CompTIA A+ Core Certification: The Basics of Hardware and Network Troubleshooting - Troubleshooting Motherboard, RAM, CPU, and Power Issues

220-1101: Objective 5.2: Given a scenario, troubleshoot common problems related to motherboards, RAM, CPUs, and power.
Many system problems are caused by bad motherboards, RAM, CPUs, and power. In the following sections, you learn about common symptoms for these problems and the most likely causes. Use this information as you track down real-life issues in your company’s systems and your clients’ systems.

POST Code Beeps
The Power-On Self-Test (POST) code beeps are used by many BIOS versions to indicate either fatal or serious errors. Beep codes vary by the BIOS maker. Although some vendors create their own BIOS chips and firmware, most major brands of computers and virtually all “clones” use a BIOS made by one of the following vendors: American Megatrends (AMI), Phoenix Technologies, IBM, Award Software (now owned by Phoenix Technologies), or Insyde Software.
As you might expect, the beep codes and philosophies used by these companies vary a great deal. For example, AMI uses beep codes for more than 10 fatal errors. It also uses 8 beeps to indicate a defective or missing video card. Phoenix uses beep codes for both defects and normal procedures (but has no beep code for a video problem), and the Award BIOS has only a single beep code (1 long, 2 short), indicating a problem with video. Insyde BIOS uses beep codes for errors, but these codes vary widely from model to model.

Note: Some vendors have switched from beep codes to blink codes with the advent of UEFI BIOS firmware. Check the documentation for the system or motherboard to determine whether beep, blink, or other reporting methods are used to indicate POST problems.
Because beep codes do not report all possible problems during the startup process, you cannot rely exclusively on beep codes to help you detect and solve system problems. In addition, beep codes can be heard only on systems with built-in speakers.
Tip: To add a wired speaker to a desktop computer, plug it into the speaker jack in the front-panel header pins.


Table: Common System Errors and Their Beep Codes
 

For additional beep codes and other BIOS support, see the following manufacturer resources:
- AMI BIOS: https://ami.com
- Phoenix BIOS: www.phoenix.com
- IBM, Dell, Acer, and other brands: www.bioscentral.com and www.wimsbios.com

Note: Don’t mix up your boops and beeps! Many systems play a single short boop (usually a bit different in tone than a beep) when the system boots successfully. This is normal.

POST Error Messages
Most BIOS versions do an excellent job of displaying POST error messages indicating the problem with the system. These messages can indicate problems with memory, keyboards, hard drives, and other components. For example, if the CMOS memory used to store system setup information is corrupt (possibly because of a battery failure or because the CMOS memory has been cleared), systems display a message such as the following:
- System CMOS Checksum Bad–Run Setup: Phoenix BIOS
- CMOS Checksum Invalid: AMI BIOS
- CMOS CHECKSUM INVALID–RUN SCU: Insyde BIOS
- CMOS Checksum Error–Defaults Loaded: Award BIOS
Some systems document these messages in their manuals, or you can go to the BIOS vendors’ websites or the third-party sites listed earlier in this guide for more information.

Note: Keep in mind that the system almost always stops after the first error, so if a system has more than one serious or fatal error, the first problem might stop the boot process before the video card has been initialized, to display error messages.

Proprietary Crash Screens (BSOD/Pinwheel)
Proprietary crash screens such as the Windows STOP error (blue screen of death [BSOD]) or the macOS pinwheel can be caused by operating system, application, or hardware errors.

BSOD Errors
If Windows is configured to reboot when a STOP error occurs, the system continuously reboots until the error is resolved. To leave a STOP error message onscreen until you decide to restart the system, clear the Automatically Restart check box in the System Failure setting in the Startup and Recovery section of Advanced System Properties (see Figure 5-1). This is accessed via Control Panel System Advanced System Settings. Under Startup and Recovery, select Settings.


Startup and Recovery Options. Note That the Option to Automatically Restart Box Under System Failure Has Been Unchecked.

