OSCILLOSCOPE GUIDE

17
BASIC ANALOG OSCILLOSCOPES
The circuitry of a simple analog oscilloscope can
be broken into three blocks as shown in Fig. 11.
These three blocks are the vertical circuitry, the
horizontal/trigger circuitry, and the display cir-
cuitry. By understanding each of these blocks on
its own, it will become much easier to under-
stand how they interact and why the oscillo-
scopes controls function the way that they do.
Vertical Circuitry
Refer to Fig. 12.
The vertical circuitry controls the vertical axis of
the display and consists of an input coupling
circuit, the input attenuator, and the vertical
amplifier (for dual trace oscilloscopes, there are
two identical vertical circuits, one for each
channel). All signals connected to the vertical
input jacks are fed through at least part of the
vertical circuitry before they are used by other
oscilloscope circuits.
The Input Coupling Circuit
The input coupling circuit allows the user to
ground the oscilloscopes input, pass only the ac
portion of the signal, or pass both the ac and dc
portions of the signal. When the input coupling
circuit is set to ground the scope input, the
attenuator input is grounded, but the input jack
is open (to prevent a short circuit at the probe).
This mode is useful for setting the trace to a
zero reference level. When the input coupling
switch is set to pass only the ac, the scope input
jack is capacitively coupled and any dc signal
component is blocked. When the input coupling
switch is set to pass both the ac and dc signal
components, the oscilloscope input jack is
directly coupled to the attenuator and both the
dc and ac (dynamic) portions of the signal are
passed.
The Input Attenuator Circuit
The input attenuator allows a wide range of
signal levels to be applied to the oscilloscope by
setting the vertical sensitivity of each channel.
The basic sensitivity of an oscilloscopes vertical
amplifier is typically 5 mV per division. Lower
sensitivities, up to 5 volts per division, are
achieved by attenuation of the input signal.
Typically, input attenuator circuits are set up in a
1-2-5 sequence and allow measurements of
signal levels from a few millivolts to tens of
volts. In other words, input sensitivity might be
switchable to 5 volts/div, 2 volts/div, 1 volt/div,
0.5 volts/div, etc.
In order to provide a calibrated voltage meas-
urement on the display, vertical input attenua-
tors must be highly accurate wideband net-
works, capable of passing all signals within the
oscilloscopes measurement range (e.g. for a 20
MHz scope, these attenuators must provide a
flat response for signals from dc all the way to
20 MHz).
In addition to the step attenuator controls, oscil-
loscopes also have a variable sensitivity control
that allows the scope sensitivity to be set
between step attenuator ranges. This control
allows waveforms to be set to occupy an exact
number of divisions, such as is necessary for
rise time measurements, where the waveform
must extend exactly from the 0% to 100% mark-
ers.
For input sensitivities greater than 5 mV per
division (for example, 1 mV per division), an
additional amplifier is switched into the circuit
instead of an attenuator. The bandwidth is typi-
cally reduced when the amplifier is selected.
The Vertical Amplifier Circuit
From the input attenuator, the signal is fed to
the vertical amplifier, where it is amplified to a
level suitable for driving the CRT vertical
deflection plates. Depending on the display
mode (single- or dual-trace, chopped, or
alternate display), the vertical amplifier stage
also handles the switching of signals to facilitate
HOW OSCILLOSCOPES WORK
Fig. 11. Simplified oscilloscope block diagram.
Fig. 12. Vertical circuitry block diagram.
OSCILLOSCOPE GUIDE
18
the display of channel 1, channel 2, dual-trace
chopped, dual-trace alternate, etc.
Also, at a preamplifier stage, a portion of the
signal is picked off and fed to the
horizontal/trigger circuitry. Position controls
adjust the dc bias of the amplifier circuit and
allow the trace to be moved about vertically.
Horizontal/Trigger Circuitry
Refer to Fig. 13.
As suggested by its name, the horizontal/trigger
circuitry controls the horizontal axis of the trace.
The horizontal/trigger circuitry can be further
broken down into three distinct sections: the trig-
ger circuit, the sweep generator, and the
horizontal amplifier.
The Trigger Circuitry
The oscilloscopes trigger circuitry plays the very
important role of telling the scopes other
circuitry when to start drawing the trace. Since
the scope display provides a graph of voltage
versus time, it is very important that the scope
starts drawing at the same point on the
waveform each time it sweeps across the
display. If the oscilloscope could not precisely
control the trigger point, it would be impossible
to measure anything related to time. The sweep
would start at a different point each time, and
therefore, the waveform would be moving to a
different position on the display each time the
CRT was swept.
