PC Repair and Maintenance: In-depth Look at Power Supply

Date: Feb 28, 2003

Article is provided courtesy of Que.

Return to the article

Student Learning Objectives

In this chapter, you will

• Learn about the various power supply form factors used in PCs today

• Study the connectors used with power supplies, including a connector used by one major manufacturer that could destroy your motherboard if you're not careful

• Learn how to understand power supply ratings and specifications

• Troubleshoot power supply problems, which are oft-overlooked culprits when tracking pesky system problems

• Learn how to test, remove, and install a power supply

• Study power protection systems, surge suppressors, and uninterruptible power supplies

Terms to Study

The following terms are just a few of the terms used in this chapter. Watch for these terms as you read. Definitions for each can be found in the UPGRADE.HLP file located on the CD-ROM or in Appendix A, "Glossary." Reviewing these terms before and after you read this chapter will deepen your understanding of this chapter.

1. Step Power good signalStep
2. Step DCStep
3. Step ACPIStep
4. Step BrownoutStep
5. Step Power supplyStep
6. Step Back probingStep
7. Step Surge protectorStep
8. Step Form factorStep
9. Step POSTStep
10. Step UPSStep
11. Step full towerStep
12. Step Mini-towerStep

Considering the Importance of the Power Supply

The power supply is not only one of the most important parts in a PC, but it is unfortunately also the most overlooked. In the words of a famous comedian, the power supply gets no respect! People spend hours discussing their processor speeds, memory capacity, disk storage capacity and speed, video adapter performance, monitor size, and so forth, but rarely even mention or consider their power supply. When a system is put together to meet the lowest possible price point, what component do you think the manufacturer will skimp on? Yes, the power supply. To most people, the power supply is a rather nondescript, unglamorous metal box that sits inside their systems, something to which they pay virtually no attention at all. The few who do pay any mind seem concerned only with how many watts of power it is rated to put out (even though no practical way exists to verify those ratings), without regard as to whether the power being produced is clean and stable, or whether it is full of noise, spikes, and surges.

I have always placed great emphasis on selecting a power supply for my system. I consider the power supply the core of the system and am willing to spend more to get a better unit. The power supply function is critical because it supplies electrical power to every other component in the system. In my experience, the power supply is also one of the most failure-prone components in any computer system, especially due to the fact that so many system assemblers use the cheapest ones they can find. A malfunctioning power supply can not only cause other components in the system to malfunction, but it also can damage the other components in your computer by delivering an improper or erratic voltage. Because of its importance to proper and reliable system operation, you should understand both the function and limitations of a power supply, as well as its potential problems and their solutions.

This chapter covers the power supply in detail. I focus on the electrical functions of the supply and the mechanical form factors and physical designs that have been used in PC systems in the past, as well as today. Because the physical shape (form factor) of the power supply relates to the case, some of this information also relates to the type of chassis or case you have.

Primary Function and Operation

The basic function of the power supply is to convert the type of electrical power available at the wall socket to the type the computer circuitry can use. The power supply in a conventional desktop system is designed to convert either 115-volt (nominal) 60Hz AC (alternating current) or 230v (nominal) 50Hz AC power into +3.3v, +5v, and +12v DC (direct current) power. Some power supplies require you to switch between the two input ranges, whereas others auto-switch.

Positive DC Voltages

Usually, the digital electronic components and circuits in the system (motherboard, adapter cards, and disk drive logic boards) use the +3.3v or +5v power, and the motors (disk drive motors and any fans) use the +12v power. Table 3.1 lists these devices and their power consumptions.

Table 3.1 Power Consumption Ratings for PC Devices

Voltage	Devices Powered
+3.3v	Chipsets, DIMMs, PCI/AGP cards, miscellaneous chips
+5v	Disk drive logic, SIMMs, PCI/AGP cards, ISA cards, voltage regulators, miscellaneous chips
+12v	Motors, voltage regulators (high output)

The power supply must deliver a good, steady supply of DC power so that the system can operate properly. Devices that run on voltages other than these must be powered by onboard voltage regulators. For example, RIMMs run on 2.5v that is supplied by an onboard regulator, and processors are supplied by a voltage regulator module (VRM) that normally is built into the motherboard as well.

Negative DC Voltages

If you look at a specification sheet for a typical PC power supply, you can see that the supply generates not only +3.3v, +5v, and +12v, but also –5v and –12v. The positive voltages seemingly power everything in the system (logic and motors), so what are the negative voltages used for? The answer is, not much! Some of the power supply designs, such as the small form factor (SFX) design, no longer include the –5v output for that reason. The only reason it has remained in most power supply designs is that –5v is required on the Industry Standard Architecture (ISA) bus for full backward-compatibility.

Although –5v and –12v are supplied to the motherboard via the power supply connectors, the motherboard normally uses only the +3.3v, +5v, and +12v. The –5v is simply routed to the ISA bus on pin B5 so any ISA cards can use it. Today, though, not many do. However, as an example, the analog data separator circuits found in older floppy controllers do use –5v.

The motherboard logic normally doesn't use –12v either; however, it might be used in some board designs for serial port or LAN circuits.

Although older serial port circuits used +/–12v outputs, today most run on only +3.3v or +5v.

The main function of the +12v power is to run disk drive motors as well as the higher-output processor voltage regulators in some of the newer boards. Usually, a large amount of +12v current is available from the power supply, especially in those designed for systems with a large number of drive bays (such as in a tower configuration). Besides disk drive motors and newer CPU voltage regulators, the +12v supply is used by any cooling fans in the system—which, of course, should always be running. A single cooling fan can draw between 100mA and 250mA (0.1–0.25 amps); however, most newer fans use the lower 100mA figure. Note that although most fans in desktop systems run on +12v, portable systems can use fans that run on +5v, or even +3.3v.

Most systems with newer motherboard form factors, such as the ATX, micro-ATX, or NLX, include another special signal. This feature, called PS_ON, can be used to turn the power supply (and thus the system) on or off via software. It is sometimes known as the soft-power feature. PS_ON is most evident when you use it with an operating system, such as Windows 9x, that supports the Advanced Power Management (APM) or Advanced Configuration and Power Interface (ACPI) specification. When you select the Shut Down the Computer option from the Start menu, Windows automatically turns off the computer after it completes the OS shutdown sequence. A system without this feature only displays a message that it's safe to shut down the computer.

The Power_Good Signal

In addition to supplying electrical power to run the system, the power supply also ensures that the system does not run unless the power supplied is sufficient to operate the system properly. In other words, the power supply actually prevents the computer from starting up or operating until all the power supply voltages are within the proper ranges.

The power supply completes internal checks and tests before allowing the system to start. If the tests are successful, the power supply sends a special signal to the motherboard, called Power_Good. This signal must be continuously present for the system to run. Therefore, when the AC voltage dips and the power supply cannot maintain outputs within regulation tolerance, the Power_Good signal is withdrawn (goes low) and forces the system to reset. The system will not restart until the Power_Good signal returns.

The Power_Good signal (sometimes called Power_OK or PWR_OK) is a +5v (nominal) active high signal (with variation from +2.4v through +6.0v generally being considered acceptable) that is supplied to the motherboard when the power supply has passed its internal self tests and the output voltages have stabilized. This normally takes place anywhere from 100ms to 500ms (0.1–0.5 seconds) after you turn on the power supply switch. The power supply then sends the Power_Good signal to the motherboard, where the processor timer chip that controls the reset line to the processor receives it.

In the absence of Power_Good, the timer chip holds the reset line on the processor, which prevents the system from running under bad or unstable power conditions. When the timer chip receives the Power_Good signal, it releases the reset, and the processor begins executing whatever code is at address FFFF:0000 (usually the ROM BIOS).

If the power supply cannot maintain proper outputs (such as when a brownout occurs), the Power_Good signal is withdrawn, and the processor is automatically reset. When the power output returns to its proper levels, the power supply regenerates the Power_Good signal and the system again begins operation (as if you had just powered on). By withdrawing Power_Good before the output voltages fall out of regulation, the system never sees the bad power because it is stopped quickly (reset) rather than being allowed to operate using unstable or improper power levels, which can cause memory parity errors and other problems.

On pre-ATX systems, the Power_Good connection is made via connector P8-1 (P8 Pin 1) from the power supply to the motherboard. ATX and later systems use pin 8 of the 20-pin connector, which is normally a gray wire.

A well-designed power supply delays the arrival of the Power_Good signal until all the voltages stabilize after you turn on the system. Badly designed power supplies, which are found in many low-cost systems, often do not delay the Power_Good signal properly and enable the processor to start too soon. (The normal Power_Good delay is 0.1–0.5 seconds.) Improper Power_Good timing also causes CMOS memory corruption in some systems.

Some cheaper power supplies do not have proper Power_Good circuitry and might just tie any +5v line to that signal. Some motherboards are more sensitive to an improperly designed or improperly functioning Power_Good signal than others. Intermittent startup problems are often the result of improper Power_Good signal timing. A common example is when you replace a motherboard in a system and then find that the system intermittently fails to start properly when you turn on the power. This can be very difficult to diagnose, especially for the inexperienced technician, because the problem appears to be caused by the new motherboard. Although it seems as though the new motherboard is defective, it usually turns out that the power supply is poorly designed. It either cannot produce stable enough power to properly operate the new board or has an improperly wired or timed Power_Good signal (which is more likely). In these situations, replacing the supply with a higher-quality unit, in addition to the new motherboard, is the proper solution.

Power Supply Form Factors

The shape and general physical layout of a component is called the form factor. Items that share a form factor are generally interchangeable, at least as far as their sizes and fits are concerned. When designing a PC, the engineers can choose to use one of the popular standard PSU (power supply unit) form factors, or they can elect to build their own. Choosing the former means that a virtually inexhaustible supply of inexpensive replacement parts will be available in a variety of quality and power output levels. Going the custom route means additional time and expense for development. In addition, the power supply is unique to the system and available only from the original manufacturer.

If you can't tell already, I am a fan of the industry-standard form factors! Having standards and then following them allows us to upgrade and repair our systems by easily replacing physically (and electrically) interchangeable components. Having interchangeable parts means that we have a better range of choices for replacement items, and the competition makes for better pricing, too.

In the PC market, IBM originally defined the standards, and everybody else copied them. This included power supplies. All the popular PC power supply form factors up through 1995 were based on one of three IBM models, including the PC/XT, AT, and PS/2 Model 30. The interesting thing is that these three power supply definitions all had the same motherboard connectors and pinouts; where they differed was mainly in shape, maximum power output, the number of peripheral power connectors, and switch mounting. PC systems using knock-offs of one of those three designs were popular through 1996 and beyond, and some systems still use them today.

Intel gave the power supply a new definition in 1995 with the introduction of the ATX form factor. ATX became popular in 1996 and started a shift away from the previous IBM-based standards. ATX and the related standards that followed have different connectors with additional voltages and signals that allow systems with greater power consumption and additional features that would otherwise not be possible with the AT style supplies.

Technically, the power supply in your PC is described as a constant voltage half-bridge forward converting switching power supply:

• Constant voltage means that the power supply puts out the same voltage to the computer's internal components, no matter what the voltage of AC current running it or the capacity (wattage) of the power supply.

• Half-bridge forward converting switching refers to the design and power regulation technique used by most suppliers. This design is commonly referred to as a switching supply. Compared to other types of power supplies, this design provides an efficient and inexpensive power source and generates a minimum amount of heat. It also maintains a small size and a low price.

Seven main power supply physical form factors have existed that can be called industry standard. Five of these are based on IBM designs, whereas two are based on Intel designs. Of these, only three are used in most modern systems; the others are pretty much obsolete.

Note that although the names of the power supply form factors seem to be the same as those of motherboard form factors, the power supply form factor is more related to the system chassis (case) than the motherboard. That is because all the form factors use one of only two types of connector designs, either AT or ATX.

For example, all PC/XT, AT, and LPX form factor supplies use the same pair of six-pin connectors to plug into the motherboard and will therefore power any board having the same type of power connections. Plugging into the motherboard is one thing, but for the power supply to physically work in the system, it must fit the case. The bottom line is that it is up to you to ensure the power supply you purchase not only plugs into your motherboard but also fits into the chassis or case you plan to use.

