Diagnose and Repair Modern Electronics: A DIY Guide: Components (part 2)

Resistors

Resistors offer opposition to current, dissipating the opposed power as heat. Limiting current is a vital function in any circuit, so every electronic product has resistors.

They come in many shapes and sizes, but those with leads are easily identifiable by their color bands indicating the resistance value in ohms. Some large resistors are marked numerically, as are some of the tiny surface-mount parts. See FIG. 10.

FIG. 10 Resistors

The basic type of resistor, found in virtually everything, is the carbon composition resistor. Made from a carbon compound, these resistors run the full range of values, from less than 1 (ohm) to 10 M-ohm (megohms, or millions of ohms). Most you'll see will be over 10 -ohm and under 1 M-ohm. The tolerance for standard carbon comp resistors is ± 5 percent. That is, the measured value should be no more than 5 percent high or low of the stated resistance.

Some applications require the use of wire-wound resistors. These look a lot like carbon comp parts, but notice the coil of wire under the paint on the body. Wire-wound resistors can dissipate more heat than can carbon comps. They also can be manufactured to very tight tolerances, but they have some inductance, since they are coils, and are ill suited to high-frequency circuits where the inductance might matter. The two types should not be interchanged.

In addition to the resistance value, the power dissipating capability of a resistor, measured in watts, is an important factor. Since power is dissipated as heat, knowing how much heat the part can take before disintegrating is vital. Standard carbon comps are rated at 1/4 watt, with some slightly larger ones able to dissipate 1/2 watt. Very tiny ones with leads are rated at 1/8 watt, while surface-mount versions typically vary from 1/8 watt down to 1/32 watt.

Other resistor formulations include carbon film, metal film and metal oxide. Carbon film types introduce a bit less noise into the circuit and are used in areas where that's a significant issue. Metal film and metal oxide parts have tighter tolerances, in the range of 1 or 2 percent of stated value. Some circuits require that for proper operation.

Symbol

Markings

The use of color-coding dates back to the vacuum tube days, when resistors got so hot that printed numbers would evaporate. To read the color code, you must first determine which end of the resistor is the start and which is the far end. Look for a gold or silver band; that's the tolerance marking, indicating the far end, and there's usually a little extra space between that band and the others. The first digit will be at the end farthest from the tolerance band.

Each number is represented by a color. The scheme is as follows:

Black 0 Green 5 Brown 1 Blue 6 Red 2 Violet 7 Orange 3 Gray 8 Yellow 4 White 9 The tolerance bands at the far end are as follows:

Brown ± 1 percent Red ± 2 percent Gold ± 5 percent Silver ± 10 percent

To determine a resistor's value, read the first two bands as numbers. The third band is a multiplier, indicating how many zeros you need to tack on to the numerical value. So, for instance, red-red-brown would be 2, 2, and one zero, or 220 ohms.

Yellow-violet-orange would be 4, 7, and three zeros, or 47,000 ohms, a.k.a. 47 K-ohm. Be careful not to confuse a black third band with a zero; it means no zeros.

This scheme works great until the resistance value gets below 10 -ohm. Try it-there's no way to mark such a value. For resistors that low, the third band is set to gold, meaning that there's a decimal point between the first two bands. So, green-blue-gold-gold would be a 5.6 -ohm resistor with a ± 5 percent tolerance. You won't see too many resistors under 10 -ohm, but you might run across one in an audio output stage or a power supply.

Some resistors have four bands plus a tolerance band. These are higher-precision parts with tighter tolerances and must be read slightly differently. With these, read the first three bands as numbers and the fourth band as the multiplier. So, the 47 K-ohm resistor would read yellow-violet-black-red.

How can you tell what type of resistor you have? Sometimes you can't. If the resistor has a gold or silver tolerance band, assume it's a standard carbon composition part, unless it's in a very low-noise circuit like an audio preamp, in which case carbon film might be a more appropriate replacement. If it has a red or brown tolerance band, indicating higher precision, it might be a metal film or metal oxide component.

