Various fault detection methods of electronic circuit boards

Using appropriate methods to find, judge and determine the specific location and cause of the electronic circuit board fault is the key to fault detection. The various fault detection methods of electronic circuit boards introduced below are effective methods summarized in long-term practice. In specific applications, it is necessary to cross and flexibly apply specific detection objects, and continuously summarize the experience methods suitable for their own work fields, in order to achieve the purpose of rapid, accurate and effective troubleshooting.

1. Observation
Observation is a method of discovering electronic circuit faults through human senses. This is the simplest and safest method, and it is also the first step in a common testing process for various instruments and equipment.
Observation method can be divided into static observation method and dynamic observation method.
1. Static observation method
It is also called the non-energized observation method. Before the electronic circuit is energized, some faults are mainly found through visual inspection. Practice has proved that the failure of solder joints, which account for a considerable proportion of electronic circuit failures, disconnected wire joints, leakage or bursting of capacitors, loose connectors, rusty electrical contacts, etc. can be found through observation. There is no need to make a big fuss on the entire circuit. Lead to fault escalation.
"Static" emphasizes meditation and concentration. Observing carefully, you can't find faults when you are so-so.
Static observation should be done step by step, first outside and then inside. Before opening the case, check the appearance of the electrical appliance for scratches, whether the buttons and sockets are damaged, whether the insurance is burned, etc. After opening the case, first look at the various devices and components inside the machine to see if there are any collisions, disconnections, burnouts, etc., and then move some components and wires with your hands or tools for further inspection. For test circuits or prototypes, check the principles for wiring errors, whether the components meet the design requirements, whether the IC pins are inserted in the wrong direction or bent, whether there are any faults such as missing soldering, bridging, etc.
When no abnormality is found in static observation, the dynamic observation method can be further used.
2. Dynamic observation method
It is also called the power-on observation method, that is, after the line is energized, the human body's sight, smell, hearing, and touch are used to check the line fault. Power-on observation, especially when larger equipment is energized, the isolation transformer and voltage regulator should be used as much as possible to gradually power up to prevent the expansion of the fault. Under normal circumstances, you should also use meters, such as ammeters, voltmeters, etc. to monitor the state of the circuit.
After the power is turned on, you need to see whether there is fire, arm smoke, etc. in the circuit; your ears should listen for abnormal sounds in the circuit; your nose should smell whether there are scorching or burning odors in the electrical appliances; your hands should touch some tubes, integrated circuits Wait for whether it is hot, (note: high-voltage, high-current circuits must be protected from electric shocks and burns) immediately cut off the power when abnormalities are found.
Power-on observation can sometimes determine the cause of the fault, but in most cases the exact location and cause of the fault cannot be confirmed. For example, if an integrated circuit heats up, it may be caused by a fault in the peripheral circuit, or an error in the power supply voltage. It may be caused by excessive load or self-excitation of the circuit. Of course, damage to the integrated circuit itself cannot be ruled out. Other detection methods must be used to analyze and judge To find out where the fault is.
2. Measurement method
The measurement method is the most widely used and most effective method in fault detection. According to the detected electrical parameter characteristics, it can be divided into resistance method, voltage method, current method, logic state method and waveform method.
1. Resistance method
Resistance is the basic feature of various electronic components and circuits. The method of using a multimeter to measure the resistance between various points of the electronic components or circuits to determine the fault is called the resistance method.
There are two basic methods for measuring resistance, "online" and "offline".
For "online" measurement, it is necessary to consider the influence of the components under test on other parallel branches, and the measurement results should be analyzed and judged against the schematic diagram.
"Off-line" measurement requires the components or circuits to be tested to be unsoldered from the entire circuit or printed circuit board. The operation is troublesome but the results are accurate and reliable.
Use resistance method to measure integrated circuits, usually first connect one test lead to the ground, use another test lead to measure the resistance of each pin to ground, then exchange the test leads to test again, and compare the measured value with the normal value (some maintenance information is given, or accumulated by yourself ) For comparison, the larger difference is often the fault. (It doesn't have to be that the integrated circuit is bad!)
The resistance method is very effective and fast for determining the on-off of switches, connectors, wires, conductive patterns of printed boards, and the deterioration of resistors, short-circuits of capacitors, open circuits of inductors, etc., but for transistors, integrated circuits and circuit units, Generally, it is not possible to directly determine the fault, and comparative analysis or other methods are required. However, since the resistance method does not need to energize the circuit, the detection risk can be minimized, so the general detection is the first to be used.
Attention
(1) When using the resistance method, it should be carried out when the circuit is powered off and the large capacitor is discharged, otherwise the result is inaccurate and the multimeter may be damaged.
