Common CPU failures are roughly the following: thermal faults, restart faults, black screen faults, and overclock faults. Since the probability of a failure of the CPU itself is very small, most of the failures are caused by the carelessness of the user. The following system home Xiaobian will introduce you to the common fault case analysis of cpu. Case 1: The CPU pin is in poor contact, causing the machine to fail to start. Symptom: A user's Athlon CPU computer has been used normally on weekdays. One day, it suddenly fails to boot, and the screen has no display signal output. It starts to recognize that the graphics card is faulty. After checking with the replacement method, it was found that the graphics card had no problem, and later it was speculated that the display was faulty. After the inspection, the display was all normal. Unexpectedly, unplug the CPU plugged into the motherboard and carefully observe that there is no burnt mark, but it is impossible to light the machine. Later, I found that the pins of the CPU are black and green, with traces of oxidation and rust (the pins of the CPU are made of copper and the outer layer is gold-plated). The CPU pins are cleaned with a toothbrush, and the computer can be powered up. It is. Fault analysis: The CPU solves the problem after rust removal, but how does rust come from? Finally, the doubts fell on the piece of refrigeration. Previously, there was an article about the condensation of the cooling film. It may be because the cooling film lowered the surface temperature of the chip too low, which was lower than the condensation point, causing the CPU to work in a humid environment for a long time. in. The exposed copper pins react with oxygen in the air to form copper rust in this environment. The rust spots caused by too many days have caused poor contact, which has caused this strange failure. In addition, there are some inferior motherboards. Due to the poor quality of the CPU socket, the contact may be poor. Users need to fix the contact between the CPU and the slot to solve the problem. Case 2: “Low temperature” can also burn down CPU failures: A friend of the author has done such a test to run the desktop Celeron II 566 processor at standard frequency (no overclocking) and heat it through a hair dryer. 55 degrees Celsius (obtained by the motherboard temperature monitoring function), as long as the program with high CPU usage runs, it will crash in a while; while the Celeron II 566 is overclocked to 850MHz, the system temperature is 50 degrees Celsius left, no power is applied, only the fan works, the host Not right, running Quake III took more than ten minutes to crash. It is estimated that the temperature has exceeded 55 degrees Celsius at this time, and the temperature of the core CPU failure has been measured and found to have reached 86.4 degrees Celsius. It was later discovered that the CPU burned down at such a low temperature. However, he found that the laptop did not have such a problem. Fault Analysis: It turns out that this is the CPU temperature detected by the motherboard that confuses us. In fact, the CPU temperature reported by the desktop board is not its core temperature at all, because the common temperature probe of the desktop board is not in contact with the CPU heat sink or the CPU at all, and only the air temperature near the CPU is measured. This caused a lot of CPUs to burn down in a seemingly low temperature situation. According to data released by Intel, the temperature limit of the Pentium III550E is 85 degrees Celsius, and the limit temperature of the Pentium III 800E is around 80 degrees Celsius. If everyone loses their vigilance and believes in the report of the motherboard, thinking that their CPU is still running at a low temperature, it is a big mistake. Why is this difference in laptops? In the original notebook, the temperature measurement of the CPU is based on the thermistor. The temperature measurement point is at the bottom of the CPU. If the reading is directly, the temperature is not so high, and the monitored temperature of the display is corrected, which is higher than the measured temperature. Closer to the core temperature of the CPU. Therefore, the CPU temperature of most notebook tests is the core temperature, and there is no case of burning C P U at low temperatures. Case 3: Suspend mode causes CPU to burn out. Symptom: The general system hang does not cause C P U to burn out. The system will automatically reduce the CPU operating frequency and fan speed to save energy. The suspend mode mentioned here causes the CPU to be burned, and both are CPUs after overclocking. Maybe you think this is a little weird. Why is the CPU after overclocking burned? This is all due to the fan's shutdown. It turns out that the monitor chip on the motherboard can monitor the fan speed, and some can automatically reduce the fan speed or even stop running when the system enters the Suspend power-saving mode. This is a good idea to save power. It also extends the life and life of the fan. In the past, when the CPU was idle, the heat was not high, so the fan did not turn, and only the heat sink could cope with heat dissipation. But now the CPU frequency is too high, even if you enter the suspend mode, when the fan does not turn, the CPU will be hot. Therefore, some people will encounter a situation where Windows 98 will crash and a blue screen will appear when it goes from suspend to normal mode. This is the error caused by overheating of C P U. In severe cases, C P U will hang due to overheating, especially Thunderbird or Duron after overclocking. Fault Analysis: This situation does not occur on every motherboard, and must meet three conditions when it occurs. First the CPU fan must be a 3pin fan so that it is controlled by the motherboard. Second, the monitor function of the motherboard must have Fan Off When Suspend, and this function is preset to On. Some motherboard presets On, and even some of the settings in Power Management have Fan Off When Suspend. You can take a look. Third, enter the suspend mode. So check your computer now. Case 4: CPU frequency common fault phenomenon: There is a computer C P U is A t h l o n X P1600+, the BIOS shows 1050MHz after booting, but the normal AthlonXP 1600+ should be 10.5 multiplier & times; 133MHz FSB = 1400MHz main frequency. In the BIOS, the FSB can only be set to 129MHz. The disassemble finds that the DIP switch of the motherboard is adjusted to 100MHz FSB, so it is adjusted to 133MHz FSB. After booting, the CPU is running normally. This is the same for several times. Later, the DIP switches on the motherboard are all set to Auto. In the default state, the system self-test is still 1050MHz. Suspected that the memory and graphics card are not synchronized, the memory CAS is changed from 2 to 2.5, and it still cannot be self-tested normally; the AGP graphics card is changed from 4X to 2X mode, and the power is restored to normal. Fault Analysis: It was later confirmed that the user's graphics card version is older. The default AGP operating frequency is 66MHz (at 100MHz, the PCI operating