Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
IEC TS 62910:2015 — Utility-Interconnected Photovoltaic Inverters Low Voltage Ride-Through Test Procedure
Technical specification for LVRT testing of PV inverters covering test circuits, voltage sag types, and compliance criteria
IEC TS 62910:2015 provides a standardized test procedure for verifying the low voltage ride-through (LVRT) capability of utility-interconnected photovoltaic inverters. As grid codes worldwide increasingly mandate fault ride-through for distributed generation, this technical specification fills a critical gap by defining reproducible test methods that manufacturers, testing laboratories, and grid operators can rely on for type approval and commissioning verification.
LVRT capability prevents large-scale PV disconnection during grid faults, which is essential for maintaining system stability in high-penetration scenarios. Without LVRT, a minor voltage sag could cascade into a widespread blackout.
1. Test Circuit Configuration and Sag Generation
The standard defines three distinct test circuit topologies for generating controlled voltage sags. The choice of topology depends on the available test equipment and the type of sag to be simulated.
1.1 Short-Circuit Simulator Topologies
| Topology | Sag Types Generated | Equipment Required | Advantages |
|---|---|---|---|
| Three-phase short-circuit | Symmetrical three-phase sag (Type A) | Three single-phase switches + reactor bank | Simple, repeatable sag depths |
| Two-phase (phase-to-phase) short-circuit | Phase-to-phase sag (Type C) | Two switches + phase reactor | Tests unbalanced fault response |
| Single-phase short-circuit with/without ground | Phase-to-ground sag (Type B) | Single switch + resistor/reactor | Most common fault type in distribution networks |
The test circuit must be connected between the grid simulator (or low-voltage grid) and the inverter under test. A series impedance (typically 0.1-5 Ω per phase) limits the fault current during the sag. The standard requires that the sag be initiated within one-half cycle of the grid frequency and maintained with a steady-state accuracy of ±2% in magnitude.
A critical implementation detail often overlooked by test engineers: the sag initiation and recovery must be synchronized to the grid voltage zero-crossing to avoid DC transients that can saturate transformers and trigger spurious protection trips. Use a precision zero-crossing detection circuit with less than 100 μs response time.
2. Voltage Tolerance Curve and Assessment Criteria
The core of LVRT testing is the voltage tolerance curve, which defines the minimum voltage magnitude that the inverter must withstand as a function of fault duration. The standard provides generic curves adaptable to national grid code requirements.
2.1 Standard Tolerance Curve Parameters
For the generic test profile, the inverter must remain connected for voltage sags as low as 15% of nominal voltage retained for up to 500 ms. For sags remaining above 80% retained voltage, the inverter shall stay connected indefinitely without tripping. The transition between the “must-ride-through” region and the “may-trip” region follows a linear slope on the log-time scale.
| Retained Voltage (% V_nom) | Required Ride-Through Duration (s) | Active Power Recovery | Reactive Current Injection |
|---|---|---|---|
| 15% ≤ V < 30% | 0.15 | N/A (priority to reactive) | 100% reactive current |
| 30% ≤ V < 50% | 0.50 | N/A (priority to reactive) | 100% reactive current |
| 50% ≤ V < 80% | 1.00 | Linear recovery starting at sag clearance | Proportional to voltage drop |
| 80% ≤ V ≤ 90% | 3.00 | Full recovery within 1 s | Optional |
| V ≥ 90% | Continuous | Normal operation | Not required |
2.2 Active Power Recovery Performance
After fault clearance, the inverter must ramp active power back to at least 80% of pre-fault output within 1 second and reach 100% within 5 seconds. The standard permits a 100 ms delay after voltage recovery before the ramp begins, allowing the phase-locked loop to re-synchronize. Measurement accuracy for power recovery tracking shall be within ±2% of full scale.
Design recommendation: Implement a software state machine with dedicated LVRT control mode that bypasses the normal MPPT algorithm during the fault. The reactive current reference should saturate at 1.1 p.u. to prevent overcurrent tripping while maximizing grid support. Test results show that feed-forward decoupling of the dq-axis currents significantly improves the transient response during voltage recovery.
3. Test Sequence and Pass/Fail Criteria
The complete test sequence comprises 24 individual sag events across the three topologies at different retained voltage levels and durations. For each test point, the inverter passes if it: (a) remains connected to the grid throughout the sag, (b) injects the required reactive current within 40 ms of sag onset, (c) does not exceed 110% of rated current during the fault, and (d) recovers active power within the specified time window.
The standard also requires three repeat tests at each operating point to account for statistical variation. If any single test fails, an investigation into the root cause is mandatory before retesting.
4. Frequently Asked Questions
Q1: Can an existing PV inverter be retrofitted for LVRT compliance?
Yes, if the inverter’s control hardware (DSP/FPGA) has sufficient computational headroom. The modification primarily involves firmware updates to the current control loop and PLL. However, hardware upgrades may be needed if the DC-link capacitor bank or power devices lack the transient energy rating.
Yes, if the inverter’s control hardware (DSP/FPGA) has sufficient computational headroom. The modification primarily involves firmware updates to the current control loop and PLL. However, hardware upgrades may be needed if the DC-link capacitor bank or power devices lack the transient energy rating.
Q2: What is the difference between LVRT and zero voltage ride-through (ZVRT)?
LVRT covers sags where retained voltage is above 0% (typically 15% or higher). ZVRT requires the inverter to withstand a complete voltage collapse to 0% for a short duration (150-300 ms). IEC TS 62910 does not mandate ZVRT, but some national grid codes (e.g., Germany’s VDE-AR-N 4120) include ZVRT requirements.
LVRT covers sags where retained voltage is above 0% (typically 15% or higher). ZVRT requires the inverter to withstand a complete voltage collapse to 0% for a short duration (150-300 ms). IEC TS 62910 does not mandate ZVRT, but some national grid codes (e.g., Germany’s VDE-AR-N 4120) include ZVRT requirements.
Q3: How does the inverter’s rated power affect the test setup?
For inverters above 100 kW, direct full-power testing may be impractical due to the capacity of the short-circuit simulator. The standard permits scaled testing at reduced power levels (minimum 30% of rated) with engineering justification that results are representative of full-power behaviour.
For inverters above 100 kW, direct full-power testing may be impractical due to the capacity of the short-circuit simulator. The standard permits scaled testing at reduced power levels (minimum 30% of rated) with engineering justification that results are representative of full-power behaviour.
Q4: Why is reactive current injection during faults important?
Reactive current injection supports grid voltage recovery by supplying leading or lagging VARs during the fault. This helps prevent voltage collapse and supports the operation of grid protection relays. German grid codes were the first to mandate this, and IEC TS 62910 adopts a similar approach.
Reactive current injection supports grid voltage recovery by supplying leading or lagging VARs during the fault. This helps prevent voltage collapse and supports the operation of grid protection relays. German grid codes were the first to mandate this, and IEC TS 62910 adopts a similar approach.
