THE

THE HANDYMAN'S
HANDYMAN'S GUIDE
GUIDE TO
TO OSCILLOSCOPES
OSCILLOSCOPES (Part
(Part 11 of
of 2)
2)
by Paul Harden, NA5N

Getting Acquainted with your Scope and making some measurements

Print as .pdf file

5 pages
8 x 11 or A4

Updated May 2004 with actual oscope waveforms

When in the course of human events, it becomes necessary to look at neat signals floating around your radio, you go to a hamfest
and buy that $50 o-scope. Now what? This two part article will attempt to explain basically how an oscilloscope works, operator
functions, basic measurements, and some advanced applications. An o-scope is a powerful tool in any shack even a real
"cheapie" with limited bandwidth.

HOW AN OSCILLOSCOPE WORKS

A block diagram of a typical o-scope is shown in Fig. 1. The test probe usually plugs into the scope via a BNC connector, then
passes through a switch to determine whether the input signal will be dc or ac coupled (to remove any dc component). Often this
switch will have a "ground" position for setting the zero-volts reference. Next is the input attenuators. The vertical input amplifier
is quite sensitive, designed for 20-50mV of input. For larger input voltages, the signal is applied to attenuators comprised of simple voltage dividers. This is the first area of concern for cheap o-scopes, as the input attenuators may not be very linear or accurate.
For example, if you appy a 10Vpp signal on the 10v/division setting, the signal should be 1 division high. Switching to 1v/div., the
signal should be 10 divisions (usually full-scale) high. If it is not exactly 10 divisions, the attenuator for that setting needs adjusting. Some scopes have internal adjustments for fine-tuning each attenuator setting.
Following the attenuators, the signal is applied to
the first vertical amplifier, which converts the
input to a differential signal. This differential signal is amplified up to high voltages for the oscilloscope deflecting plates moving the beam up and
down (in the vertical axis).
The sweep generator is usually a constant current
source charging a capacitor to make a sawtooth
waveform that eventually deflects the beam in the
horizontal axis. The frequency of the sawtooth
determines how fast the beam travels from the left
to the right side of the tube, and is controlled by the
sweep control, usually calibrated in seconds,
milli-seconds or micro-seconds per division. This
is the second area of concern for an oscilloscope
how linear the sawtooth waveform is generated.
For example, a sawtooth with a nonlinear ramp
will cause the signal displayed in the central portion of the tube to be expanded or compressed
compared to the signal at the ends.

Input
Coupling
CH-1
Vert.
Input

Volts/division

AC

VERTICAL
AMPLIFIER

INPUT
ATTEN

DC

Display tube

GND

SWEEP
GENERATOR
Trigger

Time/division
(Sec/div)
(mS/div)
(uS/div)

Trigger
Source

Level
Comparator

+
Trigger
Level

HORIZONTAL
AMPLIFIER

+V

INTernal
Line 60Hz
EXTernal

External
Trigger
Input

-V

FIG. 1 - Basic block diagram of an Oscilloscpe

The sawtooth ramp is amplified to high voltages, applied to the oscope tube, to deflect the beam from left to right. An important
task of an oscope is when the horizontal deflection begins. Normally a switch labeled "Trigger Source" determines what initiates
the sawtooth ramp. In the "Internal" position, a sample of the input signal (in the vertical amplifiers) is sampled, with a variable
resistor setting the level. When the input signal exceeds the "Trigger Level," a pulse is generated to start the sawtooth ramp and
hence the horizontal sweep. The purpose of triggering is to keep the input waveform synchronized to the sweep so it appears
stationary on each sweep. The trigger source usually has a "Line" position, which simply triggers the sweep off of 60Hz from the
power supply. This synchronizes the sweep to the AC power frequency and is useful for checking television signals, which are
synchronized to the power mains. Also, an "External" position may be present, which connects an external input signal (via a BNC
connector) to trigger the sweep generator.
Other features your oscope may have are two vertical channels for dual trace operation, various modes to display both waveforms
(alternate, chopped, A+B added, etc.), delayed sweep features, dual sweep time bases, built in calibrators, etc.

CALIBRATING YOUR OSCILLOSCOPE

The first thing you should do upon acquiring an o-scope is to check its calibration.

