Ensuring PCB reliability once a device is deployed in the field is not something to be taken lightly. In the manufacturing arena, producers take great pains to ensure high yield and reliability by following best-known practices and industry standards, as well as thoroughly qualifying their own processing capabilities. What about reliability once a device is deployed in the field?
Determining the level of reliability at the PCB level and for an entire assembly requires direct testing in the end environment, as well as simulations. Some of these simulations are simple enough to run in SPICE simulation packages inside your PCB design software, while others are detailed thermal and mechanical simulations requiring a 3D field solver. In this article, we’ll take a similar approach as we did in our SI/PI simulation and analysis cheat sheet. Keep reading to learn about some of the important tests and simulations you should perform to ensure PCB reliability.
PCB Reliability Simulations
Reliability assessments for a bare board or a PCBA usually begin with simulations in order to identify failure modes before prototyping. Sometimes, the primary failure modes will not be obvious, but a simulation can help you quickly identify a failure more or predict stresses that one would expect to cause failure. Required design changes that can prevent these failure modes can then become part of your team’s DFM checklist for the product.
Thermal and Mechanical Analysis
Mechanical and thermal failure mechanisms need to be determined with 3D field solver simulation tools. Finite element analysis (FEA) can be used to determine how heat is spread throughout the PCBA, how different regions of the PCBA react to mechanical stresses, and how heat produces mechanical stress during operation. These tools can be complex to use and configure, but they are needed to examine complex mechanical behavior that can lead to failure.
The three main points to analyze in mechanical simulations include:
- Vibration simulations: An FEA simulator is used to identify resonant modes that can produce strong vibrations when excited in the operating environment. This can lead to fatigue failure of components, solder joints, or conductors in the PCB.
- Mechanical shock: A mechanical shock test simulates a very large impulse imparted to the board, followed by an analysis of where the greatest strain reaction occurs throughout the PCBA.
- Warping or bending: This is less commonly performed than shock or vibration simulations, but it is still an important area of examination, including in flex boards that cannot be over-stressed or deformed.
An example drop test is shown below. The idea in this type of simulation is to evaluate where in the PCBA we would expect stress to be highly concentrated during the initial impulse and subsequent shock spreading throughout the PCBA.
Thermal simulations are often performed to ensure areas near components do not rise to excessively high temperatures during operation. However, one important part of high reliability electronics design is to ensure an electronic assembly can withstand thermally-induced stress due to thermal expansion and contraction. We’ve discussed this before in terms of microvias passing through reflow; the same idea applies to stressed solder balls on BGAs and solder joints generally. The ability to identify these stressed areas quickly can help to prevent device failure in the field.
Circuit Simulations
When we talk about simulations, we probably don’t think too often about circuit simulations in a SPICE package. However, some simple facts about components can create reliability problems that cannot be easily solved once a PCBA has been deployed in the end device. The issue is with component tolerances, which produce variations in electrical behavior and can affect circuit operation if tolerances are excessive.
There are three simulations that can be used to identify if component tolerances are excessive and whether they greatly impact performance:
- Monte Carlo simulations, which randomly generates component values and runs repeated SPICE simulations.
- Sensitivity analysis, where component values are varied within a specific range and specific measurements are checked for each component value.
- Worst-case analysis, where SPICE simulations are run given extreme values of components in order to determine an upper or lower limit on a performance metric.
- Time-domain and frequency domain simulations specifically focusing on unexpected stimuli and evaluating component deratings. ESD tests can fall into this category.
These same simulations can be used to vary other quantities in the system, such as the input power or other voltages throughout the design. The ultimate goal is to identify when a specific performance metric is violated due to random variations in component values, or due to instabilities in power, temperature, voltage references, or any contributor to a circuit’s electrical behavior. Once the greatest contributors to variations in circuit behavior are identified, they can be modified with the goal of making the system more stable.
PCB Reliability Testing
Reliability testing focuses on correlating measurements with simulation results, as well as evaluating device reliability against industry standards as required. There is another aspect of reliability testing that is not easily simulated: determination of device lifetime and early failures.
