In the summer of 2023, I was asked to investigate a serious reliability problem on a vertical roller mill installed in a cement and slag grinding plant. The machine had been commissioned in May 2020 and operated normally for about three years. In July 2023, the mill began to show an abrupt increase in vibration. The vibration triggered multiple severe alarms and caused several sudden stops. When the maintenance team opened the mill, they found cracks in the grinding table, which is actually a large ductile iron casting. This component is always considered a critical part of the grinding system because it directly supports the material bed and transmits the grinding forces from the rollers to the gearbox and foundation.
My role was to determine why the ductile iron casting had cracked, to assess whether the component could be used again, and to propose repair and prevention measures. I used a comprehensive analytical method that considered four important aspects: casting quality, mechanical design strength, operating conditions, and thermal damage. In this article I describe the investigation and the measures we took to return the mill to safe operation.
1. Inspection and Crack Morphology
The failed part was the disc-shaped grinding table, also called the mill table or disc body. It was manufactured as a ductile iron casting according to the European material standard EN-GJS-400-18. The first inspection was done in the mill, without removing the component. The results showed two visible cracks on the outer surface of the ductile iron casting. These two cracks were located about 90 degrees apart along the circumference. Each crack started near the outer lifting lugs and extended upward toward the upper face of the table. The visible crack length was about 600 mm, and the opening width was about 2 mm. Ultrasonic testing indicated that the cracks had not yet penetrated through the wall into the inner cavity of the table.
After the upper liner segments were removed, we examined the machined surface on the top of the ductile iron casting. Two additional crack branches were found on this machined face. These branches started from the maximum outer radius of the ductile iron casting and extended toward the center for about 350 mm. The positions of these upper-surface cracks corresponded with the outer surface cracks, meaning that both cracks lay in the same radial plane. This crack pattern is important because it suggests a common stress source, not a random casting defect.
| Crack inspection item | Finding |
|---|---|
| Number of independent cracks | 2 |
| Circumferential distribution | 90 degrees apart |
| Visible crack location | Outer lifting lug area to upper machined face |
| Visible surface length | Approximately 600 mm |
| Crack opening width | Approximately 2 mm |
| Depth penetration | Not through to inner cavity |
| Crack on upper face | From outer diameter toward center, approximately 350 mm |
| Cross-section relation | Outer and upper cracks in same radial planes |
This combination of axial and radial cracks is typical of thermal fatigue. The ductile iron casting had been subjected to high surface temperatures, followed by rapid cooling. The thermal gradient created high stresses at the outer surface of the table, where the mechanical cross-section is also interrupted by lifting lugs and liner retaining holes. Before concluding that overheating was the root cause, however, I wanted to rule out material and design deficiencies.
2. Casting Quality Review
Because the disc body is a ductile iron casting, casting quality was the first possible cause that had to be checked. A large nodular iron component of this size can contain porosity, shrinkage, inclusions, graphite degeneration, or poor nodularity. Any of these defects could reduce the load-carrying capacity of the material and lead to premature cracking.
The ductile iron casting was specified as EN-GJS-400-18 to DIN EN 1563. This is a ferritic spheroidal graphite iron with guaranteed minimum yield strength of 240 MPa and minimum elongation of 12%. The original foundry certificate and production records were available for review. I re-examined the chemical composition, mechanical properties, metallographic structure, and non-destructive testing reports.
2.1 Chemical Composition
The chemical composition of the ductile iron casting had to meet the limits shown below. The original records were compared with the measured values from retained cast-on samples. The measured values were well within the specified ranges.
| Element | Required range / max | Measured value |
|---|---|---|
| Carbon, C / % | 3.40 – 3.65 | 3.54 |
| Silicon, Si / % | 2.20 – 2.80 | 2.51 |
| Manganese, Mn / % | 0.20 – 0.35 | 0.30 |
| Phosphorus, P / % | ≤ 0.035 | 0.013 |
| Sulfur, S / % | ≤ 0.015 | 0.004 |
| Magnesium and rare-earth residuals | Specified limits | Verified from foundry records |
The measured carbon and silicon values are normal for a ferritic ductile iron casting. Manganese is low enough to avoid the formation of excessive carbides. Phosphorus and sulfur are well controlled. These results gave me confidence that the basic chemistry was not responsible for the cracking.
