Summary of PCB testing methods

advbj

Summary of PCB testing methods [Copy link]

Summary of PCB testing methods

As the carrier of various components and the hub of circuit signal transmission, PCB has become the most important and critical part of electronic information products and is widely used in all walks of life. In recent years, due to the increasing number of PCB failure cases and the great harm caused by partial failure, the "Equipment Manufacturing Industry, Standardization and Quality Improvement Plan" passed in April 2016 and the "Made in China 2025" adhere to the basic principles of "innovation-driven, quality first, green development, structural optimization, and talent-based" to form a connection, indicating that the country has higher and higher requirements for product quality and stricter control over product reliability. Therefore, the quality and reliability level of its PCB determine the quality and reliability of the entire equipment.

In the past, only military products would raise reliability requirements, but now many civilian products also raise various reliability requirements. Different types of customers and different product application fields have completely different reliability requirements for PCBs. For example, some customers require that the reliability of PCBs must still meet the PCB electrical performance requirements after being baked at 260 degrees Celsius for ten hours; some customers require that the product resistance change rate is less than 10% after IST cycles for 250 or even 1,000 times; some customers require that PCBA products meet the resonance requirements for 30 minutes under 25g acceleration, etc.

2. PCB failure cases

As we all know, product failure will cause serious economic losses and quality impacts. However, there are many modes of PCB failure, and the root causes of failure are also different, such as corrosion failure of PTH hole copper, open circuit failure caused by cracks at the bottom of HDI blind holes, delamination failure, ENIG product hole ring cracks and PCB board short circuit fire. The processing flow is cumbersome and complicated, and there may be many reasons for its failure. Therefore, how to quickly locate the root cause of PCB failure and optimize and improve the performance of the product has become one of the important topics in the PCB industry. The following are failure analysis cases related to big data computing, navigation and medical fields in recent years by the Analysis and Testing Laboratory Center:

① PCB burnout failure

In 2015, a large data center reported that one or two devices would catch fire and burn boards every month, and they needed to find the cause of the fire quickly:

Observation of the failure site showed that the burning point of the board was concentrated in the mounting hole area of the hard disk connector. The copper layer of the hole wall and the positioning tin column in the mounting hole area were completely burned. Analysis found that the NPTH hole wall was metallized, and CAF failure occurred during the power-on process, which eventually led to a short circuit and fire.

CAF growth principle : When PCBA works under certain temperature and humidity conditions and is energized, CAF may grow along the interface between the resin and the glass fiber between the two insulated conductors, eventually leading to poor insulation or even short circuit failure. The CAF generation principle is as follows:

There are three prerequisites for the generation of CAF : ① Certain temperature and humidity conditions; ② The potential difference between the two insulating conductors; ③ The presence of a channel for the migration of copper ions.

Experimental reproduction : The insulation resistance failure module of the experimental board was tested with a 12V power-on test. The test board short-circuited and caught fire the moment the power was turned on. If the power was kept on, the sample would continue to catch fire and carbonize (the test was recorded in video).

Improvement plan: ① Increase the distance between the inner and outer hole rings and the conductor to 300μm; ② Replace the high Tg substrate to ensure it has good heat resistance and CAF resistance.

② Chemical nickel-gold plate hole ring crack failure

In 2017, a multinational SMT assembly customer reported that after reflow soldering, the PTH holes of all the ENIG boards it purchased in the industry showed corrosion and blackening in the test hole rings. The probe test could not pass and the entire production line was shut down, requiring failure analysis.

In addition to affecting the contact performance of the probe, the crack defects of the PTH hole will also cause the copper layer to be corroded, reduce the corrosion resistance of the PCB, and also bring a greater risk of failure to the solderability of the product.

Mechanism of crack generation: Due to the principle of thermal expansion and contraction, the PCB board expands due to high temperature during reflow soldering and wave soldering. Since the selection of PCB board material does not match the surface treatment process, the board material will give the hole ring an upward stress, pushing the hole ring upward, causing the hole ring to deform to both sides, resulting in cracks in the hole ring.

Improvement plan: ① Replace the plate with a smaller CTE; ② Change the surface treatment process.

③ PTH hole electrochemical corrosion failure

In 2017, a customer of a certain model of 8-layer PCB board reported that the video signal was not working after the whole product was used at sea and in coastal areas for 3 years. There were black foreign objects left at the PTH hole of the abnormal signal network. It was necessary to find the root cause and locate the failure point.

Analysis of the PTH holes covered with foreign matter revealed that there were obvious cracks between the solder mask and the hole copper at the hole mouth of the plug hole, and the hole copper at the crack position had been corroded, with part of the hole copper missing.

At the same time, compared with the normal PTH holes on the PCB that are not covered by foreign matter, it is found that the solder resist in the normal holes is well bonded to the hole wall, there are no cracks, the hole copper is not corroded, and the signal conduction is good.