(2) Avoid using the lOK file of the analog multimeter when testing low-voltage integrated circuits (5V).
(3) During online measurement, the test leads of the multimeter should be tested alternately for comparison and analysis.
Second, the voltage method
When the electronic circuit is working normally, each point of the circuit has a certain working voltage, and the method of judging the fault by measuring the voltage is called the voltage method.
The voltage method is the most basic and most commonly used method among the energization detection methods. According to the nature of the power supply, it can be divided into AC and DC voltage measurement.
1. AC voltage measurement
Generally, the AC circuit in electronic circuits is relatively simple. For the 50/60Hz mains voltage after boosting or stepping down, you only need to use an ordinary multimeter to select the appropriate AC range. When measuring high voltage, pay attention to safety and develop the habit of using one-handed operation .
For non-50/60Hz power supplies, such as the measurement of the output voltage of the inverter, the frequency characteristics of the voltmeter used should be considered. Generally, the analog multimeter is 45～2000Hz, and the digital multimeter is 45～500Hz. The measurement result is out of range or non-sine wave. None of them are correct.
2. DC voltage measurement
The detection of DC voltage is generally divided into three steps:
(1) Measure whether the output terminal of the voltage stabilizing circuit is normal.
(2) Whether the voltage at the output point of the amplifying circuit and the power supply terminal of external components is normal for each unit circuit and the key "points" of the circuit.
(3) Whether the voltage of the main components of the circuit such as transistors and the pins of the integrated circuit is normal, the power supply terminal of the integrated circuit must be tested first.
A more complete product manual should provide the normal working voltage of each point of the circuit, and some maintenance materials also provide the working voltage of each pin of the integrated circuit. In addition, the voltage at each point can be measured by comparing the same kind of circuit in normal operation. The parts or components that deviate more from the normal voltage are often the parts of the fault.
This detection method requires workers to have the ability to analyze circuits and collect data about relevant circuits as much as possible in order to achieve a multiplier effect with half the effort.
Three, current method
When the electronic circuit is working normally, the working current of each part is stable, and the part that deviates from the normal value is often the fault. This is the principle of using the current method to detect the fault of the side line.
The current method has two methods: direct measurement and indirect measurement.
Direct measurement is the method of connecting the ammeter directly in series to the loop to be tested to measure the current value. This method is intuitive and accurate, but it often requires "surgery" on the circuit, such as disconnecting the wires and removing the pins of the components, etc., in order to be measured, so it is not very convenient. For the measurement of the total current of the whole machine, it can generally be measured by connecting the two test leads of the ammeter to the switch, and the measurement safety must be paid attention to for the circuit using 220V alternating current.
The indirect measurement method actually uses the method of measuring voltage to convert into current value. This method is quick and convenient, but it is not easy to accurately judge if the components of the selected side measuring point are faulty.
To determine whether the working current of the triode is normal by measuring the voltage drop of Re, such as the large deviation of the resistance value of Re itself or the leakage of Ce, it can cause misjudgment.
Fourth, the wave method
For alternating signal generation and processing circuits, using an oscilloscope to observe the waveform at each point of the signal path is the most intuitive and effective fault detection method.
The wave method is used in the following three situations:
1. The presence and shape of the waveform
In electronic circuits, the existence and shape of the waveform at each point of the circuit are generally determined. For example, the shape and amplitude of the waveform at each point are given in the standard TV schematic diagram. If the waveform at this point is not measured or the shape is quite different Larger, the fault is more likely to occur in the circuit.
When a self-excited oscillation or modulation waveform that should not appear is observed, although the fault location cannot be determined, the cause of the fault can be analyzed from the frequency and amplitude.
2. Waveform distortion
In circuits such as amplification or buffering, if the circuit parameters are mismatched or the components are improperly selected or damaged, the waveform will be distorted. The cause of the failure can be found by observing the waveform and analyzing the circuit.
3. Waveform parameters
Use an oscilloscope to measure various parameters of the waveform, such as amplitude, period, front and rear edge phases, etc., and compare them with the waveform parameters during normal operation to find out the cause of the fault.
Pay attention to the application of the wave method
(1) Be careful not to exceed the allowable voltage range of the oscilloscope for the high voltage and large pulse parts of the circuit. If necessary, use high-voltage probes or take measures such as voltage division or sampling at the circuit observation point.
(2) When the oscilloscope is connected to the circuit, the input impedance has a certain influence on the circuit. Especially when measuring the pulse circuit, a 10:1 probe with compensation function should be used, otherwise the observed waveform does not match the actual situation.
