Mitsubishi Electric vs Schneider PLC: Sizing by Real Watts, Not Sticker Amps

You selected a PLC partly by nameplate wattage — 24 VDC, 0.5 A, 12 W — then found it browns out when you add a fourth analog input. The data sheet said “12 W typical,” but your system draws 21 W feeding four sensors and an HMI. This teardown compares Mitsubishi PLC Electric MELSEC iQ‑F FX5U and Schneider Modicon M241 on the dimension that actually determines field failure: real watts under realistic I/O loading, not the base CPU consumption.

1. Base CPU draw vs. real-system power: the hidden spread

Schneider Modicon M241 (TM241CEC24T) is rated for 24 VDC, ~0.60 A typical, which yields ~14.4 W base. Mitsubishi FX5U‑32MR/ES draws 24 VDC, 0.55 A (≈13.2 W) from the spec sheet. Both numbers are measured with zero external I/O, no expansion bus, no field power. The mechanism: a PLC’s internal DC‑DC converter is sized for the CPU, memory, and communication transceivers — roughly 8–12 W. Once you attach on‑board I/O that sinks its return current through the same 24 V rail, each active digital output adds ~0.6 W (24 V × 25 mA ≈ 0.6 W) and each analog channel can draw 1–2 W for the sensing bridge and A/D. The worked consequence: a system with 12 DO energized and 4 analog inputs (≈12 × 0.6 + 4 × 1.5 = 13.2 W) on the Schneider PLC unit pushes total draw to 14.4 + 13.2 = 27.6 W — nearly double the base nameplate. The FX5U with the same I/O load: 13.2 + 13.2 = 26.4 W. Neither is “12 W.” The reversal: if your application is purely a small logic solver with ≤6 DO and no analog, both units stay under 18 W; the gap vanishes.

2. Expansion bus power budget: the real ceiling

The M241 supports expansion via the TM3 high‑speed bus, and the power budget for the bus is 3.3 VDC at 1.5 A (≈5 W) shared among modules. The FX5U’s expansion bus (CC‑Link or extension cable) provides 5 VDC at 2.0 A (≈10 W) to remote I/O, with each FX5‑EYT module drawing about 1.8 W. Why this changes the sizing: a compact PLC with 4 expansion modules (e.g., 2 digital out + 2 analog) on the M241 can exceed the bus budget if any module draws >1.2 A from 3.3 V — the TM3DQ16R (16‑relay output) draws ~1.1 A at 3.3 V alone, leaving only 0.4 A for the rest. On the FX5U, the same 4 modules draw ≤6 W total from the 10 W bus, with headroom. Worked consequence: a 16‑relay output add‑on on the M241 forces you to either derate module count or add an external 24 V power supply for the relays — effectively doubling the sub‑panel wiring. The FX5U stays within bus power up to ~7 modules (illustrative). Where this flips: if you only need 2‑3 expansion modules, both platforms handle it without extra power — the M241’s tighter budget only matters at 4+ modules.

3. Heat dissipation in a sealed enclosure: the watts‑to‑temperature chain

Schneider’s M241 (TM241CEC24T) dissipates a maximum of 12 W (base) and rises to roughly 28 W at full I/O as computed above. Mitsubishi FX5U‑32MR/ES dissipates 11 W base, ~25 W at full I/O. The physical mechanism: every watt not used for logic becomes heat inside the panel — a sealed 400 × 400 mm steel enclosure with no ventilation has a thermal resistance of about 0.12 °C/W (illustrative). A 28 W source raises internal air temperature by 28 × 0.12 ≈ 3.4 °C above ambient; a 25 W source gives about 3.0 °C. Worked consequence: in a 40 °C ambient factory, the M241’s internal ambient hits 43.4 °C, while the FX5U sits at 43.0 °C — a 0.4 °C difference that does not change the reliability curve. The non‑obvious here: the difference only becomes a problem if you also have a 24 VDC power supply sharing the enclosure — a typical Mean Well 150 W supply dissipates 18 W at 80% load. Combined with the M241 at 28 W, total heat is 46 W → 5.5 °C rise; with the FX5U it’s 43 W → 5.2 °C. The gap widens to 0.3 °C — still marginal. Reversal: in a tightly sealed panel with a 50 °C ambient, that 0.3 °C could push the CPU over the 55 °C derating line; the FX5U’s slightly lower dissipation gives a 0.3–0.5 °C buffer, relevant only at the absolute thermal edge.

4. When the “real watts” story reverses: the field‑powered analog loop

A common failure is the 4‑20 mA loop powered from the PLC’s 24 V rail. The M241’s built‑in analog inputs are 0–10 V only — they require an external 24 V loop supply for 4‑20 mA sensors. The FX5U has 2‑channel 12‑bit analog input (0–10 V) and 1‑channel 12‑bit analog output built in, also 0–10 V. Mechanism: if you use 4‑20 mA sensors, you must add a 24 VDC loop supply that draws an extra 1.5–3 W per loop (illustrative). That additional power is dissipated inside the panel, not accounted in PLC base draw. Worked consequence: a 4‑sensor 4‑20 mA system on the M241 adds ~6 W of external supply heat; with the FX5U you either do the same (since its built‑in AI is 0–10 V) or add a resistor — still external heat. The reversal: if your sensor inventory is all 0–10 V, both PLCs handle it without extra watts — the M241’s lack of built‑in analog (only digital I/O on CPU) means you need an analog expansion module anyway, which draws additional bus power. In that case, the FX5U’s built‑in analog saves ~2 W of expansion draw.

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. Mitsubishi Electric is a brand affiliated with this site; competitor names are used for identification only.