Refractories often account for less than 3% of the total installation cost of the process structure but may account for over 97% of the structure and process functionality. It is vital that these structures maintain themselves and do not under perform. Ironically, a very large number of structures that are just adequate survive because they float into the required space whilst others that have been "over designed" fail.

Modelling refractory structures can be difficult and time consuming. However, CERAM makes use of automated 2D & 3D geometry creation methods that can dramatically ease the process. The following case study examines the design of a refractory arch.

Issues Addressed

Glass furnace designs are driven by a desire to increase the performance of the glass making process but do not take into account the structural performance of the refractory required to maximise the furnace's working life. Computational modelling helps the furnace designer optimise both process performance and furnace life.

Simulation

By using automated-parametric geometry creation the size, number and height of the individual bricks and joints can be varied. The meshing of the structure can also be automated and hence the properties of the joints and the bricks can be automatically changed. Hence, initial "exploratory" models can be constructed from data sheets then more detailed models can be constructed using data generated from CERAM's in house physical and chemical testing resources.

Figure 1: Refractory Arch Showing Individual Bricks

Figure 2: Details of the Joint Structure

In the example shown, thermal boundary conditions were applied using steady state heat transfer coefficients (although if necessary radiation and transient conditions could have been taken into account).

At the working temperature, the structure expanded, but as it was unconstrained, this did not cause a problem as the peak compressive stresses remained low.

However, the arch was not free to float and the model demonstrated that the most likely cause of collapse (in this case) was an incorrect combination of lateral and vertical constraints that lead to very high compressive stresses.

Conclusions

The application of different lateral and vertical constraints dramatically changes the stress distribution in the refractory arch. In fact, the model demonstrated that an arch with more vertical freedom of movement results in a compressive stress distribution that could lead to failure.

Figure 3: Maximum Compressive Stress: 10 mm Lateral & 10 mm Vertical Expansion

Figure 4: Maximum Compressive Stress: 10 mm Lateral & 40 mm Vertical Expansion

The methods employed by CERAM enable multiple trials to be undertaken generating a range of steady state and transient profiles. Modelling gives designers a fundamental understanding of how refractory structures behave under a range of operating conditions; reducing the chance of furnace collapse and increasing furnace life.