Dismiss
InnovationQ will be updated on Sunday, Oct. 22, from 10am ET - noon. You may experience brief service interruptions during that time.
Browse Prior Art Database

Thermal Fatigue Testing of Discrete Components

IP.com Disclosure Number: IPCOM000088087D
Original Publication Date: 1977-Apr-01
Included in the Prior Art Database: 2005-Mar-04
Document File: 3 page(s) / 45K

Publishing Venue

IBM

Related People

Hathaway, JW: AUTHOR [+2]

Abstract

In power component applications, thermal fatigue failure may be activated by mechanical stresses due to differences in thermal coefficients of expansion of various materials in the component package assembly. When the machine (e.g., power supply) is turned on, the rise in case temperature of a device is mainly due to the junction power dissipation rather than to the slowly rising ambient temperature within the machine assembly. The silicon at the junction has a thermal coefficient of expansion lower than those of the packaging materials. Therefore, if the case is heated more quickly than the junction, the device is subjected to a stress much higher than in its normal use condition. In a thermal fatigue test, it is essential that the normal use condition be simulated in, at least, one test cell.

This text was extracted from a PDF file.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 53% of the total text.

Page 1 of 3

Thermal Fatigue Testing of Discrete Components

In power component applications, thermal fatigue failure may be activated by mechanical stresses due to differences in thermal coefficients of expansion of various materials in the component package assembly. When the machine (e.g., power supply) is turned on, the rise in case temperature of a device is mainly due to the junction power dissipation rather than to the slowly rising ambient temperature within the machine assembly. The silicon at the junction has a thermal coefficient of expansion lower than those of the packaging materials. Therefore, if the case is heated more quickly than the junction, the device is subjected to a stress much higher than in its normal use condition. In a thermal fatigue test, it is essential that the normal use condition be simulated in, at least, one test cell.

As shown in Fig. 1, a device under test is mounted on a movable thin metal plate 1 which is attached to a thin walled rectangular tube 2 by a hold-down nut (not shown) on a stud 3 that is welded to the tube 2. The device is mounted within an area (indicated by broken line 4) that is adjacent to a fin-type movable heat sink 5. A thermocouple is attached to the case of each device and connected to a temperature sensing logic circuit, as discussed below. When the test starts, all devices are powered up at the same time. The thermal sensing control turns off each individual device when the case temperature reaches a certain preset temperature T(max). When all devices have been turned off, the main power supply is turned off and cooling water is run through tube 2 until the case temperature of all devices reaches a preset temperature T(min). Then the main power supply is turned on. One thermal fatigue cycle has been c...