You had your Ø80 shaft quenched and the surface came back at 58 HRC. Then you had it sectioned and measured in the middle: 34 HRC. That is not a heat treatment fault – it is the limit of the steel you chose. It is called hardenability, and it is behind most complaints that “the heat treatment did not take”.
What actually happens during a quench?
Above roughly 727 °C steel becomes austenite, which can hold carbon in solution. On the way back down, what the carbon does depends entirely on the cooling rate, and there are three possible outcomes:
| Cooling | Structure formed | What it looks like | Hardness |
|---|---|---|---|
| Slow (in the furnace) | Pearlite | A lamellar (layered) mixture of ferrite and cementite | Low |
| Medium (air, salt bath) | Bainite | The same two phases, but a finer, non-lamellar mixture | Medium to high |
| Fast (water, oil) | Martensite | A forced, non-equilibrium structure where the carbon could not escape | Highest |
The aim of hardening is martensite. For that, every point in the part has to cool fast enough. And that is exactly where the problem starts.
Why does the core stay soft?
The quenching fluid only touches the outer surface. Heat from the core has to travel out through the material first, and steel does not conduct heat quickly. The result:
- The surface cools within seconds, forms martensite, and comes out hard.
- The core cools far more slowly, forms pearlite or bainite, and stays soft.
The thicker the part, the bigger the gap. A Ø20 shaft may harden right through, while a Ø100 shaft in the same material hardens only in the outer skin.
Hardenability = how deep that hardness reaches. This is set by the alloying elements.
What do the alloying elements change?
Chromium, molybdenum, nickel and manganese all delay the transformation of austenite into pearlite. Because the transformation starts later, even a slower quench is enough to form martensite.
In practice this means:
| Steel | Carbon | Alloy | Result |
|---|---|---|---|
| 1040 (C40) | ~0.40% | None | The surface hardens, the core stays soft. Needs a water quench, with a high risk of cracking. |
| 4140 (42CrMo4) | ~0.40% | Cr + Mo | The same peak hardness, but far deeper. Oil quenched, with little distortion. |
| 4340 | ~0.40% | Ni + Cr + Mo | The deepest hardening group. For thick sections under heavy load. |
All three carry the same carbon, so in theory all three can reach the same peak hardness. The difference is entirely in depth. You can look up the EN, DIN and JIS equivalents of all three in the Steel Grade Cross-Reference tool.
How is hardenability measured? The Jominy test
The standard method is the Jominy end-quench test. A cylindrical specimen is heated to austenitising temperature, then cooled by spraying water on one end only. That gives a progressively slower cooling rate along a single specimen, from the quenched end outwards.
Once cool, a flat is ground along the side and hardness is measured at set intervals from the end. The resulting curve is the character of that steel:
- A curve that falls slowly means high hardenability – the steel works in thick sections too.
- A curve that drops sharply means low hardenability – you only get full hardness in thin sections.
Material producers publish these curves in their data sheets. This is what to look at when choosing steel for a thick part.
Choosing the quenching medium
| Medium | Severity | Suits | Risk |
|---|---|---|---|
| Brine | Harshest | Unalloyed, thin parts | Highest cracking risk |
| Water | Harsh | Unalloyed steels | Cracking and distortion |
| Oil | Medium | Alloy steels | Smoke and fire; bath temperature must be controlled |
| Salt bath | Medium, controlled | Tool steels | High equipment and safety requirements |
| Air / gas | Gentle | High-alloy steels | Will not hold hardness if the alloy content is too low |
The trade-off is this: the harsher the quench, the better the hardness holds – and the greater the risk of cracking and distortion. The right approach is not to pick the fastest quench, but the gentlest quench that is still enough for that steel.
A practical approach to selection
- Identify the thickest section of the part. That is the deepest point that has to harden.
- Decide where the hardness is actually needed. Only at the surface, or right through? If the surface is enough, surface hardening is cheaper and carries less risk.
- Pick the carbon level from the hardness required. If you want 60 HRC, a low-carbon steel will never get there.
- Pick the alloy content from the section thickness. Thicker part means a more heavily alloyed steel – which also means a gentler quench, so less distortion.
- Soften the section transitions in the design. Sharp corners and sudden thickness changes are where cracks start.
Frequently asked questions
Surface 58 HRC, core 34 HRC on the same part. Did the heat treater get it wrong?
Probably not. If the part is thick and the steel is low alloy, that is exactly the expected result. The fix is not to change the heat treatment but to choose a steel with higher hardenability.
It cracks in water and will not harden in oil. What should I do?
That is the classic sign that the steel is wrong for the section. Move to an alloyed grade and an oil quench becomes sufficient – hardness holds and the cracking risk falls.
Does raising the carbon content increase hardenability?
Carbon raises the peak hardness but has very little effect on hardening depth. Depth needs alloying elements. Higher carbon also increases the tendency to crack.
When is bainite useful?
It is not as hard as martensite but it is tougher and forms with less distortion. There are controlled treatments – holding in a salt bath, for example – where bainite is produced deliberately, particularly on parts with a high cracking risk.
Can I check hardenability myself?
For a rough idea, section the part and take hardness readings at intervals from the surface towards the centre. The resulting profile shows how deep the hardness reached. The standard Jominy test itself is laboratory work.