As with any power conversion device, the LM6x4xx dissipates internal power while operating. The effect of this power dissipation is to raise the internal temperature of the converter above ambient temperature. The internal die temperature (T_J) is a function of the following:

• Ambient temperature
• Power loss
• Effective thermal resistance, R_θJA of the device
• PCB layout

The maximum internal die temperature for the LM6x4xx must be limited to 150°C. This establishes a limit on the maximum device power dissipation and, therefore, the load current. Equation 8 shows the relationships between the important parameters. Larger ambient temperatures (T_A) and larger values of R_θJA reduce the maximum available output current. The converter efficiency can be estimated by using the curves provided in the Application Curves section. If the desired operating conditions cannot be found in one of the curves, then interpolation can be used to estimate the efficiency. Alternatively, the EVM can be adjusted to match the desired application requirements and the efficiency can be measured directly. The correct value of R_θJA is more difficult to estimate. As stated in the Semiconductor and IC Package Thermal Metrics Application Report, the value of R_θJA given in Section 7.4 is not valid for design purposes and must not be used to estimate the thermal performance of the device in a real application. The values reported in Section 7.4 were measured under a specific set of conditions that are rarely obtained in an actual application.

where

• η = efficiency
• T_A = ambient temperature
• T_J = junction temperature
• R_θJA = the effective thermal resistance of the IC junction to the air, mainly through the PCB

The effective R_θJA is a critical parameter and depends on many factors (just to mention a few of the most critical parameters:

• Power dissipation
• Air temperature
• Airflow
• PCB area
• Copper heat-sink area
• Number of thermal vias under or near the package
• Adjacent component placement

Due to the ultra-miniature size of the VQFN (RNX) package, a die-attach pad is not available, requiring most of the heat to flow from the pins to the board. This means that this package exhibits a somewhat large R_θJA value when the layout does not allow for heat to flow from the pins. A typical curve of maximum output current versus ambient temperature is shown in Figure 9-3 and Figure 9-4 for a good thermal layout. This data was taken on the LM61495RPHEVM evaluation board with a device and PCB combination, giving an R_θJA of about 21.6°C/W. It must be remembered that the data given in these graphs are for illustration purposes only, and the actual performance in any given application depends on all of the previously mentioned factors.

Use the following resources as a guide to optimal thermal PCB design and estimating R_θJA for a given application environment:

• Thermal Design by Insight not Hindsight
• A Guide to Board Layout for Best Thermal Resistance for Exposed Pad Packages
• Semiconductor and IC Package Thermal Metrics
• Thermal Design Made Simple with LM43603 and LM43602
• PowerPAD? Thermally Enhanced Package
• PowerPAD? Made Easy
• Using New Thermal Metrics