The Influence of Loading Protocol in Mechanical Fatigue Tests on - - PowerPoint PPT Presentation

Vahid Tadaion

The Influence of Loading Protocol in Mechanical Fatigue Tests on Damage Development in Silica Refractories

17.06.2019

Academic Advisors - Dr. T. Tonnesen & Prof. R. Telle Industrial Advisor - Dr. Kirill Andreev (TATA STEEL)

Vahid Tadaion, EcerS Conf., 17.06.2019

Aim

Why do we do cyclic fatigue tests?

Experimental Part

• Loading protocols in Fatigue test Mode I, II and III
• Monotonic vs. Cyclic loading
• Strain amplitude vs. NO. of cycles to failure
• Damage monitoring - NDT methods - mechanical and microstructural analysis

Modelling Part

• Loading protocol
• Results
• Linear vs. Nonlinear behavior
• Models' vessels

Practical Relevance and Future Work

Vahid Tadaion, EcerS Conf., 17.06.2019

Aim

To investigate the influence of different loading protocols in cyclic loading tests on damage development and the practical relevance of different protocols with respect to service loads.

Why do we do cyclic fatigue tests?

The refractory lining in applications with batch production nature,

• ften fails after being exposed to repetitive thermal loads, so the

failure usually doesn't occur in a single run. Material selection - Long campaign life of silica lining in coke

• vens, several decennia. No sufficient in-service performance data

is available to establish a correlation with the material properties or to judge new materials.

Vahid Tadaion, EcerS Conf., 17.06.2019

Experimental Part - Fatigue Protocols

Vahid Tadaion, EcerS Conf., 17.06.2019

Mode I – Constant force limits – Stress-controlled

Drawbacks

• Refractory lining predominantly experiences strain-

controlled loads [1].​

• No strain-softening can be seen.
• No constant strain amplitude can be applied!

Strain,

• Stress,

MPa Stress, MPa Strain, -

• NO. of Cycles, -

Advantage

A well-known method to study cyclic fatigue failure

Vahid Tadaion, EcerS Conf., 17.06.2019

Strain, -

Mode II – Strain-controlled – Constant displacement limits

Advantages

• Has a simple loading protocol
• A more conventional approach

Drawback

The strain amplitude is not constant, it's diminishing.

Stress, MPa Strain,

• NO. of Cycles, -

Stress, MPa Strain, -

Vahid Tadaion, EcerS Conf., 17.06.2019

Mode III – Strain-controlled – Constant displacement amplitude

Advantages

• The strain amplitude is constant
• The method allows gradual damage accumulation

(strain softening)

Drawback

Has a complex (not self-explanatory) loading protocol

Stress, MPa Strain, - Strain,

• Stress,

MPa

• NO. of Cycles, -

Vahid Tadaion, EcerS Conf., 17.06.2019

ε Peak /ε Failure

Stress, MPa Strain, -

Results - Monotonic Loading vs. Cyclic Loading

Vahid Tadaion, EcerS Conf., 17.06.2019

Results - Strain Amplitude vs. Number of Cycles to Failure

Normallised Strain Amplitude, %

• NO. of Cycles, -

Vahid Tadaion, EcerS Conf., 17.06.2019

Results - Strain Amplitude vs. Number of Cycles to Failure

Normallised Strain Amplitude, %

• NO. of Cycles, -

Despite large spread of data, Strain Amplitude

• NO. of Cycles

Vahid Tadaion, EcerS Conf., 17.06.2019

Results - Damage Monitoring - NDT - Mechanical Characterisation

Impulse Excitation technique is more sensitive to d a m a g e . Ultra-sound Velocity can evaluate damage more progressively.

E modulus, GPa Stress, MPa Strain, -

• NO. of Cycles, -

Strain,

• Normallised

E modulus, Velocity, %

Vahid Tadaion, EcerS Conf., 17.06.2019

Results - Damage Monitoring - X-ray Microtomography

3 mm

Before Loading (0 Cycles)

3 mm

Few Cycles After Peak

3 mm

After Peak (970 Cycles)

Vahid Tadaion, EcerS Conf., 17.06.2019

Conclusion - Experimental Part

• The loading protocol significantly influences the potential to resist crack propagation.

Descending order - Mode II, Mode III and Mode I.

• Depending on the protocol, the damage accumulation can be either sigmoid (Mode I

and III) or exponential (saturation) function (Mode II).

• The strain-controlled method allows more gradual, less brittle, failure than the stress-

controlled method.

• The mechanical tests of various loading schedules can represent different service loads

and assist in selecting the optimal refractory material.

