Home/Reliability/Power Quality Analysis
Industrial Power Quality Analysis
Equipment trips and nobody can say why. Drives fail at one machine and nowhere else. A PLC resets on second shift but never on first. Before anyone replaces more hardware, it is worth measuring what the electrical system is actually delivering.
The Problem
Power quality problems are expensive precisely because they are intermittent. The failure is real, the cost is real, but the cause does not show up on a meter reading taken at 10 a.m. on a Tuesday. So the plant does what it can: it replaces the drive. Then it replaces the drive again.
The patterns are familiar once you have seen a few:
- Breakers or drives trip on no obvious overload, often at the same time of day.
- VFDs fault on overvoltage, undervoltage or DC bus faults, and fail early at one location.
- Controls and PLCs reset or behave erratically without an alarm that explains it.
- Motors run hot, or hum and vibrate on a system that measures fine at the panel.
- Transformers and neutral conductors run hotter than their loading should justify.
- Capacitor banks fail repeatedly, or fuses on them keep operating.
- Something in the plant misbehaves only when a specific large load starts.
Each of those has a measurable electrical signature. None of them can be diagnosed by inspection alone.
What Premier Does
We install a three-phase power quality analyser at the point that matters — the service, a distribution board, a specific MCC bucket, or at the affected machine — and log it continuously through enough production cycles to capture the event.
Duration is set by the problem. A daily-recurring fault may need three days. A problem that appears once a week gets a week or more. Monitoring that stops before the event happens tells you nothing, and we would rather say that up front than deliver a clean report on a plant that still has the fault.
The analysis correlates what the instrument recorded against what the plant was doing: which loads started, which shift was running, what the utility was doing. A voltage sag at 06:14 every morning is a different problem from one that follows a specific press cycle, and the recommendation is different too.
Scope note
Premier performs the monitoring, analysis and reporting. Where the finding calls for physical corrective work — a reactor at a drive, a filter, reconductoring, a neutral upsize, capacitor bank work — that work is coordinated with appropriately licensed electrical contractors, and we verify the result afterwards.
Measurement
What We Monitor
Logged continuously across all three phases and neutral, at the point of interest:
- Voltage (per phase & line-to-line)
- Current (per phase & neutral)
- Voltage imbalance
- Current imbalance
- Voltage harmonics (THD & per order)
- Current harmonics (THD & per order)
- Power factor (displacement & true)
- Real, reactive & apparent power
- Voltage sags (dips)
- Voltage swells
- Transients
- Interruptions
- Flicker
- Frequency variation
- Neutral current
- Demand & load profile
- Waveform capture on event
- Crest factor
Common causes we look for
Non-linear loads — drives, rectifiers, switching supplies, LED lighting at scale — injecting current harmonics that heat transformers and neutrals and distort the voltage everyone else sees.
A small percentage imbalance produces a much larger current imbalance in a motor. It is one of the more common reasons motors run hot on a system that otherwise looks fine.
Large motors starting across the line pulling the bus down far enough to drop out contactors and fault drives elsewhere in the plant.
Capacitor switching, utility events and inductive load switching producing voltage spikes that drives and electronics do not survive indefinitely.
Excess neutral current from triplen harmonics or unbalanced single-phase loading, and grounding arrangements that let noise onto reference conductors.
Power factor correction interacting with system harmonics to produce resonance, amplifying distortion and destroying the capacitors themselves.
Process
How a Study Runs
Define the Symptom
What fails, how often, what it is connected to, and what else is running when it happens. The single most useful thing a plant can provide is an honest timeline of the failures.
Select the Monitoring Point
Service entrance for a system-wide problem, or at the affected equipment for a local one. Sometimes both, because the answer is usually in the difference between them.
Install & Log
Analyser connected and left in place through enough production cycles to capture the event, with the connection coordinated through your electrical safety program.
Correlate
Recorded events are matched against production data, shift patterns and load starts. Data without context is a graph; data with context is a diagnosis.
Compare Against Criteria
Findings are evaluated against recognised limits and equipment tolerances — what the system delivers versus what the affected equipment is rated to accept.
Report & Recommend
A written finding on whether power quality is contributing, what is causing it, and what the practical options are — with the cost and disruption of each stated plainly.
Verify
Where corrective work is carried out, monitoring is repeated under comparable conditions to confirm the problem is actually gone.
What You Receive
- Monitoring summary — where the instrument was installed, over what period, and what the plant was doing.
- Measured results against each parameter, with the trends and the event log.
- Event analysis — waveform captures of the significant events with an explanation of what each one is.
- A direct answer to whether power quality is contributing to the reported problem, including when the answer is no.
- Root-cause assessment where the data supports one, and what further measurement would be needed where it does not.
- Recommendations, ranked, with the practical trade-offs of each stated.
- Baseline data retained for comparison, so the next study starts from something.
Why It Matters
Power quality problems get paid for either way. A plant that replaces the same drive three times has spent the cost of a monitoring study without learning anything. The difference is that the study ends with the problem understood, and usually with a fix that stops the recurring spend.
There is a documentation benefit as well. If the disturbance originates on the utility side, measured evidence is what turns that conversation from a complaint into a case.
Related
Often Runs Alongside
Get an Answer on the Trips You Cannot Explain.
Describe the symptom, how often it happens, and what equipment is affected. We will tell you where monitoring should go and how long it needs to run to catch it.