2.2 Mechanical Properties
The mechanical properties of the ductile iron casting were confirmed by tensile tests on retained cast-on samples. Three separate test coupons were taken and tested. The results were consistent and exceeded all specification requirements.
| Property | Requirement | Coupon 1 | Coupon 2 | Coupon 3 |
|---|---|---|---|---|
| Tensile strength, Rm / MPa | ≥ 370 | 402 | 402 | 402 |
| Yield strength, Re / MPa | ≥ 240 | 267 | 268 | 268 |
| Elongation after fracture, A / % | ≥ 12 | 19.5 | 18.5 | 19.0 |
| Reduction of area, Z / % | Not specified | 18 | 18 | 19 |
All tensile results were above the minimum values. The elongation values in particular show that the ductile iron casting had a predominantly ferritic matrix with good ductility. A brittle casting would have failed with much lower elongation.
2.3 Metallographic Examination
The microstructure of the ductile iron casting was also checked. The requirements were a minimum nodularity of 90%, corresponding to grade 2 in the standard classification, and a graphite size of grade 5 to 6. The retained samples showed good nodularity and a normalized graphite distribution.
| Metallographic property | Requirement | Observed result |
|---|---|---|
| Graphite form | Predominantly spheroidal, grade VI | Grade VI |
| Nodularity | ≥ 90%, grade 2 | 93%, grade 2 |
| Graphite size | Grade 5 – 6 | Mostly grade 6 |
| Pearlite content | ≤ 10% | Satisfactory, mainly ferritic matrix |
These results confirmed that the ductile iron casting had a sound ferritic microstructure with high nodularity. Graphite nodules were small and evenly distributed. There was no evidence of graphite flotation, carbide segregation, or degenerate graphite.
2.4 Non-Destructive Examination of the Original Casting
According to the production records, the upper machined face of the ductile iron casting had been subjected to 100% ultrasonic testing before leaving the foundry. The testing was performed to the requirements of EN 12680-3, level 2. No recordable defects were found at that time. This result was important because it showed that the cracks found after service were not old casting defects that had been overlooked during manufacture.
After completing this portion of the investigation, I concluded that the ductile iron casting itself did not contain any material deficiency that would explain the cracking. The next step was to verify whether the mechanical design was strong enough for the intended loads.
3. Mechanical Strength Verification
The disc body is not only a ductile iron casting; it is also a heavily loaded machine element. During grinding, the rollers push down on the material bed, and the material bed transmits the load to the table. The table must resist bending, shear, contact pressure, and local stress concentrations around liner clamps and lifting lugs.
I verified the design strength by finite element analysis. The model was based on the actual three-dimensional geometry of the ductile iron casting, including the upper liner seat, the lower hub, the outer flange, and the lifting lugs. The maximum design projection pressure was taken as 1,200 kN/m² under the roller footprint. The finite element results gave a maximum von Mises stress of approximately 33 MPa in the ductile iron casting.
The specified yield strength of the EN-GJS-400-18 ductile iron casting is at least 240 MPa. At room temperature, the design safety factor is therefore:
$$n_{\mathrm{design}}=\frac{R_e}{\sigma_{\max}}=\frac{240}{33}\approx 7.3$$
This is a very conservative safety factor for a machine element in a repeated-load application. Most industrial designs of this type would consider a safety factor above 3 or 4 to be acceptable, especially when the loading pattern is not perfectly static.
3.1 Actual Operating Loads
I then used the recorded hydraulic pressures to determine the actual load during service. The control room data showed that the upper chamber pressure of the grinding cylinder was about 10 MPa, while the lower chamber pressure was about 2.3 MPa. After taking into account the piston areas and lever geometry, the actual roller projection pressure was calculated as approximately 920 kN/m².
Using the same finite element model and scaling the stress linearly with the load, the actual maximum mechanical stress in the ductile iron casting was:
$$\sigma_{\mathrm{actual}}=33\times\frac{920}{1200}=25.3\ \mathrm{MPa}$$
The corresponding actual safety factor at room temperature was:
$$n_{\mathrm{actual}}=\frac{240}{25.3}\approx 9.5$$
| Load case | Projection pressure / kN·m⁻² | Maximum computed stress / MPa | Safety factor based on 240 MPa yield |
|---|---|---|---|
| Design condition | 1,200 | 33.0 | 7.3 |
| Actual operating condition | 920 | 25.3 | 9.5 |
These calculations proved that the mechanical design of the ductile iron casting was more than sufficient for the normal loads. At room temperature, the component could withstand about seven to nine times its design stress before yielding. This is why I ruled out conventional overload as the root cause.