Troubleshooting Windows STOP Errors (BSOD)
STOP errors (also known as blue screen of death, or BSOD, errors) can occur either during startup or after a system is running. The BSOD nickname is used because the background is normally blue (or sometimes black), with the error message in white text. Figure below depicts the Windows 10 STOP errors.


A Windows 10 STOP Error

Note: Regardless of when a STOP/BSOD error occurs, the system is halted by default. If the computer does not restart on its own, you must turn off the system and turn it back on. Before you do that, however, record the error message text and other information so that you can research the problem if it reoccurs. For more information, see the next section, “Causes of BSOD Errors.”

Causes of BSOD Errors
BSOD errors can be caused by any of the following:
- Incompatible or defective hardware or software: Start the system in Windows Recovery Environment (winRE) and uninstall the last hardware or software installed. Acquire updates before you reinstall the hardware or software. Exchange or test memory. Run SFC /scannow from Admin mode in Powershell to check for problems with system files. To enter winRE, reboot and hold the F11 key. If this doesn’t work, try a hard start by rebooting the computer three times, by holding down the power button for 10 seconds. It should open in winRE.
- Registry problems: System Restore, found in winRE, can also be used to revert the system and Registry to an earlier state.
- Viruses/malware: Scan for viruses and remove any that are discovered.
- Miscellaneous causes: Check the Windows Event Viewer and also check the system log. Research the BSOD with the Microsoft Support website.

Researching BSOD Causes and Solutions
To determine the exact cause of a STOP error, note the number or name of the error (for example, STOP 0x0000007B, HAL INITIALIZATION FAILED) and look it up on the Microsoft support website: http://support.microsoft.com. When you search for the error, be sure to specify the version of Windows in use.

Note: STOP errors are often referred to with a shortened version of the error code or by name. For example, the shortened version of a 0x0000007B error is 0x7B.
Tip: Unfortunately, you can’t take a screen capture of a BSOD for printing because a BSOD completely shuts down Windows. In this situation, you can use a digital camera or smartphone to record the exact error message.
The solution might involve one or more of the following changes to your system:
- Changing the system Registry. Sometimes you can download an automated Registry repair tool to perform these changes for you. Whether you make the changes manually or automatically, back up the registry first.
- Removing a newly added component. For example, in the error in Figure 5-1, removing a recently added memory module solved the problem.
- Replacing components such as memory.
- Upgrading an application.
- Downloading and installing a hotfix for your operating system.

On some systems, auto restart is enabled for STOP/BSOD errors, so the error messages shown in the figures below appear for only a moment before the computer restarts.

Note: Microsoft Windows 10 has a tool called Windows Troubleshooter that is available to help resolve Windows problems. Each Windows version provides slightly different access to Troubleshooter. In Windows 10, go to Settings Update and Security Troubleshoot.

macOS Pinwheel
The official name for the macOS pinwheel is the spinning wait cursor .


The Pinwheel in macOS

The pinwheel appears most often when an application or macOS itself has become unresponsive. For this reason, it is sometimes referred to as the “pinwheel of death.” It is usually caused by an application failing, but it can also indicate that the system is locked up and needs a hard reboot. If the problem is an application failure, it can usually be resolved by force-quitting the application (see the solutions in the list that follows).
You might also hear the spinning pinwheel referred to as the “spinning rainbow” or “beachball of death.”

