ZENER DIODE

Introduction to Zener Diodes

It was noted earlier that the reverse-bias saturation current of a general-purpose junction
diode is so small that it ordinarily is masked by currents associated with high-resistance
conducting paths across the junction. But currents associated with other phenomena
occurring in what is a very complex physical junction environment also can mask the
leakage current. .

If the reverse-bias voltage magnitude is increased above a threshold (the specific value
depends on the junction geometry and material parameters) one or the other (possibly
even both concurrently) of two new phenomena occur. These phenomena, which are
different from the junction phenomena described before, establish a new mechanism of
current flow, generically referred to as ‘Zener breakdown’, which masks the junction
reverse-bias leakage current. In the breakdown region of operation large current changes
occur with very small changes in reverse-bias voltage, similar to forward-bias operation
but for quite different reasons.

These phenomena occur in all semiconductor junction diodes. However the reverse-bias
breakdown voltage characteristic can be reproduced with considerable precision by
controlling doping and other manufacturing process parameters. For ordinary use diode
breakdown is characterized simply by specification of a minimum reverse-bias
breakdown voltage and current; the magnitude of the breakdown voltage is guaranteed to
be no less and the current (for a specified lower voltage) no more than specified values.
Such diodes are not intended specifically for operation in reverse breakdown, and are
expected to maintain a low-conduction state when operated in reverse-bias within the
specified breakdown voltage limit.

Diodes whose reverse-breakdown characteristics are controlled precisely during

manufacture commonly are called Zener diodes. Zener diodes command a premium
because of the special production controls and selection, and are intended specifically for
operation in the reverse-breakdown mode primarily as inexpensive voltage references.

Zener Diode Breakdown Characteristics

Ordinarily the reverse-bias blocking action of a PN junction allows only a small 'leakage' current to
flow. However if a sufficiently large reverse-bias is applied other junction phenomena develop which
dominate the leakage current, allowing in effect much larger reverse-bias currents. This is the
'breakdown' part of the diode characteristic; 'breakdown' here refers to the overshadowing of the
semiconductor junction behavior by other phenomena rather than to a destructive effect. While all
diodes display this reverse-bias breakdown phenomenon Zener diodes are manufactured specifically for
operation in the breakdown condition with guaranteed specifications. The breakdown parameters of
these Zener (or voltage reference) diodes receive special processing attention during their manufacture.

Two distinct phenomena, acting individually or concurrently depending on diode details, are involved in
the breakdown phenomena. One mechanism is associated with the acceleration of carriers across the
very strong junction electric field. Kinetic energy gained by an accelerated carrier, if sufficiently great,
can cause additional impurity atom ionization during a collision with the atom. Each additional carrier
is then also accelerated and may cause additional ionization; the ionization grows exponentially. This is
termed the ‘avalanche effect’, recalling the initiation of a massive snow slide by a small initial snowball.

Circuits Zener Diode 1 M H Miller

The second mechanism is a quantum mechanical effect more difficult to describe by a familiar analogy.
Quantum mechanics predicts the possibility of a spontaneous crossing of a semiconductor junction by
carriers subject to a strong electric field. This is called the ‘tunnel effect’; because the phenomena is not
associated with ordinary mechanics it was suggested facetiously that some sort of metaphysical tunnel
existed through which carriers somehow traveled out of ordinary sight.

While the breakdown characteristics for the two phenomena are not exactly the same they are close
enough so that the distinction largely may be ignored in general for purposes of circuit design. Thus
although the Zener effect originally referred to the quantum mechanical phenomena the label Zener
diode is applied almost universally whatever the details of the breakdown mechanism.

An illustrative breakdown characteristic is drawn to

the left; the scale is exaggerated for clarity. The
nominal Zener reference voltage of the diode is the
reverse-bias voltage at which a manufacturer-
specified 'test' current IZT flows, and typically
represents a rated maximum diode current. In
general the Zener voltage is a modest function of
temperature; a representative temperature
specification is 0.1 % change per ˚C change. The
coefficient is negative for a diode with a reference
voltage below about 5 volts, otherwise it is positive. (This is related to the dominance of one or the
other of the two phenomena producing similar terminal breakdown characteristics.)

