IEEE Std 519-2022 — Interpretation and Common Field Pitfalls

§01 What the standard is — and is not

A system-boundary standard, enforced at one point only

IEEE Std 519-2022 was approved on 13 May 2022 and published in August 2022; it supersedes IEEE 519-2014. The document was renamed from "Recommended Practice and Requirements…" to "IEEE Standard for Harmonic Control in Electric Power Systems" — a deliberate signal that its limits are written to be applied as requirements, not advisory practice.

The standard sets steady-state harmonic limits at the interface between the supply system and the user — the Point of Common Coupling (PCC). It does not address transients, and it does not set limits for individual machines, panels, or feeders inside a facility. Compliance is a property of the boundary, established by a single statistical measurement campaign at that boundary.

User responsibility

Limit harmonic current at the PCC

Keep injected current harmonics (assessed as TDD and per-order limits) within Table 2/3/4, selected by the short-circuit ratio at the PCC.

Utility responsibility

Limit voltage distortion at the PCC

Keep supply voltage distortion within Table 1 by managing system impedance — a near-sine voltage is owed to every connected user.

519 is satisfied or violated only at the PCC — never at a drive terminal, a sub-panel, or an LV main inside the plant. Anyone applying the current-harmonic limits to a single piece of equipment is applying the standard incorrectly.

Indian regulatory context

Under the CEA (Technical Standards for Connectivity to the Grid) (Amendment) Regulations, 2019, the limits of voltage harmonics by the distribution licensee, the limits of current-harmonic injection by all HV and EHV consumers and prosumers, and the point of measurement (the PCC) are to follow IEEE 519 "as amended from time to time." Because 519-2022 supersedes 519-2014, the 2022 edition is the operative reference, and TNERC implements the obligation in its power-quality framework. Measurement instruments must additionally satisfy IEC 61000-4-30 Class A (Ed. 3) with harmonic measurement per IEC 61000-4-7 — the BIS-recognised IEC measurement basis.

§02 The five quantities you must get right

Definitions, exactly as 519-2022 frames them

Most disputes are not about harmonics at all — they are about getting one of these five quantities wrong. Two of them, I_L and I_sc, are external numbers inserted into the measurement, and they decide the verdict before the meter is even read.

PCCPoint of Common Coupling

The point on the public supply electrically nearest the load at which other loads are, or could be, connected — the interface between the system owner or operator and the user, located upstream of the installation. The PCC is a technical boundary, not a property of the DISCOM: it is the consumer's power-receiving point from the DISCOM, at which the consumer carries the responsibility to share its distortion data. It should not be equated with the DISCOM's metering point — doing so would affect the consumer's rights. For an HV consumer with its own transformer it lies on the HV / incoming side; for an LV consumer fed from a shared distribution transformer it is that transformer's LV side.

I_sc · SCRShort-circuit current & ratio

I_sc is the actual available short-circuit current at the PCC — set by the upstream source MVA and the series impedance to the PCC, not merely the transformer's theoretical fault level. SCR = I_sc/I_L selects the current-limit band — a higher SCR (stronger grid) earns more relaxed limits.

iTHDTotal harmonic current distortion

RMS of the harmonic currents (orders up to the 50th, interharmonics excluded) as a percentage of the fundamental current. It rises at light load and falls at heavy load. It is not the compliance metric.

TDDTotal demand distortion

RMS of the harmonic currents (up to the 50th) as a percentage of the maximum demand current I_L. This is the current-compliance metric. iTHD is only a step on the way to it.

I_LMaximum demand load current — the 2022 definition (Clause 3.1)

Established at the PCC as the sum of the RMS currents corresponding to the 15-min or 30-min maximum demand in each of the twelve previous months, divided by 12. If twelve months are not available (short service life), sum the maximum 15/30-min apparent-power demand over the months available and divide by the number of months. For a proposed new installation, base I_L on the projected 15/30-min maximum monthly apparent-power demand over the year following operation of the harmonic-producing loads on the service application. This explicit treatment removes the 2014 ambiguity that left I_L undefined when twelve months of data did not exist.

TDD = iTHD × ( I_meas / I_L ) where I_meas ≈ fundamental current during the test

This single relationship is the engine behind most compliance disputes: the same physical emission (a fixed iTHD in amps) produces a very different TDD depending on the I_L you divide by — and I_L is supplied externally, not measured during the test.