The following are some causes of macOS unresponsiveness:
- Lack of system RAM: If a macOS device frequently displays the pinwheel and the device’s RAM can be upgraded, do so.
- Less than 10 percent free space on the macOS system drive: Free space is used as a swapfile to substitute for RAM. Remove unwanted apps and save data to external or cloud storage, to free up space. Some experts recommend keeping at least 20 percent of the macOS system drive free.
- Damaged application: It is generally quicker to redownload the app, making sure it is from a trusted source. If the system files are not current, they should be updated and the computer should be rebooted before the reinstall.
The following are some solutions for macOS unresponsiveness:
- Use the Force Quit command to terminate an application that will not respond. Force Quit is available from the Apple menu or by pressing Cmd+Option+Esc. Select the app and click Force Quit.
- If a particular application causes unresponsiveness, open the ~Library/Preferences folder, find the .plist file for the app, and drag it to the trash. The .plist file then is rebuilt.
- Use Activity Monitor to view CPU, memory, energy, disk, and networking performance stats. Activity Monitor is similar to the Windows Task Manager. You can use the Spotlight feature on a Mac to locate Activity Monitor in the Applications folder.
- Upgrade to the latest macOS version and keep it updated.
- A forced restart is performed by pressing and holding the Cmd+Ctrl buttons while pressing the power button. Press Cmd+Ctrl+Eject to quit all apps and restart.

Black Screen
A black screen on bootup can be caused by a variety of video configurations or cabling problems. Some of these can be caused by motherboard issues, or the problem might involve more complex display, monitor, or graphics driver errors. Follow these steps to troubleshoot:
- If you have only one display, plugging the video cable into an inactive video port on a system causes a blank screen. For example, some systems deactivate onboard video when you install a video card. If onboard video offers DVI and HDMI ports, typically only one can be selected (usually with motherboard jumpers).
- If a display with two or more inputs (for example, DVI and HDMI or DVI and VGA) is not configured to use the correct cable, the display will be blank. Use the display’s pushbutton controls to select the correct signal input.
- If a DVI or VGA cable is not tightly attached to the video port or display, the screen might be blank. Secure the cable.
- If an HDMI, miniHDMI, DisplayPort, or miniDisplayPort cable is not completely plugged into the video port or display, the screen might be blank. Completely insert the cable into the port.
- If input cables and display input settings check out, but the screen is still blank, shine a flashlight on the screen to see if any text or graphics is visible. If you can see text or graphics with the flashlight, the backlight on the display has failed. On an LCD-CCFL, check the inverter first. Inverter failures are much more common than backlight failures, and inverters are relatively easy to replace. On an LED display, check the LED driver board first. Keep in mind that LCD and LED display modules for laptops or complete displays for desktops are far less expensive today than they used to be; it might make sense to replace the entire display assembly.

Figure belowshows a typical inverter for an LCD-CCFL display in an all-in-one computer.


An All-in-One Computer with the Back Open for Servicing

If the cables are okay and the issue is not resolved, removing the drivers and letting Windows 10 automatically reinstall them can fix the issue. The following steps first remove the display and monitor drivers, which Windows can easily reinstall with a restart. If that does not resolve the issue, uninstalling and replacing the graphics driver could be necessary.

This needs to be done from Safe Mode. To enter Safe Mode in Windows 10, start in the Windows Recovery Environment (winRE) and uninstall the drivers. Then follow these steps:
Step 1. Check the video cable on the PC.
Step 2. Reboot and hold the F11 key to access winRE. (If this doesn’t work, try a hard start by rebooting the computer three times, by holding down the power button for 10 seconds. Then it should open in winRE.)
Step 3. In winRE, select See Advanced Options.
Step 4. From the Advanced Options menu, select Troubleshoot.
Step 5. From the Troubleshoot menu, select Advanced Options and then Startup Settings (you might need to select See More Options to see Startup settings). Then choose Restart and select item 4, Enable Safe Mode.
Step 6. In Safe Mode, press Windows+R to open the Run dialog box.
Step 7. Open the Device Manager by typing devmgmt.msc and click OK.
Step 8. Locate Display Adaptors in the list and display the installed adapters by selecting the drop arrow.
Step 9. Right-click the driver and select Uninstall Device. Selecting Uninstall Windows reinstalls fresh drivers on the next boot.
Step 10. Locate the monitor driver and uninstall it as in step 9.
Step 11. Go to the Windows menu and restart. This reboots into normal mode. If the screen works, the problem is solved. If not, repeat the process, removing and reinstalling the graphics driver.

Note: Entering Safe Mode varies by system. Consult your systems manufacturer’s documentation if the preceding procedure does not work.

