Conjugate Heat Transfer Analysis On Graphics Card

Conjugate Heat Transfer Analysis On Graphics Card.

Objective - To perform steady state CHT Analysis on the model of Graphics Card.

Introduction -

CHT Analysis -

1. CHT stands for Conjugate Heat Transfer.
2. This type of analysis allows the simulation of heat transfer between solids and fluids boundaries by exchanging thermal energy at the interfaces between them.
3. CHT analysis has an advantage over the FE thermal analysis in calculating the wall heat transfer coffecient . It is calculated directly rather than solving empirical equations.
4. CHT analysis is used to correctly predict the surface temperature and heat exchanging capabilities of engineering devices.

Benefits :

1. Effective Heat transfer
2. Energy savings
3. Interactions between solid and fluid bodies.

Applications:

1. Heat exchangers
2. electronic equipments like graphic cards,motherboards,etc
3. General heating and cooling systems.
4. turbochargers
5. Exhaust ports.

Case Description -

The Graphics card is cooled by using Forced convection by passing Air. The different parts of the graphics card are Fins,Board or Base and Processor which is the heat source. The processor is palced in contact with the board and fins. This Model is enclosed by an enclosure which acts an inlet and outlet. The air is passed from inlet which then flows over the model and exchange the heat generated by the model by convection.

Assumptions -

1. The thermal design power of the processor in graphics card is assumed as 5 W.
2. Processor is the only Heat source in the Graphics Card Model.
3. The Dimensions of processor are (8*8*1) mm.
4. The Processor is in idle condition.
5. Air flows at speed of 10 m/s and at 300K temperature.
6. The Heat generated is given by

Heat generated = TDP/Volume W/m^3

= 5/((8*8*1)*10^-9) W/m^3

= 78125000 W/m^3

Materials for the Parts -

Fins - Copper

Board - Aluminium

Processor - Gold

The Analysis on the graphics card is carried out for two mesh sizes,one baseline mesh and one Refined mesh.

Procedure -

1. Load the Model in the Spaceclaim and share the edges and faces in the model by turning On the Share topology.
2. Load the shared model in the Ansys Mesher and mesh the model with approriate mesh size. Check the mesh quality and number of elements.
3. Open Fluent. Set correct models,boundary conditions and apply proper materials for the model.
4. Click Initialization and Run for required iterations.
5. Check the Results in CFD post.

Model of Graphics Card -

CASE I - Baseline Mesh

1) Mesh -

No. of elements - 85711

No. of nodes - 16129

Element size - 10.25 mm

Mesh Quality -

2) Ansys Fluent preparation -

Steady state ,pressure-based and absolute

Turbulence model is K-epsilon and Energy equation is ON.

Zones - board - Aluminium

fins - copper

Processor - Gold - Source terms On - Constant Heat source - 78125000 w/m^3

Boundary conditions - Inlet -10 m/s and 300 K

outlet - 0 bar pressure.

Initialization - Hybrid

Run for 250 iterations

3) Results -

a) Convergence plot -

As seen from the convergence plot graph, Convergence occurs at 125 iterations.

b) Temperature Contour -

c) Velocity Contour -

d) Wall heat transfer Coefficient -

e) Board Temperature plot -

f) Fins temperature -

CASE II - Refined Mesh

1) Mesh -

No. of elements - 450049

No. of nodes - 87637

Element sizes - a) Enclosure - 6 mm

b) Processor - 0.1 mm

c) Board and Fins - 1 mm

Mesh Quality -

2) Ansys Fluent preparation -

Steady state ,pressure-based and absolute

Turbulence model is K-epsilon and Energy equation is ON.

Zones - board - Aluminium

fins - copper

Processor - Gold - Source terms On - Constant Heat source - 78125000 w/m^3

Boundary conditions - Inlet -10 m/s and 300 K

outlet - 0 bar pressure.

Initialization - Hybrid

Run for 7500 iterations

3) Results -

a) Convergence plot -

As seen from the convergence plot, the convergence occurs around 4550 iterations.

b) Temperature contour -

c) Velocity Contour -

d) Wall Heat transfer coefficient -

e) Board temperature plot -

f) Fins temperature -

Conclusion -

1. The processor,fins and Board are the parts which are focused for mesh refinement.
2. The maximum temperature attained by processor in Case I is 321.79 K and in Case II is 317.558 K .
3. The Maximum velocity attained by air in Case I is 12.3101 m/s and in Case II is 12.0177 m/s .
4. The Wall heat transfer coefficient in Case I is 1155.42 W/m^2 K and in Case II is 2665.98 W/m^2 K .
5. The potential hotspots in the graphics card are the processor, fins and the Contact region between the board and processor.