See if this can help you.
Crash is something that annoys the operator. When the computer crashes, the performance is mostly "blue screen", the system can't start, the picture is "frozen" and unresponsive, the mouse and keyboard can't be input, and the software operation is abnormally interrupted. Although there are many reasons for the crash, they are all the same, and the reasons can never be separated from both hardware and software.
Hardware-induced crash.
Heat dissipation difference
Monitor, power supply and CPU will generate a lot of heat in their work, so it is very important to keep good ventilation. If the monitor is overheated, it will lead to color and image distortion and even shorten the life of the monitor. Working too long will also lead to poor heat dissipation of power supply or monitor, leading to computer crash. The heat dissipation of CPU is an important issue related to the stability of computer operation, and it is also the "hardest hit" where heat dissipation failures occur.
Inappropriate exercise
When the computer is moving, it often makes the internal devices of the machine loose, resulting in poor contact and computer crash. Therefore, violent vibration should be avoided when moving the computer.
Dust killer
Too much dust in the machine can also lead to collapse. If the floppy disk drive head or the laser head of the CD-ROM drive is contaminated with too much dust, it will lead to reading and writing errors, which will seriously cause the computer to crash.
Device mismatch
If the main frequency of the motherboard does not match the main frequency of the CPU, the external frequency will be set too high when the old motherboard overclocks, which may not guarantee the stability of operation and lead to frequent crashes.
Software and hardware are incompatible.
Three-dimensional software and some special software may not be started or even installed normally on some microcomputers, and there may be compatibility problems between software and hardware.
Memory failure
It is mainly caused by loose memory chips, virtual welding or quality problems of memory chips themselves. The contact fault of memory stick should be eliminated according to the specific situation. If there is a problem with the quality of the memory, the memory needs to be replaced to solve the problem.
Hard disk failure
Mainly due to hard disk aging or improper use of bad tracks and sectors. This makes it easy for the machine to crash during operation. Special tools and software can be used for troubleshooting. If the damage is serious, only the hard disk can be replaced. In addition, for motherboards that do not support UDMA 66/ 100, we should pay attention to the setting of hard disk operation mode in CMOS.
CPU overclocking
Overclocking improves the working frequency of CPU, but it may also make its performance unstable. The reason is that CPU can access data in memory faster than memory and hard disk can exchange data. Overclocking makes this contradiction more prominent, exacerbates the situation that the required data cannot be found in memory or virtual memory, and will lead to "abnormal errors". The solution, of course, is also relatively simple, that is, let the CPU return to normal frequency.
Hardware resource conflict
The abnormal error is caused by the conflicting settings of sound card or video card. In addition, if other devices interrupt, DMA or port conflict, a few drivers may be abnormal, leading to crash. The solution is to start in "Safe Mode" and make appropriate adjustments in "Control Panel →→→→→→→ Equipment Management". If the driver has an abnormal error, you can modify the registry. Select "Run", type "REGEDIT", enter the registry editor, find and delete all the "primary keys" and "key values" related to the driver prefix string through the "find" function under the menu, and restart.
Insufficient memory capacity
The larger the memory capacity, the better, and it should not be less than 0.5 ~ 1% of the hard disk capacity. If there is a problem in this respect, you should replace it with a memory stick with as large a capacity as possible.
Inferior parts
When assembling compatible computers for customers, a few unscrupulous businessmen use inferior boards and memory, and some even sell fake motherboards and reused CPU and memory. Such a machine is very unstable and will inevitably crash. Therefore, users should be vigilant when purchasing a machine. They can use some newer tools and software to test the computer and test the machine continuously for a long time (for example, 72 hours) to strive for the longest warranty time.
Software-induced crash.
virus infection
Viruses will greatly reduce the efficiency of computers and lead to frequent crashes. At this time, you need to use anti-virus software such as KV300, Kingsoft Internet Security and Rising. Conduct comprehensive virus detection and anti-virus, and regularly upgrade anti-virus software.
CMOS settings are incorrect.
This kind of fault phenomenon is very common, such as improper hard disk parameter setting, mode setting and memory parameter setting, which leads to the computer being unable to start. If the memory without ECC function is set to have ECC function, it will crash due to memory error.
