Magnetic Latching Relay vs Standard Relay: Coil Power, Wiring and Selection

A smart meter designer needs a 100 A contact that holds its position for eight years on a supply measured in milliwatts. An industrial panel builder needs a relay that opens the moment a safety circuit loses power. Both engineers read the same datasheet lines: contact rating, dielectric strength, mechanical life. Yet the right part for one of them is a magnetic latching relay, and for the other it is a standard relay.

The short version: choose a magnetic latching relay when the contact must hold its state with zero coil current, and a standard relay when the contact must follow the coil and drop out once control power disappears. Price, driver cost, firmware effort and failure behavior all follow from that one property.

Comparing the Two Types Side by Side

The table below compresses the whole argument. The sections that follow explain why each row matters on a real bill of materials.

Row two settles most designs: a latching relay needs a driver that can reverse current or a second coil, and a standard relay does not.

How a Magnetic Latching Relay Holds Its Position

A permanent magnet sits in the magnetic circuit and keeps the armature parked in one of two stable positions. The coil does not hold the contact closed; it pushes the armature across and then stops working, so the contact keeps its state with no current anywhere in the control circuit.

Two coil arrangements are common, and each one changes the driver:

• Single coil with polarity reversal: one polarity sets the contact, the opposite polarity resets it, so the driver must be able to reverse the current.
• Dual coil: separate set and reset windings, each switched by its own transistor, which is simpler to drive but needs a wider package.

Pulse width causes more field failures than any other parameter. A few milliseconds is usually enough at room temperature, but coil resistance rises with heat while the drive voltage falls as a pulse capacitor discharges. The datasheet set and reset time is a floor, not a target to aim at.

A production example, the FH42L, is a 16 A to 20 A latching relay supplied as UL and CQC compliant, with defined set and reset times and certification covering the coil voltage actually ordered.

FH42L 16A/20A Latching Relay for MeteringUL/CQC-compliant latching relay switching 16 A or 20 A, with single- or double-coil options and defined set/reset times for smart home, lighting, and meter designs.View Product →

What a Standard Relay Does Differently

A standard relay holds its contact closed with coil current. Remove that current and a return spring pulls the armature back, so the contact returns to its de-energized position. That single mechanical fact creates both the strengths and the costs of the type.

The control interface is trivial: one GPIO or one driver channel, and the contact state is always a mirror of the coil pin. The relay also drops out predictably when the controller loses power, which is why safety interlocks and emergency stop chains use non-latching devices.

The cost is continuous coil current. For as long as the contact is closed, the coil draws power and generates heat inside a sealed package, eating into the margin against the rated operating temperature. That is why industrial relays are often specified with 105 °C coils and contacts in panels without airflow. A 16 A part such as the FH17 keeps the interface at its simplest: one coil, one drive pin, a defined contact arrangement.

FH17 10A/16A Power Relay for Smart SwitchesUL/TUV/CQC-compliant power relay with 10 A or 16 A switching, 200 mW coil sensitivity, and 4 kV isolation for smart switches, sockets, and appliances.View Product →

Coil Power, Heat and the Real Energy Cost

Holding current is small in absolute terms, which is exactly why it gets underestimated. A 12 V relay with a 400 mW coil held closed continuously draws 0.4 W, or about 3.5 kWh per relay per year. Multiply that by every relay on a distribution board, then by the years in service.

A latching relay pulsed for 50 ms at 1 W uses roughly 0.05 J per operation. Four operations a day for ten years comes to about 730 J, or 0.2 Wh. That gap is four orders of magnitude, and it explains why latching circuits appear wherever a device runs from a battery or a limited supply.

Coil heat also changes thermal design, because a steady 0.4 W inside a small sealed relay raises internal temperature and reduces the current the contact can carry at high ambient. Moving to a latching type often frees contact margin, since there is no coil heat to add to load heat. The trade-off is explained further in this note on how latching relay circuits cut standby power.

Wiring, Driver and Firmware Differences

The board level

• Standard relay: one transistor, one resistor, one GPIO.
• Single-coil latching relay: an H-bridge or two-output driver with a defined default state at power-up, plus a pulse timer.
• Dual-coil latching relay: two transistors and two GPIOs, with the same pulse timer.

Driver cost is real but modest, and the pulse timer is normally a firmware task rather than extra silicon.

The firmware level

A latching relay does not report where it is. After a reset, a brownout or a watchdog event, the microcontroller has to assume the contact still sits in the last commanded position and act accordingly. Two habits prevent most field problems: drive the relay to a known state during initialization, and treat a mismatch between stored state and actual state as a fault rather than a nuisance.

Where Each Type Earns Its Place

Magnetic latching relays fit best when

• The contact must stay closed for hours, days or years without external power, such as a prepaid meter load switch.
• Coil power is scarce, as in battery-powered sensors, remote valves or solar controllers.
• Standby consumption is regulated or measured, as in metering and building energy products.

Standard relays fit best when

• The load must drop out when control power disappears, as in interlocks and safety chains.
• Switching happens often enough that pulse-driving logic would dominate the code path.
• The design should stay simple: one pin in, one state out, no state tracking.

Smart metering is the clearest case for the latching type. A meter disconnect function may need to hold a 100 A contact in one position for the life of the meter, using electronics fed from a small capacitive supply. Design work in smart electricity meter applications almost always starts with a latching relay, and parts built for meter and lighting-control duty, such as the W30L, show how the type is specified in practice.

W30L 40A/50A Latching Relay for MeteringUL/TUV/CQC-compliant latching relay with 40 A/50 A switching, ≥1.5 mm contact gap, 12 kV coil-contact surge rating, and manual control for smart metering and lighting.View Product →

Selection Checklist Before You Commit a Part Number

1. Decide the required state after power loss first; everything else is downstream of that answer.
2. Check holding current against both the supply budget and the thermal budget.
3. Confirm the minimum set and reset pulse width at temperature extremes and at the lowest drive voltage.
4. Verify the contact rating against inrush, since lamp and motor loads draw several times rated current for a few milliseconds.
5. Check the certifications the end market needs, usually UL, TUV and CQC, and confirm they cover the exact variant ordered.
6. Add driver parts, pulse timing code and extra test steps to the cost comparison before declaring a winner.

FAQ: Magnetic Latching Relays and Standard Relays

The Decision in One Line

Choose the magnetic latching relay when the contact must remember its position without power. Choose the standard relay when the contact must follow the coil and drop out when control power disappears. Once that requirement is written down, the driver circuit, the firmware, the thermal budget and the certification path all fall into place, and the two types stop looking like competitors.