THE COMPANY
PROVEN EXCELLENCE
GUIDING MISSION
GUARANTEE OF COMMITMENT
OUR QUALITY MANAGEMENT SYSTEM
RELEX WORLDWIDE
CAREERS
DIRECTIONS
CONTACT US
RELEX RELIABILITY STUDIO
ENTERPRISE EDITION
DEMONSTRATIONS
LITERATURE
WHAT'S NEW IN STUDIO
SYSTEM ARCHITECTURE
CUSTOMER TESTIMONIALS
FAULT TREE/EVENT TREE
FMEA/FMECA
FRACAS CORRECTIVE ACTION
HUMAN FACTORS RISK ANALYSIS
LIFE CYCLE COST
MAINTAINABILITY PREDICTION
MARKOV
OPTIMIZATION AND SIMULATION
RELIABILITY BLOCK DIAGRAM
RELIABILITY PREDICTION
WEIBULL
PROFESSIONAL SERVICES
RELIABILITY CONSULTING
MTBF PREDICTION SERVICES
IMPLEMENTATION SERVICES
MAINTENANCE PLANS
RELEX UNIVERSITY
CUSTOMER SUPPORT
CUSTOMER TESTIMONIALS
RELEX CUSTOMERS
AEROSPACE
AUTOMOTIVE
DEFENSE
DIVERSIFIED
ELECTRONICS
INFORMATION TECHNOLOGY
MEDICAL DEVICES
OIL & GAS
RAILROAD SIGNAL
TELECOMMUNICATIONS
CASE STUDIES
NEWS & EVENTS
RELIABILITY eFLASH
QUARTERFLASH
PRESS RELEASES
TRADE SHOWS
SEMINARS
WEBINARS
WHAT'S NEW IN STUDIO
RELIABILITY RESOURCES
IMPORTANCE OF RELIABILITY
RELIABILITY 101
RELIABILITY DICTIONARY
RELIABILITY ARTICLES
RELATED WEB SITES
RECOMMENDED BOOKS
FAULT TREE/EVENT TREE/PRA
FMEA/FMECA
FRACAS CORRECTIVE ACTION
LIFE CYCLE COST
MAINTAINABILITY PREDICTION
MARKOV
OPTIMIZATION AND SIMULATION
RELIABILITY BLOCK DIAGRAM
RELIABILITY PREDICTION
WEIBULL
CONTACT US | REQUEST INFO
CUSTOMER CENTRAL
ID:  PASSWORD:
ABOUT US PRODUCTS SERVICES OUR CLIENTS WHAT'S NEW? RESOURCES DEMO
SEARCH 
ABOUT US
THE COMPANY
PROVEN EXCELLENCE
GUIDING MISSION
GUARANTEE
OF COMMITMENT
OUR QUALITY
MANAGEMENT SYSTEM
RELEX WORLDWIDE
CAREERS
DIRECTIONS
CONTACT US
PRODUCTS
RELEX RELIABILITY
STUDIO
FAULT TREE/EVENT TREE
FMEA/FMECA
FRACAS
CORRECTIVE ACTION
HUMAN FACTORS
RISK ANALYSIS
LIFE CYCLE COST
MAINTAINABILITY
PREDICTION
MARKOV
OPTIMIZATION
AND SIMULATION
RELIABILITY
BLOCK DIAGRAM
RELIABILITY PREDICTION
WEIBULL
ENTERPRISE EDITION
DEMONSTRATIONS
LITERATURE
WHAT'S NEW IN
STUDIO
SYSTEM ARCHITECTURE
CUSTOMER
TESTIMONIALS
PROFESSIONAL SERVICES
RELIABILITY CONSULTING
MTBF PREDICTION SERVICES
IMPLEMENTATION
SERVICES
MAINTENANCE PLANS
RELEX UNIVERSITY
CUSTOMER SUPPORT
CUSTOMER TESTIMONIALS
OUR CUSTOMERS
RELEX CUSTOMERS
AEROSPACE
AUTOMOTIVE
DEFENSE
DIVERSIFIED
ELECTRONICS
INFORMATION
TECHNOLOGY
MEDICAL DEVICES
OIL & GAS
RAILROAD SIGNAL
TELECOMMUNICATIONS
CASE STUDIES
WHAT'S NEW?
NEWS & EVENTS
RELIABILITY eFLASH
QUARTERFLASH
PRESS RELEASES
TRADE SHOWS
SEMINARS
WEBINARS
WHAT'S NEW IN
STUDIO
RESOURCES
RELIABILITY
RESOURCES
IMPORTANCE OF
RELIABILITY
RELIABILITY 101
FAULT TREE/
EVENT TREE/PRA
FMEA/FMECA
FRACAS
LIFE CYCLE COST
MAINTAINABILITY
PREDICTION
MARKOV
OPTIMIZATION
AND SIMULATION
RELIABILITY
BLOCK DIAGRAM
RELIABILITY PREDICTION
WEIBULL
RELIABILITY DICTIONARY
RELIABILITY ARTICLES
RELATED WEB SITES
RECOMMENDED BOOKS
SUPPORT
ONLINE SUPPORT
CUSTOMER CENTRAL
LOBBY
MY RELEX
ONLINE CUSTOMER SUPPORT
RELEX UNIVERSITY ONLINE
STUDIO DOWNLOAD
SERVICE PACKS
PARTS LIBRARIES UPDATES
HELP & DOCUMENTATION UPDATES
RETAIN SUPPORT SESSIONS
KNOWLEDGE BASE AND TIPS FROM TECH SUPPORT
SUGGESTION BOX
LOG OUT
SEARCH
FILE NOT FOUND
COPYRIGHT AND DISCLAIMER
PRIVACY
WEB SITE FEEDBACK
SITE MAP
Calculating Junction Temperature Variations

