Focus keyphrase: HSS composite roll spalling analysis
Meta description: What causes large-scale spalling on HSS composite rolls? This article reveals the root causes through chemical composition, hardness testing, and microstructure analysis.
Introduction: The Spalling Accident
High-speed steel (HSS) is widely used for hot strip mill work rolls due to its high hardness, good wear resistance, and excellent strength-toughness matching.
A steel plant’s 1450mm hot continuous rolling line finishing mill used HSS composite rolls with specifications of φ805mm × 700mm.
This roll was put into service in August 2021. In May 2022, while rolling DC01 low-carbon steel (3.5mm thick), large-scale spalling occurred on the drive side surface opposite the rolling direction. This caused a serious steel piling accident.
To prevent recurrence, this article analyzes the spalling causes and proposes corrective measures.
1. Macroscopic Fracture Analysis
1.1 Spalling Scale
The spalling area was extremely large, extending from one edge to half the total working face length.
| Parameter | Measurement |
| Spalling width | Approx. 1/3 of roll circumference |
| Total spalling area | Approx. 1/6 of working face surface |
| Consequence | Complete roll报废 (scrap) |
1.2 Crack Source Characteristics
The crack source was located in an elliptical area within the spalled region. Radiating crack lines were clearly visible around the source.
Layered cracks caused by radial expansion tearing appeared above the crack source.
The deepest spalling layer reached the inner ductile iron layer. The spalled surface broke into multiple fragments with no obvious deformation, indicating high brittleness.
1.3 Layer Structure
After cleaning, the spalled fragments clearly showed the HSS layer and intermediate layer. Some fragments exhibited a distinct three-layer structure.
Key finding: The crack source was mainly located at the interface between the intermediate layer and the inner ductile iron.
Cracks propagated:
- Axially along the interface.
- Radially through the intermediate layer.
- Upon reaching the HSS interface, cracks extended axially until complete roll surface cracking.
2. Chemical Composition Analysis
Samples containing the HSS layer, intermediate layer, and inner layer were cut radially from spalled fragments.
Test Results
| Layer | Result vs. Standard | Result vs. Design |
| Surface HSS | Meets GB/T 25825-2010 | Consistent with design |
| Intermediate layer | No specific standard | Consistent except lower carbon |
| Inner ductile iron | No specific standard | Consistent except lower carbon |
Conclusion: The chemical composition basically met design requirements and was not the direct cause of failure.
3. Hardness and Strength Analysis
Hardness Test Results
| Layer | Average Hardness (HS) | GB/T 25825-2010 | Manufacturer Spec |
| Surface HSS | 80.0 | Meets (75-95) | Meets (≥80) |
| Intermediate | 58.5 | N/A | N/A |
| Inner | 33.0 | Meets (32-45) | Below (36-43) |
Strength Test Results
Due to size limitations, non-standard small tensile specimens were used.
| Fracture Location | Tensile Strength vs. Manufacturer Spec |
| Intermediate layer | Below requirement |
| Surface layer | Met requirement |
Finding: The intermediate and inner layers showed strength and hardness deficiencies.
4. Microstructure Analysis
This revealed the root cause of spalling.
4.1 Porosity and Inclusions
Multiple porous cavities were found near all layer interfaces.
Dense, irregular non-metallic inclusion bands (approx. 4.5mm wide) appeared near the graphite structure on the inner side of the intermediate-inner interface.
D-type spherical oxide inclusions were rated above Class 3.0 (severely exceeding the standard).
EDS analysis identified inclusions as Mg-Al-Si nitrides, Mg-Ca sulfides, and Ti-V sulfides/carbonitrides.
4.2 Microstructure of Each Layer
| Layer | Microstructure | Defects Observed |
| Surface HSS | Tempered martensite | Rod/block grain boundary carbides (10-50μm); triangular porous cavities (>200mm) |
| Intermediate | Network ledeburite | Closed network grain boundary carbides (max ~200μm); coarse fish-bone ledeburite (~400μm near inner interface); porous cavities along grain boundaries |
| Inner | Pearlite + spheroidal graphite | Small porous cavities (~100μm) at carbide/pearlite interfaces |
Critical finding: Cracks propagated mainly along the intermediate layer, where coarse carbides exist exclusively.
5. Fracture Morphology Analysis (SEM)
SEM examination of spalled fragments revealed:
| Feature | Observation |
| Fracture type | Brittle quasi-cleavage |
| Maximum carbide size | 328μm |
| Defects | Porous cavities at carbide-matrix interfaces; crystal face protrusions inside cavities |
Tensile specimen fracture surfaces showed identical characteristics.
Conclusion: Under stress, cracks easily initiate and propagate at these coarse carbides and porous cavities.
6. Conclusions and Corrective Measures
6.1 Root Causes of Failure
Severe internal defects: Coarse, closed-network carbides and extensive porous cavities in the intermediate layer.
Excessive interface inclusions: Non-metallic inclusions (D-type) at the intermediate-inner interface severely exceeded standards, weakening bonding strength.
Crack initiation and propagation: Cracks originated in the intermediate layer or at layer interfaces, then rapidly propagated through the intermediate layer under rolling stress, leading to large-scale spalling.
6.2 Corrective Actions
Optimize casting process: Strictly control centrifugal casting temperature and speed to reduce ledeburite segregation and refine carbide size.
Improve purity: Enhance molten steel refining to reduce non-metallic inclusions, especially D-type oxides.
Refine heat treatment: Apply appropriate heat treatment to improve network carbide morphology and eliminate porous cavities.
Strengthen ultrasonic testing: Conduct rigorous ultrasonic inspection of bonding interfaces before roll delivery. Rolls with internal defects must not be put into service.