The trigger level and slope controls allow the
scope user to select the exact point at which the
sweep will be triggered. The slope control allows
selection of either the positive- or negative-
going slope, and the level control allows the
selection of the exact point on that slope. Fig.
14 illustrates the way that the slope and level
controls function.
The trigger circuits also perform the switching
that selects the trigger source. Since it is
frequently desired to trigger from an event other
than the signal that is being viewed,
oscilloscopes allow a selectable trigger source.
This source can be internal (one of the channels
being displayed) or external (the signal applied
to the external trigger input jack), or line
frequency.
When an internal trigger source is selected, a
portion of the signal from one or more of the
vertical preamplifiers is fed to the trigger circuits.
This signal can be the channel that is being
viewed on the CRT, a channel that is not being
viewed, or, in the case of multiple trace display,
the trigger source can automatically switch
between each of the channels as it is displayed.
This is known as alternate triggering and is used
in conjunction with alternate display.
When an external trigger source is selected, the
source is the signal that is applied to the
external trigger input jack. Typically, external
triggering is used for such things as viewing
logic signals with reference to a known timing.
Line triggering uses the line frequency for
triggering and is used for work on power
supplies, or circuits that must be synchronized
with line voltage.
Trigger coupling is also selected within the trig-
ger circuitry. Typical trigger coupling modes are
ac (trigger signal is capacitively coupled and all
ac signal components are used), high frequency
reject (trigger is capacitively coupled and a low-
pass filter rejects all high-frequency signals),
and low frequency reject (trigger signal is
capacitively coupled and a high-pass filter
rejects all low-frequency signals). The cutoff
point for high and low pass filters varies
between oscilloscope models and brands. Most
scopes also have TV trigger coupling (using a
sync separator) to trigger from horizontal or
vertical TV sync pulses. This mode is useful for
viewing frames (vertical sync pulses) or lines
(horizontal sync pulses) of video signals.
The Sweep Generator (Time Base)
After the selected trigger event occurs, a linear
sawtooth sweep generating circuit is turned on
and produces the waveform shown in Fig. 15.
The first part of the waveform is the linear
sweep ramp that causes horizontal movement
of the trace. As voltage is increased, the
electron beam moves further to the right of the
CRT. When the voltage reaches its peak level
HOW THE OSCILLOSCOPES WORK
Fig. 13. Horizontal/trigger circuitry block diagram.
Fig. 14. Function of slope and level controls.
OSCILLOSCOPE GUIDE
19
(the top of the ramp), the electron beam is at
the right edge of the CRT. At this point, the
electron beam is turned off (this is called retrace
blanking) and the sweep voltage returns to the
original level, thereby also returning the electron
beam to the left side of the CRT.
The sweep ramp is typically generated by a
circuit known as a Miller integrator. This circuit
takes a dc voltage as its input and performs the
mathematical process of integration on it.
Integration of a dc level produces a linear ramp.
Various combinations of resistance and
capacitance are used to control the speed of the
ramp, and hence, the sweep speed. When the
ramp reaches a certain level, the dc voltage is
removed from the input of the integrator,
causing the ramp to reset. The dc input is
sometimes provided by a flip-flop, and removal
of the dc level simply involves applying a reset
pulse to that flip-flop.
In Fig. 15, note the "holdoff" period that occurs
immediately after the completion of each sweep.
This is a period during which the next sweep is
inhibited. Length of the holdoff period is
controlled by the length of the reset pulse to the
flip-flop mentioned above. The holdoff period
varies with the sweep rate, but is adequate to
assure complete retrace and stabilization before
the next trigger occurs.
Since it is desirable to measure time on the
horizontal axis, it is important that the sweep
time (the time that it takes for the electron beam
to move from the left edge to the right edge of
the CRT) is linear (constant speed all the way
across) and calibrated. The step time base
control provides calibrated sweep times from
seconds to micro- or nanoseconds. As with the
step input attenuator, the time base control is
usually arranged in a 1-2-5 pattern (0.1 S/div,
0.2 S/div, 0.5 S/div, 1.0 S/div, etc.). There is
usually a variable time base control that allows
for adjustment between step time base ranges
(although use of this control causes the time
base to become uncalibrated).
Most oscilloscopes also have a sweep
magnification control that allows the entire trace
to be magnified. For example, if the main step
time base control is set to 0.5 mS/div and X10
magnification is selected, the actual time base
becomes 0.05 mS/div (50 nS/div).
The Horizontal Amplifier
The horizontal amplifier simply boosts the signal
to a level capable of driving the CRTs horizontal
deflection plates. Since the sweep speeds can
range from seconds to micro- or nanoseconds,
this amplifier must have a relatively wideband
width and must faithfully reproduce the linear
sawtooth sweep waveform. Position controls
adjust the amplifiers dc bias and allow the trace
to be moved about horizontally.
