of them and violates three. Source: MIMAH engineering practice.

Frequently Asked Questions

Do I really need a pumping test if the driller gave me a yield figure? Yes. A driller's yield figure is typically an airlift estimate taken at completion, which is not a controlled test and does not give drawdown against rate. It is a useful indication and it is not a design input.

How much margin should I leave between safe yield and pump capacity? Enough that the pump at full irradiance cannot exceed the safe yield, so the system is physically unable to over-abstract. The exact figure depends on the aquifer, the seasonal range and the confidence in the test data, and it is a hydrogeological judgement rather than a fixed percentage.

Does a solar pump need a flow sleeve? It needs one wherever the water velocity past the motor at the minimum operating speed would fall below the manufacturer's requirement. That includes most large-diameter boreholes and every open-water installation. Solar systems need the check done at minimum speed, not rated flow.

Can I set the pump inside the well screen? Avoid it where the well design allows. Setting within the screen increases entrance velocity at the screen, encourages sand production and can damage the formation. Above the screen is the correct default.

The Well Decides, the Equipment Follows

Everything in a solar borehole scheme is downstream of one number: what the well will sustainably give. Get that right and the rest of the design is arithmetic. Get it wrong and no quality of pump, panel or controller will rescue the scheme, because the failure is in the ground rather than in the equipment.

That is why the pumping test is the cheapest insurance in this entire field, and why the systems that run for fifteen years without intervention are almost always the ones where somebody insisted on it before ordering anything. Across water and solar infrastructure work in Sudan and Nigeria, the pattern is consistent: schemes fail at the borehole and get diagnosed at the pump.

Planning a solar borehole scheme, or troubleshooting one that under-delivers? Talk to our engineering team. We will work from your pumping test data, or tell you what test you need, and size the system to a yield the well can actually sustain.