Accelerated Life Tests
There are tests that are used to examine long-term reliability of individual components, sub-assemblies, and PCBAs: these are accelerated life tests. One important side-effect of accelerated life tests with a batch of PCBAs is that it allows defective products to be identified before they are put into the field. Once these are identified, the root cause of failure can be investigated in the lab, and design alterations can be proposed based on these investigations.
Accelerated life tests are often referred to collectively as “burn-in testing”. In fact, there are multiple types of accelerated life tests:
- Highly accelerated life testing (HALT): This type of stress test pushes a device until failure by mimicking over-stressed operation in the end product’s intended environmental conditions.
- Highly accelerated stress testing (HAST): This type of test is similar to HALT, but it is intended to determine the device’s stress limits that result in total failure, rather than to estimate the device’s lifetime.
- Highly accelerated stress screening (HASS): This type of test uses the same environmental stresses as are used in HAST, but this is typically performed after a complete HALT test is completed for a PCBA.
- Burn-in testing: This set of tests is used to identify the minimum required environmental stress that will weed out early failures from a batch of electronics assemblies. The remaining boards will be found fit to deploy in the field.
- Environmental stress screening (ESS): Designed to mimic extreme conditions in the intended deployment environment in order to determine failure modes resulting from environmental exposure. This testing modality can be combined with other types of tests (e.g., functional).
Extreme environmental stresses that might lead to early failure include pressure testing, submersion testing, exposure to humidity, or other factors that might be encountered during operation. These tests become important benchmarks for simulations and failure analysis as they can be used to determine root failure causes before mass production.
Shock and Cycle Tests
Impulse responses in a PCBA should be tested against target industry standards and correlated with simulation results. Two important sets of tests used to evaluate mechanical and thermal ruggedness are shock and cycle tests. Shock tests follow the same idea as in simulations: a large thermal excursion or mechanical stress is applied very quickly, and the device is inspected for damage or failure. These tests could be applied while the device is in operation and while its electrical behavior is being monitored.
Cycle tests specifically refer to thermal or mechanical cycling, where a thermal excursion or mechanical stress is applied repeatedly. The device is then periodically checked for failure, consistent operation, or some critical performance metric. Thermal cycling tests are most common in products that are likely to experience repeated thermal excursions during operation. Typical failure modes from repeated thermal cycling include delamination, solder fatigue, via fatigue at the barrel or butt plating, or failure of components.
There are several mechanical and thermal testing standards that are used to qualify PCBs, including:
- IPC-9701A
- IPC-TM-650 2.6.7
- IPC/JEDEC-9703
- MIL-STD-202G
- MIL-STD-810G
- MIL-STD-883
ESD Testing
Electrostatic discharge (ESD) testing is part of a larger EMC testing series. Some products will require ESD testing to specific standards before being qualified for release to market. For example, the IEC 61000-4-2 standard describes testing against kV pulses from a generator with a specific output RC network that controls pulse behavior. Other IEC standards are available that target specific types of equipment (telecommunications, office appliances, etc.). More stringent standards used to prove high PCB reliability in aerospace and defense electronics are MIL-STD-461, MIL-STD-331D, and DO-160.
High-Reliability Electronics Design
While we don’t have enough room in this article to fully explore high-reliability design, it’s important to note that the term “high-reliability” means different things in different situations. It could mean highly precise signal acquisition and measurement in a very noisy environment, or it could refer to a very rugged assembly that can withstand the demands of mechanical and thermal shocks. Make sure you contract with a design firm that understands some of the basic high-reliability design practices for various deployment environments, as well as some important design and operational standards to be followed in high-reliability electronics.
Whether you’re designing an ultra-rugged aerospace system or feature-rich embedded computing products, following these PCB reliability design guidelines helps ensure your product will be reliable and manufacturable at scale. PCBLOOP helps consumer OEMs, industrial primes, and private companies in multiple industries design modern PCBs and create cutting-edge embedded technology, including power systems for high reliability applications and precision control systems. We’ve also partnered directly with EDA companies and advanced ITAR-compliant PCB manufacturers, and we’ll make sure your next high speed digital system is fully manufacturable at scale. Contact PCBLOOP for a consultation.