Five, logic state method
For digital circuits, it is only necessary to judge the logic state of each part of the circuit to determine whether the circuit works normally. Digital logic is mainly composed of high and low level states, in addition to pulse trains and high impedance states. Therefore, a logic pen can be used for circuit detection.
The logic pen has the advantages of small size and easy to carry and use. The logic pen with simple functions can measure the logic state of a single circuit (TTL or CMOS). In addition to measuring the logic state of multiple circuits, the logic pen with more functions can also measure the number of pulses quantitatively, and some have pulse signal generators. Function, can send out a single pulse or continuous pulse for the detection circuit.
3. Tracking method
Signal transmission circuits, including signal acquisition (signal generation), signal processing (signal amplification, conversion, filtering, isolation, etc.), and signal execution circuits, occupy a large proportion of modern electronic circuits. The key to the detection of this circuit is to track the transmission link of the signal. In specific applications, there are two types of signal tracking method and signal injection method according to the type of circuit.
1. Signal tracing method
The signal tracing method is a detection method to find the signal trace for the signal flow of the signal generation and processing circuit. The specific detection can be divided into forward tracing (from input to output sequence search) and reverse tracing (from output to output). There are three types: input order search) and equal division search.
Forward tracing is a commonly used detection method, and you can use test instruments (oscilloscope, frequency meter, multimeter, etc.) to qualitatively and quantitatively detect the signal step by step to determine the fault location. Figure 3 is the circuit block diagram and detection schematic diagram of the AC millivoltmeter. We use a fixed sine wave signal to add to the input terminal of the millivoltmeter, starting from the attenuation circuit, step by step detection of all levels of the circuit, according to the function and performance of the circuit can determine whether the signal is normal, and observe step by step until it is found malfunction.
Obviously, reverse tracking detection is only different in the order of detection.
Equal tracking is an efficient method for circuits with more units. We take the time base signal generation circuit of a certain instrument as an example to illustrate this method. This circuit generates a 5MHz signal from a crystal oscillator placed in a constant temperature bath, and through a 9-stage frequency divider circuit, produces the 1Hz and 0.01Hz signals required by the test, as shown in Figure 4.
There are 10 units in the circuit. If there is a problem with the 9th unit, the forward method needs to be tested 8 times to find it. The equal division tracing method is to divide the circuit into two parts, first determine which part the fault is, and then divide the faulty part into two parts for detection. Still taking the 9th unit failure as an example, the 1kHz signal is measured with the equal division tracking method, and it is found to be normal, and the fault is determined to be in the second half; if the 1Hz signal is measured again, it is still normal, and the fault can be set to the 9, 10 unit, and the third measurement 0. 1Hz signal, you can determine the failure of the 9th unit. Obviously, the efficiency of the equal division tracking method is greatly improved.
The equal division tracing method is suitable for circuits with a multi-stage series structure, and the failure rate of the circuits at all levels is roughly the same, and the test time is similar each time. It is not applicable to circuits with branches, feedback or fewer units.
2. Signal injection method
The signal injection method is an effective detection method for signal processing circuits that do not have a signal generation circuit or a signal generation circuit is faulty. The so-called signal injection is to input a known external test signal at the input terminals of the signal processing circuit at all levels, and use terminal indicators (such as indicating meters, speakers, displays, etc.) or testing instruments to determine the working status of the circuit, so as to find There is a circuit failure.
Various radio and television receiving equipment is a typical example of using signal injection method to detect. Figure 5 is a block diagram of a typical FM stereo radio. Two kinds of signals are required for detection: an FM stereo signal is required before the discriminator, and an audio signal is required after the decoder. Usually the detection of the radio circuit is to use the reverse signal to inject people, that is, the audio signal of a certain frequency and amplitude is gradually moved forward from AR, AL, and the presence or absence of sound and the quality and size of the sound are monitored through the speaker or earphone, so as to judge the circuit failure . If the audio circuit part is normal, use the FM stereo signal source to inject people from G, H... until the fault point is found.
Pay attention to the following points when using signal injection method to detect
(1) The signal injection sequence can be forward, reverse or intermediate injection sequence according to the specific circuit.
(2) The nature and amplitude of the injection signal should be changed according to the circuit and the injection point. For example, the audio part of the radio is injected into the signal. The closer the speaker is, the stronger the signal is. The same signal injection may be normal at point B, and point D may be injected. Too much distortion causes the amplifier to saturate. The working signal of the injection point can usually be estimated as a reference for the injection signal.
(3) When injecting signals, select a suitable ground point to prevent the signal source and the circuit under test from affecting each other. Under normal circumstances, you can choose a grounding point close to the injection point.