The vertical amplfiers can be checked with a known voltage source or 9v transistor radio battery. Measure the output voltage
of the battery with an accurate voltmeter. Let's say it just happens to be +9v exactly. Set the input coupling to ground (0v) and move
the trace to the bottom division. Switch the input coupling to DC and set the attenuators to 1v/div. The deflection should be 9

divisions. Switching to 10v/div., deflection should be 0.9 divisions. Internal to the oscope (or perhaps accessible from the outside)
are adjustments for the vertical amplifier gain. Adjust this for 9 divisions of deflection in the 1v/div. range. Procedure can be
repeated with a 1.5v flashlight battery (assuming you know the exact voltage from a DVM).

The horizontal amplifiers should be checked/calibrated using a signal generator. For example, a 1MHz signal has a period of
1uS. Setting the sweep rate to 1.0uS/div., a 1MHz signal should take exactly 1 division per cycle. Set the horizontal width control
properly to ensure the beam starts at the first division and ends at the last division. If the sweep rate appears incorrect, an internal
adjustment (Sweep gain or similar) can be set for proper display of the test signal.

The main operator controls are:

Intensity - controls the brightness of the beam. NOTE: Too bright a beam can damage to the CRT tube!
Focus - adjusts the beam for the thinnest and sharpest display.
VERT & HOR Position - controls the vertical and horizontal position of the display respectively
VERT V/div - controls the vertical sensitivity of the display, i.e., how many volts (or mV) per division.
HOR Sweep Speed - sets the horizontal sensitivity, i.e., how many mS or uS per division.
VERT & HOR vernier - allows the vertical and horizontal sensitivity settings to be varied in small steps.
Other adjustments you may find on your scope are:
Astigmatism - With the scope intensity and focus properly set, this adjustment compensates for the curvature of the CRT tube by
making it in-focus across the sweep. If your trace is out-of-focus in certain areas, but in-focus elsewhere, the astigmatism needs to
be adjusted. See Fig. 2.
Trace Rotation - is a small coil around the CRT that skews the trace to ensure it is perfectly horizontal. On scopes without this adjustment, the trace is leveled by physically rotating the CRT to align the trace to the graticle grid. See Fig. 2.
DC BAL (DC Balance) - is a dc offset in the vertical amplifiers that
causes a shift in the trace baseline when changing vertical scales. It
is most obvious when measuring ac voltages. For example, you are
displaying a 10Vpp sine wave, centered on the center graticle, at
2v/div. Changing to 5v/div, the sine wave shifts off the center
graticle ... that is, it assumes a dc bias error. The DC BAL is adjusted
until the shift no longer occurs when changing vertical scales.

Fig. 2 Effects of
Astigmatism & Trace Rotation

HV ADJ. - is the high voltage that controls the intensity of the

trace. Turn up the Intensity control to its brightest position, then
adjust the HV ADJ for a trace slightly brighter than normal
intensity. The Intensity control now has the proper range. The HV
ADJ might have to be re-adjusted to acquire proper focus.

NOTE: Very bright trace displays can cause permanent

damage to the CRT, particularly on a well-used scope.
LETS MAKE SOME MEASUREMENTS
It is assumed you have your scope relatively calibrated and familiar with the front panel
controls. The sample o-scope displays are based on eight vertical and ten horizontal
divisions on the CRT screen, typical to most oscilloscopes. Most waveforms are actual
displays of the signal cited, photographed from my trusty Tektronix 475 oscilloscope.

Effects of
ASTIGMATISM
(Inconsistent focus)
Effects of
TRACE ROTATION
(Trace not level)
Fig. 3 Triggering
(+) Triggering

First ... a word on TRIGGERING.

Most oscilloscopes have a knob or two for Triggering. This tells the oscilloscope when to
start the sweep. When the Triggering Slope is placed in the (+) position, the scope will
begin its trace when the input signal goes positive. Likewise, when () triggering is
selected, the trace will begin when the input signal goes negative, as shown in Fig. 3. Often
there will be the option to chose the Triggering Source, such as CH.1 or LINE. Line
means the scope is triggered off the 60Hz line voltage, and is useful when synchronizing on
television signals or looking at 60Hz power supply noise. CH.1 or CH.2 means the scope
will trigger off the signal on channel 1 or 2 respectively. Trigger Level is at what voltage of
the input signal triggering begins. For example, if set high, triggering may not begin until
the input signal reaches several volts. When set around zero, it will trigger the moment the
signal goes positive (if set for (+) triggering). This setting can be troublesome if noise exists
on the signal. Adjust for stable triggering.

