AD6484 ,Eagle ADSL USB modemThe di
rial its
processes or ClCCtrc
2
I
high ot 10% or me
time-frames, the ID.,
C ..
AD6489BB ,Integrated Broadband Communications Processorapplications. Page 1 of 5 Communications Processor Group Last November, ADI also acquired Chipl ..
AD648AH ,Dual Precision, Low Power BiFET Op AmpFEATURESDC Performance400 mA max Quiescent Current10 pA max Bias Current, Warmed Up (AD648C)300 mV ..
AD648AQ ,Dual Precision, Low Power BiFET Op AmpSpecifications subject to change without notice.1ABSOLUTE MAXIMUM RATINGSSupply Voltage . . . . . . ..
AD648BQ ,Dual Precision, Low Power BiFET Op AmpCHARACTERISTICSInput Offset Voltage 1.0 2.0 0.5 1.0 0.2 0.4 mVInput Offset Voltage T to T 3.0/3.0/3 ..
AD648CQ ,Dual Precision, Low Power BiFET Op Ampapplications.6. Crosstalk between amplifiers is less than –120 dB at 1 kHz.The AD648 is pinned out ..
ADATE207BBP , Quad Pin Timing Formatter
ADATE207BBPZ , Quad Pin Timing Formatter
ADATE209BBCZ , 4.0 Gbps Dual Driver
ADAU1328BSTZ , 2 ADC/8 DAC with PLL, 192 kHz, 24-Bit Codec
ADAU1592ACPZ , Class-D Audio Power Amplifier
ADAU1702JSTZ-RL , SigmaDSP 28-56-Bit Audio Processor with Two ADCs and Four DACs
AD6484
Eagle ADSL USB modem
ANALOG
DEVICES
AN-348
APPLICATION NOTE
ONE TECHNOLOGY WAY It PD. BOX 9106 o NORWOOD, MASSACHUSETTS 02062-9106 0 617/329-4700
Avoiding Passive-Component Pitfalls
The Wrong Passive Component Can Derail Even the Best Op Amp or Data Converter
Here Are Some Basic Traps to Watch for
by Doug Grant and Scott Wurcer
You’ve just spent $25 or more for a precision op amp or data con-
verter, only to find that, when plugged into your board, the device
doesn't meet spec. Perhaps the circuit suffers from drift, poor fre-
quency response, oscillations-or simply doesn't achieve the accu-
racy you expect. Well, before you blame the device itself, you
should examine your passive components-including capacitors,
resistors, potentiometers, and yes, even the printed circuit boards
themselves. Subtle effects of tolerance, temperature, parasitics,
aging, and user assembly procedures can unwittingly sink your cir-
cuit. And these effects all too often go unspecified or under-
specified by manufacturers.
In general, if you use data converters having 12 bits or more of
resolution, or op amps that cost more than $5, you should pay par-
ticularly close attention to passive-component selection. To put
the problem in perspective, consider the case of a 12-bit digital-to-
analog converter (DAC). One half LSB (least-significant bit) cor-
responds to 0.012% of full scale, or only 122 parts per million
(ppm) ! The host of passive-component phenomena can quickly ac-
cumulate errors far exceeding this level.
Buying the most-expensive passive components won't necessarily
solve your problems either. Often, the correct 25-cent capacitor
will yield a better-performing, more cost-effective design than the
premium-grade $8 part. Although not necessarily easy, under-
standing and analyzing passive-component effects may prove
quite rewarding, once you understand a few basics.
CAPACITORS
Most designers are generally familiar with the range of capacitors
available. But the mechanisms by which both static and dynamic
errors can occur in precision circuit designs are easy to forget be-
cause of the tremendous variety of capacitor types, e.g.: glass,
aluminum foil, solid tantalum and tantalum foil, silver mica,
ceramic, Teflon, and the film capacitors, including polyester,
polycarbonate, polystyrene, and polypropylene types.
