Text preview for : HP-Bench-Briefs-1980-05-06.pdf part of HP HP-Bench-Briefs-1980-05-06 HP Publikacje HP-Bench-Briefs-1980-05-06.pdf
Back to : HP-Bench-Briefs-1980-05-0 | Home
Circuit Test Point Impedance select a probelscope combination While probe capacitance distorts
that has an Rin that is a t least 100 pulse shape, the flat portion of
Knowing the source impedance at times greater t h a n t h e source the pulse top (maximum
the point of measurement is critical. impedance." amplitude) can be used to make
If the source impedance is low, rise an accurate amplitude measure-
time and amplitude measurements But as frequencies rise, or pulse rise ment since i t contains low fre-
are generally no problem. For exam- time becomes very fast, scope input quency information. Conversely,
ple, batteries and power supplies capacitance becomes more and more if the pulse width is small com-
h a v e source impedances of mil- important, forcing use of an minia- pared to the measurement system
liohms. Signal generators are 25, 50 ture passive divider probe to reduce rise time, input capacitance can
or 600 ohms. The problem occurs that input capacitance. introduce errors since the source
when the source impedance is high. cannot fully charge the input
TTL h a s a source impedance of And a t the highest frequencies, if capacitance during its on time.
-2.5kR so even a t very low frequen- both amplitude and rise time are This problem becomes worse with
cies (single shot), measuring fast important in high source impedance increasing source impedance.
transition times is difficult a t best. circuits, an active FET input probe
should be used. 3. When source impedance is un-
known, the probe with the high-
If the ultimate in rise time is needed, est Zin usually yields the greatest
Basic Probe Considerations accuracy. However, for frequen-
a 500 divider probe may be used.
However, you must be careful of DC cies above 10 MHz, high probe
If t h e scope is being used a s a capacitance can reduce accuracy
monitoring device, the connection loading. A 50R divider probe with a n
input Xc of 5000 will attenuate the more than high probe resistance
between the signal source and scope can help.
is usually a direct 50R cable. How- amplitude of a signal, or upset the
ever, if the scope is being used for bias of the circuit if you probe the
4. If the source voltage is totally
wrong point (e.g., collector of a tran-
signal tracing or circuit analysis,
then some type of a n isolating device sistor), or burn up the probe if you unknown, it is wise to start with
a 1OO:l divider probe to reduce
'7
must be used to prevent the scope draw too much current.
the possibility of damaging the
from loading the circuit and a t - probe. This will also indicate
tenuating the signal. Today's mod- A current probe is useful in those
certain situations where touching whether or not there is enough
ern oscilloscopes use a probe for this signal available t o capitalize on
isolation. the circuit with any voltage probe a t
all, even one with t h e smallest the relatively low capacitance of
capacitance, changes the circuit's a 1OO:l divider probe. However,
The frequency of the signal you are in real-life situations, you proba-
measuring and source impedance at operation. It may be the collector of a
transistor where a n inductor and bly don't have a 1OO:l divider
the point of measurement influences probe. If this is the case use your
which probe to use. What you want capacitance form a tuned circuit.
standard 1 O : l divider probe.
to measure -rise time or amplitude
- is also a weighing factor. In gen-
Probe Rules for Making Probe Rules for Making Rise
eral, there are four types of probes
Amplitude Measurements Time Measurements
available for common circuit
analysis.
If have a a 1. Always try to probe the lowest
- High resistance probes minimum impedance source. For impedance point that contains
- Miniature passive divider example : emitter-to-base impe-
- Active (FET)
the waveform of interest. For
dance of a transistor is generally example: emitter-to-base impe-
- Current probes lower than the collector-to-base dance of a transistor is generally
Any voltage probe will load the cir- impedance (this implies a bal- lower than the collector-to-base
cuit you are attempting to measure. anced input measurement).
impedance (this implies a bal-
If amplitude measurements at low anced input measurement).
frequencies a r e all you a r e in- 2. Select a probe with the highest
terested in, then a passive one-to-
one 1MR non-attenuating probe
possible Zin at the frequency of
interest. When measuring pulse
2. The fastest input system will
generally have the lowest Rin
3
may be all you need. A good rule-of- amplitude, capacitance is not as and Cin. (This rule is limited only
thumb to remember is, "To keep re- important as Rin being high rela- by the maximum resistive load-
sistive loading errors below 1%, tive t o the source impedance. ing that the source can tolerate.)