Many surface-mount resistors are marked numerically, using the same idea. The last number is a multiplier. If you see an R between the numbers, that's a decimal point. You'll see that only on resistors of rather low value. For instance, 4R7 means 4.7 ohms. If you see a number with a zero at the end, don't confuse that to mean a number; it indicates no zeros. Thus, 220 means 22 -ohm, not 220 -ohm, and 220 -ohm would be marked 221. Also, a letter after the -ohm on any type of numerically marked resistor is not part of the numerical value, even if it's a K. So, 47 ohm-K is still 47 ohm, not 47 K-ohm. Used this way, the K denotes 10 percent tolerance. Pretty crazy, huh? Some tiny surface-mount resistors are too small for numerical value markings.

Instead, they sport a two-digit number that must be cross-referenced from a list. The number itself has no direct relation to the value. This marking scheme has several permutations, and you can find them on the Internet. It's highly unlikely, though, that you'll ever need to replace one of those tiny resistors, because they carry very little current and rarely fail.

Uses

Resistors are found in just about every circuit. They limit the current that can pass through other parts. For instance, transistors amplify a signal by using it as a control for a larger current provided by the power supply, somewhat like the handle on a spigot controls a large flow of water. A resistor between the supply and the transistor sets how much current the transistor has to control. Without the resistor, the transistor would have to handle all of the supply's current and would self-destruct.

What Destroys Them

Resistors rarely fail on their own. Heat caused by passing too much current burns them out, sometimes literally. Carbon composition resistors can go up in flames or become a charred lump when a short in some other part pulls a lot of current through them. It's not uncommon to see one with a burn mark obscuring the color bands.

FIG. 11 Pots and trimpots

Out-of-Circuit Testing

Use your DMM's ohms scale to see if the resistance is within the specified tolerance range. Most resistors do better than their tolerances, but expect a small difference between what's on the code and what you measure.

When the resistor is charred beyond your being able to read the color code, it can be a real problem unless you have the unit's schematic diagram. Luckily, resistance values follow a standard pattern, because there would be no point in producing resistors whose values fell within another resistor's tolerance range. So, you may be able to infer a burned-out band's value from others that can still be seen. If you ever need it, you can look up the list of standard values online.

Potentiometers

A potentiometer, or pot, is a variable resistor. A small one used for internal circuit adjustments is called a trimpot. Pots and trimpots may have two, three or (rarely) four leads, but most have three. The outer two are connected to the substrate on which the resistive element is formed, with one lead at each end of the resistor. The center lead goes to the wiper, a movable metal contact whose point touches the resistive element, selecting a resistance value that rises relative to one outer lead while falling relative to the other as you turn the knob. See FIG. 11.

A two-wire pot has no connection to one end of the resistive element but is otherwise the same. Some two-wire pots connect the free end of the element to the wiper, which slightly affects the resistance curve as you turn it, but it doesn't matter a whole lot. Two-wire pots are sometimes called rheostats.

Four-wire pots, used mostly on stereo receivers to provide the "loudness" function which increases bass at low volume levels, are like three-wire pots, but with an extra tap partway up the resistive element.

The amount of resistance change you get per degree of rotation is even from end to end on linear taper pots. On log taper pots, also called audio taper, a logarithmic resistance curve is used, so that audio loudness, which is perceived on a log curve, will seem to increase or decrease at a constant rate as the control is varied. Trimpots, whose primary use is for set-and-forget internal adjustments on circuit boards, are always linear taper.

How much power the pot can dissipate depends on its size. It's not marked on the part. In most applications, only small signals are applied to it, so it's not much of an issue.

Some pots have metal shafts, while others use plastic. Plastic provides insulation in applications like power supplies, where you might come in contact with a dangerous voltage. Also, some pots have switches built into them, and those may be rotary, operating at one end of the wiper's travel, or push-pull.

Most trimpots rotate through less than 360 degrees, just like pots. Those used in applications requiring high precision may be multi-turn, with a threaded gear inside providing the reduction ratio. Multi-turn trimpots use a small screw for adjustment, usually off to one side of the body of the component. Be wary of turning the screw, because there's no visual indicator to help you set it back where it was.

In stereo receivers, pots can be ganged together onto one shaft, so that turning it will affect both channels together. Each pot is internally isolated from the other, though you may find one end of all of them tied to ground at the terminals.

