EasyNetWorld

There are many reasons for the failure of semiconductor components, which can be mainly attributed to the following aspects

semiconductor components

1. Design of semiconductor components: At advanced feature size nodes, chip aging is an increasingly serious problem. However, so far, most design teams have no need to deal with it. With the introduction of new reliability requirements in the automotive and other markets, a comprehensive analysis of the factors affecting aging is required, which will undergo significant changes. People usually know that semiconductor devices gradually age over time, but they are unaware of the aging mechanism or the limiting factors that lead to chip failure. In addition, there are certain requirements for the shortest lifespan of devices based on different applications. For consumer devices, it may take 2 or 3 years, and for telecommunications devices, it may take up to 10 years. Due to the complex and often difficult to fully predict aging processes, many chip designs nowadays often adopt redundant design methods to ensure sufficient margin to meet the requirements of reliable lifespan operation.

Taking operational amplifiers as an example, they are the foundation of many things.semiconductor testing The operational amplifier must be correctly biased and have some margin in the overdrive voltage. Then you must ensure that sufficient margin is left, so that over time, the aging of the operational amplifier will remain within the saturation region of the transistor. The overdrive margin of transistors is decreasing because the 7nm power supply voltage is 750mV and the threshold is about 350mV, so there is almost no space to retain a large margin. With aging, the threshold voltage can shift up to 50mV. If the bias circuit of the operational amplifier is offset by 50mV, it may change from a saturated region to a linear region or a transistor region, and the transistor will become a resistor without gain. The function of an operational amplifier is to provide gain, and at that time the circuit becomes useless.

Aging and reliability are challenges faced by simulation designers. Today's designs may not run tomorrow as they may undergo degradation, and the most important thing at present is to ensure that they meet all the aging and reliability requirements of the market.

2. Manufacturing of semiconductor devices: The manufacturing of semiconductor devices involves measuring structures that are only a few nanometers in size. As a reference, the diameter of human DNA strands is 2.5nm, while the diameter of human hair is 80000 to 100000nm.semiconductor failure analysis A particle of dust can destroy several bare chips on a wafer. If the size of the bare film increases, the likelihood of random failure will increase. For mature process nodes, the yield may range from 80% to 90%. However, for newer nodes, the yield may be significantly lower than 50%, although the actual numbers are strictly confidential.

Even if the bare film is not affected by catastrophic events, it cannot be considered operable. The manufacturing steps are incomplete, and even a single atomic process change can result in significant differences. Although this may not have an impact on certain parts of the design, if process changes happen to coincide with critical timing paths, it may result in the device not meeting specifications.

As design gradually evolves to adopt advanced packaging deep submicron technology, existing simulation tools and design methods cannot reflect changes and their impact on reliability well. This can lead to loopholes in the design process, leading to some failures. Design processes increasingly allow for consideration of changes early in development to minimize their impact, while design techniques such as redundancy can reduce the number of "almost functional" chips that need to be discarded.

3. ESD protection: Typically, chips will include ESD protection. If a voltage of 0.5V is applied externally to the chip, an electric field of 0.5mV/m is generated on a 1nm medium. This is enough to cause a high-voltage arc. For individual bare chips within the package, their goal is to achieve a standard of 2kJ. If you try to minimize ESD and even eliminate it on these Wide I/O interfaces or any type of multi chip interface channel, aotomatic prober it means that you cannot truly test each chip according to the same standards you have for a single chip. They must undergo more professional testing because their ESD protection is minimal or may not be available.

Even during operation, ESD events can cause problems. In portable electronic products, ESD can cause many types of soft errors. During ESD events, noise may be caused on the power supply network (PDN) due to the sensitivity of certain ICs (oscillator ICs, CPUs, and other ICs), or the field coupling of the PDN.

4. The impact of magnetic fields on semiconductors: With the multifunctional use of smartphones and tablet terminals, the required power voltage also involves various specifications, so the number of inductors used in power circuits is showing an increasing trend. The requirement for using integrated inductors in power circuits is to have a small size and support high currents, and to have low losses in some devices that use batteries such as smartphones.

Inductors store energy in a magnetic field to perform their functions. However, inductance is not only affected by the electromagnetic energy generated by itself, but also by external magnetic flux. The inductance value of the guaranteed component refers to the value in the state of no external magnetic flux. Therefore, when packaging an inductor in the presence of external magnetic flux, it may not be able to perform its intended function.

Therefore, EMS is a new issue that people have to worry about. The energy injection test starts at 150kHz and injects 1W of energy until 1GHz. At each frequency