Figure 1 is a workable model of a non-ideal capacitor. The nomi-
nal capacitance, C, is shunted by a resistance RP, representing insu-
lation resistance or leakage. A second resistance, Rs-equivalent
series resistance, or ESR-appears in series with the capacitor and
represents the resistance of the leads and capacitor plates.' Induc-
l- _ - Ts _______ 7
I c Rs L l
l (ESRI (esu
I ---1 f- I
1 Rd. c... I
l. _-----------.--- _I
Figure 1. Capacitor equivalent circuit.
'Capacitor phenomena aren't that easy to separate out. This matching of
phenomena and models is for convenience in explanation.
Reprinted from Analog Dialogue 17-2 1983
tance, L-the equivalent series inductance, or ESL, models the in-
ductance of the leads and plates. Finally, resistance Ra, and capaci-
tance Ca, together form a simplified model of a phenomenon
non known as dielectric absorption. Dielectric absorption can ruin
the dynamic performance of both fast and slow circuits.
Dielectric Absorption
We begin with dielectric absorption, also known as "soakage" and
sometimes as "dielectric hysteresis"-perhaps the least under-
stood and potentially most damaging capacitive effect. Upon dis-
charge, most capacitors are reluctant to give up all of their former
charge. Figure 2 illustrates the effect. After being charged to V
volts at time to, the capacitor is shorted by the switch at time tr.
At time t2, the capacitor is open-circuited; a residual voltage slowly
builds up across its terminals and reaches a nearly constant value.
This voltage is due to "dielectric absorption."
-Irtc,tpr-r.," V
VW'-- t, 6 C v
to t, tz
Figure 2. Residual voltage characterizes capacitor dielectric
absorption.
Standards techniques for specifying or measuring dielectric ab-
sorption are few and far between. Measured results are usually ex-
pressed as the percentage of the original charging voltage that
reappears across the capacitor. Typically, the capacitor is charged
for more than 1 minute, then shorted for an established time be-
tween 1 and 10 seconds. The capacitor is then allowed to recover
for approximately 1 minute, and the residual voltage is measured
(see reference IO).
In practice, dielectric absorption makes itself known in a variety
ofways. Perhaps an integrator refuses to reset to zero, a voltage-to-
frequency converter exhibits unexpected nonlinearity, or a sam-
ple-and-hold exhibits varying errors. This last manifestation can
be particularly damaging in a data-acquisition system, where adja-
cent channels may be at voltages which differ by nearly full scale.
Figure 3 illustrates the case in a simple sample-hold.
V20 - S1 S2
v3 iyy,1yag, W TO ADC
v,p', T
Figure 3. Dielectric absorption induces errors in sample-
and-hold application.
The dielectric absorption is a characteristic of thé dielectric mate-
rial itself, although it can be affected by inferior manufacturing
processes or electrode materials. As a percentage of the charging
voltage, dielectric absorption specifications range from a low of
0.02% for Teflon, polystyrene, and polypropylene capacitors to a
high of 10% or more for some aluminum electrolytics. For some
time-frames, the DA. of polystyrene can be as low as 0.002%.
Common ceramic and polycarbonate types display typical dielec-
tric absorptions of 0.2%; this corresponds to one-half of an LSB
at only 8 bits! Silver mica, glass, and tantalum capacitors typically
exhibit even larger dielectric absorptions, ranging from 1.0% to
5.0%, with those of polyester devices falling in the vicinity of
0.5%. As a rule, if your capacitor’s spec sheet does not discuss
dielectric absorption in your time frame and voltage range, exer-
cise caution.
Dielectric absorption can produce long tails in the transient re-
sponse of fast-settling circuits, such as those found in high-pass ac-
tive filters or ac amplifiers. In some devices used for such applica-
tions, Figure I's Ra-Cd, model of dielectric absorption can have
a time constant of milliseconds.' In fast-charge, fast-discharge ap-
plications, the dielectric absorption resembles "analog memory";
the capacitor tries to remember its previous voltage.
In some designs, you can compensate for the effects of dielectric
absorption if it is simple and easily characterized, and you are
willing to do custom-tweaking. In an integrator, for instance, the
output signal can be fed back through a suitable compensation net-
work, tailored to cancel the circuit equivalent of the dielectric ab-
sorption by placing a negative impedance effectively in parallel.