No Power
No power when you turn on the system can be caused by several issues.
- Power supply failure: A power supply that has stopped working prevents the system from starting. Use a multimeter or a power supply tester to determine whether a power supply has failed.
- Incorrect front panel wiring connections to the motherboard: The power switch is wired to the motherboard, which, in turn, signals the power supply to start. If the power lead is plugged into the wrong pins on the motherboard or has been disconnected from the motherboard, the system will not start and you will not see an error message. Check the markings on the front panel connectors, the motherboard, or the motherboard/system manual to determine the correct pinouts and installation.
- Loose or missing power leads from the power supply: Make sure both the ATX and ATX12V or EPS12V power leads from the power supply are connected firmly to the motherboard. The connectors lock into place.
- Surge suppressor or UPS failure: If the surge suppressor or uninterruptible power supply (UPS) unit connected to the computer has failed, the computer cannot start. Replace the defective surge suppressor or UPS unit, or replace the battery in the UPS unit.

Overheating
If you touch the power supply case and it feels too hot to touch, it is overheated. Overheated power supplies can cause system failure and possible component damage. They can result from any of the following causes:
- Overloading
- Fan failure
- Inadequate airflow outside the system
- Inadequate airflow inside the system
- Dirt and dust

Use the following sections to figure out the possible effects of these problems in any given situation.

Overloading
An overloaded power supply is caused by connecting devices that draw more power (in Watts) than the power supply is designed to handle. Consider upgrading the hard drive when you add more card-based devices to expansion slots, use more bus-powered USB and Thunderbolt devices, and install more internal drives in a system. This reduces the odds of having an overloaded power supply, which can cause various performance problems.
If a power supply fails or overheats, check the causes listed in the following sections before you consider replacing the power supply. If you determine that you do need to replace the power supply, purchase a unit that has a higher wattage rating and a higher +12V rating.

Fan Failure
The fan or fans inside the power supply cool it and are partly responsible for cooling the rest of the computer. If fans fail, the power supply and the entire computer are at risk of damage. Fans also might stop turning as a symptom of other power problems.
A fan that stops immediately after the power comes on usually indicates incorrect input voltage or a short circuit. If you turn off the system and turn it back on again under these conditions, the fan will stop each time.
To determine whether a fan has failed, listen to the unit; it should make less noise if the fan has failed. You can also see the fan blades spinning rapidly on a power supply fan that is working correctly. If the blades are not turning or are turning very slowly, the fan has failed or is too clogged with dust to operate correctly.
To determine whether case fans have failed, look at them through the front or rear of the system—or, if they are connected to the motherboard, use the system monitoring feature in the system BIOS/UEFI to check fan speed.

Figure below illustrates a typical example.


The System Fan (Case Fan) Either Has Failed Or Was Never Connected to the Motherboard Power/Monitor Header

Note: If a fan has failed because of a short circuit or incorrect input voltage, you will not see any picture onscreen because the system cannot operate.
If the system starts normally but the fan stops turning later, this indicates a true fan failure instead of a power problem.

Inadequate Airflow Outside the System
The power supply’s capability to cool the system depends partly on free airflow space outside the system. If the computer is kept in a confined area (such as a closet or security cabinet) without adequate ventilation, power supply failures due to overheating are likely.
Even systems in ordinary office environments can have airflow problems; make sure that several inches of free air space exist behind the fan outputs for any computer.

Inadequate Airflow Inside the System

As you saw in previous guides, the interior of the typical computer is a messy place. Data cables (particularly wide ribbon cables on older systems), drive power cables, header cables, and expansion cards can create small air dams that block airflow between the heat sources—such as the motherboard, CPU, drives, and memory modules—and the fans in the power supply and the case. Figure below illustrates a typical system with a lot of cable clutter, which can interfere with airflow.


