Pump Selection and Sizing for Ugandan Farms

The direct answer

A pump is not specified by horsepower and it is not specified by price. It is specified by a duty point: a flow rate and a total head, both of which you calculate before you talk to anyone selling anything.

Flow is how many litres per hour or cubic metres per hour the farm needs, set by the area you irrigate, the crop and the hours you can run the pump.

Total head is the total resistance the pump must overcome, in metres. FAO's irrigation pumping manual defines head as "the expression of the potential energy imparted to a liquid to move it from one level to another", and breaks it into components: static suction lift, "the vertical distance from the surface of the water to the eye of the impeller"; static discharge head, "the vertical distance or difference in elevation between the point at which water leaves the impeller and the point at which water leaves the system"; friction head, because "When water flows through a pipe, the pressure decreases because of the friction against the walls of the pipe"; and pressure head, because "a certain head to operate an irrigation system is required. For example, in order for a sprinkler system to operate, a certain head is required."

Once you have flow and total head, the power follows. FAO's manual gives it as BP = Q x TDH / (C x E pump), where BP is brake power, Q is discharge, TDH is total dynamic head in metres, C is 102 when Q is in litres per second or 360 when Q is in cubic metres per hour, and E pump is pump efficiency.

Two things catch Ugandan farms out specifically.

One: altitude reduces how far a surface pump can suck. FAO's water lifting guidance states that "centrifugal pumps are limited to a maximum in practice of about 4-5m suction lift at sea level (reducing to around 2m suction lift at an altitude of 2 000m", and that such a pump "depends on atmospheric pressure to raise the water, so it is limited to pumping lifts of no more than 5-6m." Most of Uganda sits well above sea level. If your water surface is more than a few metres below the pump, a surface pump will not lift it here, whatever the seller says.

Two: the wrong size costs money in both directions. On oversizing, US engineering guidance states that it "results in higher-than-necessary system operating and maintenance costs" and that "oversized pumps typically require more frequent maintenance than properly sized pumps. Excess flow energy increases the wear and tear on system components, resulting in valve damage, piping stress, and excess system operation noise." On running a pump away from its design point, FAO states that "The use of pumps off their design point is a common cause of gross inefficiency and wasted fuel."

This article gives you the calculation. It does not choose your water source, design your irrigation layout, size a solar array or recommend a supplier.

Who this article is for

Farmers about to buy or replace an irrigation pump. Farm managers comparing quotations that do not look comparable. Investors checking whether a proposed system is specified or merely priced. Installers who want a common language with their customers. Students learning the calculation.

You should have already chosen your water source and know its water level in the dry season. If you have not, Falcon's article on on-farm water sources comes first, because head cannot be calculated without it.

The problem this solves

Walk into any Ugandan agricultural equipment supplier and you will be asked one question: what is your budget. That is the wrong first question and it produces two predictable outcomes.

Undersized pumps cannot deliver the flow at the head required, so the farmer runs them longer, irrigates less area than planned, or discovers the far end of the field never gets water.

Oversized pumps cost more to buy, more to run, and wear faster. The engineering guidance above is explicit about the maintenance and component-damage consequences.

Both happen because the transaction starts with a price band rather than a duty point. A duty point makes three quotations comparable. A budget makes them incomparable.

There is a second, national dimension. Uganda's National Irrigation Policy records that "The Ugandan irrigation potential is 3.03 million hectare" and that "Currently, the ratio of cultivated area under irrigation to the irrigation potential is only 0.5%". A World Bank account of Uganda's Micro-scale Irrigation Program notes that "Currently, less than 2% of the irrigation potential is utilized, with only 77,000 ha of irrigated land on record." As irrigation expands, a great many pumps will be bought by people buying their first one. The specification habit is worth establishing now.

The decision you are making

You are producing a single specification line that you can hand to three suppliers:

"I need [X] cubic metres per hour at [Y] metres total head, drawing from [source type] with a dry-season water level [Z] metres below the pump position, powered by [power source]. Please quote a pump whose performance curve puts that duty point near its best efficiency, and show me the curve."

Everything below produces the numbers in that sentence.

Uganda-specific context

Three Ugandan conditions change pump selection and are the reason a specification copied from elsewhere often fails here.

Altitude. Uganda sits high, and atmospheric pressure falls with elevation. Because a surface pump relies on atmospheric pressure to push water up the suction pipe, the practical suction lift available on a Ugandan farm is materially less than the sea-level figure printed in most catalogues. FAO's own bracket runs from "about 4-5m suction lift at sea level (reducing to around 2m suction lift at an altitude of 2 000m". This is the single most common cause of a Ugandan surface pump that will not draw.