Such compensation has been shown to improve sample- and-hold
circuit performance by factors of 10 or more (Reference 7).
Parasitics and Dissipation Factor
In Figure 1, a capacitor's leakage resistance, RP, the effective series
resistance, Rs, and effective series inductance, L, act as parasitic
elements which can degrade an external circuit's performance. The
effects of these elements are often lumped together and defined as
a dissipation factor, or DF.
A capacitor’s leakage is the small current that flows through the
dielectric when a voltage is applied. Although modeled as a simple
insulation resistance (RP) in parallel with the capacitor, the leakage
actually is nonlinear with voltage. Manufacturers often specify
leakage as a megohm-microfarad product, which describes the
dielectric's self-discharge time constant, in seconds. It ranges from
a low of ls or less for high-leakage capacitors, such as aluminum
and tantalum devices, to the 100's of seconds for ceramic
capacitors. Glass devices exhibit self-discharge time-constants of
1,000 or more; but the best leakage performance is shown by Tef-
lon and the film devices (polystyrene, polypropylene), with time
constants exceeding 1,000,000 megohm-microfarads. For such a
device, leakage paths-created by surface contamination of the de-
vice's case or in the associated wiring or physical assembly-can
overshadow the dielectric's leakage.
Effective series inductance, ESL (Figure l) arises from the induc-
tance of the capacitor leads and plates, which, particularly at the
higher frequencies, can turn a capacitor's normally capacitive
mactance into an inductive reactance. Its magnitude depends on
I Much longer time constants are also quite usual. In fact, some devices can be
modeled by several paralleled Ra-Ce, circuits, with a wide range of time
:unstants.
construction details within the capacitor. Tubular wrapped-foil
devices display significantly more lead inductance than molded ra-
dial-lead configurations. Multilayer ceramic and fiIm-type devices
typically exhibit the lowest series impedances, while tantalum and
aluminum electrolytics typically exhibit the highest. Con-
sequently, electrolytic types usually prove insufficient for high-
speed local bypassing applications.
Manufacturers of capacitors often specify effective series induc-
tance by means of impedance-versus-frequency plots. These show
graphically, and not surprisingly, that the devices display predo-
minantly capacitive reactance at low frequencies, witlrrising im-
pedance at higher frequencies because of their series inductance.
Effective series resistance, ESR (resistor Rs of Figure 1), is made
up of the resistance of the leads and plates. As noted, many man-
ufacturers lump the effects of ESR, ESL, and leakage into a single
parameter called dissipation factor, or DF. Dissipation factor
measures the basic inefficiency of the capacitor. Manufacturers de-
fine it as the ratio of the energy lost to energy stored per cycle by
the capacitor. The ratio of equivalent series resistance to total
capacitive reactance---at a specified frequency-approximates the
dissipation factor, which turns out to be equivalent to the recip-
rocal ofthe figure of merit, Q.
Dissipation factor often varies as a function of both temperature
and frequency. Capacitors with mica and glass dielectrics gener-
ally have DF values from 0.03% to 1.0%. For ceramic devices, DF
ranges from a low of 0.1% to as high as 2.5% at room tempera-
ture. And electrolytics usually exceed even this level. The film
capacitors usually are the best, with DF's ofless than 0.1%.
Tolerance, Temperature, and Other Effects
In general, precision capacitors are expensive and-even then-
not necessarily easy to buy. In fact, choice of capacitance is limited
by the the range of available values and tolerances. Tolerances of
11% for some ceramics and most film-type devices are common,
but with possibly unacceptable delivery times. Most film
capacitors can be made available with tolerances of less than
t 1%, but on special order only.
Most capacitors are sensitive to temperature variations. Dissipa-
tion factor, dielectric absorption, and capacitance itself are all
functions of temperature. For some capacitors, these parameters
vary approximately linearly with temperature; in others they vary
quite nonlinearly. Although not usually important for sample-
and-hold applications, an excessively large temperature coefficient
(ppm/°C) can prove harmful to the performance of precision inte-
grators, voltage-to-frequency converters, and oscillators. NPO
ceramic capacitors, with temperature-drift as low as 30 ppm/°C,
usually do the best. On the other hand, aluminum electrolytics'
temperature coefficients can exceed 10,000 ppm/°C.