A Cluttered System with Many Unsecured Cables Blocking Airflow

The use of Serial Advanced Technology Attachment (SATA) drives and the elimination of internal floppy drives means that the wide ribbon cables used on the old PATA and floppy drives are no longer used. Disorganized systems can still cause overheating, however. You can do the following to improve airflow inside a computer:
- Use cable ties to secure excess ribbon cable and power connectors out of the way of the fans and the power supply.
- Replace any missing slot covers.
- Make sure that case fans and CPU fans are working correctly.

The Figure below illustrates a different system that uses cable management (using cable ties, bundling cables between the drive bays and the outer case wall, and routing cables behind the motherboard) to improve airflow.


A System with Good Airflow Due to Good Cable Management

Dirt and Dust
Except for a few of the early ATX power supplies, most power supplies use a cooling technique called negative pressure. With this technique, the power supply fan works like a weak vacuum cleaner, pulling air through vents in the case, past the components, and out through the fan. Vacuum cleaners remove dust, dirt, pet hair, and so on from living rooms and offices; the power supply’s weak impression of a vacuum cleaner works the same way.
When you open a system for any kind of maintenance, look for the following:
- Dirt, dust, hair, and gunk clogging the case vents
- A thin layer of dust on the motherboard and expansion slots
- Dirt and dust on the power supply vent and fans
For the most thorough check, remove the computer’s front panel. You never know what you will find inside a PC that has not been cleaned out for a year or two.

As you can see in the figure below, you might discover a system with almost completely clogged air vents. A system in this condition can fail catastrophically at almost any time.


A System with Extremely Dirty Air Vents

Use a vacuum cleaner specially designed for computer use or compressed air to remove dirt and dust from inside the system. If you are using compressed air, be sure to spread newspapers around the system to catch the dirt and dust. If possible, remove the computer from the computer room so the dust does not spread to other equipment.

Installing/Replacing Case Fans
If an overheating system has failed fans or empty fan bays, replace the failed fans or add new ones. To replace a fan, follow these steps:
Step 1. After removing all power to the system and opening the case, locate any failed fans.
Step 2. Disconnect the fan from the motherboard or the power supply.
Step 3. Remove the fan from the case. A fan is held in place by four screws inserted from the outside of the case.
Step 4. (Start here to add a new fan.) Determine the size of fan needed (typical sizes are 120mm, 140mm, and 200mm), and hold the fan inside the case as you attach screws to the fan from the outside.
Step 5. Connect the fan to a system fan header (use the same one as before if you are replacing a fan) on the motherboard. If no system fan header is available, use a Molex power supply connector (or a splitter if you don’t have an unused Molex connector).

No Power
A computer or device that is losing power while operating is usually caused by a battery being drained of its charge. The easy and obvious fix is to plug the device into a wall socket. But what if a computer does not boot even with a charged battery or while plugged in? Although this situation is less common, it can happen. Solutions can vary, depending on the device.
If a booting computer has a black screen and is emitting a series of beeps, check the manufacturer’s documentation to decode this. Sometimes it is simply a loose peripheral cable, such as a keyboard or monitor, or an unplugged fan or disk. It can also indicate complete fan failure.
- Laptop: Unplug the computer from the wall and remove the battery. Wait about 30 seconds, to let any extra static drain away. Then plug the computer back into the wall and boot. Because it would not boot without the battery, it might need replacing.
- Desktop: If the issue is not a loose peripheral cable/fan failure and the computer remains unresponsive, the problem could be a failed power supply. See how to test power supplies in the upcoming section “Burning Smells.”
Unplug your computer and plug it directly into a wall outlet that you know is working (instead of a power strip or a battery backup that could be failing). Make sure the power switch on the back of your power supply is flipped on; if the outlet is connected to a light switch, make sure that switch is turned on, too. Check/test your power supply with a tester for symptoms of failing power supply, including these:
- Random computer crashes
- Random blue screen crashes
- Extra noise coming from the PC case
- Recurring failure of PC components
- PC won’t start but your case fans spin