Seasonal water level movement. Head is not a fixed property of a site. In the dry season the water surface in a shallow well, a stream or a borehole falls, static suction lift rises, and the pump moves further from its design point exactly when it is most needed. Every measurement in this article should be taken in the driest month, not the wettest. Falcon's water sources article sets out the same discipline for choosing the source.

Crop water demand varies enormously across the country. A peer-reviewed Ugandan study using FAO Penman-Monteith within CROPWAT 8 quantified grain amaranth water requirements at 467.5 millimetres per season in the Karamoja sub-region and 174.1 millimetres per season at Tororo. That is one crop, one method, and nearly a threefold difference inside one country. A demand figure borrowed from another district is not a demand figure for your farm.

Alongside those, the national picture matters commercially. Uganda's National Irrigation Policy records that "The Ugandan irrigation potential is 3.03 million hectare" and that "Currently, the ratio of cultivated area under irrigation to the irrigation potential is only 0.5%", while a World Bank account of Uganda's Micro-scale Irrigation Program notes that "Currently, less than 2% of the irrigation potential is utilized, with only 77,000 ha of irrigated land on record." Most Ugandan pumps sold in the next decade will be somebody's first pump, which is precisely why the specification habit is worth establishing now.

Step 1: work out the flow you need

Flow follows from crop water requirement, area and running hours.

The principle. FAO Irrigation and Drainage Paper 56 defines reference evapotranspiration as "The evapotranspiration rate from a reference surface, not short of water", denoted ETo, and crop evapotranspiration under standard conditions as ETc, related by "Experimentally determined ratios of ETc/ETo, called crop coefficients (Kc), are used to relate ETc to ETo or ETc = Kc ETo."

The honest limitation. Falcon could not verify a published reference evapotranspiration figure in millimetres per day for Uganda and does not state one. What it can show you is how much the requirement varies within the country. A peer-reviewed Ugandan study using FAO Penman-Monteith within CROPWAT 8 quantified grain amaranth water requirements at 467.5 millimetres per season in the Karamoja sub-region and 174.1 millimetres per season at Tororo. That is one crop, one method, two locations, and nearly a threefold difference inside one country.

What to do instead of guessing. Three options, in order of reliability:

1. Ask your district production office or an irrigation specialist for a local crop water requirement figure for your crop and zone, and record who gave it and when.
2. Use CROPWAT or an equivalent tool with local climate data, which is what the Ugandan study did.
3. Measure. Irrigate a small measured area for one season, record how much water you applied and when the crop showed stress, and use that as your own local figure. This is slower and it is the only figure that will be genuinely yours.

Then convert to a flow rate. Daily volume in cubic metres divided by the hours you can actually run the pump gives cubic metres per hour. Be honest about the hours. A solar system runs when the sun is up; a fuel pump runs when someone is there to watch it.

Add the other demands. Livestock drinking water, spraying, washing and household use all come off the same pump on many farms.

Step 2: calculate total head

This is the number suppliers most often ignore and it decides everything.

The components

FAO's manual gives the static component sum as "Total Static Head = Static Discharge Head + Static Suction Lift", and adds friction and pressure heads on top of that.

Total head is the sum of all four. Miss one and the pump underperforms.

Friction head, and why pipe diameter matters more than people think

FAO's manual states that the friction head "can be calculated using either hydraulic formulae or tables and graphs", and Falcon does not reproduce a formula it could not verify character-exactly. What it can give you are the velocity limits that make friction manageable, which is the practical control.

• FAO's manual, citing Keller and Bliesner, recommends that "for centrifugal pumps the diameter of the suction pipe should be selected such that the water velocity V < 3.3 m/s".
• FAO's pressurised irrigation handbook states that "The manufacturers of PVC and PE pipes recommend that the maximum flow velocity in the plastic pipes should not exceed 1.5 m/s."

On layout, the manual is direct: "The diameter of the suction pipe should not be smaller than the inlet opening of the pump and it should be as short and direct as possible. If a long suction pipe cannot be avoided, then the diameter should be increased", and "Like the suction pipe, the discharge pipe should be as short and free of elbows as possible, in order to reduce friction."

The practical rule that follows: buying a larger pipe is almost always cheaper than buying a larger pump. Friction losses fall sharply as diameter rises, and pipe is a one-off cost while pump energy is a recurring one.

The altitude problem, stated plainly

This is the single most Uganda-relevant section of this article.