WWW.HPARCHIVE.COM
c 3. A t high frequencies, the 50R di-
vider probe (500R at lpf) is the
capabilities. If it doesn't, select the
slowest sweep speed possible. This
best bet for accurate rise time will present a very slow-moving dot
measurements. However, you which you can use for adjustments.
must be careful of DC loading. To adjust astigmatism and focus, set
The 500Q input Xc will attenuate the beam intensity to a low level.
the amplitude of a signal, or Position the spot to center screen
upset the bias of the circuit if you and then adjust t h e focus and
probe the wrong point (e.g., col- astigmatism controls for the small-
lector of a transistor), or burn up est round dot.
the probe if you draw too much
current. How many of you are guilty of pick-
ing up a divider probe, connecting it
T I m
HOW 1.0 bet
n . mee Aavice on
- . 1 .
to your scope and t a k i n g meas-
urements without first checking the adjustments. Overshoot means the
Signal Source Loading and compensating capacitance is too
probe's compensation?
Probes large and the high frequencies are not
attenuated enough. Undershoot
One of the most common "pilot er- means the capacitance is too small
Application Note 152, titled "Prob- rors" is using an un-compensated and the high frequencies are
ing In Perspective," is available probe to make measurements. An attenuated too much.
free of charge from Hewlett-Packard un-compensated probe will cause er-
(write to the address at the rear of rors in the display which will be un-
this issue). AN152 describes i n detected Some kind Of a
detail all aspects of signal source should be recalibrated using the
standard waveform is checked. To be main vertical amplifier gain ad-
loading and probes. There a r e safe, You should always check probe
graphs, formulas, and lots of good justment (check your scope's service
compensation: manual for the proper procedure).
information - much more than can
be briefly described here. - at the beginning of each work
day With the scope checked and t h e
- whenever you re-connect a probe Probe compensated, YOU a r e now
Probe Compensation and to a different input connector ready to make some measurements.
Calibrating Your Scope - whenever you change probes
After you have gone through the To compensate the probe, connect it Observing Two Signals at
rigors of selecting the right probe, to the calibrator squarewave signal, the Same Time
you're ready to make some select DC coupling, and adjust the
measurements. scope's controls for a stable display. There are two techniques oscillo-
Select the lowest VOLTSLDIV set- scope manufacturers use to display
Let's begin by making sure your ting possible and center the top por- more than one signal at a time; dual
scope is operating properly. You tion of the squarewave on the screen. beam and dual trace. The dual beam
should check its trace alignment This provides a more precise adjust- scope has two independent deflection
astigmatism and focus adjust- ment method (if your scope is ad- systems within its CRT; hence two
ments, and finally, if required, probe justed properly). Adjust the probe beams a r e displayed simultane-
compensation. until you get a flat-topped square ously. The dual trace scope incorpo-
wave with no rounding or overshoot rates electronic switching to alter-
Trace alignment may be needed if of the signal's corners. Refer to nately connect two input signals to a
your scope is operated near a strong Figure 2. single deflection system; hence two
magnetic field. To make this ad- traces are displayed alternately by a
justment, ground the input and ad- After probe compensation, check the single beam. The switching rate is
just the TRACE ALIGNMENT con- scope's vertical accuracy against the usually in the 250-500 kHz range.
the best trace alignment internal calibrator square wave.
orizontal graticule line. With the vernier in the CAL posi- Most dual beam scopes are used in
tion, set the VOLTSDIV control to applications where two events that
way to adjust astigmatism obtain a display that is nearly full occur simultaneously would not be
3 is with a dot displayed on scale. The displayed square wave displayed correctly on a dual trace
m. Of course this assumes should match the p-p value of the scope as i t is switching between
ir scope has X-Y display calibrator output. If not, the scope signals.
