November 18, 2019

    Do you work with active and passive microelectronic circuitry and design? If so, you’re very familiar with the negative effects that high temperatures can have on these devices. High operating temperatures can cause performance degradation and decreased service life. As such, engineers should constantly consider thermal implications to mitigate potential problems in device designs.

    Gallium nitride (GaN) is still considered a relatively new semiconductor technology. Because of its superior properties, it is being used more and more in market areas previously dominated by legacy technologies. One area where GaN rises above the incumbent technologies is thermal management. In short, GaN “can take the heat” much better than other semiconductor technologies.

    This blog post and our video titled “Understanding GaN Thermal Analysis” provide practical tips about how to properly design using GaN with thermal considerations in mind.
     

    How to Determine GaN Reliability

    Semiconductor reliability is partly deduced by estimating the device maximum channel temperature (TCH,MAX) so an estimated lifetime can be determined. These values are gathered by measuring and modeling thermal resistances, power dissipations, and heat transfer.
     

    Understanding GaN Thermal Analysis Video

    You will learn:

    • GaN reliability and thermal resistance
    • Calculating device power dissipation
    • Estimating device channel temperature and lifetime

    Infrared (IR) microscopes are widely used for determining fault location by searching for hot spots on the semiconductor device. However, spatial resolution limits, difficulty imaging reflective surfaces, and chip surface structures (air bridges) limit the usefulness of IR imaging for measuring GaN channel temperatures. Furthermore, even if perfectly accurate numbers are obtained from IR testing, the true maximum channel temperature actually resides somewhere below the device gate.

    Even though there are limitations with using IR technologies, it remains a popular technique for measuring device temperature. At Qorvo, we take a few more steps beyond IR imaging to accurately determine GaN TCH,MAX. We use 3D thermal modeling, also called finite element analysis (FEA) in conjunction with micro-Raman thermography and then compare those results with RF testing and IR imaging. Using this combined data set, we develop an FEA model for a packaged part providing an accurate value for TCH,MAX.
     

    IR Imaging vs Micro-Raman Thermography

    The Two Main Steps to Determining the Device Maximum TCH and FEA Reliability Estimates

    (1) Finding the Device IR Surface TCH

    • First, the base temperature of the device must be determined.
    • Whether the part is in die or in packaged form – the base temperature is measured similarly – typically using a thermocouple.
      Where to Measure Device Temperature for Package (L) or Die (R)
    • Once the base temperature is known, the next step is to calculate your power dissipation.
    • Use the Qorvo online PAE/Pdiss/Tj calculator to help you determine the power dissipation of your device.
    • You can also use this same calculator to identify your maximum channel temperature.
      View of Qorvo's PAE/Pdiss/Tj Online Calculator

    (2) Determining the FEA Device Median Time to Failure

    • As mentioned previously, the actual maximum TCH for a GaN device occurs below the surface under the device gate as described in GaN Device Channel Temperature, Thermal Resistance and Reliability Estimates.
    • By knowing your IR surface temperature value, you can determine the FEA modeled TCH and GaN device lifetime estimation by using the FEA-Modeled Temperature Estimates versus IR Estimates graph, as shown below.
    • Then take the FEA modeled TCH value and use the Qorvo GaN Device Reliability graph to determine your lifetime prediction.

    Temperature Estimates Graph (L) and Device Reliability Graph (R)

    We encourage you to review the short tutorial video “Understanding GaN Thermal Analysis” as well as the application note GaN Device Channel Temperature, Thermal Resistance and Reliability Estimates for a more detailed explanation on this topic.

     

    Have another topic that you would like Qorvo experts to cover? Email your suggestions to the Qorvo Blog team and it could be featured in an upcoming post.

    Explore the Qorvo Blog

    About the Author

    Qorvo GaN Experts

    Thanks to the authors for this blog and the associated application note: Mark Woods, Matthew Irvine, and Dylan Murdock – all mechanical design engineers at Qorvo. They actively work on GaN technology and bring extensive thermal management knowledge, each from a different perspective: modeling, packaging, and throughput – to name a few. The work they do has a direct impact on Qorvo’s efforts to make GaN’s unique properties more effective in thermal management of device design.