frequency is 100 ÷ 3 = 33.3MHz, AGP is PCI × 2 = 66.6MHz) At 133MHz FSB, the AGP frequency is 133 ÷ 3 × 2 = 88.7MHz). Because of the 133MHz FSB used by AthlonXP, the operating frequency of AGP is increased to 88.7MHz. Therefore, the black screen of the display is obviously the result of the graphics card. After the video card is reduced in operating frequency, the system returns to normal. I also often see faults on the network due to abnormal CPU frequency. Some of the early Pentium III or Athlon motherboards defaulted to 100MHz FSB, and now the new core CPUs are 133MHz FSB. In this way, in the case of automatic detection of the motherboard, the CPU is used for frequency reduction, and it is often not found by people. In this case, just adjust the FSB and the asynchronous work of the graphics card or memory. Case 5: The mystery of computer performance degradation: A Pentium 4 computer was extremely stable at the beginning of its use, but it appeared to be infected with a virus, and its performance dropped dramatically, occasionally with a crash. First use anti-virus software to kill and find nothing. Then I suspected that there was too much disk fragmentation. I used Windows's disk defragmenter to clean up the problem. Also think that there is a problem with Windows, formatting and reinstalling the system, still no effect. Open the chassis and find that there is a problem with the fan of the CPU heat sink. Replace the heat sink and solve the problem. Fault Analysis: The core of the original Pentium 4 processor is equipped with a thermal monitoring system that continuously detects the temperature. As long as the core temperature reaches a certain level, the system will reduce the operating frequency of the processor until the core temperature returns below the safe limit. This is the real reason for the performance degradation of the system. At the same time, this also explains the importance of the radiator. It is recommended to give priority to some brand radiators, but they also have grades. When purchasing, you should pay attention to the maximum CPU frequency that you can support, and then take the medicine according to your own C P U. Case 6: The host failure phenomenon that is constantly restarting: The fixture of the CPU heat sink is removed by mistake. Then install it as it is. Re-install the fan power-on evaluation, the result is just boot, the computer will automatically restart. Checking other parts is fine. According to conventional experience, it should be a problem of heat dissipation. It is possible that the motherboard automatically detects the CPU overheating and automatically protects it. However, it is no problem to repeatedly check the thermal grease and the heat sink. Reinstall it or restart it repeatedly. After replacing the cooling fan, everything is OK. Is there a problem with the heat sink? After repeated comparisons, it was finally found that the direction of the buckle was reversed, causing a gap between the heat sink and the core of the CPU, and the CPU overheated, causing the motherboard to be restarted. The original CPU cooling fan is not properly installed, which will cause Windows to automatically restart or fail to boot. Fault analysis: With the continuous improvement of process and integration, CPU core heating is a serious problem, so the current CPU requirements for cooling fans are getting higher and higher. The problems caused by improper installation of the cooling fan are quite common and frequent. If you are using a CPU such as a Pentium 4 or Athlon, choose a CPU fan with a good quality and be sure to pay attention to the correct installation method. Otherwise, the lighter is the machine restart, and the heavy CPU is burned. Conclusion: In fact, the above mentioned faults are not terrible, most of them are caused by carelessness. Common faults are mainly concentrated on heat dissipation and frequency.. Just be careful to avoid similar problems. No power, is the motherboard broken? In the process of repairing the computer, we always encounter such a problem: after a computer presses the power switch, there is no movement, the fan does not turn, and the power indicator does not light. In this case, we have encountered a lot of problems. First check the power supply, then check the motherboard, and use minimal systemization to add the components one by one. When the component is checked, the above situation occurs, which indicates that there is a problem with the component. Some time ago, the author encountered a problem with the motherboard not charging, check it out, the power is good, the motherboard is good, even the CPU is bad. If the CPU is broken, it can also cause the motherboard to not be powered. The situation is this, a customer sent a computer to the company, saying that the computer is broken, can not open the machine. Power on the host, indeed can not boot. First remove all power plugs, short circuit green line and black ground, the power supply works normally. It is also normal to test the voltage output of each channel with a multimeter. There is no problem with the power. Then use the minimum system method, unplug the memory, CPU and graphics card, sound card, hard disk, optical drive, floppy drive, only power on the motherboard. As a result, the motherboard did not respond after pressing the power switch. Forced to power up, or add power, the motherboard is broken, too simple. Fortunately, the customer bought the machine for a short time, and went to the warehouse to replace a new board for the customer. I went to the second, and then I installed the machine, connected the power supply, and turned it on. Unexpectedly, and the situation just now, there is nothing moving. The new motherboard can't be bad? How is this going? I double-checked the CMOS discharge jumper on the motherboard, in "Normal", there was no problem. If the CMOS discharge jumper is in the “Clear” position, most motherboards cannot be powered up and there is no response, just like the motherboard is broken, except for some 3D motherboards. No way, I will re-open the machine and then eliminate it one by one according to the minimum system method. Powering the motherboard separately, normal, there is no problem with the motherboard. Plug in the CPU, power up, no response. The CPU is broken? Plug the CPU into another motherboard and you cannot power it up. The computer's fault is really strange, the CPU can also be short-circuited! Tip: During the process of repairing the computer, we sometimes encounter the indicator light on the motherboard, but when the machine is not open, you can force the motherboard to power up, use a wire with a plastic sleeve, directly short the green black, so that the motherboard does not It can be powered directly after the power-on circuit. If the host starts normally, it indicates that there is a problem with the boot circuit on the motherboard. We can directly short-circuit the green and black emergency method through the external power switch without having to replace the motherboard, which can save a lot of expenses. This article comes from [System Home] www.xp85.com