WWW.HPARCHIVE.COM
`Since the greater majority of oscillo- against Channel B on the "X"axis. mode at the lower sweep rates. If /7
scope users have the dual trace mod- This was discussed i n detail i n your oscilloscope does not have this
els, we will confine this article to Part 1. automatic feature, the general rule
those types. Most of the following is to use the Alternate mode for fast
discussion is confined to the input Alternate Mode sweep speeds and the Chop mode for
switching controls on the front panel slow sweep speeds. On some occa-
In the Alternate mode, the A and B sions, fast sweeps might require the
and how they interact to provide the channels are alternately displayed,
dual trace capability. Chop mode if the signal rep-rate is
one channel per sweep. At fast sweep low, or even single-shot.
speeds, the alternate traces will ap-
pear t o be displayed a t the same Algebraic Sum
Dual Trace Input Controls time. However, as the sweep speed is
slowed, t h e traces will begin t o When both channels A and B are
There are many various ways to flicker showing the alternating selected (or added), you're in the A
manipulate two signals through two pattern. plus B mode. The CRT screen will
separate vertical input amplifiers display the algebraic sum of the two
and apply them to a single deflection Chop Mode input signals.
system CRT. Front panel controls
allow you to view the two inputs at In the Chop mode, both A and B One use of the A plus B mode is the
what appears to be the same time in channels are alternately displayed dual channel display of single-shot
either the Alternate or Chop modes. by switching between channels at a events. Another use is checking bal-
And you can add or subtract the fixed high-speed rate (250-500 kHz). anced or push-pull type amplifiers.
channels so that you can view the Even a t slow sweep speeds, both Balanced signals should have equal
algebraic sum or difference between channels seem to be displayed at the amplitude and be 180 degree out of
the two signals, Some oscilloscopes same time. Some oscilloscopes have phase. Since the sum of these signals
allow you to switch a channel to the the Chop mode connected to the is zero volts, you would expect to see
horizontal axis so you can view sweep control so the scope auto-
matically switches into the Chop
a straight line. If the signals do not
have equal amplitude or are not 180
7
Channe1.A on the "Y"axis plotted
The 50-Ohm Input Versus
Editor's Note: The following infor- input starts out with low impedance Problems of "High
mation about 50-ohm and HF in- and has essentially a constant input Impedance" Scope Inputs
puts is a small segment edited from impedance over the oscilloscope ver-
one of HP's application notes. For tical amplifier bandwidth, and virtually 0 Capacitive loading is much higher
more information about probes, eliminates the effects of capacitive than with 50-0hm inputs.
signal source loading, rise time loading. These input characteristics 0 Input impedance is highly variable
measurements, and phase meas- dictate the applications for which each with frequency.
urement rules, send for Probing in input is best suited and the choice of 0 There is a tendency to have confi-
Perspective, Application Note 152. probe to do the job. dence that there is no loading be-
Use the address on the last page of cause R is high, when in fact
`Bench Briefs'. capacitive loading is extremely
Benefits of "High high.
In recent years, there has been a lot Impedance" Scope Inputs
0 Does not offer a aood termination
of discussion over the merits and
demerits of these two types of for fast 50-ohm signal sources.
0 Passive probes (refer to Application Even when a 50-ohm termination is
oscilloscope inputs. The key issue Note can be used where high
in making a comparison is input im- used to shunt the high input resist-
input resistance is required. No ance, the VSWR caused by the
pedance versus frequency. The need for an active probe unless sig-
"high impedance" input is only high remaining capacitance is high.
nal levels are small relative to verti-
impedance for frequencies below cal sensitivity.
approximately 1 MHz. Above 1
MHz, the shunt capacitance takes 0 Can tolerate much greater input Benefits of 50-ohm
voltages than a 50-Ohm input. Oscilloscope Input
over and there is a fair amount of
1 uncertainty as to what the input im-
pedance actually is. The 50-ohm
0 Can be used with high voltage
probes.
