How difficult is it to make a car chip? Can computer chips be used in cars?

Li Shufu used to say that building a car is nothing more than a sofa with four wheels; Now, Apple says that building a car is a smart phone with four wheels.

In fact, both views are correct, but the times are different.

In the era when Li Shufu began to build cars, cars were mainly hardware, and the intelligence of the whole car was very low. Semiconductors (chips, sensors, etc. ) accounts for less than 65,438+0% of the vehicle cost. Now, the automobile has changed from a hardware-led era to an era based on hardware and software enablement. Today's semiconductors (chips, sensors, etc. ) has accounted for about 35% of the cost of the whole vehicle in smart cars, which is a world of difference.

Based on this, many people ask editors, China's semiconductor industry is already very weak, is it difficult to make chips on smart cars? If not, can the chip on the computer be used in the car?

In fact, it is more difficult to make car chips, and civilian chips cannot be used in cars.

First of all, the working environment of automobile chips is even worse.

The working temperature requirements of chips and components in automobiles are relatively wide, and there are different requirements according to different installation positions, but they are generally higher than those of civilian products, such as engine compartment-40℃-150℃; Body control requirements -40℃- 125℃.

And conventional consumer chips and components only need to reach 0℃-70℃. In addition, other environmental requirements, such as humidity, mildew, dust, saline-alkali natural environment (seaside, snow water, rain water, etc. ), EMC and harmful gas erosion are all higher than the requirements of consumer chips.

Second, automotive electronic components are not allowed to make mistakes, which requires extremely high operational stability.

Cars will encounter more vibration and impact during driving, and the semiconductor of automobile track class must continuously ensure stable operation in various changes such as high and low temperature alternation, vibration and wind impact, waterproof and sun protection, high-speed movement and so on. In addition, automobiles require extremely high anti-interference performance of devices, including anti-ESD, EFT group pulse, RS conducted radiation, EMC, EMI and so on. Under these interferences, the chip can neither work uncontrollably, nor interfere with other devices in the car (control bus, MCU, sensor, stereo, etc.). ).

Third, the design of automobile chips requires longer service life and lower failure rate.

The design life of ordinary cars is about 654.38+500,000 kilometers in 2005, which is much longer than the life requirements of consumer electronic products.

The second is the failure rate requirement. The zero-kilometer failure rate is one of the most important indicators for automobile manufacturers, and to ensure the reliability of the whole vehicle, the requirements for all parts of the system are very high.

Because semiconductor is the primary problem in fault arrangement of automobile manufacturers, the basic requirement of automobile manufacturers for fault rate is single-digit PPM (one in a million), and most automobile manufacturers require PPB (one in a billion). It can be said that the failure rate requirement of automotive grade semiconductors is often, "zero? There is zero tolerance for errors with defects.

In contrast, the failure rate requirements of industrial chips are

Fourth, pay attention to safety and require high product consistency.

Automobile-grade semiconductors should also ensure high product consistency when mass production is realized. For cars with complex components, it is definitely unacceptable that the semiconductor components with poor consistency lead to safety hazards in the whole vehicle, so strict yield control and complete product traceability system management are needed, and even the traceability of packaging raw materials for semiconductor products is needed.

Fifth, the long-term effective supply cycle.

The life cycle of automotive semiconductor products usually takes more than 15 years (that is, the life cycle of the whole vehicle can work normally), and the supply cycle may be as long as 30 years. Therefore, there is a high demand for the supply chain configuration and management of automotive semiconductor enterprises, that is, the supply chain should be reliable and stable, which can support the whole life cycle to deal with any sudden crisis.

Compared with the PC chip, the chip in the car is very difficult, in addition to controlling the cost, energy consumption and so on.

In the end, even if we develop a chip that meets the specifications of the car, it is also a ticket. The R&D cycle of automobile industry generally reaches about 2 years, which is impossible for many small enterprises to support such a long cycle.

In addition, automotive chips also need to bear huge responsibilities. The damage of the chip on the PC side causes the computer or mobile phone to crash at most, but the car is closely related to safety. In the next ten years or hundreds of thousands of kilometers of automobile cycle, if it is caused by the design of automobile chips, it means taking responsibility.

For small and medium-sized enterprises with relatively weak financial strength, it is likely that they will get into trouble and thus cannot enter the automobile supply chain again. The shared responsibility of automotive semiconductors in terms of safety and reliability will also make many manufacturers cautious when choosing to enter the automotive regulatory market.

Due to the high standards and thresholds of the above-mentioned automotive semiconductor industry, a large number of semiconductor manufacturers who lack financial strength and industrial supporting resources and want to quickly put chips on the market to achieve benefits will be rejected. The lack of new players also makes the existing automobile semiconductor enterprises (Tier2), parts suppliers (Tier 1) and automobile manufacturers (OEM) form a strong supply chain relationship, which constitutes a solid industry barrier for new enterprises.

Write at the end:

In the 1950s, semiconductor products (including chips, sensors, etc. ) used in automobile manufacturing accounts for less than 65,438+0% of the total manufacturing cost. Today, its cost can be as high as 35% of the total cost, and it is expected to increase to 50% by 2030.

The development of smart cars has also promoted the rapid growth of the global automotive semiconductor market. Statistics show that at present, the output value of European automotive semiconductors in 20 19 reaches1508.8 billion US dollars, accounting for 36.79% of the total output value of global automotive semiconductors, ranking first in the world.

The United States contributed the second largest revenue scale of automobile chips in the world, reaching $654.38+03.387 billion, accounting for 32.64% of the world.

The output value of Japanese automotive semiconductors in 20 19 reached 10677 billion USD, accounting for 26.03%. However, in Chinese mainland, the sales revenue of automotive semiconductors in 20 19 years is only about 10 billion US dollars, accounting for less than 3%. Compared with Europe, America and Japan, there is a big gap.

But don't be discouraged. Sino-US trade war makes China automobile industry pay more attention to the upstream construction of its supply chain. At present, a number of chip design companies such as CAMBRIAN, Horizon and Xijing Technology have emerged in China automobile market. In the future, I hope these enterprises can break through the tight encirclement and contribute to China becoming a smart car power.

This article comes from car home, the author of the car manufacturer, and does not represent car home's position.