You pick a relay by contact rating, mount it, and move on. Then the panel temp rises, the load has a mild inrush, and the relay welds or drops out at 70 % of its sticker. That’s not a defect—it’s a mismatch between the rated “amps” and the real watts the relay has to dissipate. Below I compare three Omron relay families—G2R, MY, and G7J—on the dimensions that determine whether a relay survives its first year. No generic advice; only numbers you can verify and a mechanism you can apply.
1. Contact Rating vs. Real Wattage: The AgCdO / AgSnO₂ Trade-off
The G2R-1 and G2R-2 are both rated 10 A at 250 VAC, with AgCdO contacts. At face value they handle 2500 VA (about 2500 W at unity power factor). But the mechanism that kills them is contact erosion from arcing, which scales with the square of the interrupted current and the arc duration. AgCdO gives good arc extinguishing at moderate currents but erodes faster under sustained high-inrush loads (like motor starts or capacitor banks). The MY2 and MY4, also AgCdO, are rated only 5 A at 250 VAC — half the contact rating. The worked consequence: if you have a 1200 W heater cycling on/off (resistive, no inrush), both the G2R-1 and MY2 will close fine, but the G2R-1 runs cooler because its contact resistance is lower relative to its rating. Over 100,000 cycles, the MY2’s contacts will pit sooner at that same 1200 W because it operates at 60 % of its AC rating vs. 48 % for the G2R-1. The reversal: for purely resistive loads under 500 W, the MY2 is cheaper and perfectly adequate—you gain nothing by oversizing.
2. Dielectric Strength and Operating Temperature: The Thermal Ceiling
The G2R and MY series both have a dielectric strength of 1500 VAC. The G7J series jumps to 2500 VAC. That matters not for routine operation but for transient overvoltage (e.g., lightning surge on a long cable run). The mechanism: when a surge exceeds the dielectric breakdown, an internal arc can carbonize the coil insulation or weld the armature. The worked consequence: if your relay sits near a compressor or VFD with frequent voltage spikes, the G7J’s higher dielectric margin gives you about 1.7× the headroom (2500/1500) before flashover. But the more immediate constraint is operating temperature. The G2R and MY are rated -40 °C to 70 °C; the G7J goes to 85 °C. At 70 °C ambient, a G2R-1 carrying 8 A (80 % of rating) will have its coil temperature rise roughly 30 °C above ambient (typical for a 0.9 W coil in still air) → 100 °C internal, which is above the 70 °C ambient limit—effectively derating the relay. The reversal: for panels with active cooling staying below 55 °C, the G2R handles the same load as the G7J at half the cost. Only use the G7J above 70 °C or where surges are common.
3. Mounting and Heat Dissipation Path: The Silent Derater
The G2R-1 is PCB-mount; the G2R-2 is socket-mount. The MY2 is PCB, MY4 is socket. The G7J-4A is panel-mount. The mechanism: a socket adds contact resistance (roughly 0.5–1 mΩ per pin) and reduces the thermal path from the relay base to the PCB copper. For a G2R-2 carrying 10 A, the socket pin loss is about 0.05–0.1 W per contact, negligible. But the bigger effect is that socket-mount relays often sit in a plastic base with no metal heatsink; PCB-mount relays can wick heat through the solder pads into the board copper. The worked consequence: if you crowd four G2R-2 relays on a socket rail inside a 40 °C panel, each relay’s internal temperature can be 10–15 °C higher than an equivalent PCB-mount version. That derates the contact life by roughly 20–30 % (Arrhenius-type acceleration). The reversal: if you need field-replacement without desoldering, socket-mount is mandatory—the thermal penalty is a trade-off you accept. But if the panel is already hot (> 60 °C), choose PCB-mount and add a small fan.
4. Coil Voltage Tolerance and Dropout: The Logic-Side Gotcha
The G2R-1 is available in 5, 12, and 24 VDC coils; the MY2 in the same voltages; the G7J-4A in 12 and 24 VDC. The mechanism: a relay’s pick-up voltage is typically 70–80 % of nominal; dropout is about 10–20 % lower. If your control supply sags (e.g., a 24 VDC bus drops to 18 V during a battery recharge), a 24 V coil relay may chatter or drop out. The G2R and MY series have a dropout voltage around 2.4–3.6 V (for 5 V coil) and 6–9 V (for 24 V coil) – roughly 30 % of nominal. The worked consequence: in a system with a large battery charger causing a 5 V dip on a 24 V bus, the relay stays pulled in if the coil voltage stays above ~16 V. But if you accidentally use a 12 V coil on a 24 V bus (reverse polarity or miswire), the coil burns out in seconds. The reversal: for long cable runs (> 50 m) between controller and relay, use a 24 V coil to minimize voltage drop—the coil current of the G2R-1 at 24 V is about 37 mA, so a 1 V drop over 50 m of AWG18 is tolerable. A 5 V coil would lose 20 % of its voltage.
Quick-Comparison Table (Sizing by Real Watts)
| Family | Contact Rating (AC) | Contact Material | Dielectric | Temp Range | Mount | Best For |
|---|---|---|---|---|---|---|
| G2R-1 | 10 A / 250 VAC | AgCdO | 1500 VAC | -40 to 70 °C | PCB | Resistive loads ≤ 8 A, moderate temp, no high inrush |
| G2R-2 | 10 A / 250 VAC | AgCdO | 1500 VAC | -40 to 70 °C | Socket | Same but with field-replaceability |
| MY2 | 5 A / 250 VAC | AgCdO | 1500 VAC | -40 to 70 °C | PCB | Low-power signal switching, ≤ 3 A |
| MY4 | 5 A / 250 VAC | AgCdO | 1500 VAC | -40 to 70 °C | Socket | Multi-pole signal or low-current interlocking |
| G7J-4A | 40 A / 250 VAC | AgSnO₂ | 2500 VAC | -40 to 85 °C | Panel | High inrush, hot ambient, surge-prone lines |
Topology/standards per the cited standards; all product ratings are manufacturer-stated values from the cited datasheets, current to 2026-06; derived/illustrative figures are labelled as such. This is not an independent head-to-head test. Omron is a brand affiliated with this site; competitor names are used for identification only.