Energy Meter Accuracy Class: Class 0.5S, 1 and 2 Explained

An energy meter's accuracy class is the maximum permissible error, in percent, that the meter may show under the reference conditions of its standard. Class 2 allows ±2 %, class 1 allows ±1 %, and class 0.5S allows ±0.5 % over an extended current range. The class is not a marketing grade — it is a measured limit, and the meter has to hold it at several current points and power factors, not just at full load.

The class you need is decided by three things: who is billed, how much energy is at stake, and whether the meter works through current transformers. Buying a tighter meter than the application needs is one of the most common ways to waste money in a metering project — especially when the current transformer next to it is the real limit.

Where the numbers come from

Static (electronic) meters for active energy are covered by IEC 62053-21 for the classes 1 and 2, and by IEC 62053-22 for the higher-accuracy classes 0.2S and 0.5S. The class index equals the maximum error at reference conditions, which is why the numbers map directly onto a percentage.

Reactive energy has its own class scale, so a meter that is class 1 for active energy may carry a different class for reactive energy. Check both columns if the tariff charges for reactive power.

What the "S" actually changes

The S marks the extended-range classes. On a non-S meter the accuracy is only guaranteed above a fairly high fraction of rated current — the low-load region is outside the specification. On an S-class meter the accuracy band is pushed down to roughly 1 % of rated current or below.

That matters because a CT-operated feeder does not sit at full load. Overnight, at weekends, or on a lightly loaded submain, the secondary current can fall into exactly the region where a non-S meter is unspecified and an S meter is still accurate. Two meters that look identical on a full-load test can therefore disagree by far more than their class difference once the load drops.

Why the current transformer usually decides the result

A CT-operated measurement has three error sources in series: the CT's ratio and phase error, the burden imposed by the secondary leads and the meter's current circuit, and the meter's own error. They add, so the weakest element dominates.

The practical rule: the meter should be at least one class better than the CT, so that the combination is limited by the CT rather than by the meter. Beyond that, more meter accuracy buys very little until the CT and the wiring are also improved.

The test conditions that decide pass or fail

• Current points — the meter is tested at several fractions of rated current, not one.
• Power factor — usually at unity and at 0.5 inductive, sometimes 0.5 capacitive. A meter can pass at unity and fail at 0.5 inductive; the datasheet must show both.
• Reference temperature — errors drift outside the reference band, which is why the operating temperature range is printed separately.
• Voltage and frequency tolerance — the class applies within a defined band, not across the full supply variation of a weak grid.
• Warm-up and no-load conditions — starting current (when the meter must begin registering) and creep (no registration with voltage applied and no current) are separate limits from the class.
• Distortion — the standards test under defined waveform conditions. A feeder dominated by variable-speed drives can push real-world error beyond the class figure, which is a site problem rather than a meter defect.

Reading a meter datasheet: seven checks

Before comparing prices, pull these seven items from each datasheet. A quotation that omits them is not comparable.

Verified YOMIN meters and CTs for each class

YOMIN builds both sides of the measurement, which makes the pairing easier to control:

• Class 1 three-phase meters — prepaid IC-card meters with LCD display, RS485 communication and DIN-rail mounting, available in three-phase four-wire configurations.
• Class 1 single-phase meters — split prepayment keypad STS meters with active energy metering, for prepaid sub-metering.
• Class 0.5 current transformers — split-core metering CTs including 2000/5 A and 1000/5 A models rated 720 V AC, and a 250/5 A 50-turn version for smaller feeders.
• Class 1.0 current transformers — MSQ series low-voltage CTs for energy-meter inputs, including 75/5 A versions.

See the energy meter range and the current transformer range for the pairing that matches your class target. For the CT side in detail, read CT accuracy classes explained; for the panel side, the DIN rail guide covers module widths and mounting, and how to choose a three-phase energy meter covers the selection sequence.

What a better class does not fix

• Wrong CT ratio or reversed polarity — no meter class recovers a measurement that started wrong at the CT.
• Lead burden — long, thin secondary leads add error that the meter cannot see or correct.
• Phase imbalance and harmonics — a class figure measured on a clean waveform does not transfer to a distorted supply.
• Unread data — a 0.5S meter with no working communication path still has to be read by hand.

Specify the class for the commercial risk, then spend the next increment on the CT and the installation. That is the order that actually reduces measurement error.

Frequently asked questions

Is class 0.5 more accurate than class 1?

Yes. The class index is the maximum permissible error, so class 0.5 allows up to ±0.5 % while class 1 allows up to ±1 %. A lower number always means a tighter meter.

What does the S mean in class 0.5S?

The S marks the extended-range classes. Accuracy is held down to a much lower fraction of rated current than on the non-S classes, which matters on CT-operated metering where the current is small for long periods overnight or at light load.

Do I need a 0.5S meter for tenant billing?

Usually not. Class 1 is normally sufficient for tenant and sub-metering, and money spent on a better current transformer reduces the total error far more than upgrading the meter alone.

Can I use a class 1 meter with a class 0.5 current transformer?

Yes, but the combination is then limited by the meter. The total error of a CT-operated measurement is the sum of the CT error, the lead burden effect and the meter error, so the weakest element sets the result.

Does the accuracy class cover the current transformer?

No. The class printed on the meter applies to the meter alone, measured at its own terminals under reference conditions. The CT carries its own separate class, and the installation adds lead and burden errors on top.