How to use this calculator
- Enter the composition. Carbon, manganese, nickel, chromium and molybdenum in weight percent, from the mill cert or grade spec.
- Read Ms. The Andrews equation returns the temperature where martensite first forms on cooling.
- Set the quench temperature. Room temperature for a normal quench, lower for sub-zero or cryogenic treatment.
- Check retained austenite. High retained austenite flags a temper or cryo step before the part is put in service.
How it works
Martensite only starts forming once a quenched steel cools past a temperature set almost entirely by its chemistry — carbon most of all:
Ms(°C) = 539 − 423·C − 30.4·Mn − 17.7·Ni − 12.1·Cr − 7.5·Mo · f = 1 − exp(−0.011·(Ms − Tq))
The first relation is Andrews' 1965 linear fit; the second is Koistinen-Marburger, giving how much austenite has turned to martensite by the time the part reaches the quench temperature. The rest of the heat-treat picture lives in the carburizing case depth calculator and the Larson-Miller tempering calculator.
Worked example
Verified against the live calculator
AISI 4140 — 0.40 C, 0.875 Mn, 0.95 Cr, 0.20 Mo —
quenched to room temperature:
Ms = 539 − 423(0.40) − 30.4(0.875) − 12.1(0.95) − 7.5(0.20) ≈ 330°C · RA = exp(−0.011·310) ≈ 3.3%
That 330°C Ms matches the published figure for 4140, and quenching
to 20°C leaves only about 3% retained austenite — low
enough that a normal temper handles it. Push the carbon to 1.0% and
Ms collapses toward 70°C: now a room-temperature quench leaves a
large austenite fraction, and a sub-zero soak earns its place.
Frequently asked questions
How do you calculate the martensite start temperature?
The Andrews (1965) linear equation: Ms(°C) = 539 − 423·C − 30.4·Mn − 17.7·Ni − 12.1·Cr − 7.5·Mo, with each element in weight percent. AISI 4140 (0.40C, 0.88Mn, 0.95Cr, 0.20Mo) comes out at about 330°C — matching the published value.
Why does carbon lower Ms so much more than the alloying elements?
Its coefficient is by far the largest — 423 versus 30.4 for manganese and 12.1 for chromium. Every 0.1% carbon drops Ms about 42°C, which is why high-carbon steels have low Ms temperatures and tend to retain more austenite after quenching.
What is retained austenite and how is it estimated?
The austenite that has not yet transformed when the part reaches the quench temperature. The Koistinen-Marburger relation gives the transformed fraction as 1 − exp(−0.011·(Ms − Tq)); the balance is retained austenite. For 4140 quenched to room temperature that is about 3% — small, but high-carbon and high-alloy steels can retain far more.
When should I use a cryogenic or sub-zero treatment?
When retained austenite is high enough to hurt dimensional stability or hardness. Lowering the quench temperature (Tq) drives the Koistinen-Marburger fraction up — that is exactly what a sub-zero or cryo soak does: it converts more of the leftover austenite to martensite before it transforms unpredictably in service.
Method & assumptions
- Andrews (1965) linear equation (verified 2026-06-13 against multiple metallurgy references); calibrated for low-alloy steels to roughly 0.6% C. Co, Si, W, V and the high-alloy/stainless range fall outside this form.
- Retained austenite by Koistinen-Marburger with α = 0.011 /°C — the widely-used value (strictly derived for low-carbon, applied broadly as an engineering estimate; adjustable here).
- Composition-only screening: it does not capture cooling rate, austenitizing temperature, grain size, segregation or prior microstructure. A dilatometer or measured hardenability test is the real answer.
- No alloy or grade tables are embedded — enter the weight percents from your mill cert or specification.