0 Minimizes input capacitance and
the problems that it causes.
r degrees out of phase, then the signal
- -
you see will be a small sine wave.
Trigger Controls for Dual
Trace Oscilloscopes
As an example, when you're looking
at dual trace presentations, you may
want to see the correct time re-
Algebraic Difference The purpose of the trigger circuit is lationship between two pulses (i.e.,
When both channels A and B are to produce a stable display on the how much a pulse on channel A
selected and one channel is inverted, CRT. This is accomplished by syn- leads or trails a pulse on channel B).
you're in the A minus B mode. The chronizing the scope's sweep signal Or, maybe you only want to compare
CRT screen will display the alge- with the signal to be viewed. Several the shape of two signals, but their
braic difference between the two controls allow you to select the time separation makes comparison
input signals. source, positive or negative mode, difficult. The ability to select various
and level of the synchronizing trig- trigger functions from t h e front
One use of the A minus B mode is to ger signal. panel enhances the scope's useabil-
measure the voltage across an un- ity. Most modern dual trace oscillo-
grounded component without upset- When you're looking at just one sig- scopes feature controls that allow:
ting (or loading the circuit). This is nal on a single channel scope, trig- - trigger selection from either
called a balanced o r ungrounded gering is normally simple and input channel (shows time
input. For example, to measure the straightforward. However, when relationship)
voltage across the base-emitter junc- dealing with complex digital signals, - trigger selection from both
tion of a transistor, set both chan- or RF, or two asynchronous signals, channels (used for pulse shape
nels to the same volts-per-division, you need all the help you can get in comparison)
then connect channel A to the base the form of additional trigger con- - delayed triggering (called
and channel B to the emitter of the trols. You need to be able to tell the delayed sweep)
transistor. Connect the ground clips scope exactly which signal, and even - trigger holdoff
to circuit ground. This allows you to which portion of the signal, to trig- - trigger view (allows you to
view the small base-emitter voltage ger the sweep on. display the trigger signal)
on the CRT without upsetting or
grounding the circuit.
The "High Impedance" Input
0 Presents a better termination for b) Active probes are generally The high impedance oscilloscope
high speed 50-Ohm sources. required to increase the input input is more general purpose than
Minimizes pulse shape distortion, resistance to the 100kR to the 50-ohm input. However, it is
VSWR, reflections. 10MR area. Active probes are generally not as capable for making
0 When an appropriate probe is expensive but generally offer a accurate high speed pulse meas-
added to the 50-Ohm input, the input more flexible general probing urements, phase shift meas-
impedance can be considerably solution. urements, and high frequency
higher than that of a "high impe- c) 50-0hm inputs are not compati- amplitude measurements, even
dance" input scope. The source when a probe has been carefully
frequency for which this is true de- ble with high voltage probes.
selected.
pends on the particular probe
selected. 0 Does not have ac coupling for sig-
nal input. Most oscilloscope manufacturers
offer selectable high impedance
Problems with 50-ohm Input and 50-ohm inputs in the same
0 Limited maximum input voltage. mainframe or plug-in vertical ampli-
Typically, the maximum voltage
Summary fier. The choice of both inputs plus
which can be applied directly is less the various probes offered allow the
than +1OV. To summarize, the 50-ohm input of- versatility required to make most
fers superior measurement capability waveform measurements.
Requires a probe to increase the in many situations. However, it cannot
input resistance: be considered to be a general purpose
a) Passive probes can be used to solution because a probe is required
increase the input resistance to to increase the input resistance, and
5kQ if 1OOX division ratios can be ac coupling is not available without an
used. active probe.