Sluggish Performance
Troubleshooting sluggish performance can be tricky because the problematic behavior can vary while the computer is operating. However, several common culprits can sag a computer’s performance:
- Maintenance: Keeping software and antivirus/anti-malware protection up-to-date is a great first step. Operating system updates work to control security issues, so make sure the best practice of updating is in place. Malware can also eat up processing power, and outdated drivers can slow performance when using peripherals. Also consider removing rarely used programs from the startup, to free up space.
- Hard drives: Cleaning up storage disks is important. A disk that has to look for data can slow operation. Running disk cleaning tools can help. Upgrading an HD to an SSD also can make a big difference in performance.
- RAM: Few solutions fix a sluggish computer like adding RAM. Remember that the amount of RAM determines how much processing space is available for open applications. Most computers have minimum RAM standards that can meet basic needs, but if a user is simultaneously performing tasks in multiple applications, RAM space can be used up just by having applications open. If closing applications helps the sluggishness, RAM is a likely factor.
- Video cards: If a user is working in processor-intensive applications (for example graphics, video editing, or 3D rendering), a separate graphics card is essential. The CPU on the motherboard can process only so much at once: If regular tasks are being shared with video processing, sluggishness can result. Adding or upgrading a video card with a robust graphics processing unit (GPU) can alleviate the sluggishness issues.
- CPU: An older computer could be asked to do tasks it was not designed for. If it is an older motherboard with only one or two cores doing the processing, it might be time for an upgrade.

Overheating
The most common cause for computer overheating is poor ventilation. Remember that the computer case is designed with very specific airflow patterns to best carry in cool air and then to carry away air heated by the CPU. It is common for dust to collect at the air intake vents and slow or even block the airflow. This leads to overheating in the case and can trigger a shutdown until the CPU cools down.

Burning Smells
The following are typical causes for overheating and burning smells:
- Dead short caused by loose screws, slot covers, or cards: Shut down the system and secure all metal components.
- CPU overheating: Check the fan speed for the CPU heat sink. Clean the fan if it is dirty. Replace the fan if it has failed or is turning too slowly. Check power management settings and CPU drivers in the operating system to make sure that thermal throttling is working.
- Power supply overheating: Check the power supply fan and clean it, if possible. Replace the power supply with a higher wattage–rated unit if the problem persists.
- Power supply failure: Test the power supply to verify proper operation.
If a power supply fails any of these measurements, replace it and retest the new unit.

Power Supply Tester
You can use a power supply tester to determine whether a power supply is working. The power supply does not need to be removed from the computer for testing. However, the 24-pin (or, on older systems, 20-pin) ATX power supply cable and the 4-pin ATX12V or 8-pin EPS12V connectors must be disconnected from the motherboard for testing. The power supply must also be plugged into a working AC outlet or surge suppressor.
Figure below illustrates two types of power supply testers. One tester is a simple go/no-go tester. When you plug it into a power supply’s 20-pin or 24-pin motherboard connector, the power supply starts if it is working, and the green LED turns on. If the power supply does not work, the green LED stays off.


A Simple Power Supply Tester (Top) Compared to a Deluxe Model That Tests Voltages and Can Also Test Other Components

The second tester has its own power switch and checks the major voltage levels, including Power Good, when you turn it on. The display turns a light blue if the power supply tests okay; however, if any voltage level is out of range, the display turns red, as in the Figure below.

Step-by-Step Power Supply Troubleshooting
Use the procedure outlined next to find the actual cause of a dead system. If one of the test procedures in the following list corrects the problem, the item that was changed is the cause of the problem. Power supplies have a built-in safety feature that shuts down the unit immediately in case of a short circuit.

The following steps are designed to determine whether a power problem is caused by a short circuit or another problem:
Step 1. Smell the power supply’s outside vent. If you can detect a burning odor, the power supply has failed. Replace it.
Step 2. Check the AC power to the system; a loose or disconnected power cord, a disconnected surge protector, a surge protector that has been turned off, or a dead AC wall socket will prevent a system from receiving power. If the wall socket has no power, reset the circuit breaker in the electrical service box for the location.
Step 3. Check the AC voltage switch on the power supply; it should be set to 115V for North America. If the switch is set to 230V, turn off the power, reset the switch, and restart the system. Note that many desktop computer power supplies no longer require a switch selection because they are autoswitching.