§03 The real system

From the grid source to the PCC — and where the analyzer sits

A typical 33 kV industrial connection. The limits apply at the PCC, and that is exactly where a Class A power-quality analyzer is installed — tapped through the metering CTs and VTs, alongside the DISCOM energy meter.

FIG. 1 — Single-line: utility source → DISCOM substation → PCC (with Class A analyzer + energy meter) → consumer transformer → LV loads. The blue values at each stage update live from the §06 calculator.

★ PCCthe contractual boundary; the only place limits are assessed.

CT / VTmetering current & voltage transformers feeding the analyzer.

Class Aanalyzer to IEC 61000-4-30 Ed.3 — legally defensible aggregation.

Dashed zoneequipment terminals — 519 limits are not applied here individually.

FIG. 2 — Typical non-linear loads (variable-frequency drives, UPS, rectifiers, EV chargers, LED drivers) draw non-sinusoidal current; the resulting current harmonics flow upstream and combine at the PCC, where IEEE Std 519-2022 sets the limits. Compliance is assessed on the aggregate current at the PCC — not at any single load.

§04 The limits

Voltage, current, and what 2022 changed

Voltage limits are unchanged from 2014. The current table gained two real changes — even-order harmonics and inverter-based resources — and the column structure was adjusted to carry the even-harmonic rule.

Table 1 — Voltage distortion limits (at the PCC)

IEEE Std 519-2022, Clause 5.1 — unchanged from 2014
Bus voltage V at PCC	Individual harmonic (%)	THD (%)
V ≤ 1.0 kV	5.0	8.0
1 kV < V ≤ 69 kV	3.0	5.0
69 kV < V ≤ 161 kV	1.5	2.5
161 kV < V	1.0	1.5

Table 2 — Current distortion limits, 120 V through 69 kV (the band that applies to 11 / 22 / 33 kV consumers)

The highlighted row tracks the §06 calculator below — change I_sc or I_L there and watch which limit band applies.

IEEE Std 519-2022, Clause 5.2 — maximum harmonic current distortion in percent of Iₙ
I_sc/I_L (SCR)	2 ≤ h < 11 ^a	11 ≤ h < 17	17 ≤ h < 23	23 ≤ h < 35	35 ≤ h ≤ 50	TDD
< 20	4.0	2.0	1.5	0.6	0.3	5.0
20 – 50	7.0	3.5	2.5	1.0	0.5	8.0
50 – 100	10.0	4.5	4.0	1.5	0.7	12.0
100 – 1000	12.0	5.5	5.0	2.0	1.0	15.0
> 1000	15.0	7.0	6.0	2.5	1.4	20.0

^a 2022 even-harmonic rule: for h ≤ 6, even harmonics are limited to 50% of the value shown; even harmonics of order h > 6 are limited to 100% of the value shown. The band header begins at h = 2 (2014 began at h = 3). Tables 3 (69–161 kV) and 4 (> 161 kV) carry the same structure with progressively tighter values — for Indian 11/22/33 kV industrial consumers, Table 2 is the relevant one.

Change 1 — Even-order harmonics relaxed

Even harmonic order	IEEE 519-2014	IEEE Std 519-2022
h ≤ 6 (i.e. 2, 4, 6)	25% of the odd limit	50% of the limit (relaxed ×2)
h > 6 (i.e. 8, 10 …)	25% of the odd limit	100% of the limit (relaxed ×4)

The change acknowledges that inverter-based resources, active front-end drives, and active filters can produce some even-order content that the stricter 2014 limits flagged unnecessarily.

Change 2 — Inverter-based resources are now in scope

In Clause 1.2 (Purpose) the word "passive" was removed: any user equipment that alters system impedance and raises voltage distortion — including inverter-based resources (IBR) and distributed energy resources (DER) — is now within the standard's purview. Per Clause 5.2, where IBR/DER generation (e.g. rooftop or captive solar) exceeds 10% of the annual average demand, the current-harmonic limits of IEEE 1547 or IEEE 2800 apply at the PCC instead of Table 2; at or below 10%, the Table 2 limits apply. The 2014 edition gave no specific treatment for grid-interactive equipment.

§05 How a compliant measurement is made

A full week, business-as-usual, on a Class A instrument

519 compliance is statistical, not a single reading. A meter "showing 14% TDD right now" tells you nothing about compliance until it is processed over the assessment window against the percentile rules.