Symbols

Markings

Pots are numerically marked with their resistance values using the same scheme as resistors. A B in the code indicates a linear pot, while a C means a log pot. An A, though, can mean either, as the codes have changed over the years. Some pots have LIN or LOG printed on them. Most don't, though.

Uses

Pots are used to adjust operating parameters for analog signals and power supply voltages. Once, they were the primary method of setting just about everything. In this digital age, volume, treble, bass, brightness, contrast and such are more often selected from a menu or adjusted with up/down buttons.

Trimpots on circuit boards are less common too, but you'll still find some, especially in power supplies, including switchers.

What Destroys Them

Most failures are caused by wear or dirt where the wiper makes contact with the resistive element. The symptom is scratchiness in audio, flashes on the screen in video, or inability to set the pot to specific spots without the wiper's losing contact with the element. Contact cleaner spray will usually clear it up, but if the element is too worn, replacement is the only option.

Occasionally, the resistive element cracks, resulting in some weird symptoms because the wiper is still connected to one end but not the other. Plus, which end is connected reverses as you turn the control over the broken spot. Audio may blast through part of the control's range and disappear below the break point.

Out-of-Circuit Testing

Test the outer two leads on a three-lead pot with your DMM's ohms scale. It should read something close to the printed value. If the element is cracked, it'll read open.

To check for integrity of the wiper's contact, connect the meter between one end of the pot and the wiper. Slowly turn the pot through its range, observing the change in resistance. This is one test better done with a good old analog VOM; the meter needle will swing back and forth with the position of the wiper, and it's easy to interpret.

A two-lead pot should read something close to its stated resistance at one end of its control range and 0 ohms at the other.

To verify if a pot is linear taper or log taper, measure the resistance at one-third of the rotation and again at two-thirds. See if those values are about one-third and two-thirds of the total resistance from end to end. If it's a log taper part, they won't even be close.

Relays

Relays are switches controlled by putting current into a coil. When the coil is energized, the resulting magnetic field pulls a metal plate toward it, pressing the attached switch contacts against opposing contacts. In this way, a small current can control a much larger one, just as a transistor does in switch mode.

FIG. 12 Relays

Relays may have any number of contacts for switching multiple, unconnected circuits at the same time. As with switches, each set is called a pole. Also, some contacts may be normally open, or NO, meaning that they are not touching until the relay is energized, and some may be normally closed, or NC, meaning the opposite.

Each direction is called a throw. Thus, a double-pole, double-throw, or DPDT, relay would have two switches, each with three contacts: the NO side, the movable contact, and the NC side. See FIG. 12.

Symbols

Markings

Relays may be marked schematically for their internal construction-that is, where the coil connections are and what kinds of switches are inside. You may find a voltage rating; that specifies the coil voltage, not the maximum voltage permitted on the switch contacts. A resistance rating is for the coil too. If you find one, you can calculate the coil's pull-in current with Ohm's Law. It also helps you when testing the coil with your DMM.

Some relays include an internal diode across the coil to prevent the reverse voltage it generates when power is removed from feeding back into the circuitry and damaging it. If the relay shows polarity markings (+ and -), it has a diode.

The maximum current the switch contacts can handle usually isn't shown, but it might be. If the markings read 12 VDC, 3 A, that indicates a 12-volt coil intended to be driven with DC power, with switch contacts capable of switching 3 amps. Some relays are made specifically for AC coil operation, too, with the appropriate markings.

FIG. 13 Switches

Uses

Relays were once widely used to switch large currents with smaller ones. These days, semiconductors usually do that job, but some applications still employ relays. Power supply delay circuits, which prevent power from reaching the circuitry for a few moments after the supply is turned on, often have relays. Speaker protection circuits in high-power audio amplifiers use them too, because the high-current audio signal is not impeded at all by a relay, but it would be by a semiconductor.

Most relays make an audible click when they switch, giving their presence away.

What Destroys Them Relay troubles usually involve the switch contacts. Corrosion from age and oxidation, and pitting from arcing when large currents are switched, cause resistance or flaky contact. Once in awhile, a relay will become sticky, not wanting to open back up after power is removed from the coil. This condition can be caused by arcing in the contacts making them stick to each other, and by weakening of the spring used to pull the plate away from the coil's iron core.

