Sub-Turbine Losing Capacity? The Pump May Not Be at Fault
The replacement assembly is picked out, and the decision is close to being made. The turbine in the deep well has been moving less water year after year, the sets run longer than they used to, and the far end of the system never quite comes up the way it once did. A new pump is the obvious answer, and the one most operators reach for. It is also the answer that is ordered before anyone establishes that the pump is actually what changed.
An aging submersible turbine losing capacity slowly is one of the few pump problems where the equipment is not automatically the suspect. The pump is one candidate among several, and on North Florida karst, it is not always the strongest. Ruling candidates in and out in a sensible order is what keeps a sound assembly on the ground, and what ensures a replacement, if one is needed, suits the well as it is now rather than the well as it was.
Start with What Actually Changed
Before any candidate can be ruled out, there has to be something to rule it out against: what the system used to deliver, at what discharge pressure, and how long a set used to take to finish.
Replacing a pump because output fell is like replacing a truck's engine because it climbs hills slower, without first checking whether the trailer got heavier. The engine may well be tired. It is also the most involved thing to change, and the last thing worth assuming.
The shape of the decline matters as much as its size: a steady drift over years points at things that themselves change slowly, like wear, incrustation, and ground movement.
Note roughly when the output first seemed different and what the discharge gauge read then. A rough timeline of when the change started narrows the candidate list faster than any single reading taken today.
Three Candidates That Get Named First
Three explanations come up in nearly every conversation about a well that has slowed down. Each is real, each has its own tell, and none should be assumed without the evidence that fits it.
The water level in the well: If demand has grown beyond what the well supplies at that rate, the pumping level falls lower than it used to, and the pump starves for part of the run. Added zones, longer run times, or neighboring wells behaving the same way point here.
Restriction at the intake: Mineral incrustation and iron bacteria narrow the open area water has to pass through. The tell is a well drawing down harder than it once did to produce the same rate, often alongside staining or a change in odor.
Sand and grit production: Solids carry their own signature. Grit collecting in strainers, filters, or the bottom of a tank confirms it quickly, and its absence removes it from the list just as quickly.
Gather the one piece of evidence each depends on, then move on.
Karst Ground That Does Not Stay Still
Here is the candidate that rarely makes the list at all, and the one North Florida makes real. The Floridan Aquifer sits in limestone, and limestone dissolves. Wells across this region typically finish somewhere between 100 and 400 feet, drawing from rock that is not a uniform sponge but a system of solution features, seams, and voids that carry water at very different rates.
That ground moves. Subsidence and the slow enlargement or collapse of solution features shift how water reaches a borehole and how far the level falls when a large pump pulls on it. Over enough years, a well can end up working against a different drawdown profile than the one it was set for, without a single component of the pump having changed. The assembly is doing what it always did. The conditions it does it in are not the same.
What points here: a pumping level under load sitting materially different from what older records show at the same rate, with no grit, no fouling evidence, and no change in demand to explain it. What rules it out: a pumping level that has stayed where it always was. If the well still holds level, attention belongs inside the equipment.
This candidate also changes what a replacement should look like. Lowering an identical assembly into a well whose behavior has shifted reproduces the old performance on paper and not in the field.
Wear Inside the Bowl Assembly
Now the pump itself. A submersible turbine builds pressure in stages, and each stage depends on tight running clearances between the impeller and the bowl it turns inside. Grit passing through year after year rounds the vane edges and widens those clearances. Water that should be handed to the next stage slips backward inside the assembly instead.
The result is a pump turning at the same speed and delivering less water. Discharge pressure falls first, and flow follows. Nothing about it is dramatic, which is why it goes unnoticed until a full set stops finishing on time.
What confirms it: the pump producing measurably less than its rating at a lift that has not changed. What weakens it: a pumping level that has moved a long way, which means something is loading the pump that the pump did not cause.
The Motor and the Water That Cools It
A submersible motor is cooled by the very water it pumps. Flow past the motor carries heat away, so when the well delivers less water, that cooling flow falls with it. A motor that ran comfortably for years can start running hot for a reason unrelated to the motor.
Supply belongs in the same conversation. Voltage sagging under load, or imbalance across the legs of a three-phase feed, makes a motor work harder and run hotter for the same job. Neither is a common explanation for a slow decline in capacity on its own. Their place on the list is different: this is the candidate that turns a gradual problem into a failed one while everyone studies flow. Trip history, run temperatures, and readings taken at the panel under load either raise or clear the flag in one visit.
