Solar Well Pumps: What Actually Sets the Size

Search for a solar well pump and you will be handed a list of pumps. That is the wrong end of the problem. Nearly every decision here is settled before a model is chosen, by the well and by where the water has to go.

There are really two questions wearing the same words, and they have different answers.

Putting in a pump to run on solar is the straightforward one. Purpose-built DC pumps start gently, run slowly through the day, and fill a tank. Most of this page is about sizing one.

Running the AC pump you already have on solar or a generator is the harder one, and it fails for a reason that has nothing to do with head. That is the last section.

Head is the answer, not the pump

Lifting water is work, and the work is set by how much water and how far up. A pump is only ever a more or less efficient way of doing that work. Get the head right and choosing a pump is reading a curve; get it wrong and no pump will save you.

Total dynamic head is five things added together:

• Depth to standing water. How far down the water sits when nothing is running.
• Drawdown. How much that level falls while the pump runs. Your well log gives this at a stated pumping rate.
• Vertical rise. From the wellhead up to wherever the water is delivered.
• Friction loss. Resistance in the delivery pipe, read off a friction table. Small on short fat pipe, large on long thin pipe, and routinely ignored on exactly the long runs where it matters.
• Pressure head. Any pressure the pump has to deliver at the far end. See below, because this one is usually the biggest.

Well depth is not depth to water

This is the mistake that wastes the most money, and it is easy to see why people make it. The depth of the well is the number an owner knows. It is on the drilling report, it is what the driller quoted, it is the number you say out loud. The depth to water is a different number that most people have never looked up.

A 400 foot well with water standing at 80 feet is an 80 foot lift. Treating it as a 400 foot lift multiplies the head by five, and the pump, the array and the cost follow it up.

The figure you want is the static water level, and it is on the well log. If you do not have the log, a water level meter will tell you, and some counties keep well records you can request.

A pressure tank is usually most of the work

Here is the part that surprises people, and it changes what you should build.

Every PSI the pump delivers costs 2.31 feet of head. That is not a rule of thumb, it is a unit conversion: a PSI is 144 pounds per square foot, water weighs about 62.3 pounds per cubic foot, and the ratio is 2.31.

A common residential pressure switch cuts out at 50 PSI. That is 115 feet of head the pump works against, and it appears on no tape measure anywhere on the property.

Put that next to a shallow well and the proportions are startling. With water standing at 20 feet, a few feet of drawdown, a short rise and modest pipe friction, the lift is around 35 feet and the pressure is 115. The pressure tank is roughly three quarters of the total work. That share barely moves across any plausible water level, so you do not need to know your well precisely to know it is true of yours.

Which makes the open storage tank the single largest saving available. Pump to a tank at atmospheric pressure, then pressurize from the tank separately, or gravity feed if the site allows it. Same water, same well, and on the numbers above the daily energy falls by about three quarters. This is most of the reason solar water pumping works economically at all, and it is why nearly every purpose-built solar pump installation you will see has a tank in it.

The tank also does something a bigger array cannot: it stores water instead of electricity, and a water tank is far cheaper per unit of storage than a battery. A few days of water in a tank costs a fraction of a few days of autonomy in a bank. Size the tank first and the electrical system gets easier.

A well pit buys suction, not energy

If your pump sits in a pit below grade, it is there to solve a suction problem. A jet pump can only pull water up so far before atmospheric pressure runs out, so dropping the pump closer to the water brings it back inside that limit. That is a real and sometimes necessary fix.

It saves no energy at all. Whatever depth the pit takes off the suction side, it adds straight back on the discharge side, because the water still has to climb out of the pit on the way to the tank. Total dynamic head is the net rise from water to point of use, and moving the pump somewhere inside that path does not change either end of it.

Worth saying plainly because the intuition runs the other way: a pit looks like it should help the running cost, and it does not.

A well pit is a confined space, and that is not a figure of speech. Heavier-than-air gases collect in pits, oxygen can be displaced, and people are killed in them every year. Never run an engine near an open pit, never treat it as somewhere to put equipment, and never put batteries or an inverter down there. If you are working in one, that is confined-space procedure, not a ladder and a hopeful attitude. The carbon monoxide guidance that applies to generators applies here with less margin, because a pit does not ventilate.

Measure your casing before you shop

Most DC solar submersibles have bodies between three and four inches across. A two inch casing, which is what a packer well normally has, will not accept them, and no amount of correct sizing gets round it.

This rules out most of the market before any other specification matters, and almost nothing written on the subject mentions it. Measure the casing, or find it on the well log, before you read a single product page.

The spec that decides it, and nobody publishes it

If you are trying to run an existing AC pump, from a generator or an inverter, one number decides whether it works: the current the motor draws at the instant it starts. That is locked rotor current, and it is several times the running figure.

For residential pumps it is generally not published. Not on the nameplate, and often not in the manufacturer’s literature either. Nameplates usually give full load amps at each wiring voltage and nothing about starting at all.

That is not an oversight, and understanding why makes the situation clearer. A jet pump is sold as a plumbing product. Its literature is pages of flow against pressure and lift, with a single line of electrical. On a utility service the omission costs nothing, because the supply is effectively unlimited, the motor starts regardless, and the branch circuit is sized from full load current rather than from starting current. Nobody in the normal installation path ever needs the number.

It only becomes the deciding specification when the source is limited, which is exactly a generator, an inverter or a solar controller. That is the case the manufacturer is not writing for.

So every answer you find is a rule of thumb. Three times running watts, two times for a capacitor start, and so on. None of them are derived from a specification, because there is no specification to derive them from, and they get repeated until they read like data. Treat them as the guesses they are.

What you can actually do:

• Measure it. A clamp meter that captures inrush will give you the real figure in a few minutes, and then you have something the datasheet does not.
• Provide margin and accept the uncertainty, which is what most people do, while knowing that is what you are doing.
• Remove the question. A soft starter, or a purpose-built solar pump that ramps up instead of slamming on, makes starting current stop being the constraint.

When a pump will not start, suspect the capacitor first

A pump that hums, will not turn, or trips a breaker looks exactly like an undersized generator, and usually is not one. On a capacitor start motor the start capacitor is a wearing part, and a failing one produces precisely those symptoms.

The expensive version of this mistake is buying a larger generator to fix a part that costs a few dollars. Check it before you conclude anything about your power source.

What is behind the numbers here

The head arithmetic is unit conversion and geometry, and it is the same method any pump manufacturer’s sizing guide uses. The 2.31 feet per PSI figure is derived above rather than asserted.

The starting current position is a finding rather than an omission. We looked for locked rotor current on a 1 HP Goulds J10S, a common residential jet pump, and it is on neither the nameplate nor the manufacturer’s technical brochure, and there is no NEMA code letter to derive it from either. A multiplier on full load amps would have produced a number, and publishing that number would have meant presenting arithmetic on an assumption as a specification. It is left out on purpose.

This page names no products, for the same reason the generators page does not. Pump performance is published as curves in datasheets, a recommendation is only as good as the curve it was read from, and casing diameter rules out most candidates before performance is even relevant. Naming models needs that verification done properly rather than quickly.

Once you have a head figure and a daily gallon figure, the electrical side is the same problem as any other load. The system sizing calculator turns a daily watt-hour figure into an array and a bank, and the appliance power budget will tell you what the rest of the system is already carrying. If you are weighing a generator against more storage, generators for off-grid systems covers why that machine charges a battery rather than running a house.