Instrument and basis

Power-quality analyzer compliant with IEC 61000-4-30:2015 Class A (Edition 3) — Edition 3 cancels and replaces the 2008 Edition 2, so a legally defensible measurement requires an Ed.3 Class A instrument, type-tested as a power-quality instrument (PQI) under IEC 62586.
Under IEC 61000-4-30 Class A, every quantity is built from a 200 ms basic measurement (10 cycles at 50 Hz / 12 cycles at 60 Hz), which is then aggregated up to the 3-second very-short-time (VST) interval and the 10-minute short-time (ST) interval; harmonics are measured to IEC 61000-4-7.
The 3-second interval is mandatory for IEEE Std 519-2022. The daily 99th-percentile limit is evaluated on 3-second VST values, so an instrument that does not produce the 3-second aggregation cannot generate that statistic and must not be used for an IEEE Std 519-2022 assessment.
Measurement located at the PCC, captured under normal, business-as-usual facility operation.

Statistical compliance assessment — the percentile rules (minimum one-week window)

Daily 99th percentile VST (3-second) harmonic currents should be less than 2.0× the Table 2 values.
Weekly 99th percentile ST (10-minute) harmonic currents should be less than 1.5× the Table 2 values.
Weekly 95th percentile ST (10-minute) harmonic currents should be less than (≤) the Table 2 values.

This three-tier rule allows brief peaks while holding the sustained emission to the table limit. A few-hour snapshot cannot produce any of these percentiles and is therefore not a valid 519 assessment — only a continuous campaign over the assessment week can.

IEEE Std 519-2022 is satisfied or violated only at the PCC — never at an individual drive, panel, or the LV main.

§06 See it yourself

The compliance calculator — from fault level to verdict

Enter the system data and the tool first derives the available fault current I_sc at the PCC (impedances combined per IS 13234 / IEC 60909). That I_sc then drives the IEEE Std 519-2022 SCR band and the PASS/FAIL on your measured emission. Change only I_L, or only the assumed fault level, and the verdict moves while the physical harmonics stay put.

Stage 1 · Short-circuit current at the PCC I_sc = U_n / (√3 · |Z|) · Z = Z_source + Z_tx + Z_line

PCC voltage U_n (line-to-line) kV

Upstream source fault level S_k MVA

Power transformer rating S_T MVA

Transformer impedance u_k %

Line / cable length km

Line reactance x Ω/km

Line resistance r Ω/km

Source reactance X_S = U_n²/S_k—

Transformer reactance X_T = (u_k/100)·U_n²/S_T—

Line impedance (R + jX)—

Total |Z| at PCC—

Available fault current I_sc—

Source and transformer modelled as predominantly reactive (R ≪ X at HV); line R and X are entered explicitly. Voltage factor c = 1.0 — IS 13234 / IEC 60909 apply c = 1.10 for the maximum short-circuit current above 1 kV, which would raise I_sc by ~10%.

↓ the computed I_sc carries into the IEEE Std 519-2022 check below — you can still override it by hand

Stage 2 · IEEE Std 519-2022 TDD check TDD = iTHD × I_meas / I_L · limit from SCR = I_sc / I_L

Measured iTHD % of fundamental

Operating current during test I_meas A

Maximum demand current I_L (12-mo avg) A

Available fault current at PCC I_sc kA · auto from Stage 1

Resulting TDD—

SCR = I_sc / I_L—

Applicable band (Table 2)—

Applicable TDD limit—

—

Enter values to evaluate.

Demonstration presets

A · The I_L error (same system & I_sc; only I_L changes)

B · The fault-level error (same emission & I_L; only source S_k changes)

Preset A reproduces a documented educational-institution case: the same system gives I_sc ≈ 1.5 kA, and iTHD 9.8% at 33 kV becomes TDD 12.45% against an 8% limit (FAIL) when I_L is the pandemic-suppressed 32.96 A — yet the identical emission gives 7.97% (PASS) at the correct I_L of 51.5 A. Preset B holds a 9% emission fixed and changes only the assumed upstream fault level: the true 5000 MVA source puts SCR just above 50 (12% limit, PASS), while an understated 1500 MVA drops SCR just below 50 (8% limit, FAIL) — the verdict turns on the fault level at a Table-2 band boundary. Note: this single-window check illustrates the I_sc/I_L effect — a field verdict applies the §05 daily/weekly percentile rules over a full week.

