SRC Forum - Message Replies

Forum: Reliability & Maintainability Questions and Answers

Topic: Reliability & Maintainability Questions and Answers

Topic Posted by: Reliability & Maintainability Forum ( )
Organization: System Reliability Center
Date Posted: Mon Aug 31 12:47:36 US/Eastern 1998

Back to message list Topics List About this forum
Original Message:

Posted by: Richard Smith ( )
Date posted: Mon Jan 7 15:00:26 US/Eastern 2002
Subject: Reliability Testing Comparison (RDGT/HALT/HASS/PRAT etc.)
Does anybody know of a website or reference that discusses or compares the benefits or disadvantages of the following types of reliability testing: - Reliability Development Growth Testing (RDGT) - HALT/HASS (Highly Accelerated Life Test/Highly Accelerated Stress Screening) - Production Reliability Acceptance Testing (PRAT) - etc. What I am looking for is something that would help decide when one method would be more advantageous or cost effective over another, and would possibly yield meaningful results.


Subject: Re: Reliability Testing Comparison (RDGT/HALT/HASS/PRAT etc.)
Reply Posted by: Gary Sunada ( )
Organization: Reliability Analysis Center
Date Posted: Mon Jan 14 16:23:08 US/Eastern 2002
For testing and screening methodologies such as Reliability Development Growth Testing (RDGT), Production Reliability Acceptance Testing (PRAT), Highly Accelerated Life Test (HALT), and Highly Accelerated Stress Screening (HASS), most of their relevance depends on the stage of the productís lifecycle and really cannot compare to each other.
RDGT is also known as Test, Analyze And Fix (TAAF), is usually done development and preproduction units to identify failures and implement corrective actions before the product is finalized. The cost of design change is minimized at this point and results in improved production reliability and customer satisfaction.
PRAT is used after production starts as a quality check on manufacturing processes and vendor controls, determining if the production equipment continues to perform to specifications and maintains its reliability requirements. The cost of design change at this point is quite high, due to re-design and retrofitting costs.
For PRAT and RDGT, MIL-STD-781 revision D and MIL-HDBK-781 revision A provides an extensive discussion on these two.
HALT (for use during development or supplier / part changes) and HASS (during production), are relatively new methodologies that compete with the more traditional RDGT and PRAT, respectively. They differ because of their use of extremely high-level thermal, humidity, shock and vibration stresses (often beyond the given specifications of the product) in order to ferret out latent defects that may not otherwise be found by conventional testing. Testing is relatively expensive as higher stresses are required, but could be offset due to shorter length of the test (calendar time). Again, the costs of change are minimized during development (HALT) but are significantly higher during production (HASS).
HALT and HASS are performed by a number of facilities that are familiar with the underlying philosophy involved, and could inform you more effectively on your specific requirements. RACís website lists various testing laboratories which might be of help at


Subject: HALT and HASS
Reply Posted by: Kirk Gray ( )
Organization: AcceleRel Engineering
Date Posted: Fri Feb 8 16:00:11 US/Eastern 2002
Gary is correct about the process of HALT and HASS. But It is a major change in perspective and philosophy of reliability testing of electronic hardware. It does save cost overall in its ability to rapidly illuminate flaws and weakness. These defects are the main cause of failures of electronics. For most applications, the inherent "life" of defect free electronics far exceeds the technological usefullness lifetimes by decades. The most efficient testing occurs when you are testing based on material strengths, not trying to quantify "lifetimes". Electronics are inherently robust and the main reason there are field failures is because of special cause defects in design, manufacture, and not because they are "worn-out".

Back to message list Topics List About this forum
Reply to this message