Silicon Valley Computer SHUGART 706 User Manual Page 54

  • Download
  • Add to my manuals
  • Print
  • Page
    / 108
  • Table of contents
  • BOOKMARKS
  • Rated. / 5. Based on customer reviews
Page view 53
62
FIG. 2- MEASURE
INPUT
OFFSET
VOLTAGE
by
grounding
the non -inverting input.
With a bi-
polar
power
supply, the inverting
input is held
at
virtual ground
by the op -amp's output.
FIG.
3- OFFSET VOLTAGE AND
CURRENTS
may
be measured with this
circuit, depending
on the state of
S1
and
S2.
As before.
IB,
refers
to the inverting input
and IB, refers to the non -inverting input.
Input
offset
voltage
Another result
of mismatched differen-
tial-
amplifier inputs is offset
voltage,
Voti.
The error is reflected as
a
voltage
at
the output
when
both of
the inputs are
grounded.
A test circuit for measuring Vos
is
shown in Fig. 2.
Looking at the circuit,
however. we
see that only one input is
directly tied to ground; the other is at a
virtual ground.
That concept is important
in understanding how op -amps work.
One
of the basic properties of an ideal op -amp
is that its differential input
voltage
is zero.
Hence, when
the non -inverting input is
tied low,
the sum of the
voltages
at the
inverting input must
also be zero, or at
ground. That sum
consists of the input
signal and the feedback.
In an ideal
device.
with
no input signal
there
would
be no output.
But, in a real
device mismatched base -to- emitter (or
gate -to-
source,
as the case may be)
volt-
ages between the input transistors
will
cause a DC offset
voltage
at the output.
Input offset
voltage
is
defined
as the
volt-
age
necessary
at the
inputs
to bring the
output to zero.
Returning to
Fig.
2, in that configura-
tion, the op -amp attempts to balance its
inputs by forcing the inverting input
volt-
age high or low to compensate for
any
voltage
difference between the two in-
puts.
The result
(Vos) can be read directly
with
a
voltmeter.
A circuit that allows you to measure all
three characteristics (offset
voltage
and
the input bias
currents) is shown in
Fig. 3.
Vos
is measured
by closing both
switches
SI
and
S2. The
value
is read
directly from
the
meter.
The
input
bias currents
are measured by
alternately opening
each switch.
With
SI
open
and S2 closed, we
read an offset
voltage
that
we'll
call
VI.
Closing SI and
opening
S2 gives
us
V2.
IBI is
calculated
from
this formula:
IBt
=
V1
-
Vos
IB2
is calculated in
a
similar manner:
IB2 = V2 Vos
The calculated bias
currents are displayed
in
microamps. IB is
derived by averaging
1BI
and IB2;
Ios
is the difference between
them (in microamps).
Variations
in
offset parameters
The values
of
Ios
and
Vos
are depen-
dent on the
gains of the input transistors.
Generally,
the lower the
gain of a tran-
sistor,
the less the offset
parameter
will
be
affected. Lower input
gains, however,
tend
to increase input
bias currents, and
that
reduces input impedance.
In addi-
tion, open -loop gain
of an amplifier
also
has a pronounced
effect on
the overall
influence
of the offset parameters.
Temperature
effects
Offset
differences can
be affected
greatly by temperature
change. It's not
the
offset
itself that is the
culprit. Offset
volt-
age and current
can be compensated
for in
a given
circuit simply
by using trimmer
potentiometers.
The problem is
in the tracking
of the
offset
values.
Even "perfectly"
matched
transistors
have slightly
different tem-
perature
curves. The differences
arise be-
cause
of slight temperature
gradations
within
the semiconductor
substrate of the
device itself, and because
of impurities
that are introduced
during the manufac-
ture of
the device.
The result is that one
transistor's leak-
age current or offset voltage
can change at
a rate that is faster
than that of
the other
transistor, thereby
creating mis- matched
outputs. For
example, an offset voltage
that is handily
compensated for at room
temperature
may prove too much
to han-
400
350
300
250
200
150
100.
50
0
-75
-50 -25 0 +25 +50
+75 +100
+125
TEMPERATURE-
°C
FIG.
4 -INPUT BIAS CURRENT is
a
function
of
temperature.
dle at 125 °E. The
overall effect is
called
temperature drift.
All op -amps
are designed to minimize
the effects
of temperature
drift -some
just do it better than
others. The effects
of
temperature
drift can be measured
using
the circuit shown
in Fig. 3.
With
both SI
and S2 closed, record
the
value
of Vos
at
room
temperature. Now
raise the tern -
perature
and note how Vos
changes.
With
properly tracking
transistors, the
change
should
be slight, if it's
noticeable at all.
The table in
Fig.
4
shows how
input bias
current
of a typical
device
varies
with
changes in temperature.
Temperature
drift is normally
ex-
pressed
numerically in
terms of tem-
perature coefficient.
The temperature
co-
efficient of offset voltage
is calculated
from
this formula:
TEMPERATURE COEFFICIENT =
ovos
ST
and
the temperature
coefficient of
offset
current
is calculated:
TEMPERATURE
COEFFICIENT
=
abs o
ST
Temperature
coefficient usually is spec-
ified in
units of microvolts per degree
Centigrade (p.V
/ °C)
and in
units of
micro-
amps
, nanoamps, or picoamps per
degree
Centigrade (µA / °C, nA / °C, or pA / °C),
depending on the type of
device.
Remember, we're not
concerned
with
the actual change in input
bias -it
will
fluctuate. What
concerns us is that the
individual input currents change
together.
The smaller
the temperature coefficient,
the better the
two
inputs
track.
Aging
Another
factor that influences how an
op -amp
works
is
aging. Op -amp charac-
teristics have a tendency to change as op-
erating time increases.
Change
is
especially noticeable in the offset
values.
Devices that change much can be incon-
venient
to use, because the
equipment
in
which
they're used may have to calibrated
fairly
often.
Aging is
accelerated
with
continuous
exposure to elevated temperatures -such
as that frequently experienced inside an
enclosure.
Fortunately,
aging
is not a
con-
tinuous process. Most
changes
take
place
within
the first 100 hours of operation.
Once
stabilized,
you can expect an
ampli-
fier
to perform
very
reliably over the next
10,000 hours or so.
In fact. forced aging is one
way
to in-
crease
life
and enhance
the
performance
of a part, as
we
shall learn later on in this
series.
Common
-mode rejection ratio
Another
consequence of
imperfect
matching is common -mode
voltage.
Nor-
Page view 53
1 2 ... 49 50 51 52 53 54 55 56 57 58 59 ... 107 108

Comments to this Manuals

No comments