subject: Laptop CPU Failure Analysis and Solution [print this page] Laptop CPU Failure Analysis and Solution Laptop CPU Failure Analysis and Solution
CPU fault common to the following categories: cooling failure, restart failure, black failure and overclocking failure. Failure of the CPU itself, the probability is very small, so most of the failures are caused due to user carelessness.
Case I: CPU pins is bad, cause the machine to not start
Symptom: Athlon CPU a user of a computer, daily use has been normal, one day suddenly will not boot, no screen display signal output, began to identify graphics card failure. After checking with the substitution and found no problem card, and later speculated that display failure, after inspection, the display also all normal. Apart from wondering, unplug the plug on the motherboard's CPU, carefully observe the no burn marks, but just could not lit machine. CPU's pins are later found black hair green, with traces of oxidation and rust (CPU's pins for the copper material, the outer gold), then on the CPU pins with a toothbrush to do the cleaning work, computer work and can power the.
Failure Analysis: CPU rust solve the problem, but how come what rust. The last piece fell into doubt chip cooling, refrigeration talked about my previous article, the phenomenon of film condensation, probably because of chip cooling down the chip's surface temperature too low, the dew point lower than the end, lead to long-term work in a damp environment CPU in. The bare copper pins in this environment, oxygen in the air in the reaction of the patina. Too bad cause rust spots over time, giving rise to the peculiar problems. There are also some bad motherboard, the CPU slot of poor quality, can also cause bad contact, users need their own CPU and socket contacts fixed before the problem.
Case II: "cold" work can also be burned CPU
Symptom: I have a friend do such a test, the desktop Celeron 566 processor running at standard frequencies (no overclocking), heated to 55 degrees Celsius through the hair dryer (with the motherboard temperature monitoring functions are), just run CPU occupancy rate of the procedure, one will crash; while the Celeron 566 overclocked to 850MHz, the system temperature is 50 degrees Celsius, ten minutes before running Quake III crashes. At this time the temperature has been estimated that more than 55 degrees Celsius, while the core temperature by measuring, that has reached 86.4 degrees Celsius. Later found in such low CPU running was almost burned. But he found that laptops are no such "duplicity" of the problem.
Failure Analysis: It turned out that the motherboard detects CPU temperature confuse us. In fact, the Desktop Board reported its CPU core temperature, the temperature is not as common desktop board temperature sensor and the CPU did not touch the heat sink or CPU, measure only the air temperature near the CPU. This appears to cause a lot of CPU temperature in case of fire. Intel released data from the point of view, Pentium 550E at 85 degrees C temperature limit, Pentium 800E of the limit temperature at 80 degrees Celsius. If we lose our vigilance, partial report of the letter board, thinking that their CPU is also running at low temperature, it would be wrong.
Why this difference does not appear laptop? The original notebook is used on the CPU temperature thermistor temperature measurement point at the bottom of the CPU, if the direct reading, in fact, is not so high temperature, and its display monitor temperature after a correction, than measurements of the temperatures are high, so closer to the CPU's core temperature. So most of the test notebook CPU core temperature is the temperature, low temperature burning CPU will not be the case.
Symptom: general system hangs and will not cause CPU burn, the system will automatically reduce the CPU operating frequency and fan speed to save energy. The suspend mode mentioned here was burned causing CPU are all overclocked CPU. Perhaps you will find this a bit weird, why the overclocked CPU burned? This is all caused because the fans stop. It turned out that the monitoring chip on the motherboard to monitor fan speed in addition, some have also entered in the system Suspend (suspend) power-saving mode automatically reduces the fan speed or even completely stop functioning, this is a well-intentioned, can be power, but also can extend the life of the fan with the use of time. CPU is idle the past, the heat is not high, so the fan does not turn, can only cope with the heat sink cooling *. But now the CPU frequency is too high, even into the suspend mode, when the fan does not turn when, CPU will be red hot. So some people will encounter, when transferred to normal mode from suspend, Windows
98 crashes and blue screen, this is the CPU overheating error. Serious, CPU will overheat and hang up, especially after overclocking Thunderbird or Duron.
Failure Analysis: This situation will not occur in each board, the event must meet three conditions. 3pin CPU fan to be the first fan, this will be controlled by the motherboard. Second, the board's monitoring function must have the Fan Off When Suspend (suspend mode or turn off the fan power), and this feature is default On. Some motherboards default On, and even some in the Power Management settings have Fan Off When Suspend this one option, we can pay attention to see. Third, enter the suspend mode. So, now control their own computer to check it.
