Pump care in the field: keep the motor running, stop paying for its decay

You already know how a pump fails on the bad days: a hard stop, a call from someone with no water, a scramble for parts. What's easier to miss is how it fails on the good days, quietly, over months, as bearings wear and the same gallon of water starts costing a little more power than it used to.

Small crews run to failure because nobody ever showed them the middle path between a ten-minute spot check and a full SCADA system. That middle path is a habit: look at the same few numbers every time you're at the panel, write them down somewhere that outlives you, and let the trend tell you when to act.

None of this needs new hardware. A notebook, a shared spreadsheet, or the log card taped inside the panel door will work, as long as someone keeps filling it in and someone else can find it later.

What should you check on every visit?

Two readings cover most of what you need to know: the kWh reading and the gallons reading. Add GPM when you can get it, and you already have enough to catch most problems early.

Round out the check with discharge PSI, and on a VFD, the speed or Hz it's running at. Listen and feel for bearing growl or vibration while you're there.

A gravelly sound paired with pressure swings usually means the pump is starving for suction. Fix that with a speed change or better suction piping, not a closed valve.

How does keeping the station clean protect the pump?

Housekeeping is part of the checklist, not separate from it. A motor sheds heat through its housing and, on a VFD, through its heat sink, so keeping louvers, filters, and heat sinks free of dust is part of what keeps a monthly infrared reading under 180°F, the point where a motor housing or VFD heat sink is worth investigating.

The same logic applies to the wet well or vault around the pump. Whatever collects there restricts flow, hides a developing leak, or adds load the motor has to fight against, so a five-minute look every visit, more debris than last time, standing water where there shouldn't be, catches problems while they're still cheap.

Wet-well practice specific to lift stations, grease, rags, and the housekeeping that keeps a collection-system pump from clogging, gets its own guide: see wastewater-collection-lift-stations.

How do you log a pump so the record survives turnover?

Keep the per-visit log to what fits in under two minutes: kWh start and end, gallons start and end, and whatever else you have handy, runtime, starts per day, discharge PSI, VFD percentage. If you don't have a number, write "unknown" and a note instead of leaving it blank.

Wherever you can, make a field a dropdown instead of free text. A dropdown is searchable later; a paragraph of notes usually is not.

Even a photo of a paper card taped to the panel, filed against the right asset, beats no record at all. A shared spreadsheet that calculates the change from the last entry saves the most time, since nobody has to do math at the panel or carry a log between shifts.

The point of the log isn't the log itself. It's having a baseline to check against six months from now, when someone asks whether a pump is really getting worse or just feels that way.

What do starts and run hours tell you that a spot check can't?

A spot check only shows you the ten or twenty minutes you're standing there. An hour meter tracks cumulative run time, and a start meter tracks how many times the motor has cycled on, together they show cycling behavior a single visit cannot.

One utility's booster pump ran over 180 times in a single day before a newly installed SCADA system flagged it, high enough that the vendor called to check whether something was broken. Control changes brought that down into the 75 to 100 range. Before the alert, the honest guess from occasional visits had been "50 or so."

The number matters because every start is hard on the equipment. A motor draws six to eight times its running amps at startup, which heats windings and stresses bearings, couplings, and impellers more than steady running ever does.

Fewer starts means a longer-lived motor, not just a lower power bill. Widen tank deadbands, add lag delays, or switch to a VFD soft start, then watch the starts-per-day number to see whether the fix actually held.

Can you time your pumps to the power bill?

Yes, within the limits of your storage. Make your lowest-kWh-per-thousand-gallons pump the lead, and rotate lag pumps only enough to keep them healthy, not so often that you're paying to run an inefficient one.

Use tank storage to avoid pumping during your worst rate windows, or when a well's specific capacity is already poor. If you're on a time-of-use tariff, shift heavy runtime to off-peak hours when tank levels allow it, and write the run plan down so night and weekend staff follow it too.

Two separate charges usually hide in the bill. A demand charge is tied to your single peak kW and is often locked in for the month, while a time-of-use charge is tied to when you actually used the power, and they reward different behavior.

That single monthly peak can drive half the whole bill, so knocking down that one spike, staggering which pumps start together, usually beats chasing kWh savings spread thin across the month. Shifting load overnight has trade-offs worth penciling out first: longer water age can push disinfectant demand up, and overnight backwashing creates its own residuals to handle.

When does decay call for a rebuild, and when does it call for a replacement?

Let a handful of triggers make the call instead of a gut feeling. Investigate when kWh per thousand gallons rises 15 percent or more against your baseline, and check controls and level sensors first when starts per day jump 25 percent or more without a matching change in demand.

On the well side, watch specific capacity: how many gallons a minute you get for each foot the water level falls while the pump is running. A drop of 15 to 20 percent from where you started means investigate. Let it slide 30 percent or more and start planning a rehab.

Match the rehab method to what's actually fouling the well: mechanical methods like brushing, surging, or air-lifting for slime and loose fines, acids for carbonate scale, oxidants or dispersants for biofouling and iron. A video log of the casing before you choose is cheap insurance, and recording specific capacity and kWh per thousand gallons before and after lets you reset your baseline from the new numbers.

Replacement is mostly a cost conversation. A pump's purchase price is typically under ten percent of what it costs over its life, and the energy it burns is over eighty percent, so a cheap pump that runs inefficient for years is not actually the cheap option. Correcting an oversized pump alone can recover fifteen to twenty-five percent of the energy it was wasting.

Two wells that look alike on paper will not always cost alike to run. A smaller motor pumping less water can still land at half the cost per gallon of a bigger one nearby, purely on the strength of bearing condition and control settings underneath, which is exactly why your own logged baseline matters more than comparing your number to a neighboring system's. When a pump and booster are mismatched and fighting each other on every cycle, no amount of rebuilding closes that gap; the fix is matching the motor to the job. The manager guide on energy walks through a real case and the dollar math behind that call.

Keep the trigger numbers where you can see them at the panel, not filed away. For the money math behind these same numbers, see energy-as-a-number-you-manage; for the wiring and controls behind starts and VFDs, see equipment-pumps-electrical.