TEK 475

() Triggering
TEK 475

DC Voltages.
Say you want to check the transmit-receiver (T-R) switch in your QRP rig, or other digital signal.
See Fig. 4. The key line is the input to the HCT240 inverter to form the 0v TX on key- down and
the 0v RX on key- up. This switches the rig between transmit and receive (T-R Switch). It is a
logic function, that is, a voltage to represent ON or OFF.
Place the scope lead on pin 13 at 10v/div. and you should see the waveform like the top trace in
Fig. 4 ... about +6v on key-up and 0v on key down. Move the scope lead to pin 7 and you should
see 0v on key-up and about +8v on key-down (bottom trace). If the output does not go "HI"
(+8v) on key-down, or does not go to a solid "LO" (<1v) on key-up, the inverter is not working
properly. (Its busticated). Many shortwave receivers use similar schemes for switching filters
or attenuators.

+8v

TX-

2.2K
Key
In

13
.1

7 RX-

74HCT240

Fig. 4 - 38-Special
T-R Switch

While this test could be done with a DVM, the integration time is slow, requiring long keydowns to get the voltages. A scope will also show you how clean the switching is, or if there is an
ac voltage (or RF noise) riding on the T-R voltage.
Scopes are thus good dc voltmeters, with about a 5% reading accuracy.

AC Voltages.
Here is where an oscilloscope pays for itself by making AC voltage (and frequency) measurements. You must remember, AC voltages are displayed on a scope as peak-to-peak voltages,
while a voltmeter measures in rms. RMS voltages are about 1/3 the p-p voltage read on a scope,
or specifically:

VERT: 10v/div DC
HOR: 500mS/div

VERT: 0.5v/div ACV

HOR: 0.5mS/div

Vrms=(.707 x Vpp) = 0.354 x Vpp

3.8 divisions
peak-to-peak
times 0.5v/div
= 1.9 Vpp
= 0.67 Vrms

For example, let's measure the output voltage and frequency from
the sidetone oscillator in your QRP rig. Place the scope lead on the
audio amplifier output. On key-down, you get the waveform shown
in Fig. 5. The transmit sidetone audio is 1.9Vpp.

AC Frequency Measurement.
With this waveform, we might as well see what frequency our
sidetone or transmit-offset frequency is. Most operators prefer the
sidetone to be about 700750Hz. Trigger the scope for a stable
waveform and set the time-base (sweep) to display 2 or 3 cycles, as
shown in Fig. 6. Center the waveform between two horizontal divisions so zero volts on the waveform is on a graticle line, then move
the horizontal position so the first "zero crossing" is also on a division line.
Measure the time it takes to make one complete sine wave from one
zero-crossing to the next. In this example, it is 1.5 divisions, at 1mS
per division, or 1.5mS. Frequency is simply the reciprocal of time,
such that the sidetone frequency is:

1
1
= 667 Hz
=
f=
t
1.5mS
For some, this may be about right. For others, this may be a little
low to your liking. To raise it to 700Hz, calculate the time period of
700Hz (1/700 = 1.4mS). At 1.0mS/div, you can adjust your
sidetone or transmit offset until zero-crossings for a single
sinewave is 1.4 divisions. This will be about 700 Hz. (Sidetone may
not be adjustable on some rigs).
All frequency measurements are made in this fashion, by measuring the distance between zero-crossings (or from one peak to the
next) and converting the time period to frequency. This should emphasize the importance of ensuring your sweep speed is calibrated;
as any error in the time base will cause a corresponding error in the
accuracy of your time or frequency measurements.

TEK 475

Fig. 5
Fig. 6

VERT: 0.5v/div ACV

HOR: 0.5mS/div

0v
First "Zero
Crossing"
TEK 475

3 divisions between
"zero-crossings"

=1.5mS
=667Hz
Fig. 7

Example of
Waveform Quality

VERT: 2v/div ACV

HOR: 1.0mS/div

Quality of the waveform is another feature of a scope that is unsurpassed since you are
"seeing" the waveform in real time. Two examples of waveform quality are shown in Fig. 7.

