Judging Carburized Case Depth: Why Metallographic and Hardness Methods (GB/T 9450) Differ
The metallographic method counts structure, the hardness method counts to 550 HV; on one specimen they can differ by 0.1-0.3 mm, and hardness is the arbitration method.

Definitions of the two methods
The metallographic method examines the cross-section of an annealed specimen: case depth equals hypereutectoid zone plus eutectoid zone plus half the transition zone, measured at 100x with a graduated eyepiece or imaging software. The hardness method follows GB/T 9450: Vickers indents at HV1 (9.807 N) step inward from the surface, and the case hardening depth CHD is the distance from the surface to the point reading 550 HV. The two measure different things — the metallographic value reflects the carbon gradient as structure, the hardness value reflects the actual strengthening after quench.
Sources of difference
Three factors drive the discrepancy. First, taking half the transition zone is an empirical convention that runs systematically high under steep carbon gradients. Second, hardenability affects the hardness method: in low-alloy steels with a soft core the 550 HV crossing arrives earlier, shortening the depth by 0.1 to 0.3 mm. Third, the specimen states differ — annealed for metallography, quenched for hardness — so process scatter means the two samples are never identical. GB/T 9450 states that the hardness method governs in disputes.
Hardness method procedure
The test surface must be perpendicular to the carburized face, and grinding must not temper the layer. Indent spacing is at least 2.5 diagonals with adjacent columns staggered; spacing tightens to 0.05-0.1 mm near the threshold. Find the 550 HV crossing on a smoothed hardness-depth curve rather than naive linear interpolation between two distant points. Measure at least three columns per specimen and average them; column-to-column variation beyond 0.1 mm indicates uneven carburizing.
Value of the metallographic method
The metallographic method is fast and intuitive, delivering same-shift results from furnace witness bars, ideal for process control. It also reveals defects invisible to hardness testing — carbide networks, excess retained austenite, internal oxidation — which strongly affect contact fatigue life. The recommended practice is to use both: metallography for daily furnace control, periodic hardness cross-checks to build a plant-specific correlation table between the two methods.