Table 3.2 shows modern power supply form factors, their connection types, and associated motherboards.

Table 3.2 Power Supply Connector Types and Form Factors

Modern PS Form Factors	Originated From	Connector Type	Associated MB Form Factors
LPX style*	IBM PS/2 Model 30 (1987)	AT	Baby-AT, Mini-AT, LPX
ATX style	Intel ATX, ATX12V (1985/2000)	ATX	ATX, NLX, Micro-ATX
SFX style	Intel SFX (1997)	ATX	Flex-ATX, Micro-ATX

*Note: LPX is also sometimes called Slimline or PS/2.

Each of these power supply form factors is available in numerous configurations and power output levels. The LPX form factor supply had been the standard used on most systems from the late 1980s to mid-1996, when the ATX form factor started to gain in popularity. Since then, ATX has become by far the dominant form factor for power supplies, with the new SFX style being added as an ATX derivative for use in very compact systems that mainly use Flex-ATX–sized boards. The earlier IBM-derived form factors have been largely obsolete for some time now.

→ See "Motherboard Form Factors," p. 84

LPX Style

The next power supply form factor to gain popularity was the LPX style, also called the PS/2 type, Slimline, or slim style (see Figure 3.1). The LPX-style power supply has the exact same motherboard and disk drive connectors as the previous standard power supply form factors; it differs mainly in the shape. LPX systems were designed to have a smaller footprint and lower height than AT-sized systems. These computers used a different motherboard configuration that mounts the expansion bus slots on a "riser" card that plugs into the motherboard. The expansion cards plug into this riser and are mounted sideways in the system, parallel to the motherboard. Because of its smaller case, an LPX system needed a smaller power supply. The power supply designed for LPX systems is smaller than the Baby-AT style in every dimension and takes up less than half the space of its predecessor.

Figure 3.1 LPX form factor power supply.

As with the Baby-AT design in its time, the LPX power supply does the same job as its predecessor but comes in a smaller package. The LPX power supply quickly found its way into many manufacturers' systems, soon becoming a de facto standard. This style of power supply became the staple of the industry for many years, coming in everything from low-profile systems using actual LPX motherboards to full-size towers using Baby-AT or even full-size AT motherboards. It still is used in some PCs produced today; however, since 1996 the popularity of LPX has been overshadowed by the increasing popularity of the ATX design.

ATX Style

One of the newer standards in the industry today is the ATX form factor (see Figure 3.2). The ATX specification, now in version 2.03, defines a new motherboard shape, as well as a new case and power supply form factor.

Figure 3.2 ATX form factor power supply, used with both ATX and NLX systems.

The shape of the ATX power supply is based on the LPX design, but some important differences are worth noting.

One difference is that the ATX specification originally called for the fan to be mounted along the inner side of the supply, where it could draw air in from the rear of the chassis and blow it inside across the motherboard. This kind of airflow runs in the opposite direction as most standard supplies, which exhaust air out the back of the supply through a hole in the case where the fan protrudes. The idea was that the reverse flow design could cool the system more efficiently with only a single fan, eliminating the need for a fan (active) heatsink on the CPU.

Another benefit of the reverse-flow cooling is that the system would run cleaner, more free from dust and dirt. The case would be pressurized, so air would be continuously forced out of the cracks in the case—the opposite of what happens with a negative pressure design. For this reason, the reverse-flow cooling design is often referred to as a positive-pressure-ventilation design. On an ATX system with reverse-flow cooling, the air would be blown out away from the drive because the only air intake would be the single fan vent on the power supply at the rear. For systems that operate in extremely harsh environments, you can add a filter to the fan intake vent to further ensure that all the air entering the system is clean and free of dust.

Although this sounds like a good way to ventilate a system, the positive-pressure design needs to use a more powerful fan to pull the required amount of air through a filter and to pressurize the case. Also, if a filter is used, it must be serviced on a periodic basis—depending on operating conditions, it can need changing or cleaning as often as every week. In addition, the heat load from the power supply on a fully loaded system heats up the air being ingested, blowing warm air over the CPU, reducing overall cooling capability. As newer CPUs create more and more heat, the cooling capability of the system becomes more critical. In common practice, it was found that using a standard negative-pressure system with an exhaust fan on the power supply and an additional high-quality cooling fan blowing cool air right on the CPU is the best solution. For this reason, the ATX power supply specification has been amended to allow for either positive- or negative-pressure ventilation.

Because a standard negative-pressure system offers the most cooling capacity for a given fan airspeed and flow, most of the newer ATX-style power supplies use the negative-pressure cooling system.

The ATX specification was first released by Intel in 1995. In 1996, it became increasingly popular in Pentium and Pentium Pro–based PCs, capturing 18% of the motherboard market. Since 1996, ATX has become the dominant motherboard form factor, displacing the previously popular Baby-AT. ATX and its derivatives are likely to remain the most popular form factor for several years to come.

The ATX form factor addressed several problems with the power supplies used with Baby-AT and mini-AT form factors. One is that the power supplies used with Baby-AT boards have two connectors that plug into the motherboard. If you insert these connectors backward or out of their normal sequence, you will fry the motherboard! Most responsible system manufacturers "key" the motherboard and power supply connectors so that you cannot install them backward or out of sequence. However, some vendors of cheaper systems do not feature this keying on the boards or supplies they use. The ATX form factor includes different power plugs for the motherboard to prevent users from plugging in their power supplies incorrectly. The ATX design features up to three motherboard power connectors that are definitively keyed, making plugging them in backward virtually impossible. The new ATX connectors also supply +3.3v, reducing the need for voltage regulators on the motherboard to power the chipset, DIMMs, and other +3.3v circuits.

Besides the new +3.3v outputs, another set of outputs is furnished by an ATX power supply that is not normally seen on standard power supplies. The set consists of the Power_On (PS_ON) and 5v_Standby (5VSB) outputs mentioned earlier, known collectively as Soft Power. This enables features to be implemented, such as Wake on Ring or Wake on LAN, in which a signal from a modem or network adapter can actually cause a PC to wake up and power on. Many such systems also have the option of setting a wake-up time, at which the PC can automatically turn itself on to perform scheduled tasks. These signals also can enable the optional use of the keyboard to power the system on—exactly like Apple systems. Users can enable these features because the 5v Standby power is always active, giving the motherboard a limited source of power even when off. Check your BIOS Setup for control over these features.

NLX Style

The NLX specification, also developed by Intel, defines a low-profile case and motherboard design with many of the same attributes as the ATX. In fact, for interchangeability, NLX systems were designed to use ATX power supplies, even though the case and motherboard dimensions are different.

As in previous LPX systems, the NLX motherboard uses a riser board for the expansion bus slots. Where NLX differs is that it is a true (and not proprietary) standard. See Chapter 4, "Motherboards and Buses," for more information on the NLX form factor.

For the purposes of this discussion, NLX systems use ATX power supplies. The only real difference is that the supply plugs into the riser card and not the motherboard, enabling NLX motherboards to be more quickly and easily removed from their chassis for service.

SFX Style

Intel released the smaller Micro-ATX motherboard form factor in December of 1997, and at the same time also released a new smaller SFX (Small form factor) power supply design to go with it (see Figure 3.3). Even so, most Micro-ATX chassis used the standard ATX power supply instead. Then in March 1999, Intel released the Flex-ATX addendum to the Micro-ATX specification, which was a very small board designed for low-end PCs or PC-based appliances. At this point, the SFX supply has found use in many new compact system designs.

Figure 3.3 SFX style power supply (with 90mm top-mounted cooling fan).

The SFX power supply is specifically designed for use in small systems containing a limited amount of hardware and limited upgradability. Most SFX supplies can provide 90 watts of continuous power (135 watts at its peak) in four voltages (+5, +12, –12, and +3.3v). This amount of power has proved to be sufficient for a small system with a processor, an AGP interface, up to four expansion slots, and three peripheral devices—such as hard drives and CD-ROMs.

Although Intel designed the SFX power supply specification with the Micro-ATX and Flex-ATX motherboard form factors in mind, SFX is a wholly separate standard that is compliant with other motherboards as well. SFX power supplies use the same 20-pin connector defined in the ATX standard and include both the Power_On and 5v_Standby outputs. Whether you will use an ATX or SFX power supply in a given system is dependent more on the case or chassis than the motherboard. Each has the same basic electrical connectors; the main difference is which type of power supply the case is physically designed to accept.

One limiting factor on the SFX design is that it lacks the –5v and so shouldn't be used with motherboards that have ISA slots (most Micro-ATX and Flex-ATX boards do NOT have ISA slots). SFX power supplies also won't have the Auxiliary (3.3v and 5v) or ATX12V power connectors, and therefore shouldn't be used with full-size ATX boards that require those connections.

On a standard model SFX power supply, a 60mm diameter cooling fan is located on the surface of the housing, facing the inside of the computer's case. The fan draws the air into the power supply housing from the system cavity and expels it through a port at the rear of the system. Internalizing the fan in this way reduces system noise and results in a standard negative-pressure design. In many cases, additional fans might be needed in the system to cool the processor (see Figure 3.4).

Figure 3.4 SFX form factor power supply dimensions with a standard internal 60mm fan.

For systems that require more cooling capability, a version that allows for a larger 90mm top-mounted cooling fan also is available. The larger fan provides more cooling capability and airflow for systems that need it (see Figure 3.5).

Figure 3.5 SFX form factor power supply dimensions with an internal 90mm top-mounted fan.

Motherboard Power Connectors

Every PC power supply has special connectors that attach to the motherboard, giving power to the system processor, memory, and all slotted add-on boards (ISA, PCI, AGP). Attaching these connectors improperly can have a devastating effect on your PC, including burning up both your power supply and motherboard. The following sections detail the motherboard power connectors used by various power supplies.

AT Power Supply Connectors

Industry standard PC, XT, AT, Baby-AT, and LPX motherboards all use the same type of main power supply connectors. These supplies feature two main power connectors (P8 and P9), each with 6 pins that attach the power supply to the motherboard. All standard PC power supplies that use the P8 and P9 connectors have them installed end to end so that the two black wires (ground connections) on both power cables are next to each other. Some power supplies have them labeled as P1/P2 instead. Because these connectors usually have a clasp that prevents them from being inserted backward on the pins on the motherboards, the major concern is getting the two connectors in the correct orientation side by side and also not missing a pin offset on either side. Following the black-to-black rule keeps you safe. You must take care, however, to make sure that no remaining unconnected motherboard pins exist between or on either side of the two connectors after you install them. A properly installed connector connects to and covers every motherboard power pin. If any power pins are showing on either side of the connectors, the entire connector assembly is installed incorrectly, which can result in catastrophic failure for the motherboard and everything plugged into it at the time of power-up. Figure 3.6 shows the P8 and P9 connectors (sometimes also called P1/P2) in their proper orientations when connecting.

Figure 3.6 The P8/P9 power connectors (sometimes also called P1/P2) that connect an AT/LPX power supply to the motherboard.

ATX Main Power Connector

The industry standard ATX power-supply–to–motherboard main connector is the Molex 39-29-9202 (or equivalent) 20-pin ATX style connector (see Figure 3.7). First used in the ATX form factor power supply, it also is used in the SFX form factor or any other ATX-based variations. The colors for the wires listed in Table 3.3 are those recommended by the ATX standard; however, they are not required for compliance to the specification, so they could vary from manufacturer to manufacturer. Note that I like to show these connector pinouts in a wire side view, which shows how the pins are arranged looking at the back of the connector (from the wire and not terminal side). This is because it shows how they would be oriented if you were back-probing the connector with the connector plugged in.

Figure 3.7 ATX style 20-pin motherboard main power connector, perspective view.

Figure 3.8 shows a view of the connector as if you were looking at it facing the terminal side.

Figure 3.8 ATX/NLX 20-pin main power connector, terminal side view.

