Failure analysis and maintenance of motherboard

First, the motherboard breakdown classification

1. According to the impact of the microcomputer system can be divided into non-fatal and fatal failure of

Non-fatal fault also occurs in the system during the electrical self-test, the general error message; The fatal fault occurs during the power self-test of the system, which usually causes the system to panic.

2. Depending on the scope of impact can be divided into local fault and global failure

Local fault refers to the system of one or several functions are not normal, such as the motherboard on the printing control chip damage, only caused online printing is not normal, does not affect other functions; Global failures often affect the normal operation of the entire system, causing it to lose all functions, such as the clock generator damage will cause the entire system

3. According to the failure phenomenon is fixed can be divided into stability failure and instability fault

Stability failure is due to the failure of components, circuit breakers, short circuit, the failure of the phenomenon of stable recurrence, and instability fault is often due to poor contact, component performance, so that the logic function of the chip is sometimes normal, sometimes abnormal critical state and caused. If the display card is in poor contact with the slot due to an I/O slot distortion, the display is in a variable error state.

4. Depending on the degree of impact can be divided into independent fault and correlation fault

Independent fault refers to the failure of the chip to complete a single function; correlation fault refers to a fault associated with some other faults, its fault phenomenon for many functions are not normal, and its fault in essence for the control of the common part of the function of the failure caused (for example, soft, hard disk subsystem work is not normal, On the hard disk control card, its function control is more separate, the fault is often on the motherboard's peripheral data transmission control namely DMA control circuit.

5. According to the source of the fault can be divided into power failure, bus failure, component failure, etc.

The power failure includes the +12v, +5v and +3.3v power good signal faults on the motherboard, the bus fault including the bus itself fault and the bus control power fault, and the fault of the resistor, capacitance, IC chip and other components.

Second, the main cause of the motherboard failure

1. Human failure: Live plug I/O card, and in the card and plug when the force improperly caused to the interface, chip and other damage.

2. The environment is bad: static electricity often causes the motherboard chip (especially CMOS chip) to be penetrated. In addition, the motherboard encountered power damage or power grid voltage instantaneous spike pulse, often damage the system board power plug near the chip. If the motherboard is covered with dust, it will also cause a short circuit signal and so on.

3. Device quality problem: Due to the poor quality of chips and other devices caused by damage.

Third, the main board failure inspection and maintenance of common methods

Motherboard failure is often manifested as system startup failure, screen display and so difficult to visually judge the fault phenomenon. The following list of maintenance methods have advantages and limitations, often used in combination.

1. Cleaning law

The brush can be brushed lightly to the dust on the motherboard, in addition, some cards on the motherboard, chip using the form of pins, often because of the oxidation of the pins and poor contact. Use eraser to surface oxidation layer, reseat.

2. The Observation Act

Repeatedly look at the repair of the board, to see whether the plugs, sockets are skewed, resistors, capacitance pins are colliding, whether the surface is burnt, the chip surface is cracked, the motherboard on whether the copper foil burning. Also want to see if there is a foreign matter falling into the motherboard components. Where there is doubt, you can use the multimeter to measure. Touch some of the chip's surface, if abnormal hair hot, can change a chip to try.

3. Resistance and voltage measurement method.

To prevent accidents, measure the resistance between the power +5v and the ground (GND) on the motherboard before powering up. The simplest method is to measure the resistance between the chip's power pins and the ground. When the power plug is not plugged in, the resistance should generally be 300ω, and the minimum should not be lower than 100ω. Then the reverse resistance value, slightly different, but not too large. If the positive and negative resistance is very small or close to the conduction, it indicates that there is a short-circuit, should be examined for short reasons. There are several causes of such phenomena:

(1) The chip on the system board has been penetrated. Generally speaking, such failures are more difficult to exclude. For example, the TTL chip (LS series) of the +5v connected together, can suck the +5v pin on the solder, so that its suspension, measurement, to find fault film. If the use of secant method, will inevitably affect the life of the motherboard.

(2) The board has a damaged resistor capacitor.

(3) The board contains conductive debris.

When troubleshooting the short-circuit, plug in all the I/O cards to measure whether the +5v,+12v and the ground are short. In particular, whether the +12v and surrounding signals collide. When there is a good same type of motherboard on hand, you can also use the method of measuring resistance to measure the doubt on the board, by contrast, you can quickly find the chip fault.

When the above steps are not effective, you can plug the power supply measurement. The +5v and +12v of general power supply are measured. When a voltage value is found to deviate from the standard too far, it can be separated or cut some of the lead or unplug some of the chip to test the voltage. When you cut a lead or unplug a chip, if the voltage becomes normal, then this lead leads to the components or unplug the chip is the fault.

4. Plug-and-swap method

Host systems have many causes of failure, such as the motherboard itself or I/O bus of various card failures can cause the system to run abnormally. The use of the plug maintenance method is a simple way to determine the fault on the motherboard or I/O device. The method is to shut down the plug-in board block by piece, each pull out a plate on the boot to observe the operation of the machine, once the motherboard is pulled out after a normal operation, then the fault is due to the board failure or the corresponding I/O bus slot and load circuit failure. If the system boot is still abnormal after unplugging all the boards, the failure is most likely on the motherboard. The use of the Exchange method is essentially the same type of plug-in board, bus mode, the same function of the plug-in board or the same model chip mutual exchange, according to the changes in the phenomenon of fault diagnosis. This method is used for easy to plug the maintenance environment, such as memory self-test error, can exchange the same memory chip or memory to determine the cause of the failure.

5. Static and dynamic measurement and analysis method

(1) Static measurement method: Let the motherboard suspend in a certain feature state, by the circuit logic principle or the logic relationship between chip output and input, using a multimeter or logic pen to measure the correlation point level to analyze and judge the fault reason.

(2) Dynamic Measurement Analysis method: the establishment of a special Judgment program or artificial set of normal conditions, in the course of the operation of the machine with the oscilloscope measurement and observation of the relevant components of the waveform, and the normal waveform comparison, judge the fault site.

6. Simple after complex and combined with the composition of the principle of judgment

With the wide application of LSI, the control logic is more and more integrated on the motherboard, and its logic correctness is more and more difficult to judge by measurement. It can be used to judge the logic relationship of simple chip and resistance capacity components, then the fault will be concentrated in the logical relationship is difficult to judge the LSI chip.

7. Software Diagnostic method

Through the random diagnosis procedure, the special maintenance diagnosis card and according to each kind of technical parameter (for instance the interface address), the custom special diagnosis program to assist the hardware maintenance may achieve the multiplier effect. The principle of program testing method is to use software to send data, commands, to identify the fault site by reading the status of the circuit and a chip (such as registers). This method is often used to check various interface circuit faults and various circuits with address parameters. But the premise of this method is that the CPU and the base bus run normally, can run the diagnostic software, and can run the diagnostic card installed on the I/O bus slot. The preparation of the diagnostic procedures to be rigorous, comprehensive targeted, can allow some key parts of the regular signal, can be repeated testing of occasional failures and can display record error situation.