4. Operating and Maintenance Assessment
Once the material and design were cleared, I turned my attention to the actual operating environment. The mill had been running under conditions that were very aggressive for a large ductile iron casting. I reviewed the control room history, the operator logs, and the plant maintenance procedures. The following observations were particularly important.
- The moisture content of the feed material was greater than 20%.
- The hot gas inlet temperature was generally above 600 °C during normal operation.
- The inlet temperature occasionally reached 750 °C or more, especially during attempts to stabilize the mill with very wet feed.
- The mill was started and stopped frequently because of the local peak-shaving electricity policy. Daily starts and stops were common.
- After shutdown, the operators opened the human access door in the air duct and blew compressed air directly into the mill to cool the grinding table faster.
These conditions are all relevant to the durability of a ductile iron casting. The material was never intended to operate with its surface at 700 °C or above. The direct application of compressed air after shutdown also created a sudden thermal shock on a component that was still at high temperature.
| Operating parameter | Recorded value or practice |
|---|---|
| Feed moisture | > 20% |
| Normal hot gas inlet temperature | > 600 °C |
| Peak hot gas inlet temperature | > 750 °C |
| Operating schedule | Frequent daily start/stop cycles |
| Shutdown cooling practice | Compressed air blown through access door directly onto table |
| Effect on mill operation | Severe vibration alarms and trip events |
The combination of high temperature and rapid cooling is exactly the condition that produces thermal fatigue in a ductile iron casting. I therefore examined the temperature dependence of the material properties in more detail.
5. Thermal Damage Mechanism
EN-GJS-400-18 is a ferritic ductile iron casting. Ferritic materials are normally tough and ductile at room temperature, but their strength decreases sharply as the temperature rises above roughly 400 °C. Published studies on ductile iron casting show that yield strength and tensile strength remain relatively stable below 400 °C. Above 400 °C, the strength begins to fall more quickly. At 500 °C, the remaining strength is about 50% of the room-temperature value. At 600 °C, the strength is about 30%. At 700 °C, the strength is only about 10% to 15% of the initial value.
| Metal temperature / °C | Approximate retained strength of ductile iron casting |
|---|---|
| 20 | 100% |
| 400 | Still close to room-temperature properties |
| 500 | About 50% |
| 600 | About 30% |
| 700 | About 10% – 15% |
If the table surface temperature reached 700 °C, the yield strength of the ductile iron casting would have fallen to roughly:
$$R_{e,700}\approx 0.12\times 240\ \mathrm{MPa}\approx 29\ \mathrm{MPa}$$
This is very close to the computed mechanical stress of 25.3 MPa. In other words, the safety factor, which was normally 9.5 at room temperature, dropped to nearly 1.0 at the peak metal temperature. Even a small additional stress, such as a thermal stress or a local geometric concentration, would be enough to cause yielding.
5.1 Thermal Stress and Thermal Shock
When a hot ductile iron casting is suddenly cooled by compressed air, the outer surface contracts while the interior remains hot. This creates high tensile stress on the surface. For a thin surface layer constrained by the bulk material, the thermal stress can be estimated as:
$$\sigma_{\mathrm{th}}=\frac{E\alpha}{1-\nu}\Delta T$$
where E is the elastic modulus of the ductile iron casting, α is the coefficient of thermal expansion, ν is Poisson’s ratio, and ΔT is the temperature difference between the surface and the interior.
For a typical ductile iron casting, the elastic modulus is about 169 GPa, the coefficient of thermal expansion is about 12×10⁻⁶ per °C, and Poisson’s ratio is about 0.275. Therefore:
$$\frac{E\alpha}{1-\nu}\approx \frac{169\times 10^9\times 12\times 10^{-6}}{0.725}\approx 2.80\ \mathrm{MPa/°C}$$
A temperature difference of only 100 °C would generate a thermal stress of about 280 MPa if the surface is fully constrained. That is above the room-temperature yield strength of the material. At 700 °C, when the material strength has fallen to about 29 MPa, even a modest thermal gradient can produce plastic deformation and surface cracking.