Table 3.3 ATX Main Power Supply Connector Pinout (Wire Side View)

Color	Signal	Pin	Pin	Signal	Color
Orange*	+3.3V	11	1	+3.3V	Orange
Blue	–12V	12	2	+3.3V	Orange
Black	GND	13	3	GND	Black
Green	PS_On	14	4	+5V	Red
Black	GND	15	5	GND	Black
Black	GND	16	6	+5V	Red
Black	GND	17	7	GND	Black
White	–5V	18	8	Power_Good	Gray
Red	+5V	19	9	+5VSB (Standby)	Purple
Red	+5V	20	10	+12V	Yellow

*Might have a second orange or brown wire, used for +3.3v sense feedback—used by the power supply to monitor 3.3v regulation.

ATX Auxiliary Power Connector

As motherboards and processors evolved, the need for power became greater. In particular, chipsets and DIMMs were designed to run on 3.3v, increasing the current demand at that voltage. In addition, most boards included CPU voltage regulators designed to convert +5v power into the unique voltage levels required by the processors the board supported. Eventually, the high current demands on the +3.3v and +5v outputs were proving too much for the number and gauge of the wires used. Melted connectors were becoming more and more common as these wires overheated under these loads.

Finally, Intel modified the ATX specification to add a second power connector for ATX motherboards and supplies. The criteria was that if the motherboard needed more than 18A of +3.3v power, or more than 24A of +5v power, an auxiliary connector would be defined to carry the additional load. These higher levels of power are normally necessary in systems using 250-watt to 300-watt or greater supplies.

This is a 6-pin Molex-type connector (see Figure 3.9). It is keyed to prevent misconnection.

Figure 3.9 ATX auxiliary power connector.

The pinouts of the auxiliary connector are shown in Table 3.4.

Table 3.4 ATX Auxiliary Power Connector Pinout

Signal	Color	Pin	Pin	Signal	Color
Gnd	Black	1	4	+3.3V	Orange
Gnd	Black	2	5	+3.3V	Orange
Gnd	Black	3	6	+5V	Red

If your motherboard does not feature a mating auxiliary connector, it probably wasn't designed to consume a large amount of power, and the auxiliary connector from the power supply can be left unconnected. If your power supply is rated at 250 watts or larger, you should ensure that it has this connector and that your motherboard is capable of accepting it. This eases the load on the main power connector.

ATX12V Connector

Power for the processor comes from a device called the voltage regulator module (VRM), which is built into most modern motherboards. This device senses the CPU voltage requirements (usually via sense pins on the processor) and calibrates itself to provide the proper voltage to run the CPU. The design of a VRM enables it to run on either 5v or 12v for input power. Most have used 5v over the years, but many are now converting to 12v because of the lower current requirements at that voltage. In addition, the 5v already might be loaded by other devices, whereas, typically, only drive motors use the 12v. Whether the VRM on your board uses 5v or 12v depends on the particular motherboard or regulator design. Many modern voltage regulator ICs are designed to run on anything from a 4v to a 36v input, so it is up to the motherboard designer as to how they will be configured.

Although most motherboard VRM designs up through the Pentium III and Athlon/Duron use 5v-based regulators, a transition is underway to use 12v-powered regulators. This is because the higher voltage will significantly reduce the current draw. As an example, using the same 65W AMD Athlon 1GHz CPU, you end up with the levels of draw at the various voltages shown in Table 3.5.

Table 3.5 Levels of Draw at Various Voltages

Watts	Volts	Amps	Amps at 75% Regulator Efficiency
65	1.8	36.1	—
65	3.3	19.7	26.3
65	5.0	13.0	17.3
65	12.0	5.4	7.2

As you can see, using 12v to power the chip results in only 5.4A of draw, or 7.2A assuming 75% efficiency on the part of the regulator.

So, modifying the motherboard VRM circuit to use the +12v power feed would seem simple. Unfortunately, the standard ATX 2.03 power supply design has only a single +12v lead in the main power connector. The auxiliary connector has no +12v leads at all, so that is no help. Pulling up to 8A more through a single 18ga. wire supplying +12v power to the motherboard is a recipe for a melted connector.

To augment the supply of +12v power to the motherboard, Intel created a new ATX12V power supply specification. This adds a third power connector, called the ATX12V connector, specifically to supply additional +12v power to the board. This connector is shown in Figure 3.10.

Figure 3.10 An ATX12V power connector.

The pinout of the +12v power connector is shown in Table 3.6.

Table 3.6 ATX +12v Power Connector Pinout (Wire Side View)

Color	Signal	Pin	Pin	Signal	Color
Yellow	+12V	3	1	Gnd	Black
Yellow	+12V	4	2	Gnd	Black

If you are replacing your motherboard with a new one that requires the ATX12V connection for the CPU voltage regulator, and yet your existing power supply doesn't have that connector, an easy solution is available. Merely convert one of the peripheral power connectors to an ATX12V type. PC Power and Cooling has released just such an adapter that can instantly make any standard ATX power supply into one with an ATX12V connector. The issue is not whether the power supply can generate the necessary 12v—that has always been available via the peripheral connectors. The ATX12V adapter shown in Figure 3.11 solves the connector problem quite nicely.

Figure 3.11 ATX12V adapter from PC Power and Cooling.

ATX Optional Connector

The ATX specification also defines an optional six-pin connector. This connector has two rows of three pins each to provide the signals and voltages. The computer can use these signals to monitor and control the cooling fan, monitor the +3.3v power to the motherboard, and provide power and grounding to IEEE 1394 (FireWire) devices.

This connector has gone through several revisions in pinout since first being published, and I have yet to see any motherboards or power supplies on the market that actually support it. In fact, the latest ATX/ATX12V Power Supply Design Guide published by Intel states, "Details of the 2x3 'Optional Power Connector' mentioned in the ATX 2.03 Specification are omitted from this design guide until such time as the signals on that connector are more rigidly defined."

Table 3.7 lists the pinout of the optional connector as defined in the ATX 2.03 Specification.

Table 3.7 ATX Optional Power Supply Connections

Color	Signal	Pin	Pin	Signal	Color
White/Black Stripe	1394R	4	1	FanM	White
White/Red Stripe	1394V	5	2	FanC	White/Blue Stripe
	Reserved	6	3	+3.3V sense	White/Brown Stripe

The FanM signal enables the operating system to monitor the status of the power supply's cooling fan so that it can take appropriate actions, such as shutting down the system if the fan fails.

The motherboard (under operating system control) can use the FanC signal with variable voltages to control the operation of the power supply's fan, shifting it into a low power state or shutting it off completely when the system is in sleep or standby mode. The ATX standard specifies that a voltage of +1v or less indicates that the fan is to shut down, whereas +10.5v or more instructs the fan to operate at full speed. The system designer can define intermediate voltages to operate variable-speed fans at various levels. If the power supply does not include a variable-speed fan circuit, any voltage level higher than +1v on the FanC signal is interpreted as a command to run the fan at its full (and only) speed.

The 1394 connectors are for powering an optional IEEE-1394 (FireWire or iLink) bus on a motherboard. The 1394V pin provides voltages from 8v to 40v to run FireWire peripherals off the bus, and the 1394R pin is a return or ground line for this power circuit. This separate power rail keeps the 1394 bus power separate from the system main power to prevent interference.

Dell Proprietary (Nonstandard) ATX Design

If you currently own a desktop system made between 1996 and 2000 from Dell, you will definitely want to pay attention to this section. A potential booby trap is waiting to nail the unsuspecting Dell system owner who decides to upgrade either the motherboard or power supply in his system. This hidden trap can cause the destruction of the motherboard, power supply, or both! Okay, now that I have your attention, read on....

As those of you who have attended my seminars or read previous editions of this book will know, I have long been a promoter of industry-standard PCs and components and wouldn't think of purchasing a desktop PC that didn't have what I consider an industry-standard form factor motherboard, power supply, and chassis (ATX, for example). I've been down the proprietary road before with systems from Packard Bell, Compaq, IBM, and other companies that used custom, unique, or proprietary components. For example, during a momentary lapse of reason in the early '90s, I purchased a Packard Bell system. I quickly outgrew the capabilities of the system, so I thought I'd upgrade it with a new motherboard and a faster processor. It was then that I discovered, to my horror, that LPX systems were not an interchangeable standard. Because of riser card differences, virtually no interchangeability of motherboards, riser cards, chassis, and power supplies existed. I had what I now refer to as a "disposable PC"—the kind you can't upgrade and have to throw away instead. Suddenly, the money I thought I had saved when initially purchasing the system paled in comparison to what I'd now have to spend to completely replace it. Lesson learned.

After several bad upgrade and repair experiences, I decided never again would I be trapped by systems using proprietary or nonstandard components. By purchasing only systems built with industry- standard parts, I could easily and inexpensively upgrade, maintain, or repair the systems for many years into the future. I have been preaching the gospel of industry-standard components in my seminars and in this book ever since.

Of course, building your own system from scratch is one way to avoid proprietary components, but often that route is more costly in both time and money than purchasing a prebuilt system. And what systems should I recommend for people who want an inexpensive prebuilt system but one that uses industry-standard parts so it can be inexpensively upgraded and repaired later? Although many system vendors and assemblers exist, I've settled on companies such as Gateway, MicronPC, and Dell. In fact, those are really the three largest system vendors that deal direct, and they mostly sell systems that use industry-standard ATX form factor components in all their main desktop system product lines. Or so I thought.

It seems that when Dell converted to the ATX motherboard form factor in mid-1996, it unfortunately defected from the newly released standard and began using specially modified Intel-supplied ATX motherboards with custom-wired power connectors. Inevitably, it also had custom power supplies made that duplicated the nonstandard pinout of the motherboard power connectors.

An even bigger crime than simply using nonstandard power connectors is that only the pinout is nonstandard; the connectors look like and are keyed the same as is dictated by true ATX. Therefore, nothing prevents you from plugging the Dell nonstandard power supply into a new industry-standard ATX motherboard you installed in your Dell case as an upgrade, or even plugging a new upgraded industry-standard ATX power supply into your existing Dell motherboard. But mixing either a new ATX board with the Dell supply or a new ATX supply with the existing Dell board is a recipe for silicon toast. How do you like your fried chips: medium or well-done?

Frankly, I'm amazed I haven't heard more about this because Dell has climbed to the lead in worldwide PC sales. In any case, I figure by getting this information out I can save thousands of innocent motherboards and power supplies from instant death upon installation.

If you've already fallen victim to this nasty circumstance, believe me, I feel your pain. I discovered this the hard way as well—by frying parts. At first, I thought the upgraded power supply I installed in one of my Dell systems was bad, especially considering the dramatic way it smoked when I turned on the system: I actually saw fire through the vents! Good thing I decided to check the color codes on the connectors and verify the pinout on another Dell system by using a voltmeter before I installed and fried a second supply. I was lucky in that the smoked supply didn't take the motherboard with it; I can only surmise that the supply fried so quickly it sacrificed itself and saved the motherboard. You might not be so lucky, and in most cases I'd expect you'd fry the board and supply together.

Call me a fool, but I didn't think I'd have to check the color-coding or get out my voltmeter to verify the Dell "pseudo-ATX" power connector pinouts before I installed a new ATX supply or motherboard. You'll also find that motherboard and power supply manufacturers don't like to replace these items under warranty when they are fried in this manner due to nonstandard connector wiring.

Dell's official explanation for its lack of conformance to the ATX standard was, "In the mid-90s the industry moved to a higher use of 3.3v motherboard components. Dell engineers designed a connector that supported the increased use of 3.3v current which differed from the industry proposed designs that we deemed less than robust." Unfortunately, this explanation doesn't hold much water because the standard ATX connector incorporated three 3.3v pins, allowing for up to 18A of current, and the addition of the Auxiliary Connector added two more pins with 10A of additional current. Dell's pseudo-ATX design had only three 3.3v pins in the Auxiliary Connector, which could supply only up to 15A to the board. You can see that even the main ATX Connector alone had more 3.3v current than Dell's design using two connectors!

Because its technical explanation fails to address the issue, the only other reason I can imagine it did this is to lock people into purchasing replacement motherboards or power supplies from Dell. What makes this worse is that Dell uses virtually all Intel-manufactured boards in its systems. One system I have uses an Intel D815EEA motherboard, which is the same board used by many of the other major system builders, including Gateway and Micron. It's the same, except for the power connectors, that is. The difference is that Dell has Intel custom-make the boards for Dell with the nonstandard connectors. Everybody else gets virtually the same Intel boards, but with industry-standard connectors.