If the relay has a removable cover, you may be able to pop it off and clean the contacts. Sometimes just pulling a piece of paper soaked in contact cleaner between them will spiff them up and restore proper operation. If that's not enough, very light wiping with fine sandpaper may remove the outer layer of gunk. Silver polish works too, but be sure to get it all off when you're done. Be careful not to bend the contacts, and don't sand off the plating; it's vital for their long-term survival. Whichever method you use, wipe the contacts with cleaner-soaked paper to remove residue before you put the relay's cover back on.

Out-of-Circuit Testing Use your DMM's continuity or lowest ohms scale. Check for coil continuity. If there's a diode, be sure to check in both directions. The coil shouldn't read 0 ohms; there's enough wire there for dozens to a few hundred ohms. If it reads very near zero on a relay that has a diode, suspect a shorted diode, especially if the symptom suggests that the coil doesn't want to pull in the switch. Also check that there is no continuity between the coil and its metal core. If there is any, you need a new relay.

Check the NC contacts with the meter. They should read 0 ohms or very close to it. Use your bench power supply to energize the coil. If it has a diode, be certain to get the polarity correct, with + to the diode's cathode, not its anode. (Remember, the diode is supposed to be wired backward, so it won't conduct when power is applied.) With no diode, polarity doesn't matter. Once you hear the click, check the NO contacts. They should also read 0 ohms or very close to it. When you disconnect the power supply, the contacts should return to their original state. The NO contacts should open and the NC contacts should close.

Switches---Switches permit or interrupt the passage of current. There are many, many kinds of switches, and they're used in just about everything. Toggle switches, slide switches, rotary switches, leaf switches, pushbuttons, internal switches on jacks…there's practically no end to the varieties. See FIG. 13. Many newer products don’t have "hard" power switches that actually disconnect power to the unit. Instead, a low-current switch signals the microprocessor, which then shuts down power using semiconductors to interrupt the flow.

Switches can have any number of contacts for switching multiple, unconnected circuits at the same time. Rotary and slide switches, especially, may have many sets, while toggle switches rarely have more than two or three. Each set is called a pole. Also, some contacts may be normally open, or NO, meaning that they are not connected with the switch in the "off " position, and some may be normally closed, or NC, meaning the opposite. Each direction is called a throw. Thus, a double-pole, double-throw, or DPDT, switch would have two separate sets of contacts, each with three elements: the NO side, the movable contact, and the NC side.

Symbols

Markings

If marked at all, switches may show their maximum voltage and current ratings.

Uses

Expect to find switches everywhere. From pushbuttons on front panels and remote controls to tiny slide switches on circuit boards, they handle power and information input in essentially all products. Leaf switches, with bendable, springy metal arms, sense the position of mechanisms like tape transports and laser optical heads, informing the microprocessor of the state of moving parts. Switches inside jacks sense when accessories are plugged in, altering system behavior to accommodate them.

What Destroys Them

As with relays, age, oxidation and contact pitting from arcing usually do them in.

If the switch's construction permits any access, try spraying some contact cleaner inside, and then work the switch a bunch of times.

Out-of-Circuit Testing

Test switches with your DMM's continuity or lowest ohms scale. The contacts should read 0 ohms when closed and infinity when open. There's an exception, though: Some pushbuttons, such as the kind on remotes, laptop keyboards and tiny products like digital cameras, use a carbon-impregnated plastic or rubber contact to make the connection. You can identify them by their soft feel when pressed; they don't click.

These switches are intended only for signaling, not for handling significant current, and they may have a few tens of ohms of resistance when in the "on" state. While such a reading would indicate a bad toggle switch, it's fine with these little guys.

Transistors

Along with integrated circuits, transistors are the active elements that do most of the work in circuits, amplifying, processing and generating signals, switching currents and providing the oomph needed to drive speakers, headphones, motors and lamps. See FIG. 14.

Transistors act like potentiometers, but instead of your hand's turning the shaft, a signal does. The current fed by the power supply through the pot can be much greater than that required to turn the shaft, providing gain, or amplification, as the wiggling signal molds a bigger version of itself from the power supply's steady DC.