Repeated heating and cooling cycles make the problem worse. Each cycle causes the surface to yield in tension. After many cycles, small surface cracks initiate and then grow into the body of the ductile iron casting. This is known as thermal fatigue. The crack pattern we observed, with cracks at the outer edge and at the same radial plane on the upper face, is consistent with thermal fatigue caused by repeated thermal gradients.
5.2 Metallurgical Transformation
The matrix of EN-GJS-400-18 ductile iron casting is predominantly ferrite. On the iron-carbon phase diagram, ferrite begins to transform to austenite at temperatures above about 720 °C. This transformation is accompanied by a volume change and local stresses. The original test material had a ferritic matrix, and the operating temperature of 750 °C was high enough to cause partial transformation on the surface of the ductile iron casting.
In addition, the conventional stress-relief annealing range for cast iron is about 550 °C to 600 °C. When a ductile iron casting is held at such temperatures and then cooled slowly, internal stresses can be reduced by 90% to 95%. But if the component is cooled too quickly, the stress relief becomes ineffective. The use of compressed air after shutdown was therefore particularly damaging. It not only introduced new thermal stress, but also prevented the normal stress-relief mechanism from occurring.
Another important metallurgical point is that exposure above 550 °C can cause graphitization of the matrix. This can reduce the strength and hardness of the ductile iron casting even if the original quality was perfect. The longer the exposure, the more the material degrades.
5.3 Combined Mechanical and Thermal Stress
The total stress on the ductile iron casting was the sum of the mechanical stress from the grinding load and the thermal stress from the temperature gradients:
$$\sigma_{\mathrm{total}}=\sigma_{\mathrm{mech}}+\sigma_{\mathrm{th}}$$
At normal conditions, the mechanical stress was only 25 MPa, and the thermal stress was small because the temperature was controlled below 400 °C. The total stress remained far below the yield strength. During the event, however, the mechanical stress was still about 25 MPa, the material strength was only about 29 MPa, and the thermal stress was much higher. The ductile iron casting had no margin left. It yielded locally, cracked, and then the cracks grew under repeated thermal cycles.
I also considered whether the original design could have been improved. The stress raisers at the lifting lugs and at the liner retaining holes are unavoidable in a large ductile iron casting. But even those stress concentrations would not have caused failure if the temperature had been controlled. The root cause was not the geometry; it was the thermal exposure.
6. Repair and Reinforcement Measures
After completing the analysis, I had two objectives. The first was to restore the ductile iron casting to a condition where it could continue operating for a limited period. The second was to prevent the same failure from happening again. Replacing the entire disc body would have been the best long-term solution, but the replacement would take too long because a new ductile iron casting has to be designed, cast, heat-treated, machined, and inspected. The plant needed a workable interim solution.
Because the cracks had not fully penetrated into the inner cavity, the structural integrity of the ductile iron casting was reduced but not completely lost. I decided that an emergency repair with crack arrest and local reinforcement was feasible. The repair procedure had to be very careful because welding a ductile iron casting is more difficult than welding ordinary steel.
6.1 Repair Principle
The repair was designed to achieve three goals. First, the crack tips had to be blunted so that they could not continue propagating. Second, the cracked area had to be mechanically bridged by reinforcement plates. Third, the welding process had to avoid the formation of hard and brittle phases in the heat-affected zone.
The first step was to drill a small hole at each crack tip. This is a standard crack-arrest method. Drilling removes the sharp crack tip and reduces the stress concentration. The holes also gave us a definite point of reference for later inspections.
The second step was to machine slots perpendicular to the crack direction. The slots were arranged across the crack line at carefully chosen intervals. Steel connecting ribs were placed into these slots. The ribs were then welded to the ductile iron casting so that the external load could be transferred across the crack through the steel ribs.
6.2 Selection of Welding Consumables
Welding a ductile iron casting requires a consumable that can tolerate carbon pickup from the base metal. If ordinary steel electrodes are used, carbon can migrate into the weld metal and form hard, brittle martensite or ledeburite. This can produce weld cracking and makes the repair worse than the original crack.
