Conjugate Heat Transfer Analysis for graphics card

GRAPHICS CARD:-

• Graphics card is an expansion card which generates feed of output image to display device.
• Graphics card may not be the essential component for an average or normal user but it is the most important component for high end games and people who run resource intensive graphics application or software.
• Graphics card are very advance device and are made up of complex component.
• Processor in the electronic device such as graphics card, CPU etc tend to get heat up which result in decrease in preformance or damage of the electronic device.
• To ensure proper cooling of the processor fin plate made up Aluminium material has been attached to processor made of Copper material. The base plate material of the processor card is made of Steel plate.

CONJUGATE HEAT TRANSFER ANALYSIS ON A GRAPHICS CARD:-

• For the conjugate heat transfer, device like graphics card, motherboard, data processing unit and different kind of microprocessor etc are best example in the electronic field which are needed to be simulated their thermal performance in the design stage.
• Because in supercomputer if we observe, million of element, nodes and iteration have to be performed.
• While we perform the simulation the computer temperature will predominantly increase, at the range of iteration and all, a perfectly material has to be selected .
• A high quality simulation has to run on that.
• Not only in the supercomputer but also this phenomenon should be seen on every computer. But mostly in the high performance computer (Like RAM, Graphics card, processor, etc).

OBJECTIVE:-

1. Find out the maximum temperature attained by the processor.

2. Prove that the simulation has achieved convergence with appropriate images and plots.

3. Find out the heat transfer coefficient at appropriate areas of the model.

4. Identify potential hotspots on the model.

GEOMETRY:-

The graphics card contain about following component are as follow:-

1. Base Plate:- It used to support the processor and other electronic component.
2. Processor:- It is the heat source of the graphics card.
3. Fins:- It is extended surface used for heat dissipation.

MESH:-

In this case we have run the graphics card in two different simulation with two different mesh size to obtain accurate result.

SETUP:-

General:-

1. Steady State
2. Pressure based solver
3. Energy on
4. Viscous:- k-epsilion - Standard - Standard Wall Function.

Material:-

1. Fluid:- Air
2. Solid:- Steel, Aluminium and Copper

Cell Zone:-

1. Fluid:- Air
2. Solid:- Base - Steel
3. Solid:- Fins - Aluminium
4. Solid:- Processor - Copper - Source Term (Heat source with Heat flux on 2e07 `W/m^2`)

Boundary Condition:-

1. Inlet :- Velocity (5 m/s) and Temperature (300 K)

Method:-

1. Scheme:- Coupled
2. Initialization:- Hybrid
3. `Number of Iteration :- 300`

Case 1:- (Baseline Mesh)

MESH:-

Isometric view of base mesh

Section view of base mesh

Element Size:-20.578 mm

Number of Element:- 83061

Residual Plot:-

Residual Plot for baseline mesh from here we can conclude that the solution get convergence after 150 Iteration. Hence we have convergence solution at 150 Iteration.

Temperature Contour:-

Temperature Plot of cut plane for baseline Mesh

Velocity Contour:-

Velocity plot of cut plane for baseline Mesh

Base Plate Temperature:-

Base Plate Temperature for based line Mesh

Fins Temperature:-

Fins Temperature for Based Line Mesh

Processor Temperature:-

Processor Temperature for Baseline Mesh

Wall Heat Transfer Coefficient:-

Wall Heat Transfer Coefficient for Based Line Mesh

Case 2:- (Refined Mesh)

MESH:-

Isometric view of the Refined Mesh

Section View of the Refined Mesh

Element Size :- 1.6 mm

Number of element:- 505506

Residual Plot:-

For the above residual plot we can see that we get convergence solution after 150 iteration. Hence we have convergence solution at 150 iteration.

Temperature Contour:-

Section view of Temperature contour for refined Mesh

Velocity Contour:-

Section view of velocity contour for refined mesh

Base Plate Temperature:-

Base Plate Temperature for Refined Mesh

Fins Temperature:-

Fins Temperature for refined Mesh

Processor Temperature:-

Processor temperature for refined mesh

Wall Heat transfer Coefficient:-

Wall Heat Transfer coefficient for Refined mesh

RESULT:-

CONCLUSION:-

As we have run two simulation with two different size mesh we can concluded as follow:-

• The maximum temperature attained by the processor is the baseline mesh case it is reocorded as 312.15 K. In refined mesh case, it is recorded as 310.66 K
• The scaled residual graph in both case, it is shown that the simulation converged after the iteration number is 150 and after that solution becomes stable.
• The heat transfer can happen on body. But maximum heat transfer coefficient is seen at the processor. In base line mesh, it is 1203.49 `frac{W}{m^2K}` and refined mesh it is 1193.38 `frac{W}{m^2K}`
• In fins the most heat transfer takes place but it is not hotspot on the graphics card. The potentioal hotspot of the graphics card has to be processor, it is the main heat source because it is only device which consume power.