There are thousands of subtypes of transistors, but most fall into three categories: bipolar, JFET and MOSFET.

Bipolar transistors are the standard types used in products since the 1950s. They come in two polarities, NPN and PNP, and consist of three elements joined by two junctions. The three elements are the base, emitter and collector, each with its own lead. Current passing between the base and the emitter permits a much larger current to pass between the collector and the emitter, with one of them being fed from the power supply. The greater the voltage difference between the base and emitter, the more current will pass, and a proportionally higher current can pass between collector and emitter. In most configurations, the signal is applied to the base, causing a bigger version of itself to be formed from the flow between the collector and emitter. The ratio of base-to-emitter current to collector-to-emitter current is the transistor's gain, and is inherent in the component's design.

In an NPN transistor, the base must be positive with respect to the emitter for collector-to-emitter current to pass. In a PNP transistor, the base must be negative. So, the two types are of opposite polarity and cannot be interchanged. Most transistors used today are NPN, but you will find circuits with some PNP parts.

JFETs, or junction field effect transistors, work on a similar principle. They label their three elements differently. Instead of the base, the controlling terminal is called the gate. The emitter is called the source, and the collector is the drain.

Instead of a current passing from gate to source, application of a voltage to the gate goes nowhere but results in an electric field that controls a channel in the transistor, permitting current to pass between drain and source. This gives the JFET a very high input impedance, which is another way of saying that it does not take much signal current to turn it on.

MOSFETs, or metal-oxide-semiconductor field-effect transistors, are similar to JFETs, and they use the same terminal names, but their internal construction is a bit different.

They have even higher input impedance and some other desirable characteristics that have resulted in their pretty much dominating FET applications. You may see JFETs in older gear, but you're much more likely to see nothing but MOSFETs in newer equipment.

Like bipolars, FETs come in two polarities, P-channel and N-channel, corresponding to PNP and NPN bipolar transistors. They also come in enhancement mode and depletion mode types, specifying what happens when the gate voltage is zero. An enhancement mode FET will be turned off with no voltage at the gate; like a bipolar transistor, it requires a bias voltage to turn it on. A depletion mode FET will be turned on with zero gate voltage. The only way to turn it off is to apply a voltage of opposite polarity to the one that will increase current flow.

Luckily, you don't need to worry too much about these arcane details. If a FET is bad, you'll look up its part number and replace it with a compatible type. Still, knowing the basics of how these parts work is essential for understanding how to troubleshoot circuits using them…which is pretty much all circuits.

As notice, transistors have many parameters, so it's not surprising that there are thousands of subtypes with different gains, power dissipation capabilities, frequency limits and so on. Some are similar enough that they can be interchanged in many circuits, but most are not. To replace one part number with another, you need a transistor substitution guide or an online cross-reference guide.

Symbols

NPN PNP N-channel JFET P-channel JFET

P-channel MOSFET enh N-channel MOSFET enh P-channel MOSFET dep N-channel MOSFET dep

Markings

Transistors are marked by part number, called a type number. There are thousands of these numbers! Some numbers indicate whether a bipolar part is NPN or PNP. If the number starts with 2SA or 2SB, it's PNP. If it starts with 2SC or 2SD, it's NPN. Sometimes the 2S will be left off, and there are plenty of type numbers that don't follow this scheme at all. A 3N indicates a FET, but some of them have numbers starting with 2N, and there are lots of other kinds of numbers for these too.

Some transistors have house numbers, which are proprietary numbering schemes used by different manufacturers to mean different things. There is no way to ascertain what the industry-standard number would be for such a part. Tiny, surface-mount transistors often have no numbers at all.

The arrangement of the leads varies with transistor type. Small Japanese parts with leads are usually laid out ECB, left to right, while American parts are often EBC. Metal-encased power transistors have only two leads and use the metal casing as the collector. Plastic power transistors are usually BCE, with the metal tab, if there is one, being C.

Uses

In discrete (nonintegrated-circuit) stages, transistors are the active elements doing the work, with passive parts like resistors and capacitors supporting their operation. You'll see this kind of construction in radio receivers, audio amplifiers and some sections of many other products. In stages where an IC is at the center of the action, transistors frequently do the interfacing between the IC and other parts of the circuitry, especially areas requiring more current than an IC can supply. MOSFETs are used as switches, permitting the microprocessor to turn power on and off to various parts of the circuitry.