Note: The A+ objectives list AC voltage as 115V or 220V. AC power is supplied at slightly different voltages in different parts of the world. The normal range of voltage is 100 to 120 volts or 200 to 240 volts. Some dual-voltage power supplies can accept either; such a supply either has a selector switch on the back or can automatically recognize the appropriate setting.

Warning: If your area uses 230V and the power supply is set to 115V, you need a new power supply and possibly other components because they have been damaged or destroyed by 100 percent overvoltage.
Step 4. Turn off the system, disconnect the power, and open the system. Verify that the power leads are properly connected to the motherboard. Connect loose power leads, reconnect the power, and restart the computer.
Step 5. Check for loose screws or other components, such as loose slot covers, modem speakers, or other metal items that can cause a short circuit. Correct them and retest.
Step 6. Remove all expansion cards and disconnect power to all drives; restart the system and use a power supply tester or a multimeter to test power to the motherboard.
Step 7. If the power tests within accepted limits with all peripherals disconnected, reinstall one card at a time and check the power. If the power tests within accepted limits, reattach one drive at a time and check the power.
Step 8. If a defective card or drive has a dead short, reattaching the defective card or drive should stop the system immediately upon powering up. Replace the card or drive and retest.
Step 9. Check the Power Good line at the power supply motherboard connector with a multimeter or a power supply tester.
An unreliable power supply can impact every aspect of PC performance. This is a long list of possible problems and solutions, but chances are good that you will track down the source of the problem before you reach the end of it.

Intermittent Shutdown
Continuous reboots can be caused by problems with the power supply or by a Windows or other operating system configuration setting.
When the Power Good line to the motherboard carries a voltage that is too high or too low, the processor resets, shutting down the system and rebooting it. Test the power supply voltage levels; replace the power supply if Power Good tests out of specifications.
Intermittent failures of USB bus-powered devices (such as mouse devices, keyboard, USB flash drives, and portable USB hard drives) usually happen because these devices draw power from the system’s power supply via the USB port. These types of failures can be an early sign of an overloaded power supply, especially for devices with low power draws, such as mouse devices and keyboards. Replace the power supply with a higher-rated unit.
Intermittent failures of other USB external devices or of internal devices can be caused by damaged data cables, power supplies or connectors, or ports.

To troubleshoot these problems, follow these steps:
Step 1. Shut down the device (and the computer, if the device is internal) and replace the data cable with a known-working replacement. If a USB device is plugged into a front-mounted USB port or a USB port on a card bracket, check the USB header cable connections to the motherboard.
Step 2. Turn on the device or computer.
Step 3. Test the device over time. If the device works correctly, the problem is solved.
Step 4. If steps 1–3 did not resolve the problem, use the original data cable and try plugging it into a different internal or external port. Repeat steps 2–3.
Step 5. Try steps 1–4 again, but this time use a replacement power connector or AC adapter.
Step 6. When you find the defective component, the problem stops. If the problem is not resolved with different data cables, connectors, or power supplies/AC adapters, the device itself needs to be replaced.

Intermittent shutdowns are often a software issue. Updating drivers is a reliable fix. Also check the settings for sleep mode in Windows 10 to make sure the computer is not simply going to sleep.

Application Crashes
Applications can misbehave or crash for a variety of reasons. Applications are written to work with operating system software, and well-written applications rarely have problems in that environment. However, OS software is being updated constantly for security and other reasons, and there is usually a lag between the OS and the application revisions. During that update lag, many problems can occur.
Microsoft is constantly updating Windows 10 with code and patches that work with specific applications. These do not necessarily install automatically. You can customize the updates to your needs in the Windows Update control panel. To access the panel in Windows 10, go to Start Settings Update & Security. The Advanced Options tab is available, if needed.
When encountering application errors, also check with the application developers to see if updates are available. Software patches are small updates that can fix known problems until a full version update is available. A patch could solve the problem. If a patch is not available and the software is essential to business, you might need to roll back the OS update to improve performance. Of course, updates happen for a reason, and if security issues arise with rolling back an update, be sure to address those in some other way, if possible.
Drivers for peripheral devices and video and graphics cards can be a source of application issues. Windows Update usually has the drivers, but the manufacturer has them as well. Uninstalling a driver and replacing it can often solve the problem.

