Article content will be supplied from the Insights CMS collection.
EL CID — Electromagnetic Core Imperfection Detection — is the primary non-destructive test for generator stator core condition. It identifies lamination insulation faults that, if left unaddressed, can progress to core hot spots, stator winding damage, and forced outages that are significantly more consequential than the fault that initiated them.
The test is well-understood in its mechanics. What is less consistently understood is what the results actually mean — and how a single numeric output (the quadrature, quad, current, measured in milliamps) should drive engineering decisions about core condition, rewind necessity, and return-to-service timing.
This article covers how EL CID testing works, how to interpret the results, and what the commonly cited action thresholds actually represent in engineering terms.
What EL CID Tests
The stator core of a large generator is constructed from thousands of thin silicon steel laminations, electrically insulated from each other. The purpose of this insulation is to limit eddy current paths in the core — if the laminations were not insulated from each other, eddy currents induced by the magnetic flux would circulate through the core mass and cause significant heating. In an intact core, eddy currents are confined to small loops within individual laminations.
When the interlaminar insulation is damaged — whether by mechanical stress, contamination, thermal cycling, or foreign object damage — adjacent laminations make electrical contact. This creates a low-impedance path for circulating currents that produces localized heating at the fault location. The heat damages the lamination insulation further, which increases the fault, which increases the heating — a progressive failure mode.
What EL CID measures. EL CID induces a low-level magnetic flux in the stator core (typically at 4% of rated flux, which avoids heating the test excitation equipment while still producing a detectable signal at fault locations). A sensing coil traversed along each slot detects the quadrature (out-of-phase) component of the induced current, which is elevated at fault locations relative to intact core areas.
The output of an EL CID test is a plot of quad current (in milliamps) along each tooth and slot traversal. Elevated readings at specific core locations indicate lamination insulation faults at those positions.
Interpreting EL CID Results The 100mA Threshold — What It Means
The most widely used EL CID action threshold is 100 milliamps of quad current. Readings above this threshold are generally considered to warrant engineering review and potentially core repair. Readings below 100mA are generally considered acceptable.
What this threshold does and does not mean. The 100mA threshold is an empirical guideline — not a regulatory requirement or an absolute engineering limit. It is derived from correlation data between EL CID readings and thermal performance during full-flux ring tests. A reading of 110mA does not mean the core is about to fail, and a reading of 95mA does not mean the core is in perfect condition.
The threshold is most useful as a trigger for investigation, not as an accept/reject criterion in isolation.
Trend matters as much as absolute value. A core with peak readings of 90mA at every outage for fifteen years is in a different condition than a core where readings were 20mA six years ago, 60mA three years ago, and are now reading 90mA. The first core is stable; the second is degrading. The EL CID result in isolation does not distinguish between these situations — the historical trend does.
Reading Distribution and Fault Location
Where elevated readings appear is as important as how high they are. Faults at the core end regions are generally more consequential than faults deep in the core, because end-region laminations operate at higher flux density and heat generated at end-region faults has less path for dissipation.
A fault pattern concentrated in a single tooth over multiple tooth and slot traversals indicates a localized fault that may be accessible for repair without a full rewind. A diffuse pattern of modestly elevated readings across many slots suggests either core contamination, a more systemic insulation degradation, or a calibration artifact — and warrants different follow-up than the localized fault pattern.
Corrective Action Thresholds
EL CID results drive engineering decisions at several thresholds:
Below 100mA, stable trend: Monitor at next outage. No action required in the current window unless trend analysis indicates acceleration.
Below 100mA, accelerating trend: Engineering review warranted. The rate of change may indicate that the fault will exceed the threshold by the next outage interval. Depending on operating schedule and risk tolerance, some plants elect to address accelerating faults before they reach the action threshold.
100-200mA, localized fault: Core repair (focal grinding, cleaning, restacking of affected laminations) may be sufficient. Engineering assessment of fault location, depth, and accessibility should precede the decision between repair and rewind.
Above 200mA, or widespread pattern: Typically indicates a condition warranting core repair or replacement evaluation. A full-flux ring test may be warranted to assess thermal performance and validate the EL CID findings before committing to a rewind.
EL CID and the Return-to-Service Decision
EL CID results are frequently obtained during an outage and factored into the return-to-service decision. The interpretation challenge at return-to-service is that the results are time-pressured — the outage schedule is open, the repair window is constrained, and the team needs to make a decision about whether to intervene now or monitor.
Two consistent errors appear at this decision point:
Applying the 100mA threshold without trend context. A core with a new reading of 92mA — just below the threshold — from a baseline of 15mA four outages ago is in a different situation than a core that has been reading 85-95mA for the past twenty years. The first warrants a hold for engineering review; the second does not. Without the trend, the threshold appears to clear both equally.
Deferring without a quantified monitoring plan. When an EL CID result warrants monitoring rather than repair, the decision to defer should include a specific re-test interval, defined trend criteria that would escalate the decision, and documentation of the basis for deferral. Deferral without a monitoring plan converts an acceptable engineering judgment into an undocumented risk accumulation.
EL CID in the Context of a Broader Stator Condition Assessment
EL CID is one element of a comprehensive stator condition assessment. It should be interpreted in the context of:
Insulation resistance (IR) and polarization index (PI) testing. IR and PI testing assesses the condition of the stator winding insulation, which is distinct from the core lamination insulation that EL CID tests. A core with acceptable EL CID results and poor PI trending has a different risk profile than a core with elevated EL CID readings and strong insulation resistance.
Partial discharge (PD) testing. Partial discharge activity in the stator winding can indicate insulation degradation that may not yet manifest in IR/PI results. PD testing combined with EL CID gives a more complete picture of stator assembly condition than either test alone.
Visual inspection findings. Core end region condition, wedge tightness, end winding support condition, and contamination observations from visual inspection supplement the electrical test data.