Different Temperature Calculation Methods Used in Predicting Component Reliability

When designing a system, many different factors affect which parts to choose. The goal is to select the parts that offer the highest reliability under the set of conditions in which they will operate. The conditions to be considered include stress, environmental, and load factors as well as the temperatures under which the system and its components operate.

Because operating temperatures are critical to part selection, reliability prediction models must provide several different methods for calculating temperature variations. The Relex Reliability Prediction module supports all of the most commonly used prediction models and the temperature variation calculations that these models offer.

To demonstrate some of the different ways in which the calculation of temperature variations are handled, this document describes the methods for modeling the temperature rise or junction temperature of a hardware component using the PRISM reliability prediction methodology, which is a newer reliability prediction model developed by the Reliability Analysis Center (RAC).1

Using the RACRates component reliability prediction models in the PRISM methodology, you can predict the reliability of the following hardware components:

  • Integrated Circuits
  • Semiconductors (Diodes, Thyristors, and Transistors)
  • Resistors
  • Capacitors

In PRISM, temperature rise is taken into account when modeling integrated circuits, semiconductors, and resistors. Capacitors only use overall system, or ambient, temperature. This document will detail the different methodologies used for assessing junction temperatures for integrated circuits, semiconductor, and resistors when using PRISM.

Temperature Calculation Overview

Sometimes the actual temperature rise or junction temperature of a component is known from testing. Other times, it can be determined from certain factors that are known about the component or supplied by the vendor of the component. In cases where little or nothing is known about a component, a standardized temperature rise for that component type might have to be used. This makes knowing how to model the temperature rise of the component just as important as knowing the temperature rise itself. Using an incorrect method of determining the junction temperature can lead not only to an incorrect junction temperature but ultimately to an incorrect reliability analysis of the system!

PRISM RACRates models allow for the use of a variety of different methods for calculating the junction temperature of a component. By providing several methods, which vary for each component category, PRISM offers the reliability analyst a full arsenal of junction temperature calculation methods. Based on knowledge about the temperature factors available for a component, the reliability analyst can select the appropriate temperature calculation method for that part.

The next several sections of this document describe the different methods for calculating junction temperatures for each RACRates component category.