Log Entries and Error Messages
Logs on a device are records kept to track the history of what has happened on the device. They record the tasks the computer has performed, people who have logged in or out, applications opened, and so on. Error messages tell when something went wrong (for example, a device failure or authentication rejection). These are helpful when an IT professional is trying to isolate and solve a problem on a system.
To access logs and error messages, go to Control Panel Administrative Tools Event Viewer. Figure below shows the Event Viewer with the system log selected. Note the many types of logs available in the different applications and system functions. In the System window, note the warning with a yellow triangle, indicating something that failed (in this case, DNS requests).


Windows 10 Event Viewer

Take time to click through the different folders of events recorded on your PC, to become familiar with the vast amount of information available. This much information can be difficult to handle, and tools are available to help search and filter the information down to a manageable amount. You can explore these tools in the right pane of the Event Viewer window, shown in Figure 5-10.

Grinding Noise
Computers usually run quietly and do not have many moving parts, so if a loud or grinding noise is coming from the computer box, pay attention—it could be serious.
If the noise is coming from the power supply, it is a sure sign of a problem. A whirring, screeching, rattling, or thumping noise while the system is on usually indicates a fan failure. If a fan built into a component such as a heat sink or power supply is failing, immediately replace the component. Any other fans in the case are usually there by design, so be sure all are working properly. If a fan is not spinning correctly, airflow could be slowed—you learned earlier that this can cause major problems, including system shutdowns.

Capacitor Swelling
Capacitors, sometimes referred to as caps, are used as part of the voltage step-down circuits that provide power to the processor. From 2002 to 2007, many motherboards were built using faulty capacitors that became distended and leaked. This problem, known as capacitor swelling, causes system failure and sometimes physical damage to the motherboard.
Figure below illustrates a motherboard with distended capacitors.


A System with At Least Two Faulty Capacitors

Some of these systems might still be in service, and the faulty capacitors can be replaced.

Note: For a detailed step-by-step tutorial on replacing bad capacitors, visitwww.itsacon.net/computers/hardware/replacing-bad-motherboard-capacitors/.
Newer systems typically use solid capacitors (see Figure below). These capacitors are much more reliable than older capacitors.


A Typical Recent Motherboard with Solid Capacitors

Inaccurate System Date/Time
Problems with BIOS/UEFI time and settings resets are typically caused by problems with either the CMOS battery on the motherboard or the CMOS chip.
If date and time settings or other BIOS settings reset to system defaults or display CMOS corrupted errors, replace the CMOS battery and reset the BIOS settings to the correct values. A CMOS battery (usually a CR2032 on recent systems) will work properly for about 3 years before it needs to be replaced. Figure below illustrates a typical CR2032 CMOS battery on a recent motherboard.


Removing Cards or Cables Might Be Necessary to Access the CMOS Battery on Some Systems

If replacing the battery does not solve the problem, the CMOS chip on the motherboard might be damaged. The CMOS chip is a surface-mounted chip that cannot be replaced, so if it is bad, the motherboard must be replaced.
If other settings, such as BIOS/UEFI passwords, have been lost or corrupted, the CMOS contents can be cleared by using a jumper on the motherboard. Depending on the motherboard, the jumper might be labeled JBAT (as in Figure 5-13), CLRTC, or CLR_CMOS. See the motherboard or system documentation for details. Turn off the system, move the jumper block, leave it in place for a few seconds, and then move it back to the normal position. The jumper is often, but not always, near the CMOS battery.