FIG. 8 Full Scale

Signal Display

The top trace shows the sidetone frequency with distortion, perhaps due to improper timeconstant on the coupling capacitors or improperly biased audio amplifiers. The bottom trace
would be a raspy sounding side tone, due to the amplifier being over-driven and in
compression (clipping). The o-scope is an invaluable tool for detecting and diagnosing such
impurities in the signal quality.

MORE NIFTY MEASUREMENTS

TEK 475

Actual display
Tektronix 475

Amplifier Gain.
The gain of an amplifier can be measured in terms of voltage or decibels
(dB). For voltage gain, it is simply Vout/Vin of the amplifier. For example, if the input is 1Vpp and the output is 4Vpp, then the amplifier has a
voltage gain of 4.

Fig. 9 Volt. vs. dB relationship

Gain in dB is often more useful and is how the gains of amplifiers are usually expressed. With dB's, every-time you double the AC voltage, you add
6dB of gain. It is the ratio of output to the input, and this ratio is easy to
measure on a scope.
It is often easier to start with the output. Set the vertical amplifier gain to
display the amplifier output as a full-scale signal as shown in Fig. 8. Now
move the scope probe to the amplifier input without disturbing the scope
gain. You will of course have a much smaller signal, and the ratio of the
input to the output will be the gain in dB. In our example of using eight
divisions for full-scale, then four divisions would be 6db, 2 divisions
12dB, etc. as shown in Fig. 9. You may want to add your own dB scale
along your scope display to remind you of this relationship. Note: this is
voltage gain (Av=20log x Vout/Vin). In this example, with 4Vpp output
and 1Vpp input (Av=4), then the gain is dB=20log(4) = 20(0.602) = 12dB,
or as shown directly on the CRT tube. Since this is a relative measurement,
the absolute Vin or Vout voltage does not need to be determined.

3dB
6dB
12dB
18dB
TEK 475

Zero
Crossings
Fig. 10
Measuring
Phase
Shifts

Insertion Loss.
In some circuits, such as filters or attenuators, the loss in the circuit needs
to be measured, and like circuit gain expressed in dB. The loss through a
circuit is called the insertion loss. It is determined in the same way as
amplifier gain just presented, except start with the input (the highest AC
voltage) as the full-scale or reference display, then measure the output AC voltage (the
lowest level). The ratio is the insertion loss in dB.

1.6 divisions
x90 =145O

Fig. 11 Phase Shift

by imposition

For example, with a signal generator connected to your receiver, you want to measure the
insertion loss through the IF crystal filter. At the filter input, you can just barely squeek out 2
divisions of input signal on your scope at its most sensitive setting. The output from the
crystal filter is 1.5 divisions. The insertion loss would be 20log(1.5/2.0 div.) = 2.5 dB. If the
output were only 1.0 division (50% reduction), the insertion loss would be 6dB.

Measuring Phase Shifts.

Phase relationships between two signals at the same frequency can be measured with 2-5O
accuracy with a scope, although more suited for a dual-trace scope. The reference signal is
applied to CH. 1 and the signal to be measured to CH. 2. For proper phase measurements,
ensure your dual trace display is in the chopped mode, not alternate mode for proper phased
referenced triggering.

TEK 475

There are many methods to do this. One is to stretch out the signal so it takes 4 horizontal divisions, such that each division is 90 of
phase, as shown in Fig. 10. By measuring from a common point on one signal (zero-crossing or from peak-to-peak) to the next, the
phase can be measured. For example, say you are making a phased-array antenna in which one feedline must cause a 90O delay.
You calculate the electrical length for
a [L=(246/f) x Velocity Factor] and cut the coax to that length. You are now working on
O
blind faith that you have exactly 90 . With a scope, you can measure it fairly accurately by injecting a signal into one end with a
signal generator (at the frequency of interest) and a 50 load on the other. Connect the scope CH.1 to the coax (signal) input and
CH.2 to the load end and measure the phase. In the Fig. 10 example, the CH.2 signal is delayed by 1.6 divisions, at 90O /div is 145..
Your delay line is too long! Cut off an inch or two at a time until the CH.2 signal is 90O from CH.1 for precise tuning of the delay