Misdeletion of system files
Command.com, Io.sys, Msdos.sys and other files are required to start Windows 9x. If these files are damaged or deleted by mistake, even if the hardware settings in CMOS are correct, it will not help. Solution: Use the startup disk of the same version of the operating system to start the computer, and then type "SYS C:" to retransmit the system files.
Initialization file is corrupted.
Because Windows 9x needs to read System.ini, Win.ini and registry files, if there are Config.sys and Autoexec.bat files, they will also be read. As long as these files have error information, they may crash, especially the four files of System.ini, Win.ini, User.dat and System.dat are particularly important.
The dynamic link library file (DLL) is missing.
Another kind of file is also very important in the Windows operating system, that is, the dynamic link library file with the extension of DLL is essentially a * * * file, which means that a DLL file may need multiple software calls at runtime. If we delete an application, the anti-installation program of the software will record the files it has installed and prepare to delete them one by one. At this time, the deleted dynamic link library file is easy to be used by other software at the same time. If the missing link library file is an important core link file, the system will crash or even crash. We can use tools such as "Super Rabbit" to delete useless DLL files, which will avoid accidental deletion.
There is too little space or too much fragmentation left on the hard disk.
If the remaining space of the hard disk is too small, some applications need a lot of memory to run, which requires virtual memory, which is provided by the hard disk, so the hard disk should have enough remaining space to meet the demand of virtual memory. At the same time, users should form a good habit of regularly sorting out hard disks and removing junk files on hard disks.
BIOS upgrade failed.
You should back up the BIOS to prevent accidents, but if your system needs to upgrade the BIOS, you'd better make sure that the BIOS version you are using is the same as your PC before upgrading. If the BIOS upgrade is incorrect or an unexpected power failure occurs during the upgrade, your system may not start. So be sure to find out the BIOS model before upgrading the BIOS. If the BIOS upgrade tool you are using can back up the current BIOS, please copy the previous BIOS on the disk. At the same time, see if the system supports BIOS recovery and know how to recover.
Incorrect software upgrade
Most people may think that there will be no problem with software upgrade. In fact, some components will be upgraded during the upgrade process, but other programs may not support the upgraded components, resulting in various problems.
Abuse of beta software
It is best to use the beta version of the software less, because the test software usually has some bugs or is not stable enough in some aspects, and after use, there will be program errors, crashes or the system can't start.
Illegal uninstallation of software
Do not directly delete the directory where the software is installed. If you delete it directly, there will be a lot of garbage in the registry and Windows directory. Over time, the system will become unstable, leading to a crash.
Use pirated software
Because these softwares may hide viruses, once executed, they will automatically modify your system, causing the system to crash during operation.
Defects in application software
This situation is more common, such as running 16-bit application software running well under DOS or Windows 3. 1 in Win 98. Win 98 is 32-bit. Although it claims to be compatible, there are many places that cannot be coordinated with 16-bit applications. There are also some cases, such as peripheral drivers that are normally used under Win 95. When the operating system is upgraded, problems may occur, causing the system to crash or fail to start normally. In this case, you should find a new version of the peripheral driver.
Too many programs have been started.
This makes the system resources exhausted, so that the data needed by individual programs can not be found in memory or virtual memory, and abnormal errors will also occur.
illegal operation
Opening or releasing related programs with illegal format or parameters can also lead to computer crash. Please remember the correct format and related parameters, and don't open and publish unfamiliar programs at will.
Abnormal shutdown of computer
Do not directly use the power button in the chassis, otherwise the system files will be damaged or lost, resulting in automatic startup or crash at runtime. For Windows 98/2000/NT and other systems, this is very important, and if it is serious, it will lead to system crash.
Conflict in memory
Sometimes it is normal to run all kinds of software, but suddenly it crashes inexplicably. It is very normal to run these applications after restarting. This is a false collapse phenomenon. Most of the reason is the memory resource conflict of Win 98. As we all know, application software runs in memory, and memory space can be released after closing the application software. However, due to design reasons, some applications can't completely release memory even after they are closed. When the next software needs to use this memory address, a conflict will occur.
The most pitiful people in the computer market are those who move faulty computers around. Although the assembly machine has her advantages over the brand machine, the lack of good after-sales service is her achilles heel.