The Display
The CRT
The CRT, or cathode-ray tube, is made of glass
and contains a vacuum. Electrons are emitted
from a heated element at the narrow end of the
tube and are accelerated by a high voltage,
typically 2000 volts or greater, towards the
display end. On the way there, they are focused
into a narrow beam whose direction is altered
slightly by vertical and horizontal deflection
plates. These deflections are controlled by the
outputs of the vertical and horizontal amplifiers
mentioned previously. Therefore, the beam
moves up and down in relation to the input
signal, and horizontally in relation to the time
base circuitry. The display end of the tube is
coated by a monochrome phosphor, usually
green, which glows where it is struck by the
electrons.
The Graticule
The oscilloscope display represents a graph of
voltage versus time. As seen in Fig. 16 the
horizontal component of the graph represents
time and the vertical component represents
voltage. Typically, oscilloscope displays have a
graph, or graticule, that is divided into 10
horizontal and eight vertical divisions. Each of
these divisions is usually broken into 5 minor
divisions (subdivisions), represented by the
"hash" marks along the center vertical and
horizontal graduations. Additionally. the graticule
HOW THE OSCILLOSCOPES WORK
Fig. 15. Sweep waveform.
Fig. 16. Typical oscilloscope graticule
OSCILLOSCOPE GUIDE
20
is marked with two dashed lines, labeled "10"
and "90" and points marked "0" and "100" are
also present. These four points are useful for
rise time and fall time measurements, since rise
time is defined as the time required for a signal
to rise from 10% to 90% of its maximum
amplitude. On almost all modern oscilloscopes,
the graticule is etched directly onto the inside of
the CRT. This eliminates parallax error that
might occur from viewing the CRT from a slight
angle.
Delayed Sweep
Delayed sweep is a very important feature of
many advanced scopes. It allows a portion of
the trace to be magnified. The original trace is
sometimes referred to as the A sweep, and the
magnified trace as the B sweep.
Fig. 17 shows a block diagram of the delayed
sweep circuit and a representation of the
process. The trigger starts the A sweep
generator and several cycles of the square
wave input are displayed. The trigger is also
applied through a variable delay network, the
Delay Position control, which is adjustable by
the operator. In the Mix sweep mode, the
delayed trigger starts the B sweep generator at
a later point in time. The delay is typically
adjustable from one division to ten divisions on
the screen. When the B sweep starts, the
display is switched to view the faster B sweep
for the balance of the trace. Since the B sweep
is faster, that portion of the waveform is
expanded.
When the Sweep Mode Switch is set to the
Delay position, only the faster B sweep is
viewed. The starting point of the waveform is
still adjustable with the Delay Time control.
Variable Holdoff
As mentioned previously, all scopes have some
degree of holdoff between CRT sweeps. This is
necessary to provide time for the beam to return
to the left edge of the screen. Many advanced
units also provide a variable holdoff, which
allows this interval to be lengthened by the user.
This is useful for synchronizing on complex
pulse trains.
As discussed in "Basic Analog Oscilloscopes"
above, holdoff depends on the width of a reset
pulse applied to the flip-flop, which drives the
Miller integrator. Variable holdoff is achieved by
varying the width of that reset pulse.
Other Features
Delay Line
Many high bandwidth oscilloscopes have a
delay line in the vertical amplifier. This delay line
actually slows the signal down (by a fraction of a
microsecond) so that the oscilloscope can start
the sweep before the vertical signal starts to
deflect the CRTs electron beam. This allows the
scope to display the signal, including the trigger
edge. It is generally needed only in cases where
the rise time (or fall time, if trigger occurs on a
falling edge) of a very fast signal needs to be
observed. That is why only high bandwidth
scopes usually have delay lines; if the
bandwidth of the scope is not high enough to
view a very fast rise time, a delay line offers no
advantage.
Higher CRT Voltages
A typical 20 MHz or 30 MHz oscilloscope has a
CRT acceleration voltage of 2000 volts. Higher
voltages deliver brighter traces and are
essential for wideband scopes. As the time that
it takes for the electron beam to sweep across
the CRT is decreased (higher bandwidth scopes
have higher maximum sweep rates), the voltage
must be increased in order for the trace to still
be easily visible. Some 100 MHz units employ
voltages as high as 15 kV or 20 kV.
HOW THE OSCILLOSCOPES WORK
ADVANCED ANALOG OSCILLOSCOPES
OSCILLOSCOPE GUIDE
21
HOW THE OSCILLOSCOPES WORK
Fig. 17. Delayed sweep.