A surface-mounted centrifugal pump does not suck water up. Atmospheric pressure pushes it up, and the pump creates the low pressure that allows it. US engineering guidance explains: "Atmospheric pressure determines the maximum practical suction lift ... As the pressure on the water at sea level is one atmosphere or 34 feet, the highest theoretical suction lift for pure water at sea level is 34 feet less friction losses. This maximum theoretical suction lift can not be obtained under actual conditions. Pump manufacturers usually recommend that the design suction lift be limited to 70 percent of its theoretical value."

FAO puts the practical figure at "about 4-5m suction lift at sea level (reducing to around 2m suction lift at an altitude of 2 000m", with such pumps "limited to pumping lifts of no more than 5-6m."

Uganda sits high. Much of the country is at considerable elevation, and the derating is real, not theoretical. A pump that works at the coast will not necessarily draw the same lift in Kabale.

The design consequence. If your dry-season water surface sits more than a few metres below where you want the pump, you have two options: move the pump closer to the water, or use a submersible pump. FAO describes electro-submersible pumps as "especially suitable for installation in deep boreholes", with both motor and pump submerged. Note their characteristic failure mode from the same source: "Submersible electrically driven pumps depend on cooling via the water being pumped, and a failure of the water supply can result in serious damage to the unit."

Falcon found no source stating a numeric depth threshold at which a submersible becomes mandatory, and does not invent one. The rule is the suction lift limit above, applied to your measured dry-season water level.

Step 3: choose the pump type

Do not choose the power source before the duty point. Whether solar makes sense depends on the head and flow you have just calculated, which is exactly why FF-605 sits downstream of this article.

Step 4: size the power

FAO's manual gives brake power as BP = Q x TDH / (C x E pump), with C = 102 for Q in litres per second and C = 360 for Q in cubic metres per hour.

US engineering guidance gives the same relationship in imperial units as whp = (Q)(H)(SG) / 3,960, where whp is water horsepower, Q is flow in gallons per minute, H is total dynamic head in feet and SG is specific gravity, and explains that "The power required at the pump shaft to drive the pump is called bhp and is equal to the water horsepower ... divided by the pump efficiency (decimal) at the desired flow and pressure."

On efficiency, be careful what you assume. Falcon could not verify a published typical efficiency band for centrifugal irrigation pumps and does not print one. What is verified: FAO's manual notes that "larger pumps have higher efficiencies", that "motor efficiencies are in the range of 0.88 - 0.92", and that V-belt or gear transmission losses "do not exceed 5%". In FAO's own worked example the selected curve gave "the highest efficiency (65%) for the required discharge and head requirements", and a US worked example used 73 percent.

So use the efficiency from the actual pump curve for the pump you are being offered, not a rule of thumb. That is the whole reason to ask for the curve.

Step 5: read the pump curve

This is the step that separates a specification from a purchase.

Every genuine pump has a performance curve showing flow against head, with efficiency plotted across it. Ask for it. A supplier who cannot produce a curve for the pump they are selling is selling you a shape, not a machine.

FAO's manual gives two selection rules worth memorising.

Rule one, on the edges: "When the required Q and H combination falls outside the performance curve or when it falls at the fringes of the performance curve, that type of pump should not be selected."

Rule two, on which side to sit: "Ideally we would have preferred a pump where the required head and flow combination falls on the right-hand side of the efficiency curve. With age, the operating point will move to the left, then we would be able to operate with higher efficiency."

That second rule is a genuinely useful piece of practical wisdom: specify slightly to the right of peak efficiency, so that wear moves the pump toward its best point rather than away from it.

Cavitation, and why the suction side matters

FAO's manual explains: "In the eye of the impeller of a pump, pressure may be reduced to such a point that the water will boil. As the water is carried to areas of higher pressure in the pump, the vapour bubbles will collapse or explode at the surface of the impeller blades or other parts of the pump, resulting in the material erosion."

The condition to avoid: "If the energy available at the intake side is not sufficient to move the water to the eye of the impeller, the water will vaporize and the pump will cavitate. In order to avoid cavitation the NPSHA should be higher than the NPSHR required", where "NPSHA = atmospheric pressure - static suction lift - friction losses - vapour pressure".

Look at what is in that expression. Atmospheric pressure, which falls with altitude. Static suction lift, which rises as the water table drops in the dry season. Friction losses, which rise if the suction pipe is too narrow or too long. All three move in the wrong direction on a Ugandan farm in the dry season, at the same time. That is why cavitation shows up in February rather than in April.

US engineering guidance describes the damage: cavitation bubbles collapsing at the impeller cause the characteristic noise and erode "the impeller material, primarily from the tips of the impeller."

If a pump rattles, whines or sounds like it is pumping gravel, stop it. That is not a noisy pump; it is a pump destroying itself.

Field record: pump specification sheet

Take this to every supplier. It is what makes quotations comparable.

