Every water treatment system is really two systems working together — one moves water, one moves electricity. Nearly every troubleshooting mistake we make comes from ignoring one of them, or from jumping into the middle. This is how we diagnose at Axis: pick the lane, start at the head of it, and trace forward until the system tells you where it stops.
Two chains, one machine. Trace either one from its head and the exact link where it dies is your finding — and usually your root cause.
Video overview
9:51
Narrated overview of this module. The article below has the full procedure and the numbers.
01
Rule one: it's two systems. Pick your lane first.
The doctrineEvery water treatment system is a fluid system and an electrical system working together. Before touching anything, decide which one the symptom lives in — then diagnose that whole chain.
The symptom usually tells you which lane to open first. It's a starting bias, not a verdict — plenty of failures live at the handshake between the two (a pressure switch is a fluid device that talks electrical):
Smells hydraulic — start on the fluid rail
Pump runs, but low/no flow or pressurethe electricity is arriving; the water isn't moving right
Flows or pressures drifting from baselinefouling, scaling, plugging, valve position
Leaks, air in lines, cavitation noisephysical path problems
Quality off (TDS, hardness, chlorine through)a treatment stage not doing its job
Smells electrical — start on the power rail
Nothing runs — dead panel, no lightspower isn't arriving at all
Pump won't start, hums, or tripsstarter, overload, phase, or motor
Alarms, faults, or controls acting strangesignals, sensors, relays, PLC
Runs then stops on a patternan interlock or protection is doing its job — find which one
The handshake devices live in both worlds: pressure switches, float switches, flow switches, level transmitters, solenoid valves. When a symptom won't sort cleanly into one lane, one of these translators is usually the suspect — a fluid condition creating (or failing to create) an electrical signal.
02
Rule two: start at the beginning
The doctrineGo to the head of the chain — the well, or the disconnect — and trace forward, proving each step. The exact step where the system stops working is your finding.
Starting in the middle feels faster. It isn't. Starting at the beginning does three jobs at once:
It finds the where
Trace forward and there is a last point where everything is good and a first point where it isn't. That seam is the failure — no guessing, no "maybe it's the membrane."
It teaches you the system
Every trace is a forced tour of how this machine actually works, in order. Techs who trace from the head learn systems faster than techs who chase symptoms — and the knowledge compounds visit after visit.
It reviews the whole system
You walk past every stage on the way, so you catch the second problem — the half-closed valve, the empty day tank, the bypassed interlock — that a shortcut walks right past.
It usually finds the root
Failures flow downstream. The first broken step in the chain is very often the cause of everything after it — which is exactly why we start upstream of everything.
Trace the chain
Pick a rail, then tap each link in order — the check to make, and what a failure at that exact spot tells you.
1 · Source: is water available at all? tap ▾
well running / city supply on · storage level · source pressure
Dies here: nothing downstream can work — well pump, supply outage, or empty storage. Everything after this is a symptom, not a cause.
2 · Valves & prefilter: can water get through? tap ▾
every isolation valve position · prefilter ΔP · strainer condition
Dies here: a closed/half-closed valve or a loaded filter starving the system. The #1 "somebody worked on it last week" finding.
3 · Pretreatment: is each stage flowing and treating? tap ▾
pressure in vs. out of each vessel · hardness / chlorine tests at outlets
Dies here: a stuck control valve, a vessel in backwash/bypass, or media plugged solid. Also where quality failures (hardness or chlorine through) hide while flow looks fine.
4 · Pump: is it moving water, not just running? tap ▾
suction pressure · discharge pressure · cavitation noise · prime
Dies here: good suction + no discharge = pump problem. Bad suction = the problem is upstream — go back a link before condemning the pump.
5 · Membranes / final treatment: flows, pressures, quality vs. baseline tap ▾
Dies here: fouling, scaling, or a failed o-ring/element — and only now, with everything upstream proven, have you earned the right to blame the membranes.
6 · Delivery: tank, repressurization, point of use tap ▾
tank level & controls · delivery pump · check valves · service valves
Dies here: the treatment plant is fine — it's a delivery problem. Common when "no water" complaints arrive but the RO panel looks perfect.
1 · Supply & disconnect: is power arriving? tap ▾
meter voltage at the disconnect, all legs · disconnect actually on
Dies here: utility, a pulled disconnect, or a lost phase. Check all legs — a three-phase system missing one leg "has power" and still destroys motors.
2 · Protection: fuses, breakers, overloads tap ▾
continuity across each fuse · breaker positions · overload relay tripped?
Dies here: Why did it open? A fuse is the broken thing, never the cause — something downstream made it blow. Note it and keep tracing before you replace it.
3 · Control power: transformer & control circuit alive? tap ▾
voltage on the control transformer secondary · panel lights · HMI up
Dies here: the machine has muscle but no brain — control fuse, transformer, or a shorted control device pulling it down.
4 · Permissives & interlocks: is the logic allowing a run? tap ▾
low-pressure switch · tank floats · pretreat lockouts · E-stops · fault codes on the controller
Dies here: the most common "electrical" failure isn't electrical at all — an interlock is correctly reporting a fluid problem (low suction, full tank). Read what the interlock is protecting against, then jump to that link on the fluid rail.