OSCILLOSCOPE GUIDE
22
DSO OPERATION
Real-Time Sampling
The block diagram of Fig. 18 is meant to show
the inner workings of a DSO in simplified form
only. Also, it depicts a single channel; many of
the blocks are duplicated in a two-channel DSO.
A signal from the vertical preamplifier is fed to a
relay or switch, which routes the signal
according to whether the scope is in analog or
storage mode. In analog mode, it merely sends
the signal on to the main vertical amplifier. In
storage mode, however, it sends it through the
blocks of Fig. 18.
The signal progresses through a signal scaler,
which attenuates the signal so that it doesnt
exceed the limits of the A/D (analog-to-digital)
converter. Optimum scaling equates the A/D
limits with the upper and lower edges of the
CRT display. Thus, the scaler works in concert
with the Volts/Div setting.
The sample-and-hold switch operates under the
control of the microprocessor. It may be
sampling as quickly as possible, in real-time
sampling mode, or it may be sampling at very
specific intervals, in equivalent-time sampling
(see the next section). The capacitor on its
output is critical; it holds the sampled voltage
between samples. The output of the
sample-and-hold is fed through a unity buffer to
the A/D (analog-to-digital) converter.
The A/D converter is generally a flash type
which operates at at least 10 MHz. Its
conversions are controlled by the
microprocessor in sync with the sample-and-
hold switch. It converts the analog voltage
applied into a digital quantity eight or twelve bits
wide (or more). In the diagram, the wide gray
arrows indicate digital buses.
The digital words are stored in a memory which
is controlled by the microprocessor, which
increments the address counter as each sample
is acquired. The memory is generally twice as
long as the number of samples required to fill
the display. This is because the scope must be
able to show not only the data after the trigger
event, but as much as a full screen of
"pretrigger" data, if specified by the operator.
During data acquisition, the microprocessor
places the memory in "write" mode via the R/W
(read/write) control line.
In order to be useful, the data must be able to
be read out from the memory and displayed.
The microprocessor reads from the memory
with the R/W line set for "read", and applies the
data to a D/A (digital-to-analog) converter, which
changes it back to an analog voltage. This is
sent to the main vertical amplifier for display.
The microprocessor also sends separate binary
quantities to another D/A converter which
connects to the main horizontal (sweep)
amplifier. In this case, the microprocessor
controls the X-position of the CRT electron
beam as well as its vertical deflection.
As stated previously, this description is one of
basics only. All the blocks involved are connect-
ed through extensive electronic logic. There
may be more than one microprocessor involved,
one for data acquisition and another for display.
Equivalent Time Sampling
Equivalent time sampling enables a DSO to
digitize waveforms of a higher frequency than its
sampling rate, provided that the waveform is
repetitive. Instead of taking many samples per
cycle, the scope acquires one sample per cycle,
or perhaps even one sample per group of
cycles, of the input signal. The sampling point is
moved in time along the waveform, and after a
sufficient number of cycles have occurred,
enough samples have been stored to construct
a picture of the waveform.
The process is depicted simply in Fig. 19. The
HOW THE OSCILLOSCOPES WORK
DIGITAL STORAGE OSCILLOSCOPES (DSO)
Fig. 18. DSO block diagram
OSCILLOSCOPE GUIDE
23
sweep ramp from the horizontal circuits is used.
The process is depicted simply in Fig. 19. The
sweep ramp from the horizontal circuits is used.
This ramp resets with each trigger event; in this
case the trigger is set for the positive
zero-crossing. Instead of sampling in a free-run
fashion as before, the microprocessor and
sample-and-hold circuit now acquire a sample
only when the sweep ramp crosses a
comparison voltage which is stepping upwards
as shown.
This comparison voltage is derived by using a
binary counter to count the trigger events, and
increment by one for each trigger. The count
state is fed to a D/A (digital-to-analog) converter
which transforms it into an analog voltage. This
is the slowly stepping comparison voltage
shown in Fig. 19. The end result is that the
intersection point of this voltage and the sweep
ramp moves further to the right with each cycle
of the input, and so does the sampling point on
the input waveform.
This process is actually a controlled use of
aliasing, which is a generally undesired result of
sampling a waveform too slowly. Aliasing often
occurs accidentally in real-time sampling mode,
usually due to operator error. The user may not
even be aware that the display is totally false,
because it appears as a valid waveform. In the
case of equivalent-time operation, however, the
process is monitored and controlled by the
sophisticated logic and software of the DSO and
becomes a useful tool.
HOW THE OSCILLOSCOPES WORK
Fig. 19. Equivalent time sampling.