A capacitor’s maximum working temperature should also be con-
sidered. Polystyrene capacitors, for instance, melt near 85°C, com-
pared to Teflon's ability to survive temperatures up to 200°C.
Sensitivity of capacitance and dielectric absorption to applied volt-
age can also hurt capacitor performance in a circuit application.
Although capacitor manufacturers do not always clearly specify
voltage coefficients, the user should always consider the possible
effects of such factors. For instance, when maximum voltages are
applied, some high-density ceramic devices can experience a de-
crease in capacitance of 50% or more!
Similarly, the capacitance and dissipation factor of many types
vary significantly with frequency, mainly as a result of a variation
in dielectric constant. In this regard, the better dielectrics are poly-
styrene, polypropylene, and Teflon.
Assemble Critical Components Last
The designer's worries don't end with the design process. Com-
monly used printed-circuit-board assembly techniques can prove
ruinous to even the best of designs. For instance, some commonly
used p-c board cleaning solvents can infiltrate certain electrolytic
capacitors-those with rubber end caps are particularly suscepti-
ble. Even worse, some of the film capacitors, polystyrene in par-
ticular, actually melt when contacted by some solvents. Rough
handling of the leads can damage still other capacitors, creating
random-or even intermittent circuit problems. Etched-foil types
are particularly delicate in this regard. To avoid these difficulties,
it may be advisable to mount especially critical components as the
last step in the board assembly process-if possible.
Designers should also consider the natural failure mechanisms of
capacitors. Metallized film devices, for instance, often self-heal.
They initially fail due to conductive bridges that develop through
small perforations in the dielectric films. But the resulting fault cur-
rents can generate sufficient heat to destroy the bridge, thus return-
ing the capacitor to normal operation (at slightly lower capaci-
tance). Of course, applications in high-impedance circuits may not
develop sufficient current to clear the bridge.
Tantalum capacitors also exhibit a degree of self-healing, but-un-
like film capacitors-the phenomenon depends on the temperature
at the fault location rising slowly. Therefore, tantalum capacitors
self-heal best in high impedance circuits which limit the surge in
current through the capacitor':; defect. Use caution, therefore,
when specifying tantalums for high-current applications.
Electrolytic capacitor life often depends on the rate at which
capacitor fluids seep through end caps. Epoxy end seals perform
better than rubber seals, but an epoxy sealed capacitor can explode
under severe reverse-voltage or overvoltage conditions.
RESISTORS AND POTS
Designers have a broad range of resistor technologies to choose
from, including carbon composition, carbon film, bulk metal,
metal film, and both inductive and non-inductive wire-wound
types. As perhaps the most basic-and presumably most trouble-
free--of components, the resistor is often overlooked as a potential
source of errors in high-performance circuits. Yet, an improperly
selected resistor can subvert the accuracy of a 12-bit design by de-
veloping errors well in excess of 122 ppm, (V2 LSB). When did you
last take the time to actually read a resistor data sheet You’d be
surprised at what can be learned from an informed review of the
Consider the circuit of Figure 4, which amplifies a o-to-loo-mV
input signal 100 times for conversion by a 12-bit ADC with a O-to-
10-volt input range. The gain-setting resistors can be bought in in-
itial tolerances ofas low as t 0.001% (10 ppm) in the form of pre-
cision bulk metal-film devices. Alternatively, the initial tolerance
"-- woo OMKQ
0 - IMV
Figure 4. Temperature changes can reduce amplifier
accuracy.
of the resistors may be corrected through calibration or selection.
Consequently, the initial gain accuracy of the circuit can be set to
whatever tolerance is required, limited perhaps by the accuracy of
calibration instrumentation.