The inverse of the slope of the diode characteristic (typically at the test point) is called the 'dynamic
resistance' of the diode, and is a parameter noted in the manufacturers' specifications. The slope of the
characteristic does not vary greatly for currents in the range (roughly) between 0.1 IZT and IZT, a usual
range of operation of a Zener diode. (Note again that the scale in the figure is distorted for illustrative
purposes.) The minimum usable current is conditioned by the necessity of operation above the knee,
i.e., in the breakdown region, and the general desirability of avoiding the rapid change of slope in the
immediate vicinity of the knee.

The following figure displays PSpice computations of breakdown characteristics using a nonlinear

Circuits Zener Diode 2 M H Miller

model of the 1N5231 Zener diode (Manufacturer’s specifications at 27oC are: Vz = 5.1 @ 20mA,
rz = 175W @ 1mA, rz = 8.2W @ 5mA, rz = 2.2W @ 20mA.). The breakdown characteristics are plotted
for a broad temperature range. Note the small breakdown voltage sensitivity to temperature changes for
a given diode current. Note also the narrow range of diode voltages at a given temperature
corresponding to large diode current changes (above the 'knee' of the characteristic).

Since a forward-biased diode already provides large current changes for small changes in bias voltage
why a special interest in breakdown operation? The simple answer is that the breakdown voltage can be
manufactured to precise specifications over a very large range of voltages. For example, in a nominal
5 volt range the specified breakdown voltages for several diodes are: [email protected], [email protected],
[email protected], and [email protected]. Zener diodes operated in the breakdown region are widely used as
reliable and inexpensive voltage references.

Diode-Zener Comparison
An idealized-diode equivalent circuit for a Zener diode,
allowing both for forward- and reverse-bias, is drawn to
the right. (Verification of the characteristic shown is left
as an exercise.) For reverse-bias voltages less than the
Zener voltage the diode behaves (roughly) as a voltage
source in series with a small resistor; for larger voltages it
behaves (roughly) as an idealized diode.

A test circuit as shown to the left is analyzed numerically

using the PSpice computer analysis program. A detailed
idealized-diode analysis is left as an exercise. However it
should be clear that for the positive half-cycle of the
sinusoidal input voltage the diode is reverse-biased and Vo
will 'stick' at the breakdown voltage. For the negative half-
cycle the diode is forward-biased and Vo will be small.

The plot shows the output voltage 'clipped' by the forward-bias characteristic of the diode. Note that Vo
is slightly negative (diode threshold) for forward-bias operation. For reverse-bias operation (Vo > 0) the
voltage again is clipped, this time at the Zener voltage.

Circuits Zener Diode 3 M H Miller

Zener Diode ‘Shunt’ Regulator
A 'shunt voltage regulator' provides an informative and
relatively uncomplicated circuit application of a Zener
diode; the circuit diagram is drawn to the left. VS and RS
represent the Thevenin equivalent circuit as seen looking
back into the terminals of a power supply, and RB and the
Zener diode serve as a control devices to regulate the voltage
across the load RL. Note the standard icon used to represent
the Zener diode.

Absent the regulating elements (RB -> 0 and Zener diode

removed) an increase in load current, for example by
reducing RL, produces an 'internal' power supply voltage drop across RS, and consequently a lower
power supply terminal voltage. This variability of the power supply terminal voltage is described by the
'regulation' of the power supply, defined formally as the change in terminal voltage between 'no load'
and 'full load' conditions, divided by the 'no load' voltage. It is essentially a measure of the effect of the
internal resistance of the supply on the terminal voltage, as the load current changes from one to the
other extreme of its specified operating range.

To obtain an improved regulation the power supply will be made to provide a (essentially) constant
current, large enough to provide at least the maximum load current needed. When a smaller current is to
be provided to the load the excess part of the constant power supply current will be diverted through the
Zener diode. Both these actions are obtained by adding the Zener diode and ballast resistor RB to
modify the load as seen by the supply. Thus with the Zener diode operating in brealdown the supply
current will be approximately (VS – VZ)/(RS + RB). This current then divides, with an amount VL/RL
flowing through RL and the remainder flowing through the Zener diode. (Note: Because the Zener
resistance rZ is not exactly zero the Zener voltage increases slightly as the diode current increases and
this causes the supply current to decrease slightly.) The essential idea, as noted before, is to shunt
'excess' current through the Zener diode when RL is a maximum, and then decrease the amount of this
shunted current as load current demand increases. Since the power supply itself tends to 'see' a fixed
current, its terminal voltage changes little. The Zener diode provides an approximation to a constant
voltage source over a large current range, and RB provides a corrective voltage drop between the supply
voltage (generally larger than the Zener voltage) and the Zener regulated load voltage.