line. (While departing from o-scopes for a moment, the sharp null of a phased array is astounding when exactly 90 delay is
achieved. More than 10-15O in error causes a very mushy null with little difference over a single vertical antenna. Most errors in
achieving exactly 90 by the "measure-and-cut" method are due to uncertainties in the stated velocity factor of the coax).
Another method is to superimpose the two signals on top of eachother. Make one signal larger than theO other so you know which
one is what, as shown in Fig. 11. In this example, the smaller signal lags the larger signal by about 100 , estimated by where they
cross. For more accurate determination, use the time base to measure the time period of one cycle (T1), then the time period one
O
signal lags (or leads) the other (T2). The phase shift is then =(T2/T1)x360 .
Phase measurements can be made on a single trace scope as well. First, connect the reference signal, uisng a BNC "T," to both the
external trigger and the normal vertical input. Adjust the trigger level so the zero-crossing occurs at the beginning of the trace
(left-hand graticle). Remove the reference from the vertical input, but not the external trigger, and apply the signal to be tested to
the vertical input without altering the time base or trigger level. The distance of zero-crossing of the test signal is from the lefthand graticle can now be measured to determine the phase, though with slightly less accuracy than using a dual-trace scope.
An interesting experiment is to measure the phase shift of the audio signal at different frequencies as it travels through the stages in
a CW, SSB or AM active filter. What is the phase shift of the wanted vs unwanted frequencies?

Measuring Rise and Fall Times.

In digital circuits, it is sometimes important to know the rise and fall times of a signal through a
gate. In amateur radio transceivers, this same interest could be applied to how fast the T-R
switch switches. On key-down, if the transmitter turns on slightly before the receiver is turned
off, it can produce an annoying "thump" in the receiver. Rise and fall times are measured by
triggering on the edge of the signal of interest, then increase to a faster sweep speed to measure
the time it takes the signal to reach 90% of its final level. The signal to be measured is shown in
Fig. 12 on the top trace, and the expanded version on the bottom. For proper rise times, the
signal being measured should be well within the bandwidth of your scope and using a low
capacity probe.

Fig. 12
Rise and Fall Times

For example, in Fig. 12 (bottom trace), the rise time is about 1/4th of a division. If the sweep
speed is 100nS/division, the rise time would then be about 25nS.

USING LIMITED BANDWIDTH SCOPES

Today's scopes have 200500MHz bandwidths. Likely your scope is much less than that. A limited bandwidth scope is still very
useful to the amateur or homebrewer. Say the bandwidth of your scope is 5MHz. This does not mean you can't see 7MHz signals.
It just means the peak-to-peak value has lost meaning, and will likely be very weak, since it is beyond the bandwidth of the scope.
(Like other bandwidth measurements in electronics, the bandwidth of a scope is usually based on the 3dB bandwidth. That is,
at the maximum bandwidth, you are already at the 3dB point, or a 25% reduction in the peak-to-peak voltage display). You can
still resolve individual cycles higher than the cited bandwidth to a certain degree and make gain and phase measurements, since
they are based on ratios.
Most of the examples in this article explore many regions of a communications receiver or ham transceiver without the benefit of
any great bandwidth. Experiment with your scope to learn its limitations. Use a good scope probe and make measurements with a
good ground to get the most out of the bandwidth you have.
For the homebrewer building circuits in the HF bands, a 50 MHz scope with good calibration will yield fairly accurate
measurements up to 30 MHz with little concern for accuracy. The old 465 or 475 series of Tektronix scopes, with 100/200 MHz
bandwidths, make an excellent oscilloscope for the amateur or experimenter. They can often be found at hamfests today for
$100150, and tend to maintain a fairly good calibration almost regardless of how much use they have seen.
In Part 2 - we'll probe (bad pun) into some advanced measurement techniques, even with a simple scope ... such as measuring
sideband rejection, filter responses, VCO phase noise, etc. (and what it all means).
72, Paul NA5N
[email protected]
[email protected]

Colophon:
Article prepared by NA5N using CorelDraw 11