Temperature changes, however, can limit the accuracy of the am-
plifier of Figure 4 in several ways. The absolute temperature coeffi-
cients of the resistors are unimportant, as long as they track. Even
so, carbon composition resistors, with temperature coefficients of
approximately 1,500 ppm/°C, would not suit the application. Even
if the tempcos could be matched to an unlikely 1%, the resulting
15 ppm/°C differential would prove inadequate-a shift of as little
as 8°C would create a V2-LSB error of 120 ppm.
Manufacturers do offer metal film and bulk metal resistors with
absolute temperature coefficients ranging between t 1 and t 100
ppm/°C. Beware, though; temperature coefficients can vary a great
deal, particularly among resistors from different batches. To avoid
this problem, expensive matched resistor pairs are offered by a few
manufacturers, with temperature coefficients that track one
another to within 2 to 10 ppm/°C. Low-priced thin-film networks
are good and are widely used.
Unfortunately, even matched resistor pairs cannot fully solve the
problem of temperature-induced resistor errors. Figure 5a illus-
trates error-inducing through self-heating. The resistors have iden-
tical temperature coefficients but dissipate considerably different
amounts of power in this circuit. With an assumed thermal resis-
tance (data sheet) of 125°C/W for Vs-watt resistors, resistor RI's
temperature rises by 0.0125°C, while resistor R2's temperature
rises by 124°C. With a temperature coefficient of 50 ppm/°C, the
result is an error of 62 ppm (0.006%).
Even worse, the effects of self-heating create nonlinear errors. In
the example of Figure Sa, with half the voltage input, the resulting
error is only 15 ppm. Figure 5b graphs the resulting nonlinear
transfer function for the circuit of Figure fa. This is by no means
a worst-case example; smaller resistors would give even worse re-
sults due to their higher thermal resistance.
v, F-- vo " a TO TOV
0 TO loanv
R, . tam, 10.0tn%.t60pe/"C
a. R, " 9.9m, 20.0015‘, Ware
5FA . 125‘C/W
AT V. IOV. Vo " tN. Vx I tN
Po INR, name, .0
AT v. " wmvyo " mv, Vx . ioomv
Po IN R, . %I-O.IMW
Pg m n. - .2; " 93am
n. WILL HEAT up omzs'c
Rt WILL HEAT ur urc
tTa6'Cltsopvm/'m - um " om“ ERROR
Figure 5. Resistor self-heating leads to nonlinear amplifier
response. (a) Anatomy oftemperature-induced nonlineariW,
(b) Nonlineartransferfunction (scale exaggerated).
The use of higher-wattage resistors for those devices that dissipate
the greatest power can minimize the effects of resistor self-heating.
Alternatively, thin- or thick-film resistor networksminimize the ef-
fects of self heating by spreading the heat more evenly over all the
resistors in a given package.
Often overlooked as a source of error, the temperature coefficient
of resistance of typical wire or pc-board interconnects can add to
a circuit's errors. Metals used in p-c boards and for interconnecting
wires (e.g., copper) have a temperature coefficient as high as 3,900
ppm/°C. A precision 10-ohm, 10 ppm/°C wirewound resistor, with
0.1-ohms of interconnect resistance, for instance, effectively turns
into a 45 ppm/°C resistor. The temperature coefficients of inter-
connects play a particularly significant role in precision hybrids,
where thin-film interconnects have non-negligible resistance.
One final consideration applies mainly to designs that see widely
varying ambient temperatures: a phenomenon known as tempera-
ture retrace describes the change in resistance which occurs after
a specified number of cycles of exposure to low and high ambients
with constant internal dissipation. Temperature retrace can ex-
ceed 10 ppm, even for some ofthe better metal-film components.
In summary, to design resistance circuits for minimum tempera-
ture-related errors, consider the following (along with their cost):
. Closely match resistance-temperature coefficients.
. Use resistors with low absolute temperature coefficients.
. Use resistors with low thermal resistance (higher power rat-
ings, largercases).
. Tightly couple matched resistors thermally; (use standard re-
sistance networks or multiple resistors in a single package).
. For large ratios, consider using stepped attenuators.
Resis tor Parasitics
Resistors can exhibit significant levels of parasitic inductance or
capacitance, especially at high frequencies. Manufacturers often
specify these parasitic effects as a reactance error, in % or ppm,
based on the ratio of the difference between the impedance
magnitude and the dc resistance, to the resistance, at one or more
frequencies.
