• Used legacy OEM parts
  • 30-day warranty
  • Returns within 30 days
  • Worldwide shipping
Cover illustration for "Solid-State Relays vs Electromechanical Relays": line drawing of a panel-mount solid-state relay on a heatsink beside an electromechanical relay.

A relay lets a small control signal switch a larger load. The classic electromechanical relay (EMR) does it with a coil and moving contacts; a solid-state relay (SSR) does it with semiconductors and no moving parts. Both are common in control panels, and neither is better in every case. This article compares how they work, how they fail and when to use each. For the construction of electromechanical relays, see How It Works: Relays.

How a solid-state relay works

An SSR has two isolated sides:

  • Input: the control signal (for example 3-32 V DC or a 120 V AC signal, depending on the model) lights an LED inside an optocoupler. The light crosses an insulating gap, so the control circuit stays electrically separated from the load.
  • Output: the light triggers the switching semiconductor. For AC loads this is usually a triac or two thyristors (SCRs) connected back to back; for DC loads it is usually a MOSFET or a transistor.

An AC output built with a triac or thyristors turns off only when the load current passes through zero. That is why an AC-output SSR cannot switch a DC load: the current never reaches zero and the output would not turn off. Choose an SSR with an output designed for the type of load.

Block diagram of an AC solid-state relay: control input through a current limiter to the LED of an optocoupler across an isolation gap, then a zero-cross detector, trigger circuit, triac output with an RC snubber across the load terminals.

Inside an AC-output solid-state relay.

Zero-cross or random turn-on

AC SSRs come in two main versions:

  • Zero-cross (zero-voltage) turn-on: the output waits until the line voltage is near zero before turning on. This limits inrush and electrical noise, and suits resistive loads such as heaters.
  • Random (instant) turn-on: the output turns on as soon as the input is energized, at any point of the AC wave. It is used for some inductive loads and for phase-angle control, where the turn-on point is varied on purpose.

Computed waveforms: after the control input is applied, a random turn-on SSR switches on immediately at any point of the AC wave, while a zero-cross SSR waits for the next voltage zero.

Random and zero-cross turn-on compared.

Heat and heatsinks

The biggest practical difference is heat. A closed relay contact has a very low resistance; a conducting semiconductor has a voltage drop, often around 1 to 1.6 V for thyristor-type AC outputs (check the data sheet). The power lost in the SSR is roughly that voltage drop multiplied by the load current:

Load current Loss at 1.2 V drop Loss at 1.6 V drop
5 A 6 W 8 W
10 A 12 W 16 W
25 A 30 W 40 W
40 A 48 W 64 W

Values for guidance, calculated as voltage drop x current; use the manufacturer's on-state voltage and thermal data for real sizing.

That heat must go somewhere. Most panel-mount SSRs above a few amperes need a heatsink sized from the manufacturer's derating curves, a thin layer of thermal compound or a thermal pad, and free airflow. The current rating printed on an SSR usually assumes a specific heatsink and ambient temperature; in a hot cabinet the usable current is lower.

Leakage current and isolation

An SSR that is off is not a true open circuit. A small leakage current flows through the output, partly through the built-in snubber network. With small loads (a contactor coil, a small relay, a PLC input), that leakage can keep the load energized or make it hum. The output terminals can also carry a voltage relative to ground when the SSR is off. An SSR is therefore never an isolating device: always use a disconnect switch before working on the load.

Failure modes

  • SSR: semiconductors damaged by overcurrent, overvoltage or overheating often fail short, so the load stays on. In heating applications this is why a separate over-temperature cut-out is standard practice. Fast semiconductor fuses (selected by their I²t rating) and a varistor across the output help protect the SSR.
  • EMR: contacts wear with every operation under load, especially with inductive loads and DC, through arcing. The contacts may weld closed after a high inrush, or become high-resistance and fail to conduct. The coil can also burn out.

Switching life and noise

With no moving parts, an SSR has no contact wear, no contact bounce and no audible click, and it can switch many times per second. That makes it the usual choice for time-proportional temperature control, where a heater is switched on and off every second or so. An EMR's mechanical life is usually far higher than its electrical life under load, and the electrical life falls as the load current and the inductance rise; the data sheet gives curves for both.

Electrically, an EMR switching an inductive load produces arcing and noise; a zero-cross SSR produces very little switching noise. A random turn-on SSR, on the other hand, creates sharp current steps that can disturb nearby electronics if not filtered.

Side-by-side comparison

Point Solid-state relay Electromechanical relay
Switching element Triac, thyristors or MOSFET Moving metal contacts
Voltage drop / heat when on Noticeable; heatsink often needed Very low
Off-state leakage Yes, small Practically none (air gap)
Typical failure Often fails short (stays on) Contact wear, welding or open contacts
Switching frequency and life High; no contact wear Limited by contact wear under load
Noise Silent; low EMI with zero-cross Audible click; arcing on inductive loads
AC and DC Output must match the load type Many models switch AC or DC within their ratings
Contact arrangements Usually single NO output NO, NC, changeover, several poles
Purchase cost Usually higher, plus a heatsink Usually lower for the same current

For guidance only; ratings depend on the exact model.

Which one to use?

  • Choose an SSR for frequent switching, heater control, silent operation or vibration-prone locations.
  • Choose an EMR for occasional switching, multiple poles or changeover contacts, very small loads where leakage matters, mixed AC/DC duty, or where you want no heat in the panel.
  • When you replace a failed relay, use the same type as the original unless the circuit is reviewed: swapping an EMR for an SSR changes leakage, heat and failure behavior.

Work on electrical equipment must be done by qualified personnel following the applicable electrical code and the manufacturer's instructions.

Key takeaways

  • SSRs switch with semiconductors through an optocoupler; EMRs switch with moving contacts.
  • Zero-cross SSRs suit resistive loads; random turn-on suits some inductive loads and phase control.
  • SSRs produce heat (voltage drop x current) and need proper heatsinking.
  • SSRs leak a small current when off and often fail short; never use them for isolation.
  • EMRs offer low loss, no leakage and flexible contacts, but their contacts wear.

Related parts at 3JIndustry

Browse used solid-state and electromechanical relays in our Relays collection, and timing relays in our Timers collection. All parts are used and come with a 30-day warranty.

Looking for a discontinued part? Send the full part number to sales@3jindustry.com or use our contact page.

Request a quote

Fill in this form: you will get a case number right away and a confirmation e-mail.

This information is used only to answer your request. Privacy policy