Most computer failures are software failures, so there is no need to move the whole machine around until it is determined that it is a hardware failure. Even if the hardware fails, you don't need to move the whole machine, just take the faulty parts. Therefore, it is necessary for us to understand the diagnosis and testing methods of hardware faults, so that we don't have to move the whole machine when the computer fails in the future (of course, it is best not to fail! )。
Faults and countermeasures
If you want to make a comprehensive analysis of computer failures, I'm afraid you have to write a book, and it's not easy. Software failures have been introduced in various newspapers and periodicals. Here, only the symptoms, causes and solutions of hardware failures are introduced.
First, the computer startup process
I, first of all, let's learn about the computer startup process.
1. The boot loader and self-diagnosis program stored in ROM are moved to ram.
2. With the boot loader running, the operating system stored in the auxiliary storage device sends the system file to RAM.
3. Execute the system files Io.sys and Msdos.sys, and the message "Starting Windowsn98…… ..." will appear on the screen.
4. If there is Config.sys, execute it.
5. Realize the Command.com file of the system file.
6. If Autoexec.bat exists, execute it.
7. Read the initialization files "System.ini" and "Win.ini" of Windows, and then read the registry file.
8. After startup, the initial screen appears and the operating system is running.
In this process, monitor whether the hardware in the ROM BIOS of the motherboard is abnormal, including hardware failure, wiring, installation of various cards, etc. If an error occurs, nothing is displayed on the screen and the startup stops.
In this case, it is probably a hardware failure.
Ⅱ. System startup sequence
1, PC power on-the lights on the monitor, keyboard and chassis are flashing.
2. Check the video card-a short video card message appears on the screen.
3. Check the memory-with the beep, the memory capacity information appears on the screen.
4. Execute BIOS- a short BIOS message appears on the screen.
5. Detect other devices-information of other devices (CPU, hard disk, MEM ...) appears.
6. Execute the initialization file of OS (operating system)-start Windows 98, etc.
ⅲ. The motherboard makes a sound when it starts, and you can judge what kind of error it is through the sound.
According to the different forms of the motherboard, the expression of sound is also different.
AMI BIOS:
1 Short: Memory refresh failed.
2 short circuit: memory check error
3 Short: Basic memory error
4 Short circuit: System clock error.
5 short: CPU error
6 Short: Keyboard error
7 Short: Real Mode Error
8 Short: Memory display error
9 short: ROM BIOS check error.
1 Long 3 Short: memory error.
Bonus BIOS:
1 Short: Normal startup
2 Short: Non-fatal error
1 long 1 short: display error.
1 long 2 short: keyboard error.
Other BIOS can be found in the BIOS error tone, which will not be described in detail here.
Second, confusing software failures.
You must clearly distinguish between hardware failure and software failure, otherwise it will take a lot of effort to move the computer to the computer market, and how do you feel when the shopkeeper tells you that it is a software failure? In particular, startup failure may also be caused by software failure. Let's take a look at the startup exception caused by software failure.
Ⅰ. Errors in Ⅰ.CMOS Settings
If the hard disk in CMOS settings is set incorrectly, the computer can't start with the operating system (Windows) in the hard disk because it can't recognize the hard disk. When the screen appears but can't start, you should check the contents of CMOS settings. To correctly identify the hard disk, you can use the "IDE hard disk automatic detection" option in CMOS settings.
Ⅱ. Errors in system files
Command.com, io.sys, msdos.sys and drvspace.bin are required when Windows starts. If these files are damaged, even if the hard disk is recognized, it cannot be started. At this point, you can use the Sys.com file to recover these files. After starting with the startup disk, just type "Sys c:".
Ⅲ. Error initializing file
Windows should read six files at startup: Autoexec.bat ","Config.sys, System.ini, win.ini, user.dat and system.dat. But when reading, if there is wrong information in it, it will fail to start. These files are difficult to recover, so methods such as Windows reset should be used. But this is not a hardware failure, there is no need to take the computer to the computer market.
Fourth, Windows errors Most of the failures after the initial Windows screen appear are software failures. Conflicts between programs or problems with drivers, etc. Such problems can be solved by reading books and so on.
Third, a hardware fault that is not a fault.