Integrated Circuits

PRISM provides four different methods of determining junction temperature for integrated circuits:

  • Ambient Temperature + Default Temperature Rise
  • Ambient Temperature + Actual Temperature Rise
  • Ambient Temperature + Thermal Resistance (Theta-JA) * P
  • Case Temperature + Thermal Resistance (Theta-JC) * P

Ambient Temperature + Default Temperature Rise

This method is appropriate to use when no specific information on the component is available. The default temperature rise is a RAC-supplied standard temperature rise for the selected part type. For example, a non-hermetically sealed digital integrated circuit has a default temperature rise of 13 degrees Celsius.

Ambient Temperature + Actual Temperature Rise

If the part has been tested and the temperature rise is known, this method can be used. Because the temperature rise is already determined, there is no need to enter any other information.

Ambient Temperature + Thermal Resistance (Theta-JA) * P
Case Temperature + Thermal Resistance (Theta-JC) * P

If the part has not been tested but the thermal resistance and the operating power of the component is known, the temperature rise can be approximated. Choosing the calculation method to use depends on the type of thermal resistance. Theta-JA, or the "junction-to-ambient," is a measurement of the temperature difference between the component and the ambient atmosphere when 1 watt of power has been consumed by the component. Theta-JC, or "junction-to-case," is similar except that it is a measurement of the temperature difference between the component and the component's case or packaging.

The choice between ambient and case is determined by whether or not the component can have a different temperature than the board to which it is attached. If the component is cooled in some way (i.e., forced air, heat sink, or liquid cooling), then junction-to-case should be used. When using junction-to-case, however, the actual case temperature, thermal resistance, and operating power must all be known. The case temperature is a separate temperature measurement from the actual ambient or system temperature and must be kept independent of the system temperature.

Many companies provide information on thermal resistance and operating power for their components. An accurate temperature for a component can be determined when this value is multiplied by the operating power and then added to the case or system temperature.

Semiconductors (Diodes, Thyristors, and Transistors)

PRISM provides five different methods for determining the part temperature of semiconductors:

  • Ambient Temperature + Default Temperature Rise
  • Ambient Temperature + Actual Temperature Rise
  • Ambient Temperature + Thermal Resistance (Theta-JA) * P
  • Case Temperature + Thermal Resistance (Theta-JC) * P
  • Ambient Temperature + Delta-T * Stress

The first four temperature calculation methods are identical to the ones described for integrated circuits. The fifth method provides for modeling the temperature in a semiconductor using the electrical stress and temperature difference of the component as factors. This method can be used when the temperature differential between the component and its environment is known, taking into account such factors as the closeness of other heated components and heat sinks. The temperature of the part becomes the function of the difference in the highest and lowest temperature areas and the stress it undergoes. The stress is considered to be the ratio of rated and applied forward current.

Resistors

PRISM provides five different methods for determining the part temperature of resistors:

  • Ambient Temperature + Default Temperature Rise
  • Ambient Temperature + Actual Temperature Rise
  • Ambient Temperature + Thermal Resistance (Theta-EA) * P
  • Case Temperature + Thermal Resistance (Theta-EC) * P
  • Ambient Temperature + Delta-T * Stress

The five temperature calculation methods for resistors are similar to the ones described for semiconductors. However, for resistors, thermal resistance is a measurement of element-to-ambient and element-to-case rather than junction. In the fifth calculation above, the stress factor is power and not electrical stress. The stress is the ratio of the rated and actual power dissipation of the component.

Conclusion

When designing a system, you may not always have knowledge in advance about the junction temperatures of each component. Therefore, it is important that reliability prediction models allow flexibility in how junction temperature is calculated. All of the many reliability prediction models supported by Relex provide this flexibility. When appropriate, they include different temperature calculation methods. Based on the information known about a component, the reliability analyst can select the best method to calculate its junction temperature.

If you would like to receive more information about how the temperature calculation methods for different predictive modeling methodologies have been incorporated into Relex Reliability Prediction, please e-mail info@relex.com.

1The Reliability Analysis Center (RAC) is a Department of Defense (DoD) chartered Information Analysis Center (IAC) sponsored by the Defense Technical Information Center (DTIC), the central facility for the collection and dissemination of scientific and technical information for the U.S. Department of Defense.

© Copyright 2008 Relex Software Corporation REQUEST INFO | PRIVACY | FEEDBACK | SITEMAP
© Copyright 2008 Relex Software Corporation