Tables 3.8 and 3.9 show the nonstandard Dell main and auxiliary power supply connections. This nonstandard wiring is used on Dell's pseudo-ATX systems.

Table 3.8 Dell Proprietary (Nonstandard) ATX Main Power Connector Pinout (Wire Side View)

Color	Signal	Pin	Pin	Signal	Color
Gray	PS_On	11	1	+5v	Red
Black	Gnd	12	2	Gnd	Black
Black	Gnd	13	3	+5v	Red
Black	Gnd	14	4	Gnd	Black
White	–5v	15	5	Power_Good	Orange
Red	+5v	16	6	+5VSB (Standby)	Purple
Red	+5v	17	7	+12v	Yellow
Red	+5v	18	8	–12v	Blue
KEY (blank)	—	19	9	Gnd	Black
Red	+5v	20	10	Gnd	Black

Table 3.9 Dell Proprietary (Nonstandard) ATX Auxiliary Power Connector Pinout

Pin	Signal	Color	Pin	Signal	Color
1	Gnd	Black	4	+3.3	Blue/White
2	Gnd	Black	5	+3.3	Blue/White
3	Gnd	Black	6	+3.3	Blue/White

At first I thought that if all Dell did was switch some of the terminals around, I could use a terminal pick to remove the terminals from the connectors (with the wires attached) and merely reinsert them into the proper connector positions, enabling me to use the Dell power supply with an upgraded ATX motherboard in the future. Unfortunately, if you study the Dell main and auxiliary connector pinouts I've listed here and compare them to the industry-standard ATX pinouts listed earlier, you'll see that not only are the voltage and signal positions changed, but the number of terminals carrying specific voltages and grounds has changed as well. You could modify a Dell supply to work with a standard ATX board or modify a standard ATX supply to work with a Dell board, but you'd have to do some cutting and splicing in addition to swapping some terminals around. Usually, it isn't worth the time and effort.

If you do decide to upgrade the motherboard in any Dell system purchased between 1996 and 2000, a simple solution is available—just be sure you replace both the motherboard AND power supply with industry-standard ATX components at the same time. That way nothing gets fried, and you'll be back to having a true industry-standard ATX system. If you want to replace just the Dell motherboard, you're out of luck unless you get your replacement board from Dell. On the other hand, if you want to replace just the power supply, you do have one alternative. PC Power and Cooling now makes a version of its high-performance 300W ATX power supply with the modified Dell wiring for about $100. The internals are identical to its industry-standard, high-performance 300W ATX supply (which it sells for about 30% less)—only the number and arrangement of wires has changed.

Fortunately, starting in 2000, Dell switched to using industry-standard ATX power connections in its Dimension 4300, 4400, 8200, and newer systems. That means barring any other unforeseen glitches, these systems should be more easily upgradable by just replacing either the power supply or the motherboard alone. I, for one, am glad to see Dell moving back toward industry standardization because its systems are now more appealing to purchase as a starting point for a system that will be user upgradable and repairable in the future.

Power Switch Connectors

Three main types of power switches are used on PCs. They can be describes as follows:

• Integral Power Supply AC switch

• Front Panel Power Supply AC switch

• Front Panel Motherboard Controlled switch

The earliest systems had power switches integrated or built directly into the power supply, which turned the main AC power to the system on and off. This was a simple design, but because the power supply was mounted to the rear or side of the system, it required reaching around to the back to actuate the switch. Also, switching the AC power directly meant the system couldn't be remotely started without special hardware.

Starting in the late '80s systems began using remote front panel switches. These were essentially the same power supply design as the first type. The only difference is that the AC switch was now mounted remotely (usually on the front panel of the chassis), rather than integrated in the power supply unit, and connected to the power supply via a four-wire cable. The ends of the cable are fitted with spade connector lugs, which plug onto the spade connectors on the power switch. The cable from the power supply to the switch in the case contains four color-coded wires. In addition, a fifth wire supplying a ground connection to the case might be included. The switch was usually included with the power supply and heavily shrink-wrapped or insulated where the connector lugs attached to prevent electric shock.

This solved the ergonomic problem of reaching the switch, but it still didn't enable remote or automated system power-up without special hardware. Plus, you now had a 120v AC switch mounted in the chassis, with wires carrying dangerous voltage through the system. Some of these wires are hot anytime the system is plugged in (all are hot with the system turned on), creating a dangerous environment for the average person when messing around inside her system.

The four or five wires are color-coded as follows:

• Brown and blue. These wires are the live and neutral feed wires from the 110v power cord to the power supply. These are always hot when the power supply is plugged in.

• Black and white. These wires carry the AC feed from the switch back to the power supply. These leads should be hot only when the power supply is plugged in and the switch is turned on.

• Green or green with a yellow stripe. This is the ground lead. It should be connected to the PC case and should help ground the power supply to the case.

On the switch, the tabs for the leads are usually color-coded; if not, you'll find that most switches have two parallel tabs and two angled tabs. If no color-coding is on the switch, plug the blue and brown wires onto the tabs that are parallel to each other and the black and white wires to the tabs that are angled away from each other. If none of the tabs are angled, simply make sure the blue and brown wires are plugged into the most closely spaced tabs on one side of the switch and the black and white wires on the most closely spaced tabs on the other side.

See Figure 3.12 as a guide.

Figure 3.12 Power supply remote push button switch connections.

As long as the blue and brown wires are on the one set of tabs and the black and white leads are on the other, the switch and supply will work properly. If you incorrectly mix the leads, you will likely blow the circuit breaker for the wall socket because mixing them can create a direct short circuit.

All ATX and subsequent power supplies that employ the 20-pin motherboard connector use the PS_ON signal to power up the system. As a result, the remote switch does not physically control the power supply's access to the 110v AC power, as in the older-style power supplies. Instead, the power supply's on or off status is toggled by a PS_ON signal received on pin 14 of the ATX connector.

The PS_ON signal can be generated physically by the computer's power switch or electronically by the operating system. PS_ON is an active low signal, meaning that the power supply voltage outputs are disabled (the system is off) when the PS_ON is high (greater than or equal to 2.0v). This excludes the +5VSB (standby) on pin 9, which is active whenever the power supply is connected to an AC power source. The PS_ON signal is maintained by the power supply at either 3.3v or 5v. This signal is then routed through the motherboard to the remote switch on the front of the case. When the switch is pressed, the PS_ON signal is grounded. When the power supply sees the PS_ON signal drop to 0.8v or less, the power supply (and system) is turned on. Thus, the remote switch in an ATX-style system (which includes NLX and SFX systems as well) carries up to only +5v of DC power, rather than the full 115v–230v AC current like that of the older AT/LPX form factors.

Peripheral Power Connectors

In addition to the motherboard power connectors, power supplies include a variety of peripheral power connectors for everything from floppy and hard drives to internal case fans. The following sections discuss the various types of connectors you're likely to find in your PC.

Peripheral and Floppy Drive Power Connectors

The disk drive connectors on power supplies are fairly universal with regard to pin configuration and even wire color. Figure 3.13 shows the peripheral and floppy power connectors.

Figure 3.13 Peripheral and floppy power connectors.

Table 3.10 shows the standard disk drive power connector pinout and wire colors, whereas Table 3.11 shows the pinouts for the smaller floppy drive power connector.

Table 3.10 Peripheral Power Connector Pinout (Large Drive Power Connector)

Signal	Color	Pin	Pin	Signal	Color
+12V	Yellow	1	3	Gnd	Black
Gnd	Black	2	4	+5V	Red

Table 3.11 3 1/2'' Floppy Power Connector Pinout (Small Drive Power Connector)

Signal	Color	Pin	Pin	Signal	Color
+5V	Red	1	3	Gnd	Black
Gnd	Black	2	4	+12V	Yellow

Note that the pin numbering and voltage designations are reversed on the two connectors. Be careful if you are making or using an adapter cable from one type of connector to another. Reversing the red and yellow wires will fry the drive or device you plug into.

To determine the location of pin 1, look at the connector carefully. It is usually embossed in the plastic connector body; however, it is often tiny and difficult to read. Fortunately, these connectors are keyed and therefore difficult to insert incorrectly. Figure 3.14 shows the keying with respect to pin numbers on the larger drive power connector.

Figure 3.14 A peripheral female power supply connector.

Early power supplies featured only two large style drive connectors, normally called peripheral connectors today. Later power supplies featured four or more of the larger peripheral (drive) connectors and one or two of the smaller 3 1/2'' floppy drive connectors. Depending on their power ratings and intended uses, some supplies have as many as eight peripheral/drive connectors.

If you are adding drives and need additional disk drive power connectors, Y splitter cables (see Figure 3.15) and large to small drive power connector adapters (see Figure 3.16) are available from many electronics supply houses (including Radio Shack). These cables can adapt a single power connector to service two drives or enable you to convert the large peripheral power connector to a smaller floppy drive power connector. If you are using several Y-adapters, be sure that your total power supply output is capable of supplying the additional power.

Figure 3.15 A common Y-adapter power cable.

Figure 3.16 A peripheral–to–floppy-drive power adapter cable.

Power Supply Loading

PC power supplies are of a switching rather than a linear design. The switching type of design uses a high-speed oscillator circuit to convert the higher wall-socket AC voltage to the much lower DC voltage used to power the PC and PC components. Switching type power supplies are noted for being very efficient in size, weight, and energy in comparison to the linear design, which uses a large internal transformer to generate various outputs. This type of transformer-based design is inefficient in at least three ways. First, the output voltage of the transformer linearly follows the input voltage (hence the name linear), so any fluctuations in the AC power going into the system can cause problems with the output. Second, the high current-level (power) requirements of a PC system require the use of heavy wiring in the transformer. Third, the 60Hz (hertz) frequency of the AC power supplied from your building is difficult to filter out inside the power supply, requiring large and expensive filter capacitors and rectifiers.

The switching supply, on the other hand, uses a switching circuit that chops up the incoming power at a relatively high frequency. This enables the use of high-frequency transformers that are much smaller and lighter. Also, the higher frequency is much easier and cheaper to filter out at the output, and the input voltage can vary widely. Input ranging from 90 volts to 135 volts still produces the proper output levels, and many switching supplies can automatically adjust to 220v input.

One characteristic of all switching-type power supplies is that they do not run without a load. This means that you must have the supply plugged into something drawing power for the supply to work. If you simply have the power supply on a bench with nothing plugged into it, either the supply burns up or its protection circuitry shuts it down. Most power supplies are protected from no-load operation and shut down automatically. Some of the cheap clone supplies, however, lack the protection circuit and relay. They are destroyed after a few seconds of no-load operation. A few power supplies have their own built-in load resistors, so they can run even though no normal load is plugged in.

According to IBM specifications for the standard 192-watt power supply used in the original AT, a minimum load of 7.0 amps was required at +5v and a minimum of 2.5 amps was required at +12v for the supply to work properly.

Because floppy drives present no +12v load unless they are spinning, systems without a hard disk drive often do not operate properly. Some power supplies have a minimum load requirement for both the +5v and +12v sides. If you fail to meet this minimum load, the supply shuts down.

Because of this characteristic, when IBM used to ship the original AT systems without a hard disk, they plugged the hard disk drive power cable into a large 5-ohm, 50-watt sandbar resistor, which was mounted in a little metal cage assembly where the drive would have been. The AT case had screw holes on top of where the hard disk would go, specifically designed to mount this resistor cage.

This resistor would be connected between pin 1 (+12v) and pin 2 (Ground) on the hard disk power connector. This would place a 2.4-amp load on the supply's +12v output, drawing 28.8 watts of power (it would get hot!) and thus enabling the supply to operate normally. Note that the cooling fan in most power supplies draws approximately 0.1–0.25 amps, bringing the total load to 2.5 amps or more. If the load resistor were missing, the system would intermittently fail to start up or operate properly. The motherboard would draw +5v at all times, but +12v would normally be used only by motors, and the floppy drive motors would be off most of the time.