Because the diode is a nonlinear circuit element an analytical

examination would be an involved one. Hence, as a simplifying
measure more than adequate to provide an appreciation of the
regulating action and a good estimate of performance, the Zener
breakdown characteristics are approximated as shown to the right
of the circuit diagram above. The piecewise linear representation
of the Zener characteristic, which is applicable over the range of
normal operation of the Zener diode (i.e., in reverse-bias
breakdown), is used for approximate calculations. Do not confuse the idealized diode used in the model
with the Zener diode whose characteristic is being modeled; the Zener diode characteristic is being
approximated over a limited range of operation by a combination of idealized circuit element models.
(Verify the model does have the indicated characteristic.)

The regulating action is observable in the curves obtained by an analysis of the circuit using the
piecewise-linear Zener diode approximation; these are drawn below.

Circuits Zener Diode 4 M H Miller

The unity slope reference (dashed) line is the voltage transfer characteristic of the power supply alone,
i.e., when the regulating elements are removed VL=VS. This provides a reference against which to
display the effect of regulation. The other (solid line) curve describes the circuit with the regulating
elements inserted. When the power supply voltage is low (so that VL < VZ) the Zener diode is in
reverse-bias, but not yet in breakdown; the diode state is more or less open-circuit and has negligible
effect on circuit operation. RB and RL form a resistive voltage divider, making VL somewhat smaller
than VS, and so the regulation curve is a line segment of slope somewhat less than 1.

As VS is increased the breakdown voltage VZ of idealized diode model is reached; because of the
voltage-divider action VS will be somewhat larger than VZ when this occurs (see figure). This causes
the ‘constant’ Zener voltage (with some variation because of the finite Zener resistance) to appear across
the load. Note that quality of the regulation is measured by the extent to which RL||rZ << RS+RB.

An important implicit requirement not always recognized explicitly is that the regulating action depends
on the assumed operation of the Zener diode in its breakdown region. This is not something that
happens automatically; it must be designed to be so by proper choice of element values. It means, for
example, enough current must be drawn by the Zener diode to maintain proper operation even in the
'worst case' situation when the maximum load current has been siphoned off from the supply current,
i.e., for the minimum diode current. Hence at full-load current one should design the circuit to provide
at least a minimum Zener 'keep-alive' current of roughly 0.1 IZT. On the other hand when the load
draws the minimum current the increased current through the Zener (the source current will not change
much) should not exceed the rated IZT. Between these operating requirements, and of course knowing
the (nominal) Zener diode voltage, an appropriate value of RB can be determined.

This calculation is particularly noteworthy here because it is rather different from the more familiar case
of solving for specific element values common in introductory courses. Two extreme ('worst-case')
conditions are involved in the form of inequalities, not equalities. The result of the calculation is not the
value of the resistance to use but rather inequalities that specify a range of acceptable resistances, greater
than some value but less than another value. The details of the circuit behavior will depend to some
extent on the choice made.

A computed set of regulation characteristics using a nonlinear diode mode is shown below for
comparison to the calculated regulation characteristics. Regulation computations are shown for several

Circuits Zener Diode 5 M H Miller

choices of load resistance (and so load current). Detailed comparison between the computed and
analytical curves is left as an exercise, e.g., calculate the theoretical value of the source voltage at which
regulation begins and compare to the computed value. Thus using VZ= 5.1v, RL = 250W, RS = 220W,
and RB = 47W, the voltage across the Zener diode reaches the nominal breakdown voltage of 5.1v when
the source voltage is approximately 10.6v.

As a matter of some interest note that there is no regulating action for 0 £ VS £ 20 when RL = 25W,
because not enough current is available from the supply over and above the required load current to
enable Zener diode breakdown to occur.

The regulated regulation curve also was computed, and is drawn below. Note that absent the regulation
provided by the Zener diode the load voltage VL @ IL =30 milliampere would be 7 volts. The data are
shown using two scales, one to provide detail on the computed load voltage values and the other to
provide some perspective on the overall effect of the regulation. The ‘cost’ of the regulation in this
illustration is associated with the 10 volt (approx.) drop across RS+ RL.

Circuits Zener Diode 6 M H Miller