WWW.HPARCHIVE.COM
~- - - _ I
______I_ -
Selectable Triggering Composite Triggering Trigger Holdoff
Selectable triggering is a conven- Composite triggering is the only way Trigger Holdoff is a variable control
ience feature. It allows you to look at to show two asynchronous signals. It used in conjunction with the Trigger
t h e display and then select the works like this. In the Alternate Level control. Trigger Holdoff
proper trigger source a t the push of a mode, Channel A sweeps once, then increases the time between sweeps
button. Selectable triggering allows Channel B, etc. The trigger selection and helps stabilize the display when
you t o trigger t h e display from controls cause t h e sweep to be triggering off complex digital sig-
either one of the input channels. triggered by the displayed signal; nals. On scopes without this control
therefore when Channel A is being you would use the Sweep Vernier
A typical set-up might be a signal displayed, it is the trigger source control as a holdoff, but then your
pulse into Channel A and its trigger and when Channel B is being dis- sweep is no longer calibrated.
pulse into Channel B. The correct played, it is the trigger source.
time relationship between the pulses Trigger View
is obtained when t h e sweep is A typical set-up might be two asyn- Some oscilloscopes have a feature
triggered by Channel B's signal in chronous pulses with nanosecond called trigger view. Basically it
the Alternate mode with Internal rise tiwes but separated in time by allows you to simultaneously display
trigger selected. Figure 3 shows how microseconds. You don't care about the external trigger signal on the
the time relationship between the the time relationship between the CRT in addition to the input signals.
two signals changes when the trig- two signals but want to compare the This can be quite valuable in verify-
gering is changed from Channel B to pulse shapes. If a fast sweep is used, ing the time relationship of the trig-
Channel A. only one of the pulses can be dis- ger signal to the displayed
played at a time.
1 In this situation, the pulses can be
compared by selecting Composite
triggering in the Alternate mode.
Figure 4 shows how the time re-
lationship between the two pulses is
lost when composite triggering is
used.
Delayed Triggering , ...
I -
I
I
I."
Delayed triggering is directly tied to
Figure 4. Composite trigger example
Figure 3. Trigger example showing Delayed Sweep. Delayed Sweep showing how to compare two asyn-
time relationship between signal allows easy location and expansion chronous signals connected to CHAN
of a small portion of the display, A and CHAN B.
connected to CHAN B.
permitting detailed analysis of that
View ` A shows the display (sweep) portion of the waveform. Delayed
being triggered on the positive-going Sweep can be triggered after a pro- "A' = Internal trigger (correct
AIternate display time
edge of CHAN B trigger. View `6' grammed delay, eliminating any Trigger on A or B rela-
shows the display being triggered on waveform jitter from the expanded tionship)
the positive-going edge of CHAN A
signal.
display.
"B" = Internal trigger (incorrect
How the sweep is triggered in the Alternate display time rela-
` A = Internal trigger Delayed Sweep mode will be de- Composite Trigger tionship)
Alternate display
Trigger on CHAN B (correct)
scribed in the Delayed Sweep por- In Alternate mode and Composite
Positive slope tion of this article. Trying to explain Trigger, each signal is its own trigger
it now may cause some confusion. source. Effective for signal shape
'B = Internal trigger comparison.
Alternate display
Trigger on CHAN A (incorrect)
Positive slope
I WWW.HPARCHIVE.COM
r waveforms. In Trigger View, the
point where the center horizontal
delayed sweep speed control is then
set to a faster sweep speed than the
graticule line and t h e trigger main sweep (the delayed sweep is
waveform intersect is the trigger triggered after t h e main). This
point. By varying the Trigger Level causes a small part of the main-
and Slope controls, you can select sweep trace to become intensified or
any point on the positive or negative brightened, depending on the setting
edge of t h e displayed trigger of the delayed sweep speed control.
waveform to trigger the sweep cir- The slower the setting, the larger
cuit, and measure how it affects the the intensified portion becomes. This
input signals. intensified marker can be moved
along the signal by rotating the
- - .-.. Limit Control
Banawiatn
.. - T . 1 1
DELAY control. Then, if we switch
the mode to Delayed Sweep, AUTO sweep sawtooth waveforms. Delay
time (twtl) is set by DELAY control,
mode, only the intensified portion
and sweep speed (t3-t'~)is set by
The bandwidth of some scopes can be will be displayed over the full DLY'D TIME/DIV control. tl-12 is the
reduced to minimize interference in screen. In other words, we have intensified part of the waveform.