Although it is not a fault, it often happens that the user assembles incorrectly or the socket is loose. At this time, you can turn on the computer to check wiring, sockets and other errors. When purchasing new EIDE devices, such as hard disks and CD-ROMs, it is important to set the middle connected device as "SlaveE" and the side connected device as "Master". If the settings are incorrect, you may not be able to start or make an error when using the corresponding device.
Fourth, hardware fault detection methods
Let's take a look at the basic test method of hardware failure. When there is no image on the display, the following methods can be used to test the faulty component.
First, prepare a workbench.
Second, pull out the motherboard from the chassis, and then pull out all the components on the motherboard, leaving only CPU and RAM. Then put the motherboard on the workbench.
Third, connect the regulated power supply to the motherboard.
Fourth, insert the graphics card into the AGP slot. Of course, if it is a PCI graphics card, it is inserted in the PCI slot. When inserting, pay attention to completely insert the gold-plated part of the graphics card into the slot.
5. After connecting the power socket of the monitor, connect the video card to the monitor.
Six, turn on the monitor power supply, and then turn on the chassis power switch. Then touch the power switch of the motherboard with a metal bar.
The power switch of the motherboard is the part connected with the chassis power switch, which is generally marked as "PWR SW" or "POWER SE".
7. If the BIOS version information appears on the screen and the screen is normal, it means that the CPU, motherboard, RAM, graphics card and power supply are all normal. Usually, the components that are prone to failure are "graphics card", "motherboard" and "hard disk" in turn.
Eight, then connect the hard disk and soft area for testing. Then connect the optical drive for detection, and then the sound card. Modems and other connections are tested one by one. If the screen does not appear, it means that the components connected later are faulty or have compatibility problems. Just deal with the failed component.
Nine, the chassis problem
Sometimes there is a problem when the motherboard is loaded into the chassis, which leads to the failure of startup. Therefore, if there is no problem with the above components, the motherboard can be installed on the chassis for testing. If there are no errors in the test, there are software problems such as CMOS setting errors and drivers.
A Simple Method for Detecting Computer Faults
If the "false fault" is excluded, then there is a real fault! If you check the appearance of the accessories again, including opening the chassis, see if there are any signs or obvious scars on the surface of the components inside the host. If there are, the faulty components will be clear. If not, try the following treatment first.
1. Dust removal
Dust floating in the air is a big killer of computers. After using it for a period of time, it may fail because of too much dust on key components such as the motherboard, even in a special computer room. Therefore, computers that have been used for a long time should be cleaned first, and the dust on the motherboard and peripherals should be gently brushed off with a brush. If the dust has been removed, or there is no dust, the fault still exists, indicating that there are other problems with the hardware.
In addition, because some cards or chips on the circuit board exist in the form of pins, vibration, dust and other reasons often cause pin oxidation and poor contact. Eraser can be used to wipe off the oxide layer on the surface. After reconnecting, start the machine to check whether the fault is eliminated.
By the way, when the keyboard is used for a long time, there will often be failures such as missing keys and stuck keys, which should be dealt with in time, otherwise some words will be typed incorrectly when inputting files. When handling, the keyboard should be supported by a bracket, with the keys facing down. Open the back cover of the keyboard, clean the contact between the circuit board and the keys with alcohol, and pry up the shrapnel under the stuck keys properly to restore its original elasticity.
Note: When a dirty or scratched floppy disk is inserted into the floppy drive, it will scratch the read-write head and damage the floppy drive. Be very careful when cleaning the magnetic head. Floppy drives that are not used for a long time may rust on the magnetic head. At this time, cleaning disks are not allowed. The specific method is to open the case and drop the cleaning agent on the magnetic head. After soaking for half an hour, wipe it carefully with absorbent cotton. If the cleaning disk is used blindly, it will inevitably lead to the damage of the read-write head of the floppy drive and the floppy drive will be scrapped.
2. Look, listen, smell and touch
"Look" is to observe whether the plug and socket of the system board are skewed, whether the resistor and capacitor pins collide, whether the surface is burnt, whether the chip surface is cracked, and whether the copper foil on the motherboard is burnt out. Of course, it goes without saying that you also know to check whether there are foreign objects falling between the components of the motherboard (which will cause short circuit), or whether there are burnt and discolored places on the board, whether the traces (copper foil) on the printed circuit board are broken, and so on.
