For automotive design engineers, every emerging feature, functionality and architecture introduces a variety of unique challenges, especially when these advancements are often occurring simultaneously. For instance, EVs and vehicles equipped with advanced driver assistance systems (ADAS) have significantly more complex electrical systems and sensors than conventional internal combustion engine (ICE) vehicles, and they often require new approaches to thermal management, a greater focus on battery safety and consideration of functional safety for features like wire steering and electric braking. Challenges such as these will only increase with emerging functionality, including bidirectional charging in EVs, the incorporation of more interactive infotainment systems, the continual evolution of ADAS and the eventual evolution of full self-driving (FSD).  

As the complexity of vehicle electrical systems increases, the performance requirements are placing added stress on the components used. For example, some connectors within EVs may experience near continuous operation not only when the vehicle is driving but also when charging and must be designed to support these additional operational conditions. 

The vehicles of today and tomorrow demand component manufacturers like Molex to be more cognizant than ever in ensuring parts perform to standards and regulatory requirements as well as operate reliably in the field over the life of the vehicle. Designing for Reliability (DFR) requires a fundamental shift that calls for reevaluating traditional test methods and incorporating innovative reliability prediction models that can be leveraged by artificial intelligence (AI) and machine learning (ML) to better optimize designs for performance in real-world conditions — down to the component level. 

Although vehicle functionality is becoming increasingly complicated, many automakers are working to simplify the components within and seeking single solutions that can be applied in a variety of locations with different stress conditions or duty cycles. For Molex, that means designing our connectors to withstand greater ranges of heat, vibration, ingress, corrosion and other variables. But to do so, we need to identify the design strength for reliable operation and switch the traditional test-to-pass model to test-to-failure.

While test-to-pass is the historical norm, it only tells us if we pass or fail test criteria — it doesn’t measure how far off we are when failing, nor the safety factor if passing. By contrast, test-to-failure determines the safety margins, or the difference between the design’s limit, such as product strength and the specification acceptance criteria for performance.  

This approach isn't unique to the transportation industry and reflects the challenge engineers face when accounting for more complex, feature-rich expectations. In the recent Molex Reliability and Hardware Design Survey of more than 750 design engineers and system architects, 86% of respondents revealed that they design new products to either surpass current requirements or aim to meet possible future requirements in addition to current requirements. 

How is Molex driving this transition to real-world reliability testing? 

Accelerated Life Testing (ALT) is a widely used method across industries to simulate the field life of a product. This is achieved by exposing the product to extreme environmental and use conditions over a shortened period, and then determining whether it still meets specifications. These exposures could be to stresses that exceed everyday use cases, such as very high vibration level or temperatures, or accelerated usage rates such as repeatedly connecting and disconnecting a connector. However, it is important to note that ALT is not a perfect test strategy and can lead to under or overdesign if not used appropriately.

Test-to-failure generates a more accurate understanding of product strength relative to field stress by straining the item to its point of failure. The test-to-failure data can also be used to develop the acceleration factor for the ALT. By leveraging test-to-failure methodologies, designers can optimize a product without overdesign while still ensuring performance and reliability requirements are met.

Molex uses the test-to-failure method to better understand how well our products will withstand real-world environments, improve current and future product designs and provide confidence to our customers. But we’re not just measuring the performance of a physical product or prototype — we’re predicting the reliability throughout the product design cycle. 

Predictive engineering and digital twins have long had a place in automotive design, but Molex is now applying these same methodologies to the component-level using data captured through extensive product testing, such as test-to-failure, and informed by POF models. 

Fundamentally, this provides customers and Molex alike with several major benefits, including:

1. Products are proven to be capable of withstanding the demanding conditions — with the supporting data as evidence.
2. The prototyping and testing stage can be simpler, more cost-effective, collaborative and even more experimental, enabling improvement to virtual prototypes prior to physical prototyping.

Better yet, the AI and ML models supporting these methodologies are continuously trained on the latest product test results, making them increasingly more capable and accurate over time. Modeling such as this enables more rugged, miniaturized and reliable components that can be used across a broader range of transportation requirements.  

As a leader in interconnect solutions for the transportation industry, Molex combines an extensive product portfolio with a broad range of interdisciplinary engineering expertise, from design to prototyping and testing. Our thorough history in test-to-failure, ALT and other design for reliability methodologies have laid the groundwork for our innovative, predictive approach and it provides our customers with the data needed for more informed decision making and greater confidence in the long-term reliability of our products. 

Learn more about the results of our Reliability and Hardware Design Survey here.