Most of the power supplies in use today do not require as much of a load as the original IBM AT power supply. In most cases, a minimum load of 0–0.3 amps at +3.3v, 2.0–4.0 amps at +5v, and 0.5–1.0 amps at +12v is considered acceptable. Most motherboards easily draw the minimum +5v current by themselves. The standard power supply cooling fan draws only 0.1–0.25 amps, so the +12v minimum load might still be a problem for a diskless workstation. Generally, the higher the rating on the supply, the more minimum load required; however, exceptions do exist, so this is a specification you want to check when evaluating power supplies.

Some high-quality switching power supplies have built-in load resistors and can run under a no-load situation because the supply loads. Other high-quality power supplies, such as those from PC Power and Cooling, have no internal load resistors. They require only a small load on the +5v line to operate properly. Many of the cheaper clone supplies, which often do not have built-in load resistors, might require +3.3v, +5v, and +12v loads to work.

If you want to bench test a power supply, make sure you place loads on at least one but preferably all of the positive voltage outputs. This is one reason you should test the supply while it is installed in the system, instead of testing it separately on the bench. For impromptu bench testing, you can use a spare motherboard and hard disk drive to load the outputs.

Power Supply Ratings

A system manufacturer should be able to provide you the technical specifications of the power supplies it uses in its systems. This type of information can be found in the system's technical-reference manual, as well as on stickers attached directly to the power supply. Power supply manufacturers can also supply this data, which is preferable if you can identify the manufacturer and contact them directly or via the Web.

The input specifications are listed as voltages, and the output specifications are listed as amps at several voltage levels. IBM reports output wattage level as "specified output wattage." If your manufacturer does not list the total wattage, you can convert amperage to wattage by using the following simple formula:

watts = volts x amps

For example, if a motherboard is listed as drawing 6 amps of +5v current, that would be 30 watts of power, according to the formula.

By multiplying the voltage by the amperage available at each output and then adding the results, you can calculate the total capable output wattage of the supply.

Table 3.12 shows the rated outputs at each of the voltage levels for supplies with different manufacturer-specified output ratings. To compile the table, I referred to the specification sheets for supplies from Astec Standard Power and PC Power and Cooling. Although most of the ratings are accurate, they are somewhat misleading for the higher wattage units.

Table 3.12 Typical Non-ATX Power Supply Output Ratings

Rated Output (Watts)	100W	150W	200W	250W	300W	375W	450W
Output current (amps):
+5V	10.0	15.0	20.0	25.0	32.0	35.0	45.0
+12V	3.5	5.5	8.0	10.0	10.0	13.0	15.0
–5V	0.3	0.3	0.3	0.5	1.0	0.5	0.5
–12V	0.3	0.3	0.3	0.5	1.0	0.5	1.0
Calc. output (watts)	97.1	146.1	201.1	253.5	297.0	339.5	419.5

Adding a +3.3v output to the power supply modifies the equation significantly. Table 3.13 contains data for various ATX power supplies from PC Power and Cooling that include a +3.3v output.

Table 3.13 PC Power and Cooling ATX Power Supply Output Ratings

Rated Output	235W	275W	300W	350W	400W	425W
Output current (amps):
+5V	22.0	30.0	30.0	32.0	30.0	50.0
+3.3V	14.0	14.0	14.0	28.0	28.0	40.0
Max watts +5V and +3.3V:	125W	150W	150W	215W	215W	300W
+12V	8.0	10.0	12.0	10.0	14.0	15.0
–5V	0.5	0.5	0.5	0.3	1.0	0.3
–12V	1.0	1.0	1.0	0.8	1.0	1.0

If you compute the total output using the formula described earlier, these power supplies seem to produce outputs that are much higher than their ratings. The 300W model, for example, comes out at 354.7 watts. However, notice that the supply also has a maximum combined output for the +3.3v and +5v of 150 watts. This means you cannot draw the maximum rating on both the 5v and 3.3v lines simultaneously but must keep the total combined draw between them at 150W. This brings the total output to a more logical 308.5 watts.

Most PC power supplies have ratings between 150 and 300 watts. Although lesser ratings are not usually desirable, you can purchase heavy-duty power supplies for most systems that have outputs as high as 600 watts or more.

The 300-watt and larger units are recommended for fully optioned desktops or tower systems. These supplies run any combination of motherboard and expansion card, as well as a large number of disk drives and other peripherals. In most cases, you cannot exceed the ratings on these power supplies—the system will be out of room for additional items first!

Most power supplies are considered to be universal, or worldwide. That is, they also can run on the 220v, 50-cycle current used in Europe and many other parts of the world. Many power supplies that can switch from 110v to 220v input do so automatically, but a few require you to set a switch on the back of the power supply to indicate which type of power you will access.

Power Supply Specifications

In addition to power output, many other specifications and features go into making a high-quality power supply. I have had many systems over the years. My experience has been that if a brownout occurs in a room with several systems running, the systems with higher-quality power supplies and higher output ratings are far more likely to make it through the power disturbances unscathed, whereas others choke.

High-quality power supplies also help protect your systems. A power supply from a vendor such as Astec or PC Power and Cooling will not be damaged if any of the following conditions occur:

• A 100% power outage of any duration

• A brownout of any kind

• A spike of up to 2,500v applied directly to the AC input (for example, a lightning strike or a lightning simulation test)

Decent power supplies have an extremely low current leakage to ground of less than 500 microamps. This safety feature is important if your outlet has a missing or an improperly wired ground line.

As you can see, these specifications are fairly tough and are certainly representative of a high-quality power supply. Make sure that your supply can meet these specifications.

You can also use many other criteria to evaluate a power supply. The power supply is a component many users ignore when shopping for a PC, and it is therefore one that some system vendors might choose to skimp on. After all, a dealer is far more likely to be able to increase the price of a computer by spending money on additional memory or a larger hard drive than by installing a better power supply.

When buying a computer (or a replacement power supply), you always should learn as much as possible about the power supply. However, many consumers are intimidated by the vocabulary and statistics found in a typical power supply specification. Here are some of the most common parameters found on power supply specification sheets, along with their meanings:

• Mean Time Between Failures (MTBF) or Mean Time To Failure (MTTF). The (calculated) average interval, in hours, that the power supply is expected to operate before failing. Power supplies typically have MTBF ratings (such as 100,000 hours or more) that are clearly not the result of real-time empirical testing. In fact, manufacturers use published standards to calculate the results, based on the failure rates of the power supply's individual components. MTBF figures for power supplies often include the load to which the power supply was subjected (in the form of a percentage) and the temperature of the environment in which the tests were performed.

• Input Range (or Operating Range). The range of voltages that the power supply is prepared to accept from the AC power source. For 110v AC current, an input range of 90v–135v is common; for 220v current, a 180v–270v range is typical.

• Peak Inrush Current. The greatest amount of current drawn by the power supply at a given moment immediately after it is turned on, expressed in terms of amps at a particular voltage. The lower the current, the less thermal shock the system experiences.

• Hold-Up Time. The amount of time (in milliseconds) that a power supply can maintain output within the specified voltage ranges after a loss of input power. This enables your PC to continue running without resetting or rebooting if a brief interruption in AC power occurs. Values of 15–30 milliseconds are common for today's power supplies, and the higher (longer), the better. The ATX12V specification calls for a minimum of 17ms hold-up time.

• Transient Response. The amount of time (in microseconds) a power supply takes to bring its output back to the specified voltage ranges after a steep change in the output current. In other words, the amount of time it takes for the output power levels to stabilize after a device in the system starts or stops drawing power. Power supplies sample the current being used by the computer at regular intervals. When a device stops drawing power during one of these intervals (such as when a floppy drive stops spinning), the power supply might supply too high a voltage to the output for a brief time. This excess voltage is called overshoot, and the transient response is the time that it takes for the voltage to return to the specified level. This is seen as a spike in voltage by the system and can cause glitches and lockups. Once a major problem that came with switching power supplies, overshoot has been greatly reduced in recent years. Transient response values are sometimes expressed in time intervals, and at other times they are expressed in terms of a particular output change, such as "power output levels stay within regulation during output changes of up to 20 percent."

• Overvoltage Protection. Defines the trip points for each output at which the power supply shuts down or squelches that output. Values can be expressed as a percentage (for example, 120% for +3.3 and +5v) or as raw voltages (for example, +4.6v for the +3.3v output and +7.0v for the +5v output).

• Maximum Load Current. The largest amount of current (in amps) that safely can be delivered through a particular output. Values are expressed as individual amperages for each output voltage. With these figures, you can calculate not only the total amount of power the power supply can supply, but also how many devices using those various voltages it can support.

• Minimum Load Current. The smallest amount of current (in amps) that must be drawn from a particular output for that output to function. If the current drawn from an output falls below the minimum, the power supply could be damaged or automatically shut down.

• Load Regulation (or Voltage Load Regulation). When the current drawn from a particular output increases or decreases, the voltage changes slightly as well, usually increasing as the current rises. Load regulation is the change in the voltage for a particular output as it transitions from its minimum load to its maximum load (or vice versa). Values, expressed in terms of a +/– percentage, typically range from +/–1% to +/–5% for the +3.3, +5, and +12v outputs.

• Line Regulation. The change in output voltage as the AC input voltage transitions from the lowest to the highest value of the input range. A power supply should be capable of handling any AC voltage in its input range with a change in its output of 1% or less.

• Efficiency. The ratio of power input to power output, expressed in terms of a percentage. Values of 65%–85% are common for power supplies today. The remaining 15%–35% of the power input is converted to heat during the AC/DC conversion process. Although greater efficiency means less heat inside the computer (always a good thing) and lower electric bills, it should not be emphasized at the expense of precision, stability, and durability, as evidenced in the supply's load regulation and other parameters.

• Ripple (or Ripple and Noise, or AC Ripple, or PARD [Periodic and Random Deviation]). The average voltage of all AC effects on the power supply outputs, typically measured in millivolts peak-to-peak or as a percentage of the nominal output voltage. The lower this figure is, the better. Higher-quality units normally are rated at 1% ripple (or less), which if expressed in volts would be 1% of the output. Consequently, for +5v that would be 0.05v or 50mv (millivolts). Ripple can be caused by internal switching transients, feed through of the rectified line frequency, and other random noise.

Power Factor Correction

Recently, the power line efficiency and harmonic distortion generation of PC power supplies have come under examination. This generally falls under the topic of the power factor of the supply. Interest in power factor is not only due to an improvement in power efficiency, but also because of a reduction in the generation of harmonics back on the power line. In particular, new standards are now mandatory in many European Union (EU) countries that require harmonics be reduced below a specific amount. The circuitry required to do this is called power factor correction (PFC).

The power factor measures how effectively electrical power is being used, and is expressed as a number between 0 and 1. A high power factor means that electrical power is being used effectively, whereas a low power factor indicates poor utilization of electrical power. To understand the power factor, you must understand how power is used.

Generally, two types of loads are placed on AC power lines:

• Resistive. Power converted into heat, light, motion, or work

• Inductive. Sustains an electromagnetic field, such as in a transformer or motor

A resistive load is often called working power and is measured in kilowatts (KW). An inductive load, on the other hand, is often called reactive power and is measured in kilovolt-amperes-reactive (KVAR). Working power and reactive power together make up apparent power, which is measured in kilovolt-amperes (KVA). The power factor is measured as the ratio of working power to apparent power, or working power/apparent power (KW/KVA). The ideal power factor is 1, where the working power and apparent power are the same.

The concept of a resistive load or working power is fairly easy to understand. For example, a light bulb that consumes 100W of power generates 100W worth of heat and light. That is a pure resistive load. An inductive load, on the other hand, is a little harder to understand. Think about a transformer, which has coil windings to generate an electromagnetic field and then induce current in another set of windings. A certain amount of power is required to saturate the windings and generate the magnetic field, even though no work is being done. A power transformer that is not connected to anything is a perfect example of a pure inductive load. An apparent power draw exists to generate the fields, but no working power exists because no actual work is being done.

When the transformer is connected to a load, it uses both working power and reactive power. In other words, power is consumed to do work (for example, if the transformer is powering a light bulb), and apparent power is used to maintain the electromagnetic field in the transformer windings. In an AC circuit, these loads can become out of sync or phase, meaning they don't peak at the same time, which can generate harmonic distortions back down the power line. I've seen examples where electric motors have caused distortions in television sets plugged into the same power circuit.