Some of their very sensitive varieties are used to amplify and detect radio signals. Many audio power amplifier output stages are made from bipolar transistors. It's hard to find any function that transistors don't do. After electrolytic capacitors and bad connections, transistors will be the focus of much of your repair work.

What Destroys Them

Transistors are not especially fragile, but they work hard in many circuits and fail more often than do most components. Overheating due to excessive current will burn them out, as will too high a voltage. MOSFETs are particularly prone to shorts from static electricity. Sometimes the internal structure of a transistor develops a tiny flaw, and the thing self-destructs with no apparent cause. In fact, many random product failures occur for precisely this reason. You change the part and the unit works again, and you never find any reason for the dead transistor.

Out-of-Circuit Testing

If you have a transistor tester, use it! Nothing's easier than hooking up the leads and getting the test result. If you don't have one, you can use your DMM's diode test or, lacking that, the ohms scale to check for shorts. If you get near-zero ohms between any two leads, check in the other direction. If it's still near zero, the part is shorted.

Checking for opens is a bit more complicated, because some combinations of terminals should look open, depending on to what the control terminal is connected.

Connecting the base of a bipolar transistor to its collector should result in its turning on, showing measurable resistance between the emitter and collector in one direction.

Connecting the base to the emitter should turn it off.

Similarly, connecting the gate of a FET to its drain should turn it on, and connecting it to the source should turn it off. However, some FETs are symmetrical and will turn on with the gate connected to either of the other terminals, as long as the polarity of the applied voltage is what the gate needs. And the whole situation is complicated by the enhancement/depletion mode issue, because a depletion mode FET will stay on. FETs are not easy to test with an ohmmeter!

Voltage Regulators

Voltage regulators take incoming DC and hold their output voltage to a specific value as the incoming voltage fluctuates or the load varies with circuit operation. Yesteryear's simple products often had no voltage regulators, but practically everything made today does.

Linear regulators act like automatic variable resistors, passing the incoming current through a transistor called the series pass transistor and setting the base current to keep the output voltage constant. When the load changes and the voltage increases or decreases, the regulator detects that and adjusts the base current to compensate, altering the transistor's resistance and permitting more or less current to go through it. The power lost in the resistance of the transistor is wasted and dissipated as heat.

Switching regulators chop the incoming current into fast pulses whose width can be varied. Those pulses are then applied to a capacitor, which charges up, converting the pulses back to smooth DC. By monitoring the output voltage, the regulator detects changes and alters the pulse width. The wider each pulse, the more current can charge the capacitor, raising the output. Narrower pulses lower it. While more complicated, this approach, called pulse-width modulation (PWM), supplies only the current required to keep the output voltage constant, without wasting the excess as heat. Thus, it’s much more efficient.

PWM regulation is an inherent feature of switching power supplies, and some products use switching regulators internally as well. They are complicated, though, and also generate a fair amount of RF noise, so they aren't suitable for all uses.

Linear regulation, while wasteful, is still very common in low-current applications because the amount of power wasted is trivial. The linear approach is a lot simpler and cheaper, too, making it attractive.

While both types of regulators once took a bunch of components to implement, they can be had in chip form today, requiring just a few external parts to support their functions. The three-terminal linear voltage regulator, available in both fixed- and variable-voltage varieties, is the most common type you'll find. It looks like a transistor but is really an IC, with one terminal for input, one for ground and one for output. See FIG. 15. Hang a capacitor or two on it and it's ready to rock. Some products have several regulators supplying separate sections of their circuitry.

Symbol

IN OUT

REF

FIG. 15 Voltage regulators

Markings

Voltage regulators are marked by part number, like transistors. Some standard markings tell you the voltage, which is handy. In particular, the widely used 7800/7900 series of linear regulators offers useful marking information. All regulators starting with 78 are positive regulators with negative grounds. A 7805 is a 5-volt regulator, a 7812 is 12 volts, and so on. All parts starting with 79 are negative regulators with positive grounds. They use the same voltage numbering scheme.