Vahid Tadaion, EcerS Conf., 17.06.2019

Modelling Part Fatigue Protocol Mode II

Vahid Tadaion, EcerS Conf., 17.06.2019

Modelling Part

Goal - Is it possible to model fatigue degradation? Two DEM Models

• Environment Itasca PFC5 2D
• Contact bond Flat-joint
• Model shape Cylinder (d=30, H=50mm)
• Loading setup Unconfined compressive fatigue
• Loading protocol Strain-controlled fatigue - constant displacement limits (Mode II)
• εT = 40% εPeak
• εT = 95% εPeak

Vahid Tadaion, EcerS Conf., 17.06.2019

εT = 95% εPeak εT = 40% εPeak

• Extensive strain-hardening occurs
• Significant irreversible strain is developed
• Fatigue degradation
• No strain-hardening occurs
• No irreversible strain is developed
• No fatigue degradation

Results - Stress-Strain Curves

Model A

Non-linear behavior

Model B

Linear behavior

Vahid Tadaion, EcerS Conf., 17.06.2019

Model B Model A

Before Loading

Results - Models' vessel at different loading stages

Vahid Tadaion, EcerS Conf., 17.06.2019

Results - Models' vessel at different loading stages

Model B Model A

at Peak** Before Loading After 10 Cycles*

*εT = 95% εPeak **Monotonic loading

Vahid Tadaion, EcerS Conf., 17.06.2019

Results - Models' vessel at different loading stages

Model B

Well-packed

Model A

Loosely packed at Peak**

at Failure***

Before Loading After 10 Cycles*

*εT = 95% εPeak **Monotonic loading ***10% σPeak

Vahid Tadaion, EcerS Conf., 17.06.2019

Conclusion - Modelling Part

• It is possible to model fatigue degradation with DEM.
• For given loading programs, the degradation patterns are relatively similar to the

stress-strain curves observed in Lab experiments.

• The degradation process can be judged from the irreversible strains developed in the

programs with a similar maximal displacement amplitude.

Vahid Tadaion, EcerS Conf., 17.06.2019

What's the Practical Relevance

• f the Fatigue Protocols?

Vahid Tadaion, EcerS Conf., 17.06.2019

Practical Relevance of Fatigue Mode III

Bending Mode III

Repetitive thermal shock events in an unconstrained refractory lining,

e.g. silica lining of Coke oven walls

Vahid Tadaion, EcerS Conf., 17.06.2019

Practical Relevance of Fatigue Mode III

Bending Mode III

Repetitive thermal shock events in an unconstrained refractory lining,

e.g. silica lining of Coke oven walls

2015-2016 (TATA STEEL in collaboration with KU Leuven & TU Delft)

We conducted cyclic fatigue tests on Silica and Fused Silica [2,3,4].

2018 (ATHOR project - RWTH Aachen in collaboration with TATA

STEEL)

We conducted cyclic thermal shock tests [5].

Vahid Tadaion, EcerS Conf., 17.06.2019

Practical Relevance of Fatigue Mode II

Comp. Mode II

Compaction of the back-up lining/joint opening in a constrained refractory lining,

e.g. Insulation lining in Steel Ladle

Vahid Tadaion, EcerS Conf., 17.06.2019

Practical Relevance of Fatigue Mode II

Comp. Mode II

Compaction of the back-up lining/joint opening in a constrained refractory lining,

e.g. Insulation lining in Steel Ladle

2019 (ATHOR project - RWTH Aachen in collaboration with TATA STEEL and other partners)

We will conduct fatigue tests on Insulation fire-clay brick classes 26 and 28.

• Fatigue Mode II - the insulation lining is frequently squeezed between the working lining and ladle shell.
• Fatigue Mode III and cyclic thermal shock tests [5,6] - what if the insulation materials are employed as the hot face

layer - attractive to reduce energy losses.

Vahid Tadaion, EcerS Conf., 17.06.2019

References

[1]Schacht, C. (2004). Refractories handbook. CRC Press. [2]Tadaion, V. (2016). Cyclic fatigue resistance of silica based refractories. (master's thesis). Aalto University, Helsinki, Finland. [3]Andreev, K., Wijngaarden, M., Put, P., Tadaion, V., & Oerlemans, O. (2017). Refractories for Coke Oven Wall–Operator’s Perspective. BHM Berg-und Hüttenmännische Monatshefte, 162(1), 20-27. [4]Andreev, K., Tadaion, V., Koster, J., & Verstrynge, E. (2017). Cyclic fatigue of silica refractories–effect of test method on failure process. Journal of the European Ceramic Society, 37(4), 1811-1819. [5]Andreev, K., Tadaion, V., Zhu, Q., Wang, W., Yin, Y., & Tonnesen, T. (2019). Thermal and mechanical cyclic tests and fracture mechanics parameters as indicators of thermal shock resistance–case study on silica

• refractories. Journal of the European Ceramic Society, 39(4), 1650-1659.

[6]Manson, S. S. (1966). Thermal stress and low-cycle fatigue.

Advanced THermomechanical multiscale mOdelling of Refractory linings Advanced THermomechanical multiscale mOdelling of Refractory linings

Thank you for your attention

27

Acknowledgments: this work was supported by the funding scheme of the European Commission, Marie Skłodowska-Curie Actions Innovative Training Networks in the frame

• f the project ATHOR - Advanced THermomechanical multiscale modelling of Refractory

linings 764987 Grant.

www.etn-athor.eu

Tadaion@ghi.rwth-aachen.de