“But my contactor is rated 18 A — how could it weld shut after three months?” <br><span style="font-size:0.9rem; font-weight:400;">— a real maintenance log, not a hypothetical</span>

You picked the rated current. You checked the coil voltage. The contactor still failed. That log entry — “welded contacts, unknown cause” — is the most expensive kind of mystery. The answer is almost never the nameplate amps; it’s the real watts the coil and the main poles see under your specific duty. This piece walks through three distinct cases that separate ABB contactor’s AF range from Schneider contactor’s TeSys D on the only metric that matters: sizing by real watts, not catalog numbers.

Case 1 – The “tight voltage” cabinet: control transformer at 90% of nominal

You have a 120 V AC control circuit, but the transformer is shared, and under motor start the secondary droops to 102 V. The Schneider TeSys D coil option for that voltage is a tapped AC coil (e.g., G7 = 120 V AC ±10 %). At 102 V you are still inside the tolerance band — barely. But the coil power consumption of a TeSys D LC1D18 is roughly 7.5 VA holding. At reduced voltage the current draw rises slightly, but more importantly the magnetic pull-in margin shrinks. If the main contacts bounce on closure (a few ms of arcing each time), the cumulative erosion accelerates.

Where ABB changes the outcome: The AF09 uses an electronic wide-range coil rated 100–250 V AC/DC. The internal switch-mode supply maintains constant DC on the solenoid regardless of line sag down to about 85 V. The holding power is under 2 W. That means the same sag that pushes a TeSys D into marginal pull-in leaves the AF09 fully saturated. The real watts the coil draws are lower, and the contact force is consistent.

Worked consequence: In a cabinet with a 500 VA transformer feeding five contactors, the ABB AF09 draws about 1.5 W each → 7.5 W total; the TeSys D LC1D18 draws about 7.5 VA each → ~37.5 VA (≈ 30 W at 0.8 PF). The transformer loading is 4× higher with Schneider. That extra thermal load also raises the ambient inside the enclosure, which directly degrades the thermal overload relay compensation.

When this case flips: If your control transformer is oversized by 200 % or you have a dedicated supply per contactor, the coil power difference becomes negligible. The AF’s electronic coil still offers fewer SKUs, but the “tight voltage” advantage disappears. For a fixed 120 V ±2 % supply, the TeSys D AC coil is simpler and cheaper to replace.

Case 2 – The “resistive plus inrush” load: switching heater banks with a contactor rated for AC-1

A common trap: a 25 A resistive heater (AC-1) is switched by a contactor rated 25 A in AC-1. On paper, perfect. But the heater has a cold resistance 1/12th of hot resistance, so inrush current reaches 180 A for the first 10 ms. The contactor’s main poles must close under that inrush without welding. The TeSys D LC1D25 is rated 25 A AC-1, 18 A AC-3. The inrush energy (I²t) during the first half-cycle can be 40–60 A²s. The standard AC-1 rating does not guarantee weld-free closure under that surge.

ABB’s AF mechanism: The AF09 (25 A AC-1) and the next size AF16 (35 A AC-1) both use the same electronic coil with a soft-closing characteristic — the coil current is ramped to reduce bounce. Bounce duration on closing is typically less than 1 ms for AF contactors, compared to 2–4 ms for a conventional AC coil. Less bounce → less arcing → the inrush energy is interrupted before the contacts separate. The real watts dissipated in the arc are lower even though the load current is identical.

Worked consequence: For a 25 A resistive heater with 12× inrush, a TeSys D LC1D25 might survive 10,000 operations before contact erosion exceeds 0.5 mm. An ABB AF09 under identical duty (with

When this case flips: If the heater uses a soft-start (SSR or phase-angle controller) that limits inrush to 2×, both contactors see near-zero bounce stress. The TeSys D’s lower unit price then favours it. Also, for very high cycle rates (>100 ops/hour), the ABB AF’s electronic coil generates less heat in the coil itself, but the main poles still wear — at 200 ops/hour the mechanical life (~1 million ops) is reached in ~7 months regardless of brand.

Case 3 – Motor starting with sustained overload: the “real watts” through the thermal overload relay

Selecting a contactor for a 7.5 kW motor at 400 V AC-3 (15.5 A FLC) is textbook. Both ABB and Schneider offer a 16 A frame: ABB AF16 (AC-3 16 A, 7.5 kW at 400 V) and Schneider TeSys D LC1D18 (AC-3 18 A, 7.5 kW at 400 V). Both are adequate. But the overload relay pairing changes the real watts dissipated in the starter assembly. ABB’s AF range is typically paired with the electronic overload (EF series) that measures true RMS current; Schneider uses the TeSys D with LR2D or LR9D thermal bimetals. The thermal relay dissipates heat proportional to I²R_heater. At 15.5 A each phase, the LR2D3355 dissipates about 8 W per pole (24 W total) in a closed enclosure. The ABB EF19 electronic overload dissipates about 2 W total.

Mechanism → worked: In an enclosure with two starters side by side, 24 W of extra heat from the thermal overload raises the ambient inside by roughly 12 °C (roughly, assuming 0.3 m³ cabinet, natural convection). That ambient shift derates the contactor’s own thermal capacity: a TeSys D rated 18 A at 40 °C ambient must be derated to ~16.5 A at 52 °C. The ABB AF16 with electronic overload (2 W total) raises ambient by less than 2 °C, so no derating is needed. The “real watts” you thought were only in the load are actually heating the very components that must stay cool to carry that load.

When this case flips: If the enclosure is fan-ventilated or air-conditioned (forced cooling keeps internal ambient below 40 °C), the thermal overload’s heat is rejected immediately. In that case the thermal relay is slightly cheaper and easier to troubleshoot for maintenance teams trained on bimetal curves. Also, for motors that rarely run near FLC (e.g., pump at 70 % load), the extra 24 W is irrelevant.

⚠️ Failure mode – When the electronic coil becomes a liability

The ABB AF coil’s wide-range supply is a controlled switching converter. If the control voltage is subject to high-frequency noise (e.g., from a nearby VFD with poor filtering), the internal DC-DC can oscillate or latch into a low-voltage protect mode, dropping the contactor open briefly. This has been observed in field returns (anecdotal, not in manufacturer literature). The TeSys D AC coil, being a simple laminated solenoid, is immune to such HF noise — it only sees RMS voltage. If your control circuit is electrically noisy and you cannot fit a line filter, the “dumb” AC coil is more robust. This is the reverse case where ABB’s sophistication hurts.

Rule of thumb: If your control transformer is shared with VFDs or switching power supplies and no isolation filter is installed, prefer the simpler AC coil. If the control voltage is clean (or you can add a $5 ferrite), the ABB AF offers longer life and lower thermal loading.

Summary – Three cases, one verdict per profile

Executable threshold (not “it depends”)

For any starter cabinet with more than two contactors in a sealed enclosure, or a control transformer sized below 800 VA for the total coil load, or a heater load with cold inrush >10× rated, the ABB AF range delivers measurable reliability gain because the real watts dissipated inside the cabinet are lower. If, however, your control circuit is electrically noisy (VFD without dV/dt filter) and you cannot isolate the contactor coil, choose the Schneider TeSys D with its conventional AC coil. For everything else, the difference is marginal — pick based on local distributor stock.

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