Integrated circuit fault detection

The hardware defects of integrated circuit chips usually refer to the physical imperfections exhibited by the chips. Integrated circuit fault refers to a circuit logic functional error or abnormal circuit operation caused by an integrated circuit defect. The common factors that cause failures in integrated circuit chips include rapid performance degradation due to changes in component parameters, poor component contact, signal line failures, and equipment malfunction due to harsh working environments. Circuit faults can be divided into hard faults and soft faults. Soft faults are temporary and will not cause permanent damage to the chip circuit. It usually appears randomly, causing the chip to sometimes work normally and sometimes exhibit abnormalities. When dealing with such faults, simply reconfigure the system with the same configuration parameters when the fault occurs, and the equipment can be restored to normal. The damage caused by hard faults to the circuit is permanent and cannot be self restored without maintenance.
There are usually three modules required for fault detection in integrated circuit chips: source excitation module, observation information acquisition module, and detection module. The source excitation module is used to transmit test vectors to the integrated circuit chip to drive the chip into various working modes. Usually, it is required that the test vector set should contain as many possible input vectors as possible. The observation information collection module is responsible for collecting information for subsequent analysis and processing. The selection of observation information is crucial for fault detection, as it should include as much fault feature information as possible and be easy to collect. The detection module is responsible for analyzing and processing the collected observation information, identifying the fault features hidden in the observation information, and diagnosing the pattern of circuit faults.
The earliest circuit fault diagnosis methods mainly relied on simple tools for testing and diagnosis, which greatly relied on the theoretical knowledge and experience of experts or technicians. Among these testing methods, the four most commonly used are virtual testing, functional testing, structural testing, and defect and fault testing. Virtual testing does not require the detection of actual chips, but only tests simulated chips, which is suitable for conducting before chip manufacturing. It can detect faults in chip design in a timely manner, but it does not consider noise or differences in the actual manufacturing and operation of the chip. Functional testing determines whether a chip has a malfunction based on whether it can perform the expected function during testing. This method is easy to implement but cannot detect faults with non functional effects. Structural testing is an improvement on built-in testing, which combines scanning technology and is often used for fault detection of produced chips. Defect and fault testing is based on the actual production of completed chips, and detects whether there are faults by inspecting the production process quality of the chips. Defect and fault testing requires a high level of knowledge and experience from professional technical personnel. Chip manufacturers usually combine these four testing techniques to ensure the reliability and safety of the entire process from design to production and application of integrated circuit chips. However, for increasingly complex circuit systems, these early methods have become increasingly inadequate. Through continuous improvement and innovation, many new ideas and methods have emerged one after another.
Voltage diagnosis appeared earlier and is widely used. The observation information of voltage testing is the logical output value of the tested circuit. This method obtains the corresponding logic output value of the circuit by inputting different test vectors into the circuit, and then compares the collected circuit logic output value with the expected logic output value of the circuit corresponding to the input vector, in order to achieve the goal of detecting whether the circuit can achieve the expected logic function in the actual operating environment. This method is simple but not suitable for large-scale integrated circuits with high redundancy. If the defect appears in the redundant part, it cannot be detected. Moreover, when the circuit scale is large, the set of test vectors will also grow exponentially, which directly leads to difficulties in generating test vectors and low diagnostic efficiency. In addition, voltage diagnosis cannot detect faults that only affect circuit performance rather than circuit logic functionality. Due to the fact that the voltage logic value output by integrated circuits may not necessarily be related to all nodes in the circuit, voltage testing cannot detect non functional failure faults in integrated circuits. So, in the early 1980s, diagnostic techniques based on integrated circuit power supply currents were proposed. The power current is usually directly or indirectly related to all nodes in the circuit, so current based diagnostic methods can cover more circuit faults. However, the introduction of current diagnosis technology is not intended to replace voltage testing, but to supplement it in order to improve the detection rate and coverage of fault diagnosis. Current diagnosis technology can be divided into static current diagnosis and dynamic current diagnosis. The core of static current diagnosis technology is to compare the power supply current of the tested circuit in a stable operating state with a pre-set threshold to determine whether there is a fault in the tested circuit. It can be seen that the selection of threshold is the key to determining the detection rate of this method. Early static current diagnosis techniques used fixed thresholds, but fixed thresholds were not suitable for the development of integrated circuit chips towards deep submicron technology. So, later generations continuously improved the static current detection method, and successively proposed differential static current detection technology, current ratio diagnosis method, static current detection technology based on clustering technology, etc. Dynamic current diagnosis technology was introduced in the 1990s. Dynamic current can directly reflect the frequency of internal voltage switching in a circuit during state transition. The detection technology based on dynamic current can detect faults that cannot be detected by the previous two methods, further expanding the fault coverage range. With the development and gradual maturity of intelligent technology, integrated circuit chip fault detection technology is also moving towards the trend of intelligence