"Listening" refers to monitoring whether the working sound of power supply fan, soft/hard disk motor or seek mechanism, display transformer and other equipment is normal. In addition, the short-circuit fault of the system is often accompanied by abnormal sound, so monitoring can find some hidden dangers in time and take measures immediately before the accident.
"Smell" is to identify whether the main engine and the board have burnt smell, which is convenient for finding faults and determining the short circuit position.
"Touch" is to press the movable chip of the socket by hand to see if the chip is loose or in poor contact. In addition, when the system is running, touching or approaching the shell of CPU, monitor, hard disk and other equipment by hand can judge whether the equipment is running normally according to its temperature; Touch the surface of some chips with your hands. If it is hot, the chip is damaged.
3. Blockage detection
As mentioned above, there are many reasons for computer failure, such as motherboard failure, I/O bus failure, various card failures, etc., which may lead to abnormal system operation. Unplugging maintenance method is a simple method to judge whether the fault occurs on the motherboard or I/O equipment. This method is to pull out the patch panels one by one after shutdown, and turn on the computer to observe the running state every time one panel is pulled out. Once a board is pulled out, the main board runs normally, and the fault reason is the fault of the patch panel or the corresponding I/O bus slot and load circuit. If the system still can't start normally after all the plug-in boards are pulled out, then the fault is probably on the motherboard.
When plugging and unplugging, we can also eliminate computer faults from another aspect: some chips and boards have poor contact with slots, and correctly plugging these chips and boards back can solve the failure of microcomputer components caused by improper installation contact.
4. Exchange detection
Swap chips of the same type, the same bus mode, the same function or the same type, and judge the fault according to the change of fault phenomenon. This method is mostly used in easy-to-plug maintenance environment. For example, if there is an error in the memory self-test, the same memory stick or memory stick can be replaced to determine the fault location, and the fault phenomenon remains unchanged when the non-fault chip is replaced. If the failure phenomenon changes after replacement, it means that one of the replaced chips is bad, and the position can be further determined by replacing one by one. If we can find the same type of microcomputer components or peripherals, we can quickly determine whether it is the quality problem of the components themselves by replacing them.
5. Comparative detection
Run two or more computers of the same or similar type. According to the different performances of the normal computer and the fault microcomputer in the same operation, the fault position can be preliminarily judged.
6. Vibration knock detection
Tap the shell with your finger. If the fault is eliminated, it means that the fault is caused by poor contact or virtual welding. Then, the location of the fault point can be further checked and eliminated, but it is difficult to detect the exact location of this fault.
7. Heating and cooling detection
By artificially raising the temperature of the microcomputer operating environment, the high temperature resistance of all components, especially the CPU, can be tested, so as to find out the hidden dangers of accidents as early as possible. If the failure rate is greatly reduced after reducing the operating environment temperature, it means that the failure is caused by high temperature or components that cannot withstand high temperature, which is helpful to narrow the scope of fault diagnosis.
In fact, the heating and cooling method is based on the principle of fault initiation, and the fault location is observed and judged by creating conditions for the fault, thus promoting the frequent occurrence of faults. However, it is necessary to pay attention to controlling the heating mode in the specific implementation, and the temperature should not exceed 40 degrees Celsius.
8. Run the detection program
With the wide application of various integrated circuits, the welding process is becoming more and more complicated, and it is often difficult to find faults only by general maintenance means. However, through random diagnosis program, special maintenance diagnosis card and self-made special diagnosis program, it can often get twice the result with half the effort according to various technical parameters (such as interface address). The principle of program testing is to use software to send data and commands, and identify the fault part by reading the state of the circuit and the state of a chip (such as a register). This method is often used to check all kinds of interface circuit faults and circuits with address parameters, but the premise of application is that CPU and bus are basically running normally and can run relevant diagnostic software and diagnostic cards installed on I/O bus slots.
When selecting the diagnosis program, it should be strict, comprehensive and targeted, which can make some key parts have regular signals, test accidental faults repeatedly and display error records.
