SWRA672 May   2020 AWR6843AOP , IWR6843AOP

 

  1.   Thermal Design Guide for Antenna on Package mmWave Sensor
    1.     Trademarks
    2. 1 Introduction
    3. 2 mmWave AoP package
      1. 2.1 Thermal Characteristics of the Package
    4. 3 Salient features of AoP EVM
      1. 3.1 Thermal Challenges in Dissipating the Heat
    5. 4 Techniques for Mitigating the Heat Dissipation
      1. 4.1 Reduce the System Level Thermal Resistance
      2. 4.2 Board Size Scaling
      3. 4.3 Heatsink Options
        1. 4.3.1 Sheet Metal Heat Sink
        2. 4.3.2 Heat Sink Details
        3. 4.3.3 Mounting Options
        4. 4.3.4 Thermal Characteristics With the Sheet-Metal Heatsink
      4. 4.4 Heatsink with fins
        1. 4.4.1 Thermal Characteristics With the Heatsink
    6. 5 PCB based thermal improvements
      1. 5.1 Thermal via array
    7. 6 Application and Demos
    8. 7 Summary
    9. 8 Acknowledgment
    10. 9 References

Board Size Scaling

One of the ways to dissipate the heat is to use the board itself as a heat sink. The AoP mmWave sensor is built with FR4 PCB fabrication material designed to absorb the heat. In lot of the cases, it would satisfactorily handle the heat dissipation.

Copper traces/planes will have greater thermal conductivity than dielectric material of the PCB and readily conduct the heat through traces and planes in to larger section of the PCB. Hence, filled Cu material with plated through holes on the top/bottom and inner layers would help in spreading the heat to larger surface areas.

As a case study, various thermal simulations were performed to understand the effect of system thermal resistance on scaling of the board size at room temperature of 25 deg C. First mission side of the AOP board size is chosen and its thermal model is extracted and simulation results are extracted to different board size.

Table 1. Thermal Resistance Characteristics for 10% Duty Cycle for Various PCB Sizes

10% Duty PCB PCB Area (mm2) Device P (W) Total P (W) Tj (°C) Case Temp (°C) Board Temp (°C) Psi-jc (°C/W) Psi-jb (°C/W) Effective Rja (°C/W)
Mission section PCB 580 0.8808 1.2054 85.5 82.9 76.3 2.9 10.4 68.7
Square PCB 30 900 0.8808 1.2054 73.1 70.6 64.5 2.9 9.8 54.6
Square PCB 40 1600 0.8808 1.2054 62.0 59.5 53.3 2.9 9.8 42.0
Square PCB 55 3025 0.8808 1.2054 53.9 51.4 45.2 2.9 9.9 32.8
Square PCB 65 4225 0.8808 1.2054 51.0 48.4 42.3 2.9 9.8 29.5

Table 2. Thermal Resistance Characteristics for 25% Duty Cycle for Various PCB Sizes

25% Duty PCB PCB Area (mm2) Device P (W) Total P (W) Tj (°C) Case Temp (°C) Board Temp (°C) Psi-jc (°C/W) Psi-jb (°C/W) Effective Rja (°C/W)
Mission section PCB 580 1.2589 1.6838 105.1 102.1 92.7 2.4 9.8 63.6
Square PCB 30 900 1.2589 1.6838 88.6 85.6 77.0 2.4 9.2 50.5
Square PCB 40 1600 1.2589 1.6838 74.2 71.3 62.6 2.4 9.2 39.1
Square PCB 55 3025 1.2589 1.6838 63.7 60.8 52.0 2.3 9.3 30.7
Square PCB 65 4225 1.2589 1.6838 59.9 57.0 48.3 2.3 9.3 27.7

Table 3. Thermal Resistance Characteristics for 50% Duty Cycle for Various PCB Sizes

50% Duty PCB PCB Area (mm2) Device P (W) Total P (W) Tj (°C) Case Temp (°C) Board Temp (°C) Psi-jc (°C/W) Psi-jb (°C/W) Effective Rja (°C/W)
Mission section PCB 580 1.9 2.4949 134.3 130.0 116.4 2.3 9.4 57.5
Square PCB 30 900 1.9 2.4949 113.6 109.3 96.4 2.3 9.1 46.6
Square PCB 40 1600 1.9 2.4949 94.4 90.1 77.1 2.3 9.1 36.5
Square PCB 55 3025 1.9 2.4949 80.3 76.0 62.9 2.3 9.1 29.1
Square PCB 65 4225 1.9 2.4949 75.2 70.9 57.9 2.3 9.1 26.4

Figure 4 provides the summary of the effective thermal resistance junction to ambient for various board size and power dissipations.

PCB_scaling1.pngFigure 4. Effective Thermal Resistance vs Board Size for Various Duty Cycles

It can be seen from the graph mission side alone PCB size (580) shows effective thermal resistance Rja close to 57-68 Deg C/W depending upon the duty cycle, Hence it has limited amount of power dissipation it can handle without using additional measures such as heatsink or enclosure to take the heat out from the board. However as PCB size increases it can be seen effective thermal resistance decreases beyond 3000 sq mm there is diminishing return from the board size increase in reducing effective thermal resistance.

Figure 5 and Figure 6 shows the thermal simulation example done on the mission section of the EVM without any heatsink at 10% and 25% duty cycles, 25°C ambient temperatures.

Mission_board_1.pngFigure 5. Thermal Simulation of Mission Side EVM With 10% Duty Cycle
Mission_board_2.pngFigure 6. Thermal Simulation of Mission Side EVM With 25% Duty Cycle

It can be seen from the simulation at 25% duty cycle and above, there is very little margin to reach maximum junction temperature. Hence it’s recommended to operate under 10% duty cycle without any heatsink or other thermal measures.