VM-LSDYNA-EMAG-007

VM-LSDYNA-EMAG-007
2D Axisymmetric Inductive Heating

Overview

Reference:

Jankowski T., Pawley, N., Gonzales, L., Ross, C., & Jurney, J. (2016). Approximate analytical solution for induction heating of solid cylinders. Annual Mathematical Modelling, 4(40), 2770-2782. https://www.dynaexamples.com/em/eddycurr/2dindheat

Analysis Type(s): Electromagnetism
Element Type(s): Solid Elements ELFORM 1
Input Files:Link to Input Files Download Page

Test Case

In this test case, a heated coil is wrapped around a cylinder. The coil carries a sinusoidal current and has 11 turns. Since the model is symmetric, only a portion of it is modelled to reduce computation time as shown in Figure 196. Model dimensions are given in Table 13 and Figure 195.The material properties are shown in Table 14.

The test case replicates only the first experiment in the reference. The goal is to determine the temperature at the bottom center of the cylinder where the temperature sensor is located in the experiment.

Table 13: Model dimensions

T        293.15 KI(RMS)        %I* 1063 A
R = 4.51 cm
b = 6.65 cm
I = 19.8
N = 11
Vmax (RMS) =  47.1 V
Lcir = 0.87 μH
C = 15.41/55.7 μF
εR = 0.9

Figure 195: Problem sketch and dimensions

Problem sketch and dimensions

Table 14: Material properties of POCO grapite

Material Properties of POCO grapite
T(K)σ(1/Ωm)c(J/kg K)k(W/m K)ρ(kg/m3)
293.1575,2507211201720
393.1586,41710261081720
493.1596,5851269951720
593.15105,0011424881720
693.15111,2441549801720
793.15115,2921645751720
893.15117,4371712701720
993.15118,1241763661720
1093.15117,8071809621720
1193.15116,8601855601720
1293.15115,5491892571720
1393.15114,0351926561720
1493.15112,3961959551720

Figure 196: Problem setup: Only a portion of the model is used since it is symmetric.

Problem setup: Only a portion of the model is used since it is symmetric.

Analysis Assumptions and Modeling Notes

When considering the conductivity of the plate (eddy current diffusion), the results are time dependent. The latest time step is taken to verify the results since the problem itself is static.

LS-DYNA *EM_CONTROL card is set to 1 to activate the eddy current solver:

Intensity applied to the coils is given individually for each coil using *EM_CIRCUIT_SOURCE.

Electromagnetic properties are given using EM_MAT_002 and EM_MAT_001.

Finally, the properties are given parametrically and stored as parameters.

The mesh is formed by solid elements formulation 1, which is a constant stress solid element. There are four elements across the cylinder thickness. Figure 197 shows the mesh as well as a mesh detail.

Figure 197: Mesh

Mesh

Results Comparison

The experimental results are compared with LS-DYNA output in Figure 198. As you can see, the difference between both curves is reduced as the simulation time increases and gets very close at the end. The initial discrepancy is related to how the experiment is configured which causes the cylinder to heat up very quickly. This results in a significant temperature gradient between the heated zone and the bottom of the load, causing the discrepancies. As a result, the very first 50 seconds are disregarded when calculating the error.

Figure 198: Simulation vs experiment results for 1000 and 3000 AT for points in air

Simulation vs experiment results for 1000 and 3000 AT for points in air

The temperature is obtained from the d3plot files, specifically from node 1574 at the bottom of the cylinder where the temperature sensor is located in the experiment is (see Figure 199).

Figure 199: Node location for temperature extraction

Node location for temperature extraction

The error between the experimental results and the simulation is presented in the table below. The error curve is presented in Figure 200: Relative error between simulation and experiment. As you can see, the error decreases as the simulation progresses, leading to a very close result as the final temperature is reached.

ResultTargetLSDYNA AppError (%)
Temperature at t = 42 sec232.45271.2416.69
Temperature at t = 56 sec300.19338.0712.62
Temperature at t = 70 sec360.84398.7010.49
Temperature at t = 84 sec417.16454.829.03
Temperature at t = 98 sec469.65507.358.03
Temperature at t = 112 sec520.12556.617.02
Temperature at t = 126 sec566.88602.746.33
Temperature at t = 140 sec610.89645.955.74
Temperature at t = 154 sec653.00686.375.11
Temperature at t = 168 sec693.54723.974.39
Temperature at t = 182.3 sec732.05759.463.74
Temperature at t = 196.5 sec768.02792.163.14
Temperature at t = 210.5 sec800.05821.922.73
Temperature at t = 224.5 sec830.23849.402.31
Temperature at t = 238.5 sec860.72874.621.61

Figure 200: Relative error between simulation and experiment

Relative error between simulation and experiment