The aggregate alone is not compliance. Passing the TDD limit does not automatically imply compliance: the individual harmonic current limits specified in IEEE Std 519-2022 Table 2 shall also be verified. Equally, passing THD-V does not automatically imply compliance: the individual voltage harmonic limits specified in IEEE Std 519-2022 Table 1 shall also be verified.

§07 Myths, busted

The recurring misreadings of 519

Each of these is heard regularly in compliance discussions. Each is wrong under 519-2022, and the correct reading is short.

01"iTHD decides compliance — if iTHD exceeds the limit, the consumer fails."

Compliance is decided by TDD, not iTHD. iTHD is referenced to the fundamental and is only used to compute TDD (referenced to I_L). A high iTHD at light load can be fully compliant, and a low iTHD does not by itself prove compliance; current-harmonic compliance is decided only by the TDD referenced to I_L.

IEEE Std 519-2022, Clause 3 (definitions) & Clause 5.2

02"High iTHD at light load means there is a harmonic problem to fix."

At light load iTHD rises while the harmonic amperes fall — and it is the ampere value that heats equipment and distorts voltage. Treating a high light-load iTHD leads to oversized, sometimes harmful, filters.

Demonstrated in Janitza–Foretec live-dashboard instrument data — as load drops, iTHD rises while the harmonic amperes fall.

03"A minimum operating current must be maintained during the test."

No such requirement exists in 519. The standard's compliance metric (TDD) already accounts for facility size through I_L. Forcing artificial loading only invites artificial compliance demonstrations.

No minimum-current clause in IEEE Std 519-2022; measurement is business-as-usual.

04"The limits apply to each drive / panel inside the facility."

The limits apply only at the PCC. Applying them to individual equipment is a deliberate engineering choice some make for design margin — it is not what the standard requires for compliance.

IEEE Std 519-2022 re-emphasises application at the PCC between utility and user.

05"A short snapshot reading is enough to declare compliance."

Compliance is statistical over a minimum one-week window, evaluated against daily/weekly 99th and 95th percentile rules. A snapshot cannot generate those percentiles.

IEEE Std 519-2022, Clause 5 measurement notes; IEC 61000-4-30 Class A.

06"The PCC is the consumer's LV main or incomer."

The PCC is upstream of the installation — on the HV / incoming side for a consumer with its own transformer. Measuring at the LV main misplaces the boundary and the limits.

IEEE Std 519-2022, Clause 3 (PCC definition).

07"I_sc can be assumed, or taken from the sanctioned demand."

I_sc is the actual available fault current at the PCC, computed from source MVA and series impedance. An assumed I_sc mis-selects the SCR band and therefore the TDD limit — as Preset B above shows.
In the field, TNPDCL has applied IEC 60076-5 — the transformer short-circuit-withstand standard — to this determination (Memo No. CE/Coml/SE/EE3/AEE2/F-SC-HARMONICS/D./FLM/CM/380/2022, dt. 02-01-2023). That standard concerns a transformer's ability to withstand short circuit, not the available fault current at the PCC; using it to set I_sc or the SCR band is an unconventional and incorrect practice under IEEE Std 519-2022, which requires the actual computed fault level (IS 13234 / IEC 60909).

IEEE Std 519-2022, Clause 3 (SCR = I_sc/I_L).

08"Even harmonics are always limited to 25% of the odd limits."

That was 2014. Under 2022, even harmonics h≤6 are limited to 50% and even harmonics h>6 to 100% of the table value.

IEEE Std 519-2022, Table 2 footnote (even-harmonic rule).

09"519-2022 still ignores solar / inverter generation."

2022 brings IBR/DER into scope. Above 10% of annual average demand, IEEE 1547 / 2800 current limits apply at the PCC instead of Table 2.

IEEE Std 519-2022, Clause 1.2 ("passive" removed) & Clause 5.2.

10"Inflating I_L, or hiring a filter for the test week, achieves compliance."

Both pass the paperwork while the grid still receives the harmonics in normal operation. Neither is mitigation; both waste energy and defeat the standard's purpose.

See §09 — short-window practices.

11"If THD-V and TDD are within limits, the installation is fully compliant with IEEE Std 519-2022."

IEEE Std 519-2022 sets individual harmonic-order limits as well as the aggregate: Table 1 caps each voltage harmonic (and THD-V), and Table 2, Table 3 and Table 4 cap each current order (and TDD). A single order can exceed its own limit while the aggregate still passes — and as that order flows through the system impedance at its frequency (often raised by a parallel resonance), it can dominate the voltage distortion that neighbouring consumers receive, even those who inject nothing. Compliance with IEEE Std 519-2022 requires both the individual harmonic limits and the aggregate distortion limits to be satisfied simultaneously.