(4) The signal and the circuit under test should be coupled appropriately. For example, the AC signal should be connected in series with a suitable capacitor, and the DC signal should be connected in series with a suitable resistance to match the signal with the impedance of the circuit under test.

(5) Signal injection can sometimes be done in a simple and easy way. For example, the human body induction signal can be used as the injection signal (that is, the hand-held conductor touches the corresponding circuit part) when the radio is detected. In the same way, sometimes it is necessary to pay attention to the influence of inductive signals on the detection of external signals.

4. Replacement method
The replacement method is to replace the suspected corresponding part of the circuit with normal components, circuits or parts with the same specifications and performance to determine where the fault is. It is also one of the most commonly used and effective methods in circuit debugging and maintenance. .
In actual applications, there are three methods depending on the object to be replaced.
1. Component replacement
Except for some circuit structures that are more convenient for component replacement (such as ICs with connectors, switches, relays, etc.), it generally requires desoldering, which is troublesome and easily damages peripheral circuits or printed boards, so component replacement is common It is only used when other detection methods are difficult to distinguish, and try to avoid "major surgery" on the circuit board. For example, if you suspect that two lead components are open, you can directly solder a new component to test it; if you suspect that the capacity of a capacitor is reduced, you can try it with a capacitor.
2. Unit circuit replacement
When it is suspected that a certain unit circuit is faulty, another normal circuit of the same model or type is used to replace the corresponding unit circuit of the machine to be checked to determine whether the unit circuit is normal. Some circuits have several channels of the same circuit. For example, the left and right channels of a stereo circuit are exactly the same, which can be used for cross-replacement experiments.
The replacement test is more convenient when the electronic equipment adopts the multi-board structure of the unit circuit. Therefore, for equipment with high requirements for on-site maintenance, it is best to adopt a structure that is convenient for replacement, so that the equipment has good maintainability.
3. Parts replacement
With the development of integrated circuits and mounting technology, electronic products are rapidly developing in the direction of higher integration, more functions, and smaller sizes. Not only is it difficult to replace the components at the component level, but it is also more and more inconvenient to replace the unit circuit. The functions of a dozen or even dozens of circuits can now be completed with a single integrated circuit, and more circuit units can be accommodated on a printed circuit board per unit area. The detection and maintenance of the circuit has gradually developed to the board level and even the overall direction. Especially for the more complicated system composed of Ruqian independent functional parts, the component replacement method is mainly used in the detection.
The component replacement test should follow the following three points
(1) The replacement parts must have the same model and specifications as the original parts, or be compatible with main performance and function, and work normally.
(2) The interface of the component to be replaced works normally, at least the power supply, input and output ports are normal, and the replacement component will not be damaged. This requires analyzing the fault phenomenon and making necessary inspections of the interface power supply before replacement.
(3) The replacement should be tested separately, and do not replace more than one part at a time.
Finally, it needs to be emphasized that although the replacement method is a commonly used detection method, it is not the best method, let alone the preferred method. It is only used when there is doubt about a certain part on the basis of other methods.
For systems using microprocessors, attention should also be paid to troubleshooting software failures before hardware testing and replacement.
5. Comparative Law
Sometimes a variety of detection methods and test methods can not be used to determine the fault, but the uncomplicated comparison method can surprisingly win. Commonly used comparison methods include four methods: complete machine comparison, adjustment comparison, bypass comparison, and elimination comparison.
1. The whole machine comparison method
The whole machine comparison method is to compare the faulty machine with the normal working machine of the same type to find the fault. This method is particularly suitable for devices that lack information and are more complex in nature, such as microprocessor-based products.
The whole machine comparison method is based on the detection method. The working point measurement and waveform observation of the circuit parts that may have faults, or signal monitoring, compare the difference between good and bad equipment, and often find problems. Of course, because each device cannot be completely consistent, the test results need to be analyzed and judged. These common-sense problems require the accumulation of basic theoretical foundation and daily work.
2. Adjustment and comparison method
The adjustment comparison method is a detection method that determines the fault by comparing the changes of the circuit before and after the adjustment by adjusting the components of the whole machine or changing some of the status quo. This method is especially suitable for equipment that has been placed for a long time, or has been subjected to changes in external conditions such as handling or falling.