The three rows that most often decide whether the installation works are the dry-season water level, the site altitude and whether the vendor supplied a curve.

Costs

Falcon publishes no pump prices, installation costs or fuel prices, because none was verified and any figure would date quickly. Compare by category.

Common mistakes

• Starting with a budget instead of a duty point. It makes every quotation incomparable.
• Measuring the water level in the wet season. The dry-season level is the one the pump must work at.
• Ignoring altitude. FAO's figure falls from 4 to 5 metres of suction lift at sea level to about 2 metres at 2,000 metres. Uganda is high country.
• Forgetting friction head. It is invisible until the far end of the field stays dry.
• Forgetting the operating pressure of the application method. A sprinkler system needs pressure at the nozzle, and that is head the pump must produce.
• Undersized suction pipe. FAO says it should be no smaller than the pump inlet, as short and direct as possible, and larger if it must be long.
• Buying without seeing a pump curve. No curve, no specification.
• Specifying at the left-hand edge of the efficiency curve. Wear will move the operating point further left and efficiency will fall.
• Buying oversized "for future expansion". Engineering guidance links oversizing to higher operating cost, more frequent maintenance, valve damage, piping stress and noise.
• Running an engine pump far from its design point. FAO calls this "a common cause of gross inefficiency and wasted fuel."
• Letting a submersible run dry. FAO notes it depends on cooling from the pumped water and failure of supply can cause serious damage.
• Ignoring the noise. Cavitation noise is the sound of the impeller eroding.
• Treating abstraction authorisation as a later problem. It belongs in the same project. See FF-105.

Warning signs and escalation points

When this method works and when it does not

It works for any farm pumping from a defined source to a defined delivery point, which is nearly all irrigation pumping.

It works partially where the crop water requirement figure is uncertain, which in Uganda it usually is. Falcon could not verify a published Ugandan reference evapotranspiration figure. Where the demand side is uncertain, get the head calculation exactly right, since it is measurable, and size the flow conservatively while measuring your actual use in the first season.

It does not work if the water source is not settled. Head depends on the source and its dry-season level, so FF-105 comes first.

It does not replace an installer. This article gets you to a specification and a comparable set of quotations. Selecting the final pump against a real curve, and installing it safely, are professional tasks.

A low-risk pilot

1. Measure the dry-season water level and record the date. This single measurement is the foundation of everything else, and it is free.
2. Measure the elevation difference to your highest delivery point with a level and a tape.
3. Write the specification sheet as far as you can, leaving friction head blank.
4. Ask three suppliers to quote to that duty point, and ask each for the pump curve.
5. Compare the three curves at your duty point, not the three prices.
6. Before committing, hire or borrow a pump for one season if you can, and measure what you actually applied and how long it took. A season of real data will change at least one of your assumptions.
7. Check the abstraction position in writing before installation, per FF-105.
8. Only then buy, and keep the curve with the farm records.

Limitations and professional-support triggers

What this article does not do. It states no pump price, no installation cost, no fuel price, no efficiency band, and no numeric depth threshold for choosing a submersible, because none of those was verified. It does not reproduce a friction-loss formula, because Falcon could not verify one character-exactly from the FAO source and would rather give you the velocity limits than a possibly mis-transcribed equation. It does not size solar arrays, choose an irrigation method, design a layout or select a water source.

Two verification notes carried forward. First, no published reference evapotranspiration figure in millimetres per day for Uganda was verified, and none is stated. Second, the FAO reference crop's numeric characteristics were not confirmed and are therefore not printed. Both are gaps a local irrigation specialist can close for your site.

Bring in a professional when:

• A qualified electrician, for every electrical connection without exception, and before any submersible installation.
• An irrigation designer or agricultural engineer, to calculate friction head properly and to confirm the duty point before purchase.
• Your district production office or an irrigation specialist, for a local crop water requirement figure.
• A pump technician, at the first sign of cavitation noise, repeated priming failure or overheating.
• The Ministry of Water and Environment, for abstraction authorisation, as set out in FF-105.
• The installer, in writing, for warranty terms and spare parts availability, before money changes hands.

Clear next actions

1. Measure your source's water level in the driest month and write down the date.
2. Measure the vertical rise from the proposed pump position to the highest delivery point.
3. Get a crop water requirement figure from your district production office or an irrigation specialist, and record who gave it.
4. Work out your daily volume and divide it by the hours you can honestly run the pump.
5. Fill in the specification sheet as far as you can, and have an irrigation designer complete the friction head.
6. Send the same specification to three suppliers and ask each for the pump curve.
7. Compare the curves at your duty point and choose one that sits slightly right of peak efficiency.
8. Confirm the abstraction position in writing before installation.