How to diagnose computer faults
Many novices will have a sense of fear when they first come into contact with computers, thinking that computer failure must be an insurmountable problem. In fact, most computer failures have certain rules to follow, and this problem is like a piece of enough paper, which will be broken when poked, without much computer knowledge. Let's learn several ways to quickly diagnose computer faults.
Environmental inspection method
For some sudden hardware failures, such as boot without display. Don't think deeply yet, because we tend to overlook some details. The first thing to look at is the obvious: what if the power is turned on? Is the switch on? Is the power socket electrified? Are all the wires connected? Perhaps this is the root of the problem.
CMOS reduction method
Some users often change some settings in the motherboard CMOS out of curiosity, which is a main reason for the failure. If the computer breaks down, we can solve the problem by restoring CMOS settings. The method is simple. Press the "Delete" key on the keyboard to enter the CMOS of the motherboard, select "Load optimized default values", press the "Y" key to confirm, and then save and exit the CMOS.
Registry recovery method
Some users like to optimize or personalize the system by modifying the registry, and some users have their registry changed by malicious programs when surfing the Internet. Some failures are caused by abnormal changes in the registry. At this time, we can restart the computer, switch to MS-DOS mode, enter and execute "scanreg/Restore" in the root directory of disk C to enter the registry recovery interface, and then select a registry file and execute "Restore" to restore the registry when the computer is in good condition.
Simplified starting method
After we installed some software, some computers broke down. If the computer can still enter the operating system at this time, we can run the "msconfig" program in the start menu and close all programs except "internat.exe, Scanregistry and Systemtray" in the start menu. If the fault does not appear after restarting the computer, then the problem is mostly caused by a self-starting software.
Record tracking method
If the computer can't enter Windows or work normally after entering Windows, then we can start the computer through Logged(\Bootlog.txt), so that the generated Bootlog.txt file can record the location of the fault. The method of using login mode to start is to press F8 key on the keyboard when the system starts, and a startup menu will appear. Select to start in recording mode. After a failure occurs, restart the computer with the Windows startup disk, and then copy the Bootlog.txt file under the root directory of disk C to the floppy disk, and open the file on other computers. You will find that it records the whole process of Windows startup, from which you can find the root of the problem.
Equipment replacement method
The so-called equipment replacement means that when you suspect that there is something wrong with a certain equipment, replace it with a device with the same function (preferably the same model). If the problem disappears after replacement, it is likely that there is something wrong with the equipment.
Minimum system method
If you are not sure which hardware is out of order, you can use the least system method to judge. The least system method is to remove other hardware devices in the system, leaving only the three most basic components: motherboard, memory and graphics card, and then turn on the computer to see if there is any fault. If so, other hardware problems can be eliminated, and the fault should come from the existing three hardware. If not, add other hardware one by one, check which hardware is faulty after adding, and then deal with this hardware after finding the fault.
Program upgrade method
Many people don't pay enough attention to the driver and think that installing one will do. However, when we bought the hardware, we already had a driver. Why do hardware manufacturers keep releasing new versions of drivers? In fact, the purpose of doing this is to make the manufacturer's own products more perfect.
Because the current hardware update speed is very fast, and the hardware research and development of most hardware manufacturers precedes the software research and development, there may be some minor bugs in the drivers that match the hardware when they are just released, so it is necessary to constantly update the drivers to make up for these defects. Therefore, upgrading the driver is also an effective way to solve the hardware failure.
Software testing method
Diagnosing hardware faults usually requires knowing some hardware information, but many people don't have the habit of recording hardware information or don't know how to record it. After the computer breaks down, you may not be able to enter the system. At this time, we need a tool to test hardware under DOS, such as HwInfo for DOS, which is only 582KB in size and can be carried around on floppy disks. With its help, we can diagnose hardware faults at any time.
Change the law of resources
Many computer failures are caused by resource conflicts between hardware, and we can solve them by replacing resources. Right-click My Computer, select Properties from the drop-down menu, click Device Manager, and select View Devices by Type. If a device is found in the list marked with a yellow exclamation mark, it is likely that there is a resource conflict between hardware. The way to change resources is to double-click the hardware marked with exclamation mark with the left mouse button, select Resources, remove the check box before Use Automatic Settings, and select Change Settings to change the conflicting resources.