PFC usually involves adding capacitance to the circuit to maintain the inductive load without drawing additional power from the line. This makes the working power and apparent power the same, which results in a power factor of 1. It usually isn't just as simple as adding some capacitors to a circuit, although that can be done and is called passive power factor correction. Active power factor correction involves a more intelligent circuit designed to match the resistive and inductive loads so they are seen as the same by the electrical outlet.

A power supply with active power factor correction draws low distortion current from the AC source and has a power factor rating of 0.9 or greater. A non–power factor corrected supply draws highly distorted current and is sometimes referred to as a nonlinear load. The power factor of a noncorrected supply is typically 0.6–0.8. Therefore, only 60% of the apparent power consumed is actually doing real work!

Having a power supply with active PFC might or might not lower your electric bill (it depends on how your power is measured), but it will definitely reduce the load on the building wiring. With PFC, all the power going into the supply is converted into actual work and the wiring is less overworked. For example, if you ran a number of computers on a single breaker-controlled circuit and found that you were blowing the breaker periodically, you could switch to systems with active PFC power supplies and reduce the load on the wiring by up to 40%, meaning you would be less likely to blow the breaker.

The International Electrical Committee (IEC) has released standards dealing with the low-frequency public supply system. The initial standards were 555.2 (Harmonics) and 555.3 (Flicker), but they have since been refined and are now available as IEC 1000-3-2 and IEC 1000-3-3, respectively. As governed by the EMC directive, most electrical devices sold within the member countries of the EU must meet the IEC standards. The IEC1000-3-2/3 standards became mandatory in 1997 and 1998.

Even if you don't live in a country where PFC is required, I highly recommend specifying PC power supplies with active PFC. The main benefits of PFC supplies is that they do not overheat building wiring or distort the AC source waveform, which causes less interference on the line for other devices.

Power Supply Safety Certifications

Many agencies around the world certify electric and electronic components for safety and quality. The most commonly known agency in the United States is Underwriters Laboratories, Inc. (UL). UL standard #1950—the Standard for Safety of Information Technology Equipment, Including Electrical Business Equipment, Third Edition—covers power supplies and other PC components. You should always purchase power supplies and other devices that are UL-certified. It has often been said that, although not every good product is UL-certified, no bad products are.

In Canada, electric and electronic products are certified by the Canadian Standards Agency (CSA). The German equivalents are TÜV Rheinland and VDE, and NEMKO operates in Norway. These agencies are responsible for certification of products throughout Europe. Power supply manufacturers that sell to an international market should have products that are certified at least by UL, the CSA, and TÜV—if not by all the agencies listed, and more.

Apart from UL-type certifications, many power supply manufacturers, even the most reputable ones, claim that their products have a Class B certification from the Federal Communications Commission, meaning that they meet FCC standards for electromagnetic and radio frequency interference (EMI/RFI). This is a contentious point, however, because the FCC does not certify power supplies as individual components. Title 47 of the Code of Federal Regulations, Part 15, Section 15.101(c) states as follows:

"The FCC does NOT currently authorize motherboards, cases, and internal power supplies. Vendor claims that they are selling 'FCC-certified cases,' 'FCC-certified motherboards,' or 'FCC-certified internal power supplies' are false."

In fact, an FCC certification can be issued collectively only to a base unit consisting of a computer case, motherboard, and power supply. Thus, a power supply purported to be FCC-certified was actually certified along with a particular case and motherboard—not necessarily the same case and motherboard you are using in your system. This does not mean, however, that the manufacturer is being deceitful or that the power supply is inferior. If anything, this means that when evaluating power supplies, you should place less weight on the FCC certification than on other factors, such as UL certification.

Overloading the Power Supply

The biggest cause of power supply overload problems has historically been filling up the expansion slots and adding more drives. Multiple hard drives, CD-ROM drives, and floppy drives can create quite a drain on the system power supply. Be sure you have enough +12v power to run all the drives you plan to install. Tower systems can be especially problematic because they have so many drive bays. Just because the case has room for the devices doesn't mean the power supply can support them. Be sure you have enough +5v power to run all your expansion cards, especially PCI cards. It pays to be conservative, but remember that most cards draw less than the maximum allowed. Today's newest processors can have very high current requirements for the +5v or +3.3v supplies. When selecting a power supply for your system, be sure to take into account any future processor upgrades.

Many people wait until an existing component fails to replace it with an upgraded version. If you are on a tight budget, this "if it ain't broke, don't fix it" attitude might be necessary. Power supplies, however, often do not fail completely all at once; they can fail in an intermittent fashion or allow fluctuating power levels to reach the system, which results in unstable operation. You might be blaming system lockups on software bugs when the culprit is an overloaded power supply. If you have been running your original power supply for a long time and have upgraded your system in other ways, you should expect some problems.

Although there is certainly an appropriate place for the exacting power-consumption calculations you've read about in this section, a great many experienced PC users prefer the "don't worry about it" power calculation method. This technique consists of buying or building a system with a good-quality 300-watt or higher power supply (or upgrading to such a supply in an existing system) and then upgrading the system freely, without concern for power consumption.

Power Off When Not in Use

Should you turn off a system when it is not in use? To answer this frequent question, you should understand some facts about electrical components and what makes them fail. Combine this knowledge with information on power consumption, cost, and safety to come to your own conclusion. Because circumstances can vary, the best answer for your own situation might be different from the answer for others, depending on your particular needs and applications.

Frequently, powering a system on and off does cause deterioration and damage to the components. This seems logical, but the simple reason is not obvious to most people. Many believe that flipping system power on and off frequently is harmful because it electrically "shocks" the system. The real problem, however, is temperature or thermal shock. As the system warms up, the components expand; as it cools off, the components contract. In addition, various materials in the system have different thermal expansion coefficients, which means that they expand and contract at different rates. Over time, thermal shock causes deterioration in many areas of a system.

From a pure system-reliability viewpoint, you should insulate the system from thermal shock as much as possible. When a system is turned on, the components go from ambient (room) temperature to as high as 185° F (85° C) within 30 minutes or less. When you turn off the system, the same thing happens in reverse, and the components cool back to ambient temperature in a short period of time.

Thermal expansion and contraction remains the single largest cause of component failure. Chip cases can split, allowing moisture to enter and contaminate them. Delicate internal wires and contacts can break, and circuit boards can develop stress cracks. Surface-mounted components expand and contract at rates different from the circuit board they are mounted on, which causes enormous stress at the solder joints. Solder joints can fail due to the metal hardening from the repeated stress, resulting in cracks in the joint. Components that use heatsinks, such as processors, transistors, or voltage regulators, can overheat and fail because the thermal cycling causes heatsink adhesives to deteriorate and break the thermally conductive bond between the device and heatsink. Thermal cycling also causes socketed devices and connections to loosen, or creep, which can cause a variety of intermittent contact failures.

→ See "SIMMs, DIMMs, and RIMMs," p. 360

Thermal expansion and contraction affect not only chips and circuit boards, but also things such as hard disk drives. Most hard drives today have sophisticated thermal compensation routines that make adjustments in head position relative to the expanding and contracting platters. Most drives perform this thermal compensation routine once every 5 minutes for the first 30 minutes the drive is running, and then every 30 minutes thereafter. In many drives, this procedure can be heard as a rapid "tick-tick-tick-tick" sound. In essence, anything you can do to keep the system at a constant temperature prolongs the life of the system, and the best way to accomplish this is to leave the system either permanently on or permanently off. Of course, if the system is never turned on in the first place, it should last a long time indeed!

Now, I am not saying that you should leave all systems on 24 hours a day. A system powered on and left unattended can be a fire hazard (I have had monitors spontaneously catch fire—luckily, I was there at the time), is a data security risk (from cleaning crews and other nocturnal visitors), can be easily damaged if moved while running, and wastes electrical energy.

I currently pay 11 cents for a kilowatt-hour of electricity. A typical desktop-style PC with display consumes at least 300 watts (0.3 kilowatt) of electricity (and that is a conservative estimate). This means it would cost 3.3 cents to run my typical PC for an hour. Multiplying by 168 hours in a week means that it would cost $5.54 per week to run this PC continuously. If the PC were turned on at 9 a.m. and off at 5 p.m., it would be on only 40 hours per week and would cost only $1.32—a savings of $4.22 per week! Multiply this savings by 100 systems, and you are saving $422 per week. Multiply this by 1,000 systems, and you are saving $4,220 per week! Using systems certified under the EPA Energy Star program (so-called green PCs) would account for an additional savings of around $1 per system per week—or $1,000 per week for 1,000 systems. The great thing about Energy Star systems is that the savings are even greater if the systems are left on for long periods of time because the power management routines are automatic.

Based on these facts, my recommendations are that you power on the systems at the beginning of the workday and off at the end of the workday. Do not power the systems off for lunch, breaks, or any other short periods of time. If you are a home user, leave your computer on if you are going to be using it later in the day or if instant access is important. I'd normally recommend home users turn off the system when leaving the house or when sleeping. Servers, of course, should be left on continuously. This seems to be the best compromise of system longevity with pure economics. No matter what, these are just guidelines; if it works better for you to leave your system on 24 hours a day, seven days a week, make it so.

Power Management

As the standard PC configuration has grown to include capabilities formerly considered options, the power requirements of the system have increased steadily. Larger displays, CD-ROM drives, and audio adapters all need more power to run, and the cost of operating a PC rises steadily. To address these concerns, several programs and standards are now being developed that are intended to reduce the power needed to run a PC as much as possible.

For standard desktop systems, power management is a matter of economy and convenience. By turning off specific components of the PC when they are not in use, you can reduce the electric bill and avoid having to power the computer up and down manually.

For portable systems, power management is far more important. Adding CD-ROMs, speakers, and other components to a laptop or notebook computer reduces even further what is in many cases a short battery life. By adding new power management technology, a portable system can supply power only to the components it actually needs to run, thus extending the life of the battery charge.

Energy Star Systems

The EPA has started a certification program for energy-efficient PCs and peripherals. To be a member of this program, the PC or display must drop to a power draw at the outlet of 30 watts or less during periods of inactivity. Systems that conform to this specification get to wear the Energy Star logo. This is a voluntary program; however, many PC manufacturers are finding that it helps them sell their systems if they can advertise these systems as energy efficient.

One problem with this type of system is that the motherboard and disk drives can go to sleep, which means they can enter a standby mode in which they draw very little power. This wreaks havoc with some of the older power supplies because the low power draw does not provide enough of a load for them to function properly. Most of the newer supplies on the market, which are designed to work with these systems, have a very low minimum-load specification. I suggest you ensure that the minimum load will be provided by the equipment in your system if you buy a power supply upgrade. Otherwise, when the PC goes to sleep, it might take a power switch cycle to wake it up again. This problem would be most noticeable if you invested in a very high-output supply and used it in a system that draws very little power to begin with.

Advanced Power Management

Advanced Power Management (APM) is a specification jointly developed by Intel and Microsoft that defines a series of interfaces between power management–capable hardware and a computer's operating system. When it is fully activated, APM can automatically switch a computer between five states, depending on the system's current activity. Each state represents a further reduction in power use, accomplished by placing unused components into a low-power mode.

Advanced Configuration and Power Interface

Advanced Configuration and Power Interface (ACPI) is a newer power management and system configuration standard supported by newer system BIOS software running Windows 98 and later operating systems. If your BIOS and operating system support ACPI, full power management control is now done by the operating system, rather than by the BIOS. ACPI is intended to offer a single place for power management and system configuration control; in the past, with APM you would often be able to make power management settings in the BIOS setup as well as the operating system that often overlapped or could have conflicting settings. ACPI is supported in newer systems in lieu of APM.

Power Supply Troubleshooting

Troubleshooting the power supply basically means isolating the supply as the cause of problems within a system and, if necessary, replacing it.

Many symptoms lead me to suspect that the power supply in a system is failing. This can sometimes be difficult for an inexperienced technician to see because, at times little connection seems to exist between the symptom and the cause—the power supply.