For this reason, I selected a nickel-iron-copper coated electrode, known commercially as EZNiFeCu and classified as Z508. This electrode deposits an austenitic weld metal with high ductility. The nickel-rich deposit can dissolve carbon without forming hard phases, and it compensates for the difference in thermal expansion between the base ductile iron casting and the weld metal.
| Repair welding parameter | Recommended value |
|---|---|
| Welding electrode | EZNiFeCu, also written Z508 |
| Electrode diameter | 3.2 mm, 4.0 mm |
| Current for 3.2 mm electrode | 90 – 130 A |
| Current for 4.0 mm electrode | 130 – 170 A |
| Preheat temperature | 500 – 700 °C |
| Interpass temperature | 500 – 700 °C |
| Post-weld cooling | Slow cooling under insulating blankets |
| Post-weld heat treatment | Stress relief at 550 – 600 °C, hold 2 – 8 h, cool below 30 °C/h |
6.3 Preheating and Temperature Control
Preheating is essential when welding a ductile iron casting. The preheat temperature was controlled at 500 °C to 700 °C. This is higher than the preheat used for ordinary steel, because the goal is to slow down the cooling rate of the weld and the heat-affected zone. Slow cooling allows any carbon dissolved during welding to be controlled, and it reduces the risk of forming hard, brittle phases.
We used electric resistance heating pads and gas torches to preheat the repair area. The heated zone covered the entire welded region and the surrounding base material. Thermocouples were attached to the ductile iron casting to monitor the temperature during the whole welding operation. The temperature had to be kept as uniform as possible around the repair area to avoid creating new thermal gradients.
6.4 Welding Execution
During welding, we controlled the amperage, voltage, and travel speed carefully. The general rule was to use low heat input and short arc length. The electrode was kept as close to the workpiece as possible to prevent nitrogen pickup and porosity. After each weld pass, moderate peening was applied to the weld surface to relieve residual stresses by mechanical working. Peening must be done immediately after the weld bead has solidified but while it is still at a high temperature.
We used small-diameter electrodes for the root passes and slightly larger electrodes for the filling passes. Multiple passes were necessary to fill the slots completely. Each pass was cleaned before the next pass was deposited. This helped to reduce slag inclusions and lack of fusion.
One critical point was to avoid overheating the thin sections of the ductile iron casting. Although the preheat temperature was high, local heat input was kept low. Excessive heat input could cause the surrounding material to reach the transformation temperature and create undesirable microstructures.
6.5 Post-Weld Heat Treatment
After the welding operation was complete, the ductile iron casting was covered with thermal insulation and allowed to cool from the preheat temperature to about 200 °C at a controlled rate. The cooling rate was kept below 30 °C per hour whenever possible.
After cooling to room temperature, the repaired area was subjected to a stress-relief annealing cycle. The temperature was raised to 550 °C to 600 °C, held for two to eight hours, and then cooled slowly. This treatment reduced the residual stresses introduced by the welding process and improved the microstructure of the weld and the heat-affected zone.
Finally, the repair was inspected by ultrasonic testing and magnetic particle testing. The results showed that the weld had full fusion at the interface with the ductile iron casting and that no new cracks were present in the heat-affected zone. The repaired area was then machined back to the required profile so that the liner segments could be reinstalled.
7. Safe Operating Limits After Repair
After the repair, I made it clear that the ductile iron casting was no longer in the same condition as a new component. The repair was intended as an interim measure to allow the plant to continue operating while a new spare ductile iron casting was ordered. The operating limits had to be adjusted to avoid another thermal failure.
The most important limit is the hot gas inlet temperature. I recommended that the temperature of the hot gas in direct contact with the disc body be limited to no more than 400 °C. This is the temperature below which EN-GJS-400-18 retains most of its room-temperature mechanical strength. The temperature measurement point should be as close as possible to the table surface, not merely the gas duct inlet.
I also recommended that the mill not be cooled with compressed air through the access door. After shutdown, the mill should be purged with hot gas at controlled temperature. The cooling process should be gradual, allowing the ductile iron casting to cool naturally without creating a large temperature difference between the surface and the interior.
In addition, the feed moisture must be controlled. The plant had been using very high gas temperatures partly because the feed was extremely wet. Better raw-material management reduces the need for excessive heat input. If the moisture content cannot be reduced at the source, a separate pre-drying system should be considered.
| Post-repair control measure | Technical limit or recommendation |
|---|---|
| Maximum hot gas temperature at mill inlet | ≤ 400 °C when table is in contact with hot gas |
| Shutdown cooling method | No direct compressed air on the table |
| Cooling rate after mill stop | Gradual, slow purge with controlled gas temperature |
| Feed moisture | Reduce or pre-dry to avoid forced overheating |
| Vibration management | Continue monitoring vibration and trip levels |
| Inspection frequency | NDT after 1 month, 3 months, and 6 months |
| Spare part strategy | Order a new ductile iron casting table immediately |
The mill was restarted under these revised conditions. The first month of operation showed no abnormalities. The vibration values remained stable, and the repaired area was inspected by ultrasonic testing after the first month. No crack growth was detected. This confirmed that the repair was able to restore sufficient structural integrity for a limited service period.