Diagnosing a sub-turbine means live power at the panel and a wellhead under pressure. Leave meter readings, enclosure work, and any water level measurement to a licensed well contractor rather than doing it yourself.
What Leaves the Well and What Comes Back
Water that never reaches the field is indistinguishable, at the far end of the system, from water the pump never made.
Drop pipe joints loosen and corrode. A check valve above the pump can fail to seat, letting the column drain back between cycles so every start begins by refilling pipe instead of pushing water. Above ground, a partially closed valve, a sediment filter that has silted up, or a butterfly valve that no longer opens fully all reduce delivery while the pump does its job perfectly.
These are the easiest candidates to eliminate: they take the least disturbance to check. Pressure readings either side of a filter, a look for wet ground along the discharge run, and a valve position walk-down settle the question quickly. Skipping them is how a healthy assembly gets pulled.
Working from Least Disturbance to Most
The candidates above are not equally expensive to test, and that should determine the order in which they are tested.
A valve position walk-down, a filter pressure reading, and a look for wet ground along the discharge run cost an hour and a pair of boots. Panel readings and trip history cost a service visit. A water level taken under load costs a visit and the right instrument at the wellhead. Pulling the assembly costs a rig, most of a day, and a system that is down during the process. Each rung answers questions the rung below it cannot, and none of them get cheaper by being left until later.
Working in that order also protects the candidate this region adds, and the surface gives no sign of it. A karst well can change the job without anything happening to the pump: the level under load now sits well below what older records show, the assembly still makes what its curve promises at that new lift, and every part of it is as sound as the day it went in. Pull first, and the assembly comes up clean. The rig bill has been spent, the sets are still short, and the measurement that would have explained it has still not been taken.
If the well itself turns out to be the answer, that work belongs to a rig and a licensed well drilling contractor, not to a pump truck. Knowing which side of the line the problem sits on before anything is disassembled is what keeps the job from being done twice.
Frequently Asked Questions
Keep three numbers together: hours run per set, discharge pressure taken at the same point in the cycle each time, and the amp draw your technician records at that same point. The direction they move together is the useful part. Amps falling alongside pressure suggests the pump is simply moving less water. Amps climbing while output falls suggests drag or something binding, a different and more urgent repair.
It can, for a long time. A variable frequency drive holds the setpoint by raising motor speed, so the early part of the decline is absorbed silently while the gauge appears unchanged. The evidence lives in the drive: a controller now sitting at or near maximum speed to do what it once did at part speed has already told you capability has fallen. Drive run logs and fault history are worth pulling early.
It often is. If the well's behavior has shifted, resetting the assembly deeper can be the right answer, but it depends on enough sound casing below the current setting and on the motor still receiving flow past it. Where a setting ends up in oversized casing or above the producing interval, a flow-inducer sleeve, sometimes called a shroud, is the part that restores the cooling flow. It is a design decision, not a field improvisation.
No, and the order flips. A sudden loss points at discrete failures rather than trends: a tripped overload or protection device, a check valve stuck shut, a strainer plugged after the well passed debris, a broken shaft, or an assembly that has come apart. Electrical protection and controls are checked first because a step change almost always has a single event behind it, and events leave marks.
Frequently, and they tend to travel together. A well producing solids wears the pump that handles them, so grit damage and a changed well are a common pair rather than competing theories. The signature is a partial recovery: one issue gets corrected, output improves, and it still falls short of the old baseline. That remaining gap is the second cause.
Rehabilitation, redevelopment, screen work, and anything involving the borehole itself get referred out rather than handled as pump work. What that does to the sequence is the part worth knowing in advance: a rehabilitated well has to be yield-tested again once the rig is finished, because the numbers taken before the work no longer describe the well, and the assembly is selected against the post-work result. That means the pump decision waits on the rig's answer rather than running alongside it. Ordering an assembly off the old figures is how a well gets improved and then handed a pump matched to the condition it was rescued from.
Find out what the pumping level is doing under load before anything comes out of the ground — that reading, set against your well's history, is what tells you which one changed. Perry-Pump Repair Service LLC serves Lake Butler, Gainesville, and Alachua. Call (352) 320-2412.