5 · Starter / contactor / VFD: does the command become power? tap ▾
coil voltage present · contactor pulls in · voltage out of the contactor/VFD · drive fault log
Dies here: coil voltage present but no pull-in = bad contactor. Pulls in but no output = burned contacts or drive fault. The VFD fault log is a manual-in-miniature — read it.
6 · Motor / load: is the last device using the power? tap ▾
voltage at the motor leads · running amps vs. nameplate · winding resistance if dead
Dies here: power arrives and the motor still won't run (or draws crazy amps) — motor or a seized load. High amps often points back to a hydraulic bind: switch rails and find what's loading it.
Safety first on the electrical rail: panel work is for techs trained & authorized for it — LOTO before touching, meter every "dead" circuit, PPE per policy.
03
Rule three: there are always two-plus issues
The doctrineWhen something's broken there are at least two problems: the thing that's broken, and the thing(s) that caused it to break. Fix only the first one and you've scheduled the return trip.
Replace only the last domino and the first one is still lying there, ready to knock everything over again.
Real patterns from our world — notice the visible failure is never the whole story:
Case pattern
Symptom: no product water→Broken: feed pump burned out→ caused byran dry on low suction→ caused byRoot: plugged prefilter nobody changed
Replace only the pump and the new one dies the same death in weeks. The fix is pump + filter + a look at why the low-suction protection didn't stop it.
Case pattern
Symptom: breaker trips on start→Broken: blown fuse / tripped OL→ caused bymotor drawing high amps→ caused byRoot: seized pump from scale — a fluid problem
A fuse is never the root cause — it's the messenger. Swapping fuses until one holds just burns up the motor. And note the trail crossed rails: electrical symptom, hydraulic root.
Case pattern
Symptom: permeate flow way down→Broken: membranes scaled→ caused byantiscalant not feeding→ caused byRoot: pump lost prime after the last chemical refill
Clean the membranes without restoring the feed and you'll clean them again next quarter. The PM habit of checking day-tank drop against expected use catches this one before it costs a CIP.
The test before you leave: for every broken part you replaced, can you say out loud what made it fail — and is that fixed too? If the honest answer is normal wear, an act of nature, or a source-water condition — fine, the trail legitimately ends there. If it's "not sure," you're not done. This one habit is the difference between one trip and three.
04
Rule four: manuals are your best friend
The doctrineThe manual's entire reason for existing is to explain how this system is supposed to work. You cannot recognize "broken" until you know what "working" is defined as.
Nobody memorizes every control valve, VFD, and controller we install. You don't have to — you have to know where the answers live and go get them before guessing. The four sections that solve most calls:
Sequence of operation
What's supposed to happen, in what order, when the system starts, runs, flushes, and stops. Compare the real machine against this sequence and the deviation point jumps out — it's the manual's version of "start at the beginning."
Fault & alarm code tables
The controller is often telling you the exact answer in a code. Look it up instead of clearing it — and write the code on the work order before you reset anything, because that code is evidence.
Wiring & P&ID diagrams
The map of both rails: every interlock on the electrical chain, every valve and instrument on the fluid chain. Tracing on paper first makes tracing in the panel twice as fast — and safer.
Specifications & setpoints
Design flows, pressures, switch settings, motor amps. These are the "what good looks like" numbers you judge your readings against when there's no commissioning baseline handy.
Where to find them: the site binder, the manufacturer's website (nameplate model + serial gets you the PDF in two minutes — that's one reason we photograph nameplates on every PM), and our shared knowledge folder in Google Drive, where system O&M manuals live alongside articles like this one. On site with no manual and no signal? Call the office before you start guessing — pulling the doc for you takes five minutes.
05
Rule five: slow is smooth, smooth is fast
The doctrineSpend the first 15–20 minutes orienting yourself before touching anything — especially before jumping on the "obvious" issue.
The obvious issue is a trap in two ways: it's often the last domino, not the first — and charging at it skips the walk that would have shown you the real chain. Twenty minutes of orientation routinely saves two hours of wrench-spinning, a parts run, and a return trip. The ritual:
Get the story first. Ask the operator/customer: what happened, when, what changed recently, and what has anyone already tried? "It quit after the power blinked Tuesday" is half the diagnosis."What changed?" is the highest-value question in troubleshooting.
Walk the whole system as-found — hands in pockets. Both rails, head to tail. Look, listen, smell. Photograph the panel and anything abnormal before you disturb the evidence.
Read what the system is telling you. Gauges, flow meters, HMI screens, fault codes, pump hour meters. Compare against the baseline and PM history on the work order before forming a theory.
Pull the manual / history for anything unfamiliar. Five minutes with the sequence of operation beats forty minutes of educated guessing on a machine you've never met.
Now pick the lane, go to its head, and trace. With the story, the walk, and the numbers, your first theory is usually right — and you'll prove it in order instead of shotgunning parts.Parts-cannon troubleshooting is the most expensive kind there is.
The exception to "slow": active flooding, chemical release, electrical hazard, or anything unsafe — make it safe first, immediately. Orientation starts after the site is safe.
The one thing to remember
Pick the rail. Go to its head. Trace forward — the step where it stops is your answer, and the chain behind it is your root cause.
Axis Water Technologies · 855-443-2947 · info@axiswater.com · axiswater.com · 3462 W Loop 289, Lubbock, TX 79407Axis technician training library · content as of October 1, 2026
