Conjugate Heat Transfer CHT Analysis on a graphics card using ANSYS FLUENT

OBJECTIVE:

To perform a conjugate heat transfer analysis on a graphics card and postprocess the velocity and temperature distribution.

Conjugate heat transfer analysis:

It is defined as the variation of temperature within solid and fluid due to thermal interation between solids and fluids.Typical example of heating or cooling of object by flow of air in which it is immersed.

Application in electronic devices:

Finned surfaces of various shapes called heat sinks,are frequently used in cooling of electronic devices.Energy dissipated by these devices is transferred to the heat sinks by conduction and from the heat sinks to the ambient air by natural or forced convection,depending on the power dissipation requirements.Natural convection is usually preferred for electonic components since it is involves no moving parts.However,in the natural convection,the components are more likely to run at a higher temperature Therefore proper selection of heat sinks reduce the operation temperature and risk of failure.This is where CHT analysis plays a major role to determine the temperature distribution which will be helpful for design of electronic circuits to operate at optimum temperature.

Project description:

• A microprocessor is heated up and the energy dissipated from the processor is transferred to fins by conduction through a fin solid base.
• Air as a fluid is used for the forced convection
• Evaluate the temperature distribution for the two cases such as baseline and refined mesh with boundary conditions remains constant.
• Prove the simulation has achieved convergence with appropriate images.

METHODOLOGY:

1.Geometry-Spaceclaim:

• Enable the share topology under properties In order to avoid interpolation errors.

2.Mesh:

• Naming the selection such as inlet,outlet,symmetry wall-enclosure,chip wall-graphics card.
• Created baseline mesh and refined mesh.

3.Setup:

• Solver:Steady,pressure based,absolute velocity formulation
• Energy equation turned on
• Viscous:K epsilon
• Material:Fluid-Air ,Solid-Aluminium,Gold,Copper
• Zones:Cell zone conditions

• Solid-Fin-Aluminium
• Solid-processor-Gold-source term---> 1e8 `W/m^3`
• Solid-PCB-Copper
• Fluid-Air-Enclosure

• Boundary conditions

• Inlet-velocity inlet- 6 m/s
• outlet-pressure outlet-0(guage pressure)
• Symmetry wall.

• Hybrid initialization is done.
• 500 iterations were made to run the calculation.

4.Solution:

CASE 1: Baseline mesh:

1.Mesh (Default element size 10.289 mm):

Nodes: 15964

Elements: 85163

2.Residual plot:

• 700 iterations were made to run the calculation.
• As we see from the plot,solution starts converging after 90 iteration with little variation.

3.Temperature contour:

i)

fig.temperature distribution on a graphics card

ii)

fig.Temperature distribution on processor and PCB.

iii)

fig.Temperature contour on XZ Plane

• we see that the maximum heat is generated at processor and spread at its surrounding.
• Temperature is minimum at the inductance and chip region.

4.Velocity contour:

fig.velocity distribution along XZ plane

• Velocity of air is maximum at the top region of an enclosure and it decreases when it impinges the graphics card wall.

5.Wall heat transfer coefficient:

CASE 2: Refined Mesh:

1.Mesh:

• PCB (base)- sizing- 1mm
• Processor-sizing-0.5mm
• Fin base and fine-0.5mm
• Default- 4 mm

Nodes: 71524

Elements: 409207

2.Residual plot:

• 500 iteration were made to run the calculation.
• Solution is converged at 63 iteration.

3.Temperature contour:

i)

fig.temperature distribution on a graphics card

ii)

fig.Temperature distribution on processor and PCB.

iii)

fig.Temperature contour on XZ Plane

• we get a better results of temperature distribution compared to case 1.
• This is because more number of mesh elements provided.
• Temperature can be seen maximum at the processor and center portion of the fin base.

4.Velocity contour:

fig.velocity contour along XZ plane

5.wall heat transfer coefficient:

fig.heat transfer coefficient XZ plane

fig.heat transfer coefficient

• Heat transfer coefficient results looks more accurate than case 1.
• we notice that the heat transfer coefficient is higher where the velocity values are higher.

RESULTS:

• In both the case the temperature and residual plots are almost produce same results.Therefore grid independency test is achieved.
• Case 2 provides better results than case 1.

CONCLUSION:

Performed a conjugate heat transfer analysis on a graphics card and hence proved that the simulation has achieved convergence.

Link : https://drive.google.com/open?id=1wR0Inwb61t4lNgUWOUsV89VtA8hRFF8z