IEEE Std 519-2022, Table 1 (voltage) & Tables 2–4 (current) — individual-order limits alongside THD-V / TDD.

§08 Where the measurement goes wrong

Field practices that produce wrong verdicts

These are the practical errors that put a compliant consumer in default — or let a non-compliant one through. Most arise from the external numbers (I_sc, I_L) and from shortcuts on instrument and window.

Fault levelWrong I_sc → wrong SCR → wrong limit band

Using an assumed, rule-of-thumb, or sanctioned-demand-derived short-circuit current instead of the actual available I_sc at the PCC. Because the TDD limit is selected by SCR = I_sc/I_L, an understated I_sc tightens the band and can fail a compliant consumer; an overstated I_sc relaxes it and lets emission through.

Correct practice: compute I_sc at the PCC from the actual source short-circuit MVA and the series transformer/line impedance, agreed between licensee and consumer; then read the band from Table 2.

Demand currentWrong I_L → TDD inflated or deflated

Common forms: using sanctioned demand instead of the 12-month maximum-demand average; carrying forward a pandemic- or shutdown-suppressed I_L; not using the 15/30-min demand interval; or ignoring the partial-year and new-installation rules. When I_L is well below the operating current, TDD inflates (consumer penalised); when far above, TDD deflates (utility disadvantaged) — though the actual emission is identical.

Correct practice: compute I_L strictly per Clause 3.1 — 15/30-min monthly maximum demand averaged over twelve months (or available months, or projected demand for a new installation).

WindowShort measurement instead of a full week

Taking a few-hour or single-visit snapshot and reading instantaneous TDD off the meter. This cannot generate the daily/weekly percentiles the standard requires, and it is unrepresentative of normal operation.

Correct practice: continuous logging over the assessment week, evaluated against the 99th/95th percentile rules of §05.

InstrumentNon-Class-A meter used for a statutory check

An ordinary energy meter or basic analyzer that is not IEC 61000-4-30 Class A (Ed.3) does not aggregate or compute percentiles to the standard, leaving any result disputable.

Correct practice: a Class A Ed.3 analyzer (e.g. Janitza UMG family) with harmonics per IEC 61000-4-7, producing automated, dispute-proof results without manual calculation.

MetriciTHD compared directly to the TDD limit

Reading iTHD and testing it against the Table 2 TDD value. iTHD is referenced to the fundamental, not I_L, so this is comparing two different quantities.

Correct practice: use the analyzer's TDD (with a correctly entered I_L), or convert iTHD → TDD with I_meas/I_L before comparing.

Per-orderJudging compliance based only on TDD or THD — ignoring the individual harmonic orders

A facility can meet THD-V and TDD while a single harmonic order exceeds its individual limit. Flowing through the system impedance at its frequency — which can be high near a parallel resonance — that one order can produce a disproportionate harmonic voltage that distorts the supply reaching neighbouring consumers, including those who inject nothing. Assessing TDD only, as in the present TNPDCL practice, misses this entirely.

Correct practice: assess both the individual-order limits and the aggregate — Table 1 for each voltage harmonic, Table 2 for each current order — not TDD/THD alone. Foretec recommends that TNERC's power-quality regulations be amended to require individual-order compliance (voltage and current), not TDD alone.

LocationMeasuring at the wrong point

Connecting at the LV main, at individual loads, or downstream of a harmonic filter rather than at the contractual PCC misplaces the boundary; distortion is higher near the loads, and lower on the filtered side, than the value that actually applies at the upstream PCC.

Correct practice: measure at the PCC through the metering CTs/VTs, as in Fig. 1.

§09 The short-window approach

Temporary filters and inflated demand — passing the test, failing the grid

A weekly campaign captures only a snapshot of the year. This is precisely why power quality calls for continuous monitoring, and why a permanently installed Class A (IEC 61000-4-30 Ed.3) instrument at the consumer's PCC is the sound basis for compliance.

The shared-responsibility model — and where it currently stands

IEEE Std 519-2022 and the CEA Regulations rest on shared responsibility: the consumer limits the harmonic current injected at the PCC, and the licensee keeps the supply voltage within limits. In a sound arrangement a Class A instrument is installed at each consumer's PCC, the consumer shares its current-distortion data with the DISCOM, and the DISCOM in turn shares the voltage-quality data with the consumer.