For example, in many cases a parity check error message can indicate a problem with the power supply. This might seem strange because the parity check message specifically refers to memory that has failed. The connection is that the power supply powers the memory, and memory with inadequate power fails.

It takes some experience to know when this type of failure is power related and not caused by the memory. One clue is the repeatability of the problem. If the parity check message (or other problem) appears frequently and identifies the same memory location each time, I would suspect that defective memory is the problem. However, if the problem seems random, or if the memory location the error message cites as having failed seems random, I would suspect improper power as the culprit. The following is a list of PC problems that often are related to the power supply:

• Any power-on or system startup failures or lockups.

• Spontaneous rebooting or intermittent lockups during normal operation.

• Intermittent parity check or other memory-type errors.

• Hard disk and fan simultaneously failing to spin (no +12v).

• Overheating due to fan failure.

• Small brownouts cause the system to reset.

• Electric shocks felt on the system case or connectors.

• Slight static discharges disrupt system operation.

In fact, just about any intermittent system problem can be caused by the power supply. I always suspect the supply when flaky system operation is a symptom. Of course, the following fairly obvious symptoms point right to the power supply as a possible cause:

• System is completely dead (no fan, no cursor)

• Smoke

• Blown circuit breakers

If you suspect a power supply problem, some of the simple measurements and the more sophisticated tests outlined in this section can help you determine whether the power supply is at fault. Because these measurements might not detect some intermittent failures, you might have to use a spare power supply for a long-term evaluation. If the symptoms and problems disappear when a known good spare unit is installed, you have found the source of your problem.

Following is a simple flowchart to help you zero in on common power supply–related problems:

1. Check AC power input. Make sure the cord is firmly seated in the wall socket and in the power supply socket. Try a different cord.

2. Check DC power connections. Make sure the motherboard and disk drive power connectors are firmly seated and making good contact. Check for loose screws.

3. Check DC power output. Use a digital multimeter to check for proper voltages. If it's below spec, replace the power supply.

4. Check installed peripherals. Remove all boards and drives and retest the system. If it works, add back in items one at a time until the system fails again. The last item added before the failure returns is likely defective.

Many types of symptoms can indicate problems with the power supply. Because the power supply literally powers everything else in the system, everything from disk drive problems to memory problems to motherboard problems can often be traced back to the power supply as the root cause.

Overloaded Power Supplies

A weak or inadequate power supply can put a damper on your ideas for system expansion. Some systems are designed with beefy power supplies, as if to anticipate a great deal of system add-ons and expansion components. Most desktop or tower systems are built in this manner. Some systems have inadequate power supplies from the start, however, and cannot adequately service the power-hungry options you might want to add.

The wattage rating can sometimes be very misleading. Not all 300-watt supplies are created the same. People familiar with high-end audio systems know that some watts are better than others. This goes for power supplies, too. Cheap power supplies might in fact put out the rated power, but what about noise and distortion? Some of the supplies are under-engineered to just barely meet their specifications, whereas others might greatly exceed their specifications. Many of the cheaper supplies provide noisy or unstable power, which can cause numerous problems with the system. Another problem with under-engineered power supplies is that they can run hot and force the system to do so as well. The repeated heating and cooling of solid-state components eventually causes a computer system to fail, and engineering principles dictate that the hotter a PC's temperature, the shorter its life. Many people recommend replacing the original supply in a system with a heavier-duty model, which solves the problem. Because power supplies come in common form factors, finding a heavy-duty replacement for most systems is easy, as is the installation process.

Inadequate Cooling

Some of the available replacement power supplies have higher-capacity cooling fans than the originals, which can greatly prolong system life and minimize overheating problems—especially for the newer, hotter-running processors. If system noise is a problem, models with special fans can run more quietly than the standard models. These power supplies often use larger-diameter fans that spin more slowly, so they run more quietly but move the same amount of air as the smaller fans. PC Power and Cooling specializes in heavy-duty and quiet supplies; Astec has several heavy-duty models as well.

Ventilation in a system is also important. You must ensure adequate airflow to cool the hotter items in the system. Many processors today use passive heatsinks that require a steady stream of air to cool the chip. If the processor heatsink has its own fan, this is not much of a concern. If you have free expansion slots, you should space out the boards in your system to permit airflow between them. Place the hottest running boards nearest the fan or the ventilation holes in the system. Make sure that adequate airflow exists around the hard disk drive, especially for those that spin at high rates of speed. Some hard disks can generate quite a bit of heat during operation. If the hard disks overheat, data can be lost.

Always be sure to run your computer with the case cover on, especially if you have a loaded system. Removing the cover can actually cause a system to overheat. With the cover off, the power supply fan no longer draws air through the system. Instead, the fan ends up cooling the supply only, and the rest of the system must be cooled by simple convection. Although most systems do not immediately overheat for this reason, several of my own systems, especially those that are fully expanded, have overheated within 15–30 minutes when run with the case cover off.

In addition, be sure that any empty slot positions have the filler brackets installed. If you leave these brackets off after removing a card, the resultant hole in the case disrupts the internal airflow and can cause higher internal temperatures.

If you experience intermittent problems that you suspect are related to overheating, a higher-capacity replacement power supply is usually the best cure. Specially designed supplies with additional cooling fan capacity also can help. At least one company sells a device called a fan card, but I am not convinced these are a good idea. Unless the fan is positioned to draw air to or from the area outside the case, all it does is blow hot air around inside the system and provide a spot cooling effect for anything it is blowing on. In fact, adding fans in this manner contributes to the overall heat inside the system because the fan consumes power and generates heat.

CPU-mounted fans are an exception because they are designed only for spot cooling of the CPU. Many newer processors run so much hotter than the other components in the system that a conventional, finned aluminum heatsink can't do the job. In this case, a small fan placed directly over the processor provides a spot cooling effect that keeps the processor temperatures down. One drawback to these active processor cooling fans is that the processor overheats instantly and can be damaged if the fans fail. Whenever possible, try to use the biggest passive (finned aluminum) heatsink you can find and purchase a CPU fan from a reputable vendor.

Using Digital Multimeters

One simple test you can perform on a power supply is to check the output voltages. This shows whether a power supply is operating correctly and whether the output voltages are within the correct tolerance range. Note that you must measure all voltages with the power supply connected to a proper load, which usually means testing while the power supply is still installed in the system and connected to the motherboard and peripheral devices.

Selecting a Meter

You need a simple digital multimeter (DMM) or digital volt-ohm meter (DVOM) to perform voltage and resistance checks on electronic circuits (see Figure 3.17). You should use only a DMM instead of the older needle-type multimeters because the older meters work by injecting 9v into the circuit when measuring resistance, which damages most computer circuits.

Figure 3.17 A typical DMM.

A DMM uses a much smaller voltage (usually 1.5v) when making resistance measurements, which is safe for electronic equipment. You can get a good DMM with many features from several sources. I prefer the small, pocket-size meters for computer work because they are easy to carry around.

Some features to look for in a good DMM are as follows:

• Pocket size. This is self-explanatory, but small meters are available that have many, if not all, of the features of larger ones. The elaborate features found on some of the larger meters are not really necessary for computer work.

• Overload protection. This means that if you plug the meter into a voltage or current beyond the meter's capability to measure, the meter protects itself from damage. Cheaper meters lack this protection and can easily be damaged by reading current or voltage values that are too high.

• Autoranging. This means that the meter automatically selects the proper voltage or resistance range when making measurements. This is preferable to the manual range selection; however, really good meters offer both autoranging capability and a manual range override.

• Detachable probe leads. The leads easily can be damaged, and sometimes a variety of differently shaped probes are required for different tests. Cheaper meters have the leads permanently attached, which means you cannot easily replace them. Look for a meter with detachable leads that plug into the meter.

• Audible continuity test. Although you can use the ohm scale for testing continuity (0ohms indicates continuity), a continuity test function causes the meter to produce a beep noise when continuity exists between the meter test leads. By using the sound, you quickly can test cable assemblies and other items for continuity. After you use this feature, you will never want to use the ohms display for this purpose again.

• Automatic power off. These meters run on batteries, and the batteries can easily be worn down if the meter is accidentally left on. Good meters have an automatic shutoff that turns off the unit when it senses no readings for a predetermined period of time.

• Automatic display hold. This feature enables you to hold the last stable reading on the display even after the reading is taken. This is especially useful if you are trying to work in a difficult-to-reach area single-handedly.

• Minimum and maximum trap. This feature enables the meter to trap the lowest and highest readings in memory and hold them for later display, which is especially useful if you have readings that are fluctuating too quickly to see on the display.

Although you can get a basic pocket DMM for as little as $20, one with all these features is priced in the $100–$200 range. Radio Shack carries some nice inexpensive units, and you can purchase the high-end models from electronics supply houses, such as Newark or Digi-Key.

Measuring Voltage

To measure voltages on a system that is operating, you must use a technique called back probing on the connectors (see Figure 3.18). You cannot disconnect any of the connectors while the system is running, so you must measure with everything connected. Nearly all the connectors you need to probe have openings in the back where the wires enter the connector. The meter probes are narrow enough to fit into the connector alongside the wire and make contact with the metal terminal inside. The technique is called back probing because you are probing the connector from the back. You must use this back-probing technique to perform virtually all the following measurements.

Figure 3.18 Back probing the power supply connectors.

To test a power supply for proper output, check the voltage at the Power_Good pin (P8-1 on AT, Baby-AT, and LPX supplies; pin 8 on the ATX-type connector) for +3v to +6v of power. If the measurement is not within this range, the system never sees the Power_Good signal and therefore does not start or run properly. In most cases, the power supply is bad and must be replaced.

Continue by measuring the voltage ranges of the pins on the motherboard and drive power connectors. If you are measuring voltages for testing purposes, any reading within 10% of the specified voltage is considered acceptable, although most manufacturers of high-quality power supplies specify a tighter 5% tolerance. For ATX power supplies, the specification requires that voltages must be within 5% of the rating, except for the 3.3v current, which must be within 4%. The following table shows the voltage ranges within these tolerances.

	Loose Tolerance		Tight Tolerance
Desired Voltage	Min. (-10%)	Max. (+8%)	Min. (-5%)	Max. (+5%)
+3.3V	2.97V	3.63V	3.135	3.465
+/–5.0V	4.5V	5.4V	4.75	5.25
+/–12.0V	10.8V	12.9V	11.4	12.6

The Power_Good signal has tolerances that are different from the other voltages, although it is nominally +5v in most systems. The trigger point for Power_Good is about +2.4v, but most systems require the signal voltage to be within the tolerances listed here:

Signal	Minimum	Maximum
Power_Good (+5V)	3.0V	6.0V

Replace the power supply if the voltages you measure are out of these ranges. Again, it is worth noting that any and all power supply tests and measurements must be made with the power supply properly loaded, which usually means it must be installed in a system and the system must be running.

Specialized Test Equipment

You can use several types of specialized test gear to test power supplies more effectively. Because the power supply is one of the most failure-prone items in PCs today, you should have these specialized items if you service many PC systems.

Digital Infrared Thermometer

One of the greatest additions to my toolbox is a digital infrared thermometer. They also are called noncontact thermometers because they measure by sensing infrared energy without having to touch the item they are reading. This enables me to make instant spot checks of the temperature of a chip, a board, or the system chassis. They are available from companies such as Raytek (http://www.raytek.com) for under $100. To use these handheld items, you point at an object and then pull the trigger. Within seconds, the display shows a temperature readout. These devices are invaluable in checking to ensure your system is adequately cooled.

Repairing the Power Supply

Hardly anyone actually repairs power supplies anymore. Most of the time, it is simply easier, safer, or less expensive (considering the time and materials involved) to replace the power supply rather than to repair it. As mentioned earlier, replacement power supplies are available from many manufacturers. Even high-quality power supplies are not that expensive when compared to the labor required to repair them.

Sources for Replacement Power Supplies

Because one of the most failure-prone items in PC systems is the power supply, I am often asked to recommend a replacement. Literally hundreds of companies manufacture PC power supplies, and I certainly have not tested them all. I can, however, recommend some companies whose products I have come to know and trust.

Although other high-quality manufacturers are out there, at this time I recommend power supplies from either Astec Standard Power or PC Power and Cooling.

