SAE 4140 Plate: Why Normalizing + Tempering Beats Quench-Temper for Thin Gages

The Thin-Plate Delivery Problem

4140 plate is normally delivered in the QT condition because quenched and tempered martensite offers the best strength-toughness balance. But the quench step is also the source of its worst defect. When a thin plate – 15 mm or less – is water-quenched, the thermal and transformation stresses across the section are severe, and the plate is prone to distortion (bow) or through-thickness cracking. The mill documents a concrete case: a 13 mm plate that developed a longitudinal edge crack after quenching, making the product unusable. Thin gage concentrates the problem: less section to absorb stress, faster cooling, higher quench severity. The conventional remedies – gentler quenchants, interrupted quenching, straightening – add cost and complexity without removing the risk. The mill chose a different route: eliminate the quench entirely.

Material and the Five Test Routes

The test material was a 13 mm plate rolled from a 150 mm continuous-cast slab. Its chemistry, given in Table 1, sits inside the standard 4140 window. Five routes were compared – normalizing alone, quenching alone, and three NT tempering variants plus one QT reference – as listed in Table 2.

Normalizing Already Delivers a “Pre-Tempered” Structure

The first surprising result is how strong the normalized plate is on its own. After 850 ℃ air cooling, the structure is bainite plus a little martensite with an average hardness of 355 HBW; the water-quenched plate is lath martensite at 518 HBW. Both exceed the final specification of 269-321 HBW. The normalized plate is therefore only 37-60 HBW (after tempering) away from the required range, whereas the quenched plate must shed 213 HBW. The normalized structure is also uniform – no significant banding – and carries far less internal stress than quenched martensite, which is exactly why it does not crack. In other words, air cooling of a 13 mm 4140 plate already produces a microstructure that is a short temper away from specification. The quench step is not adding a property that the final product needs; it is adding hardness that the temper must then remove, while adding crack risk and cost on the way.

Tempered Results: NT Meets QT

Table 3 gives the tensile results. All five conditions exceed the required yield strength of 750 MPa. The NT plates follow the classical tempering trend – strength falls and elongation rises as the tempering temperature increases – but the fall is gentle between 520 and 560 ℃, which is exactly the wide process window a production line needs. The QT reference, tempered at 600 ℃, lands between the 520 ℃ and 540 ℃ NT plates.

Microstructurally, all tempered plates are tempered sorbite – ferrite with dispersed globular carbides and no lath or acicular features. The only visible difference is that the 500 ℃ NT plate retains some banding, which fades as the tempering temperature rises and the carbide distribution becomes more uniform. Hardness follows the same pattern: NT hardness falls from about 325 HBW at 500 ℃ toward the QT level as the tempering temperature rises, stabilizing in the 540-560 ℃ window where it matches the QT plate; all values sit inside the 269-321 HBW specification.

Viewpoint: Quenching Buys Strength That Tempering Gives Back

This study exposes a wasteful logic that standard practice has made invisible. The conventional sequence – quench to full martensite, then temper at high temperature to soften it – delivers a final property that sits only marginally above the NT route: the QT plate tempered at 600 ℃ yields 898 MPa, while the NT plate tempered at 520 ℃ yields 892 MPa. The quench adds roughly 163 HBW of as-quenched hardness that the 600 ℃ temper then removes, and in exchange the manufacturer carries water-quench distortion and cracking risk on every thin plate, plus the cost of the quench operation itself. For thin gage 4140 plate, the quench step is close to a round trip: it deposits stress, hardness and cost, and the temper hands most of it back. Most English-language process guides describe normalizing only as a conditioning treatment before QT and quote normalized 4140 at 200-250 HBW; they never consider it a finish. This mill’s data show that for plate up to 15 mm, bainite from air cooling is already 100 HBW stronger than that textbook figure, making NT a legitimate final condition rather than a pre-treatment.

The Production Verdict: 100 Batches, Zero Cracks

The mill then transferred the process to production: 850 ℃ normalizing followed by 540-560 ℃ tempering for all 4140 plate below 15 mm thickness. More than 100 production batches have been delivered with a 100% pass rate on mechanical properties, good flatness, and no cracking. The claim is not that NT replaces QT for every product – thick sections and components needing maximum toughness may still require the deeper hardenability of quenched martensite – but for thin plate where the customer’s real requirement is strength above 750 MPa with controlled hardness, NT delivers the specification with a process that cannot crack, at lower energy and cycle cost.

Practical Takeaways

• For thin 4140/42CrMo4 plate (≤ 15 mm), ask whether NT delivery is available before defaulting to QT – the property is equivalent and the quench-cracking risk disappears.
• Judge a supplier’s plate on the actual tensile certificate and hardness range, not on the “QT” label; the NT 540-560 ℃ product here matches the QT reference within a few MPa.
• Normalizing of thin 4140 produces bainite at roughly 350 HBW, far above the 200-250 HBW commonly quoted – do not size processes on textbook normalized hardness.
• Where distortion and cracking drive scrap, NT is also a cost lever: it eliminates the quench operation entirely and widens the acceptable process window for batch production.