Case IV: CPU frequency Troubleshooting
Symptom: There is a computer's CPU for the AthlonXP 1600, after the BIOS boot appears as 1050MHz, but the normal frequency AthlonXP 1600 should be 10.5 133MHz FSB = 1400MHz frequency. The largest in the BIOS can only be found in the FSB set to 129MHz, disassemble found that DIP switch to the motherboard 100MHz FSB, so be adjusted to 133MHz FSB, black screen after boot, CPU fan is running. Were repeated several times is so, then DIP switches on the motherboard then all the notes of Auto, the default, the system self-test is still 1050MHz. Suspected memory and graphics cards are not synchronized, reducing the memory CAS from 2 to 2.5, still not normal self; in turn change the 2X AGP 4X graphics mode from the boot back to normal.
Failure analysis: proven later, this older version of the user's card, the default AGP frequency is 66MHz (at 100MHz under, PCI operating frequency is 100 3 = 33.3MHz, AGP is a PCI 2 = 66.6MHz, in Under the AGP 133MHz FSB frequency of 133 3 2 = 88.7MHz), used as AthlonXP 133MHz FSB, AGP then the operating frequency of up to a 88.7MHz. Therefore, the display card by the apparently blank, would reduce the frequency of card, the system returned to normal.
I often see the CPU on the network caused by the abnormal frequency of failure, some early Pentium
or Athlon 100MHz FSB motherboard is the default, and now the new core of the CPU are 133MHz FSB. This automatic detection in the case of the motherboard, CPU are being down-use, generally tend to not be found. Encountered such a situation simply by adjusting the FSB and the memory card or the work can be asynchronous.
Case V: Mystery of PC performance
Symptom: a Pentium 4 computer performance unusually stable during the initial period, but then it seems infected with a virus, the performance dropped significantly, and occasionally accompanied by backing out. There is no first-use of found Avira antivirus software. Doubt caused by excessive disk fragmentation and then, with Windows Disk Defragmenter to organize, the problem remains. Also considered to be a problem with Windows, format reinstall the system, still no effect. Open the case that CPU heat sink fan problems after power does not turn. Replaced with new radiators, troubleshooting.
Failure Analysis: The original Pentium 4 processor core is equipped with a thermal control monitoring system, it will continue to test the temperature. As long as the core temperature reaches a certain level, the system will reduce the processor's operating frequency, until the core temperature to a safe limit below. This is the real reason for slow system performance. This also shows the importance of the radiator, radiator recommend some brands of priority, but they also have hierarchy, should pay attention to the purchase of its CPU can support the maximum frequency is how much CPU each other and then fill a prescription based on their own.
Case VI: The host continuously restart
Symptom: a CPU heat sink clip mistakenly knocked off. Later, as it was to install back to the heat sink clip. Re-installed fan power evaluation, the results of just getting machine, the computer automatically restarts. Check the other parts no problem, according to the experience should be a routine part of the problem of heat dissipation. May be detected by the motherboard CPU overheating, automatic protection. But the repeated inspection and heat sink thermal grease have no problem, reinstall the back or a restart. Replace the cooling fan, everything OK. Is the heat sink problem, after repeated contrast was finally discovered that the direction is the clip installed backwards. Result, the core CPU heat sink in contact with the gap, CPU overheating, motherboard detection CPU overheating, restart protection. The original CPU cooling fan installed properly, can also cause or fail to boot Windows automatically restarts.
Failure Analysis: CPU technology and integration with the continuous improvement of core fever is a more serious problem, so the current CPU cooling fan on the requirements of increasingly high. Cooling fan problems caused by improper installation is quite common and frequent. If you are using a Pentium 4 or Athlon like the CPU, please select the excellent quality of the CPU fan, and must pay attention to the proper installation methods. Noir is a light machine or reboot, re-Noir CPU burned.
Conclusion: All the above fact, these faults are not dreadful, mostly caused by user carelessness. Troubleshooting mainly concentrated on the heat and the frequencies, as long as you can be just as careful to avoid similar problems. Of course, we hope that we can learn to solve the CPU and the way the idea of failure, so that in case of CPU failure will be able to calmly deal with.
