Considerations for Test Plan Development

Posted by Greg Caswell on Apr 2, 2019 4:55:58 PM

In a previous DfR Solutions insight titled Best Practices in Test Plan Preparation, we discussed some of the most important techniques and philosophies when preparing to develop a testing plan for electronic products. What makes those techniques so powerful is that they are ubiquitous: with any design, reviewing the bill of materials, identifying use environments and assessing failure history are both applicable and crucial.

However, what that article did not discuss is that there are considerations that need to be applied in very specific ways. The following are strategies for test plan development that are dependent on specific use cases, parameters, goals, configurations and limitations. While they are just as powerful as our Best Practices, they require a thorough understanding of your product and a clear and agreed-upon set of goals throughout the supply chain.

Determining the Acceleration Factor

There are many ways to go about determining the acceleration factor for a test and selecting the acceleration factor depends on which failure mechanisms a product is most likely to see. Is it going to be solder joint fatigue? Is it going to be some form of corrosion that might occur due to exposure to different types of gases or salt or things of that nature? It is also important to consider the activation energy for the expected failure mechanism.

As the above table demonstrates, there are different activation energies for different materials. Gold is on the low end and it's got an activation energy of 0.7 to 0.9. Aluminum copper silicon composition is 0.25 to 0.86. Basic aluminum 0.35 to 0.60. These all vary and when plugged into different formulas, and you can see how they affect the actual calculation of the acceleration factor that's needed for subsequent calculations.
There are many ways to go about determining the acceleration factor for a test and selecting the acceleration factor depends on which failure mechanisms a product is most likely to see. Is it going to be solder joint fatigue? Is it going to be some form of corrosion that might occur due to exposure to different types of gases or salt or things of that nature? It is also important to consider the activation energy for the expected failure mechanism.

Different Ways of Determining Acceleration Factor 

Arrhenius equation: This is widely used for chemical reactions and is limited to temperature effects.

Eyring equation: The Eyring equation accounts for multiple stresses with synergy to temperature. It is recommended by the IPC SIR Handbook to determine acceleration factors for ECM.

Peck Equation: Peck's equation has classically been utilized as the tool to determine the acceleration factor if you're doing a temperature humidity bias type test. The Peck equation is that it was developed as a way of assessing the galvanic corrosion of aluminum bond pads in a microcircuit. People have used it for calculating the acceleration factor for a typical humidity test regardless of the failure mechanism, which makes the result that you get somewhat inaccurate.

Each of these equations have their inaccuracies; however, they serve as a good starting point.

Define the Reliability Metrics for the Product

So you've now got your acceleration factor figured out. What else can you do to come up with the proper test duration to determine the reliability of your product? You need to know some other information, such as what reliability metric you’re looking for in a product.

For example, will you measure reliability by percentage? If so, what’s a realistic number – 90%? 95%? Below are a handful of metrics by which you can set your reliability goals:

  • Confidence in Test Percentage
  • Life Expectancy of the Product in Years
  • Sample Size available for Tests – smaller sample sizes affect other variables such as test duration
  • Utilize the Beta 3 or what has been calculated based upon field history & the acceleration factor

Once Reliability Metrics are identified, that allows for the establishment of test recommendations regarding cycling, temperature-humidity-bias, vibration, shock and so on.

Understand Limitations

Of course every organization would like to have the most thorough testing possible. But many organizations have constraints put on them by their customers, and these need to be considered when mapping out a test plan. For example:

  • Sample Size
  • Test Duration
  • Chamber Capabilities
  • Monitoring Requirements

Some testing protocols may have standards based to fit customers requirements, and this has potential to greatly influence test plans. If a customer simply has a small sample size of parts to test, then test duration is a variable that can be increased to compensate for the small sample size.

A Foundation in Physics of Failure (PoF)

Physics of Failure is the use of science (physics, chemistry, etc.) to capture an understanding of failure mechanisms and evaluate useful life under actual operating conditions. At DfR Solutions, we believe a reliance on Physics of Failure generates the most accurate and efficient failure analysis testing. Incorporating Physics of Failure into your test plan means factoring it into all aspects of development, starting at the design stage and continuing through the lifecycle of the product.

Establishing testing protocols based on industry specifications is a good place to start. But Physics of Failure allows for the modification of those specifications in order to tailor test strategies specifically for the individual design and materials as well as the use environment and the reliability needs of the product.

Desire Lifetime: IC, Capacitor and Solder Wearout

IC Wearout

Since 2015, manufacturers are down to eight nanometers of feature sizes for transistors. The curve in the graphic above shows a reduction from where there might have been a life expectancy or mean service life somewhere close to 100 years for integrated circuits in the 1990s. But because manufacturers keep reducing gate geometries, there is now a scenario where there is a potential end of life characteristic.

Nowadays, we might have a five-year life expectancy with some of the 8 and 14 nanometer features that are currently being used. If you've got a requirement to last longer than that in your application, you may not be able to use some of the latest and greatest ICs that have been developed due to their increased sensitivity.

Capacitor Wearout

When choosing what features you would like to include in a test plan, you must consider the types of multilayered ceramic chip capacitors being utilized. Ceramic chip capacitors have gotten smaller and smaller in size. Meanwhile, we’re seeing higher capacitance parts at four volts, meaning the capacitance-to-voltage ratio is getting higher. As a result, a design must include significant erating or it is possible the product will experience failure in a very short period of time.

Solder Wearout

With the evolution of our industry has come a reduction in the size of integrated circuit packaging. We've gone from quad flat packages where a test could easily do more than 10,000 cycles of -40 to 125 degrees Celsius, to BGAs, QFNs and ultimately to chip scale packaging and flip chip. With each reduction in size, the cycles to failure has come down.

Looking at the packaging of your parts is a critical piece of being able to assess what you can put together in a test plan. Wearout failure mechanisms can be chemical, mechanical, electrical, overstressed, etc., meaning your list of test plan considerations needs to be exhaustive.

Is your testing plan ready to go?

While all the factors above are critical to take into consideration when developing a testing plan, the reality is that they only scratch the surface. Test factors extend down to a granular level if the test itself is to be robust, accurate and efficient.

This is why so many organizations turn to DfR Solutions for testing consultation – we are experts in reliability testing, so you don’t have to be. DfR Solutions performs hundreds of electronics reliability tests each year at our facilities, with the goal of providing our clients the most insightful testing so that they can achieve a quicker time-to-market and create the most dependable products in the industry.

To learn more about effective test preparation techniques, download our white paper on 6 Steps to Successful Board Level Reliability Testing

If you would like a consultation on your testing procedures, contact DfR Solutions for a quote today.
Contact Us

board close up


Board Level Reliability Testing: Current Challenges

board close up



Best Practices


Best Practices in Test Plan Development


White paper:

6 Steps to successful board level reliability testing

Topics: Electronics Reliability, electronics test design, Mechanical Design, Reliability Physics, Standards Based Testing

Sign me up for updates and offers from ANSYS, including DfR Solutions, and our partners. I can unsubscribe at any time.