high noise areas such as airports and rnagnified a portion of the trace. When SWEEP AFTER DELAY control
broadcast stations. On the H P is set to AUTO, sweep is triggered
1740A, the limiter effectively re- _
.r . . .. .
h e can see what really has hap-
. automatically at t l .
rlllrDc thn crnno'c hanrlur;rlth frnm
UUI.2" vvvy.2 u
Y l l U UUIlY I A U Y I I
. I I "lll
pened if we consider the signal being
100 MHz to 20 MHz. displayed by two time bases; first the
main sweep followed by the delayed, other mode, the delayed sweep is
For example, suppose you are pick- faster sweep (the intensified por- armed at the end of the delay period
ing up interference from 27 MHz tion). What we have done is to set up and requires a trigger signal (either
c' citizens band equipment. If the test
signal is less than 20 MHz, use the
a delay time from the start of the
trace to the beginning of the inten-
internal or external) to start the
delayed sweep. Since there is no way
Bandwidth Limit control to reduce sified portion of the trace. When the to know when the trigger signal will
the high frequency interference. delayed sweep is automatically occur, the delay time is
triggered, this time is equal t o the uncalibrated.
distance i n centimeters from the
Delaved SweeD Each of these methods has its own
start of the trace to the intensified
trace, multiplied by the sweep time advantages. In the AUTO mode, all
Thi m of the accumulative rate jitter that
per centimeter (i.e., it's calibrated).
mo b- has occurred since the start of the
The product is the delay time. When
ably one or tne least unaerstood delay time is displayed on the de-
we switch to Delayed Sweep (push
I capabilities of a modern oscilloscope. layed sweep. If, on the other hand,
the DLY'D button on the HP 1740A),
I n basic terms, the scope with de-
we start the main time base with an rate jitter is not desired in the dis-
layed sweep simply has two time play and a clear picture is needed,
input trigger, but we do not use it to
bases - main and delayed. then the armed mode should be used.
display the signal. Instead, we use it
as a clock that simply marks time In this mode the delayed sweep is
The controls for the two time bases retriggered after the delay time. A
until the delay period is over. Then
may be labeled and arranged in var-
the delayed time base sweeps, dis- new time reference is established,
ious ways and have various eliminating all of the jitter that has
capabilities, depending on the rnan- playing the signal. Figure 5 shows
how the delay system works in the occurred previously, providing a
ufacturer, but their purpose is basi- clear picture for accurate meas-
AUTO mode.
cally t h e same - t o expand a urements on the expanded pulse.
selected portion of the displayed sig-
nal. To accomplish this, each time There are two ways to cause the de-
base has its own complete set of layed sweep to be initiated after the How To Use Delayed Sweep
sweep and trigger controls. delay time. The first way (discussed
above), is called the AUTO mode. The delay controls on your oscillo-
In simple terms, delayed sweep func- The delayed sweep automatically scope usually will be highlighted by
tions as follows. The signal is first starts at the end of the delay period color or surrounded by lines on the
triggered by the main sweep at the with no trigger signal or other front panel. The HP 1740A sweep
speed set by the TIMEDIV dial. The external command needed. In the and delay controls are easy to find
WWW.HPARCHIVE.COM
Then move the DLY'D TIMEDIV Next move the intensified marker /7
control out of its OFF position. When alQng the waveform with t h e
this is done, a portion of t h e DELAY control until it is over the
waveform should become inten- pulse to be measured. Expand the
sified. This intensified marker is intensified portion to the full width
used t o locate the portion of the of the screen by selecting Delayed
waveform to be expanded. Adjust the Sweep (on the HP 1740A push the
Delayed Sweep Speed control so the DLY'D button).
marker is a little wider than the Adjust the DELAY control to posi-
pulse to be measured. Set t h e
tion the 50% amplitude point of the
SWEEP AFTER DELAY control to leading edge over the center vertical
the AUTO position. graticule line. Read and record the
Figure 6. Pulse width measurement DELAY dial setting. Note that some
using the delayed sweep controls. Next move the intensified marker
oscilloscopes use an LED readout for
along t h e waveform with t h e
DLY'D TlMElDlV dial = Sops DELAY control until it is over the this purpose.
