Freescale Semiconductor
Application Note
AN1955
Rev. 0, 1/2004
Thermal Measurement Methodology
of RF Power Amplifiers
By: Mali Mahalingam and Edward Mares
INTRODUCTION
This document explains the methodology used by
Freescale for thermal measurement of high power RF (Radio
Frequency) power amplifiers (RFPA). Semiconductor device
reliability heavily depends on device operating temperature so
the accurate thermal characterization of these high power
devices is crucial in establishing the reliability of the systems
that use such devices.
procedure used by the IR microscope. The emissivity
correction procedure recommended by IR microscope
manufacturers is ineffective at compensating for the
translucent nature of silicon [1]. With the IR microscopy
procedure, the maximum die surface temperature (“hot spot”)
in the measurement field can be located. The hot spot
temperature is selected as the die temperature (T
J
) for thermal
resistance
(θ
JC
). The thermal resistance calculations are
described later.
DIE SURFACE TEMPERATURE (T
J
)
MEASUREMENT
Infrared (IR) microscopy is used to determine the die
surface temperature (T
J
) during amplifier operation. Because
this IR measurement method requires a direct view of the die,
the protective ceramic lid is removed and replaced with a
modified lid that has an opening to view the die. In the case of
overmolded plastic packages, the center portion of the mold
compound is etched away until the die is sufficiently exposed.
Because the heat flow from the device to the heatsink is
dominated by conduction, the measurement error caused by
the removal of the lid or removal of the mold compound around
the die surface is negligible. The exposed die is coated with
a high emissivity coating (see Appendix) to obtain a fixed
emissivity value for IR thermal measurement. This coating
greatly improves the accuracy of the IR measurement
because it eliminates the need for any emissivity correction
Die Temperature (T
J
) Measured
with IR Microscope
CASE TEMPERATURE (T
C
) MEASUREMENT
The case temperature (T
C
) of the package is measured by
a 0.020″ diameter stainless steel sheath thermocouple (Type
J; Omega part # JMQSS- 020G - 12) that is mounted within the
heatsink of the RF circuit. It is mounted from the bottom and
protrudes through the mounting interface to contact with the
bottom surface of the package (Figure 1). A 0.032″ diameter
hole is drilled through the circuit heatsink to permit
thermocouple passage. This small hole provides minimal
disturbance to the heat flow path and interface integrity. The
thermocouple model is selected based on its sensitivity
combined with excellent durability. A spring mechanism is
added to the thermocouple to guarantee constant mechanical
contact with the bottom side of the flange. The placement for
this thermocouple is centered relative to the centermost active
transistor in the package (Figure 2).
4 Active Die
Die
Package
Heatsink
©
Freescale Semiconductor, Inc., 2004, 2006. All rights reserved.
RF Application Information
Freescale Semiconductor
ÑÑÑÑÑ
ÑÑÑÑÑÑÑÑÑÑÑÑÑ
ÑÑÑÑÑÑÑÑÑ
ÑÑÑ
Temperature of Case (T
C
) Measured with Spring−
Loaded Thermocouple Making Direct Contact
Thermocouple Centered Relative to
Centermost Active Die in Package
Figure 1. Case Temperature Measurement
Figure 2. Positioning of Thermocouple
AN1955
1
FREESCALE [ FREESCALE SEMICONDUCTOR, INC ]