8. Long-Term Prevention and Material Selection
Although the repair was successful, I advised the plant that the only fully reliable long-term solution is replacement of the ductile iron casting with a new component. The new part should be manufactured with careful quality control, including ultrasonic testing, mechanical testing, and metallographic verification. The operating envelope must remain within the limits of the material.
For future replacement, I recommended considering whether the design temperature could be improved by adding a heat shield or an insulating layer between the hot gas stream and the table surface. In some installations, a sacrificial wear liner with a thermal barrier can reduce the temperature of the structural ductile iron casting. Such modifications should be evaluated with the original equipment manufacturer because they change the thermal expansion and load path of the component.
It is also important to monitor the health of the ductile iron casting periodically. Thermocouples embedded in the table structure can provide a direct reading of the metal temperature. If the temperature reaches or exceeds 400 °C, an alarm should be triggered. If the temperature reaches 500 °C, the mill should be stopped automatically. This simple protection would prevent the type of thermal damage that caused the failure in this case.
Vibration monitoring is another essential tool. The mill was stopped many times because of vibration before the crack was discovered. If the plant had reviewed the vibration spectrum earlier, they might have noticed the growing crack pattern and shut the mill down before the damage became severe. Vibration monitoring alone is not enough, but it provides an early warning when something changes inside the mill.
Finally, the maintenance department should keep a detailed record of every inspection of the ductile iron casting. The record should include crack detection results, ultrasonic thickness measurements, temperature readings, and operating hours. This information is valuable for future life assessment.
9. Lessons Learned
This case gave me an opportunity to understand the real failure limits of a ductile iron casting in a vertical roller mill. The most important lesson is that the mechanical safety factor is only valid if the material temperature is controlled. A component can have a calculated safety factor of seven or nine at room temperature, but that safety factor can disappear in a few hours if the hot gas temperature is too high.
The second lesson is that a ductile iron casting can be successfully repaired by welding under carefully controlled conditions. Nickel-based electrodes, high preheat, slow cooling, and stress relief are all necessary. But welding cannot restore the original structure completely. The repaired zone remains a weak point, and the component must be protected by strict operating limits.
The third lesson is that operating procedure is as important as material quality. The use of compressed air to cool the mill quickly was very damaging. The operators wanted to save time, but they unintentionally created the temperature gradients that caused the cracks. The plant now understands that the ductile iron casting must never be subjected to forced rapid cooling after shutdown.
10. Conclusion
The disc body of the vertical roller mill was an EN-GJS-400-18 ductile iron casting. The failure investigation showed that the material quality, chemical composition, mechanical properties, and metallographic structure all met the design specification. The mechanical strength analysis showed that the component had a design safety factor of 7.3 and an actual operating safety factor of about 9.5 at room temperature. Therefore, the cracks were not caused by a material defect or by simple mechanical overload.
The root cause was thermal overload. The hot gas inlet temperature was frequently above 600 °C and sometimes exceeded 750 °C. At these temperatures, the yield strength of the ductile iron casting fell to only 10% to 15% of its room-temperature value. The safety factor dropped to approximately 1.0, leaving no margin for thermal stresses. The additional thermal shock caused by blowing compressed air directly onto the hot table after shutdown accelerated the formation and propagation of thermal fatigue cracks.
An emergency repair was performed by drilling crack-arrest holes, installing reinforcement ribs through slots, welding with EZNiFeCu electrode, and applying post-weld heat treatment. The repair restored sufficient structural integrity for interim operation. The mill was restarted with revised operating parameters, including a maximum hot gas temperature of 400 °C at the table surface and no forced rapid cooling after shutdown. Periodic inspection confirmed that the cracks did not grow during the initial post-repair period.
For any vertical roller mill using a large ductile iron casting as the grinding table, the central recommendation is clear: protect the component from excessive temperature. The mechanical strength of a ductile iron casting is high at room temperature, but it is drastically reduced above 500 °C. Controlling temperature is the most effective way to prevent disc-body cracking, extend equipment life, and ensure safe and reliable mill operation.