Full scale accuracy = 3% (of SOOps)
pulse to be measured. Use the hori- Re-adjust the DELAY control to posi-
7 x 50pS = 350pS
0.03 x 500ps = 1Sps accuracy zontal position control to center the tion t h e trailing edge 50%
pulse width = 3 5 0 p rt_ 15ps intensified pulse. Expand the inten- amplitude point over the center ver-
sified portion to the full width of the tical graticule line. Read and record
screen by selecting Delayed Sweep the DELAY dial setting. The pulse
because of t h e d a r k grey back- (on the HP 1740 push the DLY'D width is the difference between the
ground. But no matter which scope pushbutton). Slightly re-just the two readings times the main sweep
you have, look for the word DELAY DELAY control to make the leading TIMEDIV setting. Figure 7 shows
in the control nomenclature. edge 50% point intersect a conven- an example pulse width measure-
ient vertical graticule line. Count
Suppose you want to measure the the number of divisions between the
ment using the differential method.
A Note on Time Interval
'1
width and rise time of the 5th pulse 50% points and multiply that times Measurement Accuracy
in a pulse train. If you try and ex- the Delayed Sweep Speed control
pand the signal with the main sweep The absolute accuracy of the
setting. Figure 6 shows an example
control, the pulse moves off screen. Differential Delayed Sweep
pulse width measurement using the
You could use the horizontal mag- method relies on the princi-
delay controls.
nifier to expand the sweep time and pal that the time interval of
perform the measurements as de- Differential Delayed Sweep the pulse to be measured is
scribed in Part l.However, you want greater than lcm of the main
more accuracy than t h a t method A more accurate time interval sweep. In this case the accu-
allows. The point about accuracy to measurement can usually be made racy is X% of the reading +
remember is t h a t time interval using t h e Differential Delayed Y% of full scale. The Y% of
Sweep method. To make a differen- full scale will totally mask
measurements are LEAST accurate
using the X10 magnifier, BETTER
tial measurement, select Main Sweep out the accuracy of the
using direct delayed sweep, and and adjust the TIMEDIV control to measurement. For the HP
BEST using differential delayed expand the sweep speed to make the 1740A, the accuracy is -+ 0.5%
sweep. pulse you want to measure as wide of the reading 20.1% of full
as possible. If the time interval of scale. Therefore, the accu-
NOTE the pulse is greater than one-half racy of a lOcm (full scale)
division on the screen, the differen- measurement is 2 0.6%.
If you don't have some type of tial method will be more accurate However, as the reading is
pulse generator for the follow- than the delayed sweep method. reduced to smaller and
ing experiments, try using the smaller parts of the main
amplitude calibator output on Switch the Delayed TIMEDIV con- display, the accuracy de-
your scope. trol out of its OFF position. When creases (+ error increases).
this is done you should see the inten- At one division of main
The first step in measuring pulse sified marker as in the previous sweep the error is 2 1.5%and 1
width and rise time is to adjust the measurement. Adjust the Delayed at 1 2 division of main sweep
1
vertical controls so that pulse height TIMEDIV control so the marker is a the error is now about equal
is six divisions (Le., enough height to little wider than the pulse to be to that of the direct-from-
easily see the 50% point). measured. CRT measurement.
WWW.HPARCHIVE.COM
r pulse's rise time or even its width.
The scopes we have been discussing
usually provide a feature to elimi-
at the faster delayed-sweep rate. The
transition point between sweeps is
positioned with the DELAY control
nate this unwanted jitter - it's after the MIXED button is pressed.
called Delayed Trigger.