Under normal circumstances, the adjustable parts should not be changed randomly when testing the equipment. However, because the device is affected by external forces, it may change the factory settings and cause failures. Therefore, some adjustable capacitors, resistors, inductances and other components can be changed under the premise of making a reset mark in advance during detection, and pay attention to compare the equipment before and after adjustment. Work status. Sometimes it is necessary to touch the component pins, wires, connectors or pull out the plug-ins and reconnect them, or re-soldering the suspected printed board parts, etc., pay attention to observe and record the working conditions of the equipment before and after the status changes, find faults and eliminate them malfunction.
When using the adjustment and comparison method, it is most taboo to avoid chaos without marking. Adjustments and changes to the status quo should be changed step by step, and the status before and after the change should be compared at any time. If the adjustment is found to be invalid or a bad change, it should be restored in time.
3. Bypass comparison method
The bypass comparison method is a comparison inspection method that uses capacitors of appropriate capacity and withstand voltage to bypass certain parts of the circuit of the tested equipment. It is suitable for power supply interference, parasitic oscillation and other faults.
Because the bypass is actually an AC short-circuit test, in general, first select a capacitor with a smaller capacity and temporarily connect it between the circuit part in question and the "ground", and observe and compare the changes in the failure phenomenon. If the circuit changes in a good direction, you can appropriately increase the capacity of the capacitor and try again until the fault is eliminated. The fault location can be determined according to the bypass location.
4. Elimination comparison method
Some combined machines or combined systems often have several components with the same function and structure. When the system functions are found to be abnormal during debugging, it is impossible to determine the component that caused the fault. In this case, it is easy to identify the fault by the elimination and comparison method. The method is to insert the components one by one and monitor the whole machine or system at the same time. If the system is working normally, the suspicion of the component can be eliminated, and then another component can be inserted to test until the fault is found.
For example, a certain control system uses 8 plug-in cards to control 8 objects. During debugging, it is found that the system has interference. The comparative elimination method is adopted. When the fifth card is inserted, the interference phenomenon occurs. Confirm that the problem is on the fifth card. Replace it with another card to eliminate interference.
Attention
(1) The above method is incremental elimination, obviously the reverse direction can also be used, that is, descending elimination.
(2) This kind of multi-unit system failure is sometimes not caused by one unit component. In this case, multiple comparisons can be used to eliminate it.
(3) When adopting the exclusion comparison method, be sure to turn off the power supply every time you insert or pull out the unit components to prevent the system from being damaged by plugging or unplugging.
6. Computer intelligent automatic detection
Using the computer's powerful data processing capabilities combined with modern sensor technology can gradually automate and intelligentize circuit detection. This is more and more widely used in various computers and computer-based equipment, and the level is getting higher and higher. The following are the current common computer detection methods:
1. Power-on self-test
This is a primary detection method. Use the power-on self test program (POST, power-onself test) solidified in the computer ROM to test various hardware, peripherals and interfaces in the computer. In addition, it can also automatically test the configuration of the hardware and software in the computer. When an error (fault) occurs, a sound and screen prompt will be performed.
This startup software detects the characteristic parameters of each part of the hardware, and compares the test results with the pre-stored standard values for diagnosis, which can determine the quality of the hardware, but in general, the specific part of the fault cannot be determined, nor can it be based on the operator. Willing to conduct deep human testing.
2. Testing and diagnostic procedures
This method is that the computer runs a special detection and diagnosis program, which can be set and selected by the operator to test the target, content, and failure reporting method, and most of the failures can be located to the chip.
There are many special programs of this kind, such as QAPLUS, NORTON, PCTOOLS, etc. With the version upgrade, the functions are getting stronger and stronger. In addition, the system software generally has its own detection program, for example, DOS 6. X, WIN 3. X, and WIN. 9X all have corresponding detection functions.
Obviously, the premise of this detection method is that the computer itself is basically working normally. If the computer has a serious failure, this method will be useless.
Three, intelligent monitoring
This is the current trend of the latest technological development and the most advanced mode to ensure the normal operation of the machine. This method uses special hardware and software installed in the computer to monitor the system, such as continuous automatic testing of the CPU temperature, working voltage, and internal temperature. Once the range is exceeded, the alarm information will be displayed immediately, which is convenient for the user to take measures. Ensure the normal operation of the machine. This intelligent monitoring method can also automatically take measures to eliminate hidden troubles within a certain range. For example, if the temperature inside the machine is too high, the fan speed will be automatically increased to force the temperature to fall, or even force the machine to "sleep", and the fan speed will be reduced when the temperature inside the machine is low. Or stop to save energy and reduce noise.
Obviously, this mode of preventing trouble and automatically adjusting operation is the most ideal method for detection. Now mainstream computers and computer-based equipment mostly have this advanced function. With the development of technology, this intelligent monitoring method will be used in more products, making electronic products develop to a higher level.