Conjugate Heat Transfer Analysis on a graphics card.

JORGE LOZANO

Advanced CFD using ANSYS Fluent

Conjugate Heat Transfer Analysis on a graphics card.

Perform a steady-state conjugate heat transfer analysis on a model of a graphics card. You can use appropriate materials of your choice for the simulation. Make sure to properly define the correct solid and fluid zones. Refer the video for further clarification and the model is provided below the video.

Run the simulation for best possible mesh with combination of coarse and refined mesh in different regions. Explain the reason for choosing the mesh settings.

Your Objectives:

1. Run the simulation by varying the velocity from 1m/sec to 5m/sec for at least 3 velocities and discuss the results.
2. Find out the maximum temperature and heat transfer coefficient attained by the processor.
3. Prove that the simulation has achieved convergence with appropriate images and plots.
4. Identify potential hotspots on the model.

SIMULATION SET-UP

To perform simulations using Ansys/Fluent it’s desirable to have knowledge in the type of fluid model to adjust the condition of turbulent or laminar models according to the Reynolds number, the naiver-stokes equations, mesh characteristics, and the data postprocessing. For this case several simulations will be performed variating in the mesh, and initial conditions having as follows parametric results, duplicating several simulations and modifying the unique variables, such as the heat sink material and the inlet velocity as in F.1.

F.1 Simulation Set up

MATERIALS

The material properties will be adjusted to every volume, having some of the most common materials used in electronic graphics cards such as aluminum, gools, nickel, and silicon for some materials, havin the properties that will be important to take notice in the following table, F.2.

F.2 Material properties

The components will be known with the following names, numbers and materials applied, as in F.3.

1. Inlet
2. Outlet
3. Wall enclousure
4. Air Volume-Air
5. Electronic component 1-gold
6. Electronic component 2-gold
7. Electronic component 3-nickel
8. Electronic component 4-nickel
9. Electronic component 5-nickel
10. Electronic component 6-nickel
11. Assembly Base 1-Aluminum
12. Assembly Base 2-Aluminum
13. Heat Sink Fins-Aluminum-Steel
14. Processor-Silicon

F.3 Material Component Selection

Mesh

Fo these simulations two meshes will be perfromed pne considered as coarse, not adjusting any criteria or method to generate the mesh automatically, and the other one considering a 512k element limit due to the educational software but adding some methods and inflation generation to improve the validation.

Due to the element quantity restriction, the finer mesh could obtain has a .94 value of the mesh quality “Skewness”, having at the coarser mesh a value of “.999”, that in a validation mesh analysis represents a bad mesh, although it gives a fast, general result of what could be the possible real result.

F.4 Coarse and fine mesh

Conditions

To perform the following simulations the model of turbulence will be the K-epsilon model, adjusting the inlet velocity from 1m/s to 3m/s and 5m/s, having at first the aluminum material as the heat sink material and then steel as the material, seeing the comparison between the coarser and finer mesh using the aluminum heat sink and then with the finer mesh using the steel material, to see the effect in forces heat convection transfer from the velocities and the variation of having a material with high and low heat conductivity thermal property material.

Having the following heat sources from the graphics card:

1. 1 & 2 Electronic components: 2 W
2. 3 – 6 Electronic components: 4 W
3. Processor: 10 W

RESULTS

To perform the simulation, it’s considered that for every case the following results will be saved to generate graph, histogram, or contour, in order to visually see the results comparison, along the 1000 iterations selected, and determine the validation of the simulation with the temperature’s value, standard deviation in several elements, heat transfer within the wall and from the forced convection heat transfer, and finally the y-plus value to determine the mesh element size whether or not it need to be refined.

1. Residuals
2. Standard temperature deviation
3. Electronic components average temperature
4. Processor average temperature
5. Heat sink average temperature
6. Maximum processor temperature
7. Heat transfer coefficient processor
8. Contour temperature
9. Y plus histogram
10. Graphics card local temperature
11. Graphics card wall heat transfer coefficient
12. Fluid velocity over graphics card

Conclusion

One of the condition to consider within the post-processing of the cases are the wall heat transfer and the hot spots within every case, where the faster the velocity in the inlet, the grater the value will be, having greater values in the finer mesh, aluminum heat sink material, not determining an average value but obtaining a contour to see the variation along the processor the see the result to select an improvement or modification, as well as the hot spots, due to the variation in the flow it’s better to define a modification seeing the complete geometry results rather than just having a value.

Each simulation was named after the following system:

Simulation name description:

Fist the temperatures are obtained in G.1, where all simulations are there along several critical components, where as seen from the coarse and finer mesh there is a visual difference, where for a final result it’s desirable to use the finer cases, comparing the aluminum to the steel heat sink material, the steel cases tend to have more temperature, being the aluminum the best choice to dissipate energy due to its thermal conductivity and even though the greater velocity inlet value the lower the temperatures are, a 3 m/s velocity is great to define a reasonable temperature change and not demand more from the fan that will be use.

G.1 Components temperature

According to the dissipation through all the component being analyzed, the standard deviation of the temperatures for each case are obtained, where as the same as the temperatures, the finer meshes are considered to perform a comparison, being the heat sink one of the most important elements to analyze besides the processor, the dissipation mustn’t be more that 4 - 5 C, where as seen for the steel material it’s without a doubt a bad material to select for this system having 11 – 16 C of deviation., as seen in G.2.

G.2 Components temperature deviation

The last graph G.3 represents the forced convective heat transfer produced from the inlet velocity fluid (air @ 283 K) where as expected the higher the velocity the greater the value it will generate.

G.3 Fluid heat transfer over the graphics card

Results documentation

K-E COARSE MESH 1 M/S 283K HS: ALUMINUM

K-E COARSE MESH 3 M/S 283K HS: ALUMINUM

K-E COARSE MESH 5 M/S 283K HS: ALUMINUM

K-E FINE MESH 1 M/S 283K HS: ALUMINUM

K-E FINE MESH 3 M/S 283K HS: ALUMINUM

K-E FINE MESH 5 M/S 283K HS: ALUMINUM

K-E FINE MESH 1 M/S 283K HS: STEEL

K-E FINE MESH 3 M/S 283K HS: STEEL

K-E FINE MESH 5 M/S 283K HS: STEEL

• Simulation files can be download at the following link:

https://drive.google.com/file/d/1KqKyF3mlcLDYt3anDJ33jleEADFUoNla/view?usp=sharing