Motherboard Failures

Some motherboards will appear to fail all at once. You turn on your system, you hear the power supply engage, and maybe (but unlikely) you see some display alerting you to a problem, but nothing else happens. Many motherboards, however, begin to fail over time, affected to some degree by how many times you restart your PC, run them in extremes of heat, cold, or moisture, or press too hard when you install components such as memory and expansion boards. In my experience, I’ve seen original IBM PC motherboards that are still working today, and I’ve seen plenty of motherboards manufactured just in the past year or two that have already died.

Some of the initial and less-than-obvious symptoms of a failing or malfunctioning motherboard include

• Hesitancy or other problems when the PC is first turned on.

• Your PC may reboot itself or just go dead in the middle of a Windows session.

• Windows may be far more unstable without any seeming cause.

• Windows may begin to report ever-changing error messages or may fail to load at all, even when using Safe Mode; these problems may begin slowly and then increase.

• Devices appear and disappear at random (this can be in BIOS or Device Manager, usually the latter.

Unfortunately, most of these early symptoms can result from other failures, too, making identification of the problem more difficult. For example, a virus, overheating, or a serious disk error could spontaneously reboot your system, and as you learned in Chapter 9, “Stabilizing Your Operating System,” many different problems can contribute to serious instability in Windows. For this reason, try the barebones setup test discussed later in this chapter to see if the motherboard performs better once most of its extra components are removed.

There are also times when a motherboard is more likely to fail suddenly rather than slowly. Most of these times relate to a situation that physically stresses the motherboard, including

• After a physical move where the PC case was not well protected from being bumped or jostled (This can also affect components and how well they’re seated in the motherboard.)

• Following a difficult motherboard or motherboard component installation, where you may have flexed the motherboard too severely or shorted the motherboard or one of its components

• After an electrical event such as a nearby lightning strike or power surge while the PC was in operation and unprotected (and even if it was protected, depending on the severity of the event and the quality of the protection)

• After operating under stress for a prolonged period of time such as in extremes of temperature and humidity

• Following many years of service

Unexplained Hot or Dead Zones

Dead men may tell no tales, but dead hardware sometimes does.

Motherboards are just large printed circuit boards. Take a good look at one, and you’ll see that it’s covered with sockets and slots and ports that allow you to install hardware and wiring and communications running throughout so that the different components can talk with each other.

One of the ways a motherboard may begin to fail is when it develops dead zones, areas where hardware that is plugged into or built into the motherboard may not work any longer. A variation on this is when hardware installed to a problem zone works intermittently, yet the PC runs and there are no identified conflicts or problems except this intermittent failure. In this second situation, the same memory stick in the same memory socket in your motherboard is recognized one time, not the next time, then is recognized again…ad infinitum. Once you’ve verified that the memory is of the right type, that the stick is good (for example, it’s always recognized when it’s installed in another socket), and that you have it properly installed, you’re running out of possibilities: it’s probably a damaged motherboard.

Note 

Some people can run a damaged motherboard for years without ever seeing the full board fail. But there is a fire risk with some types of motherboard failure. If you don’t replace it immediately, make sure the PC is at least turned off when you’re not around.

Neither of these situations is good. The first situation alerts you to expect a full motherboard failure sometime down the road. The second situation not only does this, but it may also indicate a voltage problem or damage that could result in fire or the premature death or damage to other hardware installed into the motherboard.

This voltage issue brings up another problem that you may be able to spot—and avert a later disaster—by watching for certain conditions with your motherboard whenever you’re inside the PC. (Remember to disconnect power and wear an anti-static device whenever you’re inside your PC case.) This is especially true when swapping out hardware, as you’ll read in ”A Cautionary Tale: There’s Hot, and Then There’s HOT!”

If you have any of the following warning signs, turn the PC off and disconnect power, and then have the motherboard checked and/or replaced:

• Obvious signs of damage (fractures, charred areas, or melting of wires) or warping

• Sparking or smoke coming from the motherboard itself (A smoking power supply or component may not affect the motherboard unless you continue to run the PC with this problem.)

• Having to replace too much hardware over a period of a year or two (However, this is more likely to be the result of power problems; anything damaging expansion boards and other devices installed directly to the motherboard is hurting the motherboard, too.)

• Having to replace a component like a modem because of lightning or power-surge damage but your system remains affected by what may seem like odd, disconnected hardware problems

• Serious problems after an electrostatic discharge (shock or zap) within the case, such as you might have if you forget to ground yourself

• Same as above, but you notice hesitations or “dead” behavior after you install something into or remove a component from the motherboard while the PC is powered up

Warning 

Look for anything out of place with your motherboard like two pieces of metal touching one another—a situation that can produce an electrical short. A common area for such problems is the way the motherboard is positioned in the PC case; if one of its edges inappropriately touches one of the metal standoffs in the case frame, the board can short. In addition, check for foreign objects or a large accumulation of dust that can be removed with compressed air (not blowing with your mouth, because it’s moist).

A Cautionary Tale: There’s Hot, and Then There’s HOT!

As I told you in the last chapter, part of my work in preparing this book was to re-create some of the disasters we’re discussing.

One of the unwitting victims of all this research was my poor P2-350 motherboard, pressed into service on 9/11/98 and committed to a PC hardware recycling program on 8/24/02. That motherboard helped me through the writing or editing of a dozen books, thousands of hours of online research and providing technical support, a few dozen beta tests, and way too many hours playing “The Sims.” Ironically, the motherboard lasted a few years longer than the company that sold me my system.

It didn’t go out with a bang, either. Instead, I spent a week trying to resolve some really bizarre Windows stop errors (discussed in Chapter 9, “Stabilizing Your Operating System”) that I had assumed were caused by some of my testing. Then one day, I tried rearranging my memory to see if a failing memory stick was at fault.

I installed a brand new stick of working SDRAM into my system, but found I got no display—and no beeps—when I powered the PC back up. After I powered it down again and disconnected the power to the PC, I removed the new stick of RAM I had just installed. I discovered that the heat from the connector edge of the stick was severe enough to warm the entire stick.

Thinking I must have installed it improperly since less than 60 seconds of operation shouldn’t result in that level of warm up, I tried another stick of memory, being extra careful with its installation. I got the exact same result. However, this time, I could smell the overheating and was pleased to get the system disconnected again before I had an electrical fire—the situation seemed ripe for one.

When I had the board fully checked a few days later, only some of the board was getting power, while part of the board was showing a real flux in voltage. Because I stopped using the motherboard immediately upon discovering the overheating, I probably saved many of the components installed into it. So I opted to replace just the motherboard itself.

FYI, the occasion made me stop to consult my PC journal, calculate how much I had spent on the system both to purchase and to power, as well as to repair through replacement components and operating system upgrades. All total, that PC cost me about $1.58 a day, or less than the price of an Egg McMuffin or a cup of Starbuck’s coffee. Not a bad investment. Plus, since I’ve got drives and many other parts I can still use from it, the cost over the PC’s life should decrease from even that paltry $1.58.