Wirewound resistors are especially susceptible to difficulties. Al-
though resistor manufacturers offer wirewound components in
either normal or noninductively wound form, even' noninductiVely
wound resistors create headaches for designers. These resistors
still appear slightly inductive (of the order of 20 WH) for R values
below 10,000 ohms. Noninductively wound resistors that exceed
10,000 ohms actually exhibit about 5 pF of shunt capacitance.
These parasitic effects can raise havoc in dynamic circuit applica-
tions. Of particular concern are applications using wirewound re-.
sistors with values both greater and less than 10,000 ohms. Here
it is not uncommon to see peaking, or even oscillation. These ef-
fects become evident at frequencies in the low-kHz range.
Even in low-frequency circuit applications, parasitic effects in wire
wound resistors can create difficulties. Exponential settling to 1
ppm takes 20 time constants or more. Parasitics associated with
wire wound resistors can increase settling time beyond the length
of those time constants significantly.
Unacceptable amounts of parasitic reactance are often found even
in resistors that aren't wirewound. For instance, some metal-film
types have significant interlead capacitance, which shows up at
high frequencies. Carbon resistors do the best at high frequencies.
Thermoelectric Effects
The junction between any two dissimilar metals creates a thermal
EMF. In many cases, it can easily produce the dominant error in a
precision circuit design. In wire wound resistors, for instance, the
resistance wire generates a thermal EMF of 42 microvoltsf'C when
joined to the leads (A typical lead material is Alloy 180, consisting
of 77% copper and 23% nickel). If the resistor's two terminations
see the same temperature, the EMFs cancel and no net error results.
However, if the resistor is mounted vertically a temperature gradi-
ent may exist between the bottom and top of the resistor because of
air flow past the long lead and its lower heat capacity.
For a temperature difference of as little as 1°C, an error voltage
of 42 microvolts results, a level which easily overwhelms the 25-
microvolt offsets of typical precision op amps! A horizontally
mounted resistor (Figure 6) can resolve the difficulty. Alterna-
tively, some resistor manufacturers offer, on special order, tinned
copper leads, which reduce the thermal EMF to 2.5 microvolts/°C.
RESISTOR
MOUNTED
VERTICALLY
TO CONSERVE
BOARD SPACE
(WRONG)
(RIGHT)
Figure 6. Thermal gradients create significant thermoelet>
tric circuit errors.
In general, designers should strive to avoid thermal gradients on.
or near critical circuit boards. Often this means thermally isolating
components that dissipate significant amounts of power. Thermal
turbulence created by large temperature gradients can also result
in dynamic noise-like low-frequency errors.
Voltage, Failure, and Aging
Resistors are also plagued by changes as a function of applied volt-
age. The deposited-oxide high-megohm type components are espe-
cially sensitive, with voltage coefficients ranging from 1 ppm/volt
to more than 200 ppm/volt. This is another reason to exercise cau-
tion in precision applications, such as high-voltage dividers.
Resistors' failure mechanisms can also create circuit difficulties if
not carefully considered. Carbon-composition resistors fail safely
by turning into open circuits. Consequently, in some applications,
these components play a useful secondary role as a fuse. Replacing
such a resistor with a carbon-film type can lead to trouble, since
carbon-film devices can fail as short circuits. (Metal-film compo-
nents usually fail as open circuits.)
All resistors tend to change slightly in value with age. Manufactur-
ers specify long-term stability in terms of change-ppm/year.
Values of 50 or 75 ppm/year are not uncommon among metal film
resistors. For critical applications, metal-film devices should be
burned-in for at least one week at rated power. During burn-in,
R values can shift by up to 100 or 200 ppm. Metal film resistors
may need 4,000 or 5,000 operational hours for full stabilization,
especially if they are deprived of a burn-in period.
Resistor Excess Noise
Most designers have some familiarity with thermal, or Johnson,
noise in resistors. But a less widely recognized second noise
phenomenon, called excess noise, can prove particularly trouble-