Using Power-Protection Systems

Power-protection systems do just what the name implies: They protect your equipment from the effects of power surges and power failures. In particular, power surges and spikes can damage computer equipment, and a loss of power can result in lost data. In this section, you learn about the four primary types of power-protection devices available and when you should use them.

Before considering any further levels of power protection, you should know that a quality power supply already affords you a substantial amount of protection. High-end power supplies from the vendors I recommend are designed to provide protection from higher-than-normal voltages and currents, and they provide a limited amount of power-line noise filtering. Some of the inexpensive aftermarket power supplies probably do not have this sort of protection. If you have an inexpensive computer, further protecting your system might be wise.

Of course, the easiest form of protection is to turn off and unplug your computer equipment (including your modem) when a thunderstorm is imminent. When this is not possible, however, other alternatives are available.

Power supplies should stay within operating specifications and continue to run a system even if any of these power line disturbances occur:

• Voltage drop to 80v for up to 2 seconds

• Voltage drop to 70v for up to .5 seconds

• Voltage surge of up to 143v for up to 1 second

Most high-quality power supplies and systems will not be damaged by the following occurrences:

• Full power outage

• Any voltage drop (brownout)

• A spike of up to 2,500v

Because of their internal protection, many computer manufacturers that use high-quality power supplies state in their documentation that external surge suppressors are not needed with their systems.

To verify the levels of protection built into the existing power supply in a computer system, an independent laboratory subjected several unprotected PC systems to various spikes and surges of up to 6,000v—considered the maximum level of surge that can be transmitted to a system through an electrical outlet. Any higher voltage would cause the power to arc to the ground within the outlet. None of the systems sustained permanent damage in these tests. The worst thing that happened was that some of the systems rebooted or shut down when the surge was more than 2,000v. Each system restarted when the power switch was toggled after a shutdown.

I do not use any real form of power protection on my systems, and they have survived near-direct lightning strikes and powerful surges. The most recent incident, only 50 feet from my office, was a direct lightning strike to a brick chimney that blew the top of the chimney apart. None of my systems (which were running at the time) were damaged in any way from this incident; they just shut themselves down. I was able to restart each system by toggling the power switches. An alarm system located in the same office, however, was destroyed by this strike. I am not saying that lightning strikes or even much milder spikes and surges cannot damage computer systems—another nearby lightning strike did destroy a modem and serial adapter installed in one of my systems. I was just lucky that the destruction did not include the motherboard.

This discussion points out an important oversight in some power-protection strategies: Do not forget to provide protection from spikes and surges on the phone line.

The automatic shutdown of a computer during power disturbances is a built-in function of most high-quality power supplies. You can reset the power supply by flipping the power switch from on to off and back on again. Some power supplies even have an auto-restart function. This type of power supply acts the same as others in a massive surge or spike situation: It shuts down the system. The difference is that after normal power resumes, the power supply waits for a specified delay of 3–6 seconds and then resets itself and powers the system back up. Because no manual switch resetting is required, this feature might be desirable in systems functioning as network servers or in those found in other unattended locations.

The first time I witnessed a large surge cause an immediate shutdown of all my systems, I was extremely surprised. All the systems were silent, but the monitor and modem lights were still on. My first thought was that everything was blown, but a simple toggle of each system-unit power switch caused the power supplies to reset, and the units powered up with no problem. Since that first time, this type of shutdown has happened to me several times, always without further problems.

The following types of power-protection devices are explained in the sections that follow:

• Surge suppressors

• Phone-line surge protectors

• Line conditioners

• Standby power supplies (SPS)

• Uninterruptible power supplies (UPS)

Surge Suppressors (Protectors)

The simplest form of power protection is any one of the commercially available surge protectors—that is, devices inserted between the system and the power line. These devices, which cost between $20 and $200, can absorb the high-voltage transients produced by nearby lightning strikes and power equipment. Some surge protectors can be effective for certain types of power problems, but they offer only very limited protection.

Surge protectors use several devices, usually metal-oxide varistors (MOVs), that can clamp and shunt away all voltages above a certain level. MOVs are designed to accept voltages as high as 6,000v and divert any power above 200v to ground. MOVs can handle normal surges, but powerful surges such as a direct lightning strike can blow right through them. MOVs are not designed to handle a very high level of power and self-destruct while shunting a large surge. These devices therefore cease to function after either a single large surge or a series of smaller ones. The real problem is that you cannot easily tell when they no longer are functional. The only way to test them is to subject the MOVs to a surge, which destroys them. Therefore, you never really know whether your so-called surge protector is protecting your system.

Some surge protectors have status lights that let you know when a surge large enough to blow the MOVs has occurred. A surge suppressor without this status indicator light is useless because you never know when it has stopped protecting.

Underwriters Laboratories has produced an excellent standard that governs surge suppressors, called UL 1449. Any surge suppressor that meets this standard is a very good one and definitely offers a line of protection beyond what the power supply in your PC already offers. The only types of surge suppressors worth buying, therefore, should have two features:

• Conformance to the UL 1449 standard

• A status light indicating when the MOVs are blown

Units that meet the UL 1449 specification say so on the packaging or directly on the unit. If this standard is not mentioned, it does not conform. Therefore, you should avoid it.

Another good feature to have in a surge suppressor is a built-in circuit breaker that can be manually reset rather than a fuse. The breaker protects your system if it or a peripheral develops a short. These better surge suppressors usually cost about $40.

Phone Line Surge Protectors

In addition to protecting the power lines, it is critical to provide protection to your systems from any connected phone lines. If you are using a modem or fax board that is plugged into the phone system, any surges or spikes that travel through the phone line can damage your system. In many areas, the phone lines are especially susceptible to lightning strikes, which are the leading cause of fried modems and damage to the computer equipment attached to them.

Several companies manufacture or sell simple surge protectors that plug in between your modem and the phone line. These inexpensive devices can be purchased from most electronics supply houses. Most of the cable and communication products vendors listed in the Vendor List sell these phone line surge protectors. Some of the standard power line surge protectors include connectors for phone line protection as well.

Line Conditioners

In addition to high-voltage and current conditions, other problems can occur with incoming power. The voltage might dip below the level needed to run the system, resulting in a brownout. Forms of electrical noise other than simple voltage surges or spikes might travel through the power line, such as radio-frequency interference or electrical noise caused by motors or other inductive loads.

Remember two things when you wire together digital devices (such as computers and their peripherals):

• Any wire can act as an antenna and have voltage induced in it by nearby electromagnetic fields, which can come from other wires, telephones, CRTs, motors, fluorescent fixtures, static discharge, and, of course, radio transmitters.

• Digital circuitry responds with surprising efficiency to noise of even a volt or two, making those induced voltages particularly troublesome. The electrical wiring in your building can act as an antenna, picking up all kinds of noise and disturbances.

A line conditioner can handle many of these types of problems. A line conditioner is designed to remedy a variety of problems. It filters the power, bridges brownouts, suppresses high-voltage and current conditions, and generally acts as a buffer between the power line and the system. A line conditioner does the job of a surge suppressor, and much more. It is more of an active device, functioning continuously, rather than a passive device that activates only when a surge is present. A line conditioner provides true power conditioning and can handle myriad problems. It contains transformers, capacitors, and other circuitry that can temporarily bridge a brownout or low-voltage situation. These units usually cost $100–$300, depending on the power-handling capacity of the unit.

RTC/NVRAM Batteries (CMOS Chips)

All 16-bit and higher systems have a special type of chip in them that combines a real-time clock (RTC) with at least 64 bytes (including the clock data) of Non-Volatile RAM (NVRAM) memory. This chip is officially called the RTC/NVRAM chip but is often referred to as the CMOS chip or CMOS RAM because the type of chip used is produced using a CMOS (Complementary Metal-Oxide Semiconductor) process. CMOS design chips are known for very low power consumption. This special RTC/NVRAM chip is designed to run off a battery for several years.

The original chip of this type used in the IBM AT was the Motorola 146818 chip. Although the chips used today have different manufacturers and part numbers, they all are designed to be compatible with this original Motorola part.

These chips include a real-time clock. Its function should be obvious: The clock enables software to read the date and time and preserves the date and time data even when the system is powered off or unplugged.

The NVRAM portion of the chip has another function. It is designed to store basic system configuration, including the amount of memory installed, types of floppy and hard disk drives, and other information. Some of the more modern motherboards use extended NVRAM chips with as much as 2KB or more of space to hold this configuration information. This is especially true for Plug and Play systems, which store not only the motherboard configuration but also the configuration of adapter cards. This system can then read this information every time you power on the system.

These chips normally are powered by some type of battery while the system is off. This battery preserves the information in the NVRAM and powers the clock. Most systems use a lithium-type battery because they have a very long life, especially at the low power draw from the typical RTC/NVRAM chip.

Some systems have a chip that has the battery embedded within it. These are made by several companies—including Dallas Semiconductor and Benchmarq. These chips are notable for their long lives. Under normal conditions, the battery will last for 10 years—which is, of course, longer than the useful life of the system. If your system uses one of the Dallas or Benchmarq modules, the battery and chip must be replaced as a unit because they are integrated. Most of the time, these chip/battery combinations are installed in a socket on the motherboard just in case a problem requires an early replacement. You can get new modules direct from the manufacturers for $18 or less, which is often less than the cost of the older separate battery alone.

Some systems do not use a battery at all. Hewlett-Packard, for example, includes a special capacitor in some of its systems that is automatically recharged anytime the system is plugged in. Note that the system does not have to be running for the capacitor to charge; it only has to be plugged in. If the system is unplugged, the capacitor will power the RTC/NVRAM chip for up to a week or more. If the system remains unplugged for a duration longer than that, the NVRAM information is lost. In that case, these systems can reload the NVRAM from a backup kept in a special flash ROM chip contained on the motherboard. The only pieces of information that will actually be missing when you repower the system will be the date and time, which will have to be reentered. By using the capacitor combined with an NVRAM backup in flash ROM, these systems have a very reliable solution that will last indefinitely.

Many systems use only a conventional battery, which can be either directly soldered into the motherboard or plugged in via a battery connector. For those systems with the battery soldered in, normally a spare battery connector exists on the motherboard where you can insert a conventional plug-in battery, should the original ever fail. In most cases, you would never have to replace the motherboard battery, even if it were completely dead.

Conventional batteries come in many forms. The best are of a lithium design because they will last from two to five years or more. I have seen systems with conventional alkaline batteries mounted in a holder; these are much less desirable because they fail more frequently and do not last as long. Also, they can be prone to leak, and if a battery leaks on the motherboard, the motherboard can be severely damaged. By far, the most commonly used battery for motherboards today is the 2032 lithium coin battery, which is about the size of a quarter and is readily available.

Besides the various battery types, the chip can require any one of several voltages. The batteries in PCs are typically 3.0v, 3.6v, 4.5v, or 6v. If you are replacing the battery, be sure your replacement is the same voltage as the one you removed from the system. Some motherboards can use batteries of several voltages. Use a jumper or switch to select the various settings. If you suspect your motherboard has this capability, consult the documentation for instructions on changing the settings. Of course, the easiest thing to do is to replace the existing battery with another of the same type.

Symptoms that indicate that the battery is about to fail include having to reset the clock on your PC every time you shut down the system (especially after moving it) and problems during the system's POST, such as drive-detection difficulties. If you experience problems such as these, you should make note of your system's CMOS settings and replace the battery as soon as possible.

When you replace a battery, in most cases the existing data stored in the NVRAM is lost. Sometimes, however, the data remains intact for several minutes (I have observed NVRAM retain information with no power for an hour or more), so if you make the battery swap quickly, the information in the NVRAM might be retained. Just to be sure, I recommend that you record all the system configuration settings stored in the NVRAM by your system Setup program. In most cases, you would want to run the BIOS Setup program and copy or print out all the screens showing the various settings. Some Setup programs offer the capability to save the NVRAM data to a file for later restoration if necessary.

After replacing a battery, power up the system and use the Setup program to check the date and time setting and any other data that was stored in the NVRAM.

© 2025 Pearson Education, Informit. All rights reserved.

800 East 96th Street, Indianapolis, Indiana 46240