According to research, in the high-salt and high-humidity environment of the coast, a thin layer of water film will form on the surface of objects, and when the relative humidity is 65%-80% in the air, the thickness of the water film on the object is 0.001~0.1μm. When the PCB is working with electricity, the water film and the copper holes together form an electrolytic cell, an electrochemical reaction occurs, and finally black corrosion products are generated, as shown in the following figure:

Conclusion: When the solder mask plug hole is not full or there are cracks in the solder mask plug hole, the copper in the PTH hole that is not completely covered by the solder mask will form an electrolytic pool, an electrochemical reaction will occur, and the hole copper will be corroded, which will bring serious risks to the reliability of the PCB.

Improvement plan: Improve the fullness of the plug hole and improve the quality of the plug hole.

Of course, there are many types and modes of failure. The following are other typical PCB board-level failure analysis case pictures accumulated in the laboratory.

From the above cases, we can easily find that there are more and more PCB board-level failure modes, and the root causes of failure are also different. Therefore, it is necessary to summarize and refine the general failure analysis ideas and methods to form a set of methodologies that can be promoted and applied, so as to achieve twice the result with half the effort and quickly locate the root cause in the analysis of actual cases.

According to the introduction of the Chip Kaifeng Laboratory, it can implement testing work according to international, domestic and industry standards, carry out comprehensive testing work from the bottom chip to the actual product, from physical to logical, and provide chip pre-processing, side channel attack, light attack, invasive attack, environment, voltage glitch attack, electromagnetic injection, radiation injection, physical safety, logical safety, function, compatibility and multi-point laser injection and other security testing services. At the same time, it can carry out failure analysis testing services to simulate and reproduce the failure of smart products and find out the cause of failure, mainly including point needle workstation ( Probe Station), reactive ion etching (RIE), micro leakage detection system (EMMI), X-Ray detection, defect cutting observation system (FIB system) and other testing tests. It can realize the evaluation and analysis of the quality of smart products and provide quality assurance for the chips, embedded software and applications of smart equipment products.

3. PCB Failure Analysis Method

This method is mainly divided into three parts. Integrating the three parts can not only help us quickly solve failure problems and locate root causes during actual case analysis, but also train new engineers based on the framework we established, making it convenient for various departments to borrow and learn.

The following is an explanation of the analysis ideas and methods. First, the analysis ideas;

Step 1: The “Five Steps” of Failure Analysis

The failure analysis process is mainly divided into five steps: "① Collect information about defective boards → ② Confirm failure phenomena → ③ Analyze the causes of failure → ④ Verify the root causes of failure → ⑤ Report conclusions and suggestions for improvement". Among them, the first step is to understand the failure content, process flow, structural design, production status, usage status, storage status and other information of defective PCB boards, so as to prepare for the subsequent analysis process; the second step is to determine the failure location and judge the failure mode based on the failure information; the third step is to analyze the failure mode and check the root causes one by one according to the failure root cause fault tree. If the cause cannot be confirmed in the existing fault tree, it is necessary to study this type of failure problem through special project establishment and other methods, and add the research conclusions to the original fault tree, so that the fault tree is continuously enriched and improved, the root causes are exhausted, and an effective cycle mode of repeated iteration and upgrading is formed; then the reproducibility experiment is carried out through the fourth step to verify the root cause; finally, the failure analysis report is output, and a targeted improvement plan is given for the root cause of failure.

Step 2: Establishment of failure root cause fault tree

Taking the poor solderability of the gold surface of the chemical nickel-plated gold plate as an example, the method of establishing a failure analysis fault tree is explained:

In response to the common board-level failure phenomena of PCB/PCBA, we have established root cause fault trees for various failure modes, and continuously accumulated, refined, and updated them in actual combat, iteratively improving in depth and breadth, thus forming a relatively complete root cause fault tree analysis process for high-frequency failure modes such as delamination and blistering, poor solderability, poor bonding, poor conduction, and poor insulation. This can help everyone follow the failure analysis process of the fault tree in subsequent actual combat, quickly locate the root cause of failure, solve the problem, and get twice the result with half the effort.

Step 3: Create a standard library

By verifying the root causes of the fault tree, a standard library file is formed. The sources of the root cause determination standard library mainly include the following aspects: ① IPC, GJB, industry standards and other documents; ② normal product and abnormal product comparison library; ③ R&D project experience, production experience file library, etc. At the same time, the evaluation methods and evaluation criteria involved in each failure root cause in the fault tree are summarized and classified, and the common PCB standards and various abnormal data are summarized and sorted to form a PCB failure analysis standard library for reference in subsequent cases.

Nietzsche said: The world is most beautiful when viewed from a height. I believe that only by reaching the top can we live up to the support and expectations of the company's leaders and colleagues for the Analysis and Testing Laboratory Center. A journey of a thousand miles begins with a single step! In the future, the Analysis and Testing Laboratory Center will remain true to its original aspiration, forge ahead, and create brilliance together with the company.