Other regulators don't necessarily indicate their voltages in the part number, and you will have to look them up.

Uses

Voltage regulators provide stable voltage to entire devices or sections of them. In some applications, a regulator may feed a single stage or area of the circuitry. Some switching regulators can boost the voltage and regulate it at the same time. This is especially useful in battery-operated devices employing only one or two 1.5-volt cells but requiring a higher voltage.

What Destroys Them

Pulling too much current through a regulator can overheat and destroy it. This is especially true with linear regulators. Voltage spikes can cause internal shorts, and random chip failures occur too. Switching regulators are prone to blowing their chopper transistors, just like switching power supplies.

Out-of-Circuit Testing

You can use the ohms scale to check for shorts, but that's about it. To evaluate a regulator properly, it needs to be in a circuit, receiving power.

Zener Diodes Zener diodes are special diodes used in voltage regulators. In the forward direction, they conduct like normal diodes. In the reverse direction, they also act like regular diodes, blocking current. When the reverse voltage rises above a predetermined value set in manufacture, the zener breaks down in a nondestructive manner and conducts.

This results in a constant voltage drop across the part, making it useful as a voltage reference. Zeners dissipate power as heat, so they are rated in watts for how much they can take before overheating and burning out.

Zeners look much like other diodes, but many have somewhat beefier cases and thicker leads to increase heat dissipation capability. See FIG. 16.

Symbol

Markings

The band on a zener's case indicates the cathode, as with any diode. Since zeners are used for their reverse breakdown action, though, expect them to seem to face the wrong way in the circuit, with the band connected to positive. In fact, that's one way to help determine whether a diode you see on a circuit board is in fact a zener, and not just a normal diode.

Zeners are marked with part numbers, when they are marked at all. If the number begins with 1N47 and is followed by two more digits, that's a zener. Some have numbers like 5.1 or 9, which would seem to suggest their breakdown voltages. This is not always the case! Look up the numbers to determine a zener's characteristics.

FIG. 16 Zener diodes

Uses

Zeners provide a stable voltage reference in many circuits, especially power supplies and regulators. Even switching regulators may use them for reference.

Zeners can be used as linear regulators by themselves when only a small current supply is required. A resistor will be used to limit the current going to the zener, and the regulated voltage will appear at the cathode, while the anode goes to ground (in normal, negative-ground circuits).

What Destroys Them

Putting too much current through a zener will exceed its wattage rating and overheat it, destroying the part. Over time, even zeners in proper service may fail from the cumulative effects of heating.

Out-of-Circuit Testing

Test zeners for basic diode operation using the diode test or ohms scales on your DMM. Zeners can short, but most fail open, or at least they appear to do so. In fact, they may short and pass so much current that they melt inside, quickly opening.

If the zener tests bad as a normal diode, it’s bad. If it tests good, it may have lost its breakdown ability and still be bad, though. There's an easy way to tell, using your bench power supply. For this to work, the supply must be able to deliver a voltage at least a volt or so higher than the zener's expected breakdown voltage.

Take a 10 K-ohm resistor (brown-black-orange) and put it in series with the zener's cathode, using clip leads. Connect your bench power supply with its + terminal to the other end of the resistor and its - terminal to the anode of the zener. Set your DMM to read DC volts and hook it across the zener. Turn the bench supply as low as it will go, and then switch it on. Increase the supply's voltage while watching the DMM. As the indicated voltage rises, it should hit the zener's breakdown point and the DMM's reading should stop rising, even though you continue to crank up the power supply.

If the voltage keeps going up past the zener's breakdown voltage, the part is bad.

If it stops very near the rating, it's fine; standard zeners are not high-precision devices and may be off by a few fractions of a volt.

This test is also handy for characterizing unmarked zeners, as long as they are good. When you encounter a dead unmarked zener, determining what its breakdown voltage was supposed to be can be a real problem, unless you can find a schematic diagram of the product. Sometimes you can infer the breakdown voltage from other circuit clues. We'll get into that in Section 11.

There are many other kinds of less frequently used components. If you run across one that doesn't fit into any of the categories we've discussed, look up its part number to find out what the part does. An online search will usually turn up a data sheet describing the component in great detail.