With respect, we observe that this reciprocal duty has not yet been taken up by the DISCOM. In practice the obligation is placed largely on the consumer, and compliance is judged from a short, ~7-day temporary measurement. Where that measurement is also based on an incorrect short-circuit level — as in the TNPDCL practice noted in §07 — the verdict no longer reflects how the facility actually behaves across the year. That gap is what opens the door to the two short-window possibilities described below.

Why a 10-day temporary filter is not compliance

The window can be gamed; the grid cannot.

Because the assessment runs over roughly a week, there are two short-window possibilities that could, in principle, produce a "PASS" that does not reflect how the facility behaves across the year:

1. A temporary filter limited to the measurement window. There is a possibility that a harmonic filter is fitted only for the ~7–10 day campaign and removed afterwards. The report would then show compliance, while for the remaining ~355 days the unfiltered harmonics would flow into the grid as before — and neighbouring users, including those under no harmonic obligation, would still receive the distorted voltage.

2. A higher demand held during the window. Equally, there is a possibility that a high maximum demand is held through the campaign, lifting the 12-month I_L so that the operating current sits below I_L and the TDD deflates. This would waste energy and surrender demand headroom for no real benefit, and would in any case be temporary — next year's I_L resets the calculation.

Neither possibility is mitigation. The standard's intent is to limit what the facility injects in normal operation; a result that depends on temporary equipment or contrived loading would not represent that — a properly designed permanent filter, continuous monitoring, and sharing the data with the DISCOM remain the only genuine remedy. On the other hand, the DISCOM should equally install power-quality meters at its substation (SS) bus and share the voltage-quality data with the public.

Why the wrong fault level harms everyone — respectfully stated

Using an assumed I_sc instead of the actual computed fault level pulls in two opposite directions, and neither serves the grid:

1. An understated fault level penalises genuine consumers. A low assumed I_sc lowers the SCR and assigns a TDD limit stricter than the connection actually warrants. Compliant consumers are then recorded as non-compliant and asked to mitigate emissions that were within a fair limit — in effect, victimised by a number.

2. An overstated fault level on a genuinely weak grid permits over-injection. Where the real fault level is low but a higher value is assumed, the SCR and the TDD limit are inflated, and consumers are allowed to inject more harmonics than a weak network can absorb. The grid is left polluted, neighbouring users — including those under no harmonic obligation — bear distorted voltage, and the added harmonic heating of transformers and lines becomes a recurring technical and revenue loss that the licensee ultimately carries.

We submit, in good faith and as a matter of engineering record, that the present unconventional practice — short windows, an incorrect I_sc, and the absence of continuous Class A monitoring — has the unintended effect of permitting, and even encouraging, pollution of the power system, over and above the avoidable losses it creates. A single, mutually agreed method of determining I_sc and I_L, supported by permanent Class A metering and two-way data sharing, would set this right for both the consumer and the DISCOM.

Foretec position

Where a facility has a genuinely abnormal I_L (post-pandemic recovery, major expansion, prolonged supply disruption, or efficiency works that lowered load), a fair assessment should run a continuous full-month measurement and treat the maximum-week current — the week in which the highest absolute harmonic generation occurs — as the basis for I_L. This protects both the grid and the consumer from a distorted verdict.

This is Foretec's recommended interpretation and a proposed amendment — it is not present in the current IEEE Std 519-2022 text, which defines I_L on the 12-month demand basis of Clause 3.1.

§10 A fair protocol

What a defensible 519-2022 assessment looks like

A measurement is only as sound as the two external numbers and the window behind it. The checklist below is what removes discretion and dispute.

I_sc computed from actual source MVA + series impedance.
I_L per Clause 3.1 (15/30-min, 12-month / partial / projected).
SCR band and TDD limit read from Table 2/3/4.
Class A Ed.3 instrument; harmonics per IEC 61000-4-7.
Minimum one-week window, business-as-usual operation.
99th/95th percentile evaluation, not snapshots.
TDD (not iTHD) used as the compliance metric.
Even-harmonic 2022 rule applied (50% / 100%).
IBR/DER >10% routed to IEEE 1547 / 2800.
Automated report, no manual recalculation.
Any permanent filter properly designed — not a test-week loaner.