Mixed Sweep is convenient for "peel-
Delayed Trigger controls are much ing oft'' pulses one by one from a
r
the same as those that control the long train and examining them
main sweep. There is a pushbutton individually.
that selects either AUTO or TRIG
mode (which is similiar to t h e Using the Dual Trace Scope
AUTO-NORM mode). When in the to Read Propagation Delay
TRIG mode, other controls are en-
abled that allow you to select the Propagation delay in reference to dig-
delayed sweep to be triggered "in- ital circuits is the amount of time it
ternally" or "externally," divide the takes for a change a t the circuit's
external trigger amplitude by 10, input to be noticed at its output. For
AC or DC couple the trigger signal, example, when the input voltage to
and adjust the Slope and Trigger an AND gate changes from a low to a
Level to start the delayed sweep at high, the output will respond at
any point on the waveform. some later finite time. You can use
your scope t o quickly and easily
Let's use the input signal as the measure this time a n d check i t
trigger source to see how the delayed against the device's specification.
sweep is triggered. Refer to Figure 8.
Use DELAY control to center trailing It would be difficult to specify a test
edge of pulse over center horizontal Each input pulse produces a trigger circuit and all the clips and probes
graticule. pulse. The main sweep is started by required to complete such a test. By
Figure 7. Pulse width measurement the first trigger pulse. The second now you should already have your
using the differential delayed sweep trigger pulse doesn't do anything be- scope set-up, probes compensated,
method. DELAY control is used to cause it's blanked by the delay time and enough background information
center the leading edge and then set by the DELAY control. The de- to complete your own experiment.
trailing edge of pulse over center hori-
layed sweep is "armed at t l when The necessary scope control settings
zontal graticule. Pulse width is differ-
ence between the two readings times the delay time ends. The next trig- are as follows:
the main sweep TIME/DIV dial setting. ger pulse to arrive after the delay
time ends starts the delayed sweep
TIME/DIV dial = 0.2ms sawtooth which deflects the electron
beam across the CRT. Since there is
DELAY dial reading = 7.46
-5.70 no way to know when the trigger
signal will occur, the delay time is
1.76 x 0.2ms = 352ps
1.76
uncalibrated.
Ma,n
i*
I
Tilggrrs
blanked durinq
delay lime
7
1
i
:
Accuracy is 20.5% for the DELAY dial swPep I I I
I
and -t0.1% of full scale
In effect, you have eliminated all in- I I I
Delayed 1 I
terference by triggering the sweep sweep 1
0.005 x 352 = 176ps (dial) on only that portion of the waveform 1 I
`0 `1 `2
0.001 x 2ms = 2ps(fuII scale) you have selected to examine.
pulse width 352~s 4 p ~
5
Mixed Sweep Mode
Figure 8. Delayed sweep delayed
How To Use The Delayed Trigger There is another mode of delayed- trigger example. The main sweep is
started by the first trigger pulse at to.
Method To Eliminate Waveform sweep operation found on some oscil- The second trigger pulse "arms" the
Jitter loscopes, called mixed sweep. In this delayed sweep at t l . The next trigger
mode the main sweep is displayed on pulse to arrive after the delay time
Often, when you expand a signal, the screen for the amount of delayed ends starts the delayed sweep
waveform jitter becomes more pro- time desired. Then the sweep sawtooth at tq. The time between t l
and t2 is unknown which makes the
nounced. This jitter makes it dif- increases in speed part way across
delayed sweep uncalibrated.
ficult to accurately measure the the screen and finishes up the trace
WWW.HPARCHIVE.COM
- Always use identical probes (a 50 use the Algebraic Sum of two chan- If the recorder's head is adjusted
ohm passive probe is useful in nels to make sure the outputs of a properly, both signals should lie on
high impedance circuits where push-pull amplifier are 180 degrees top of one another which indicates
maximum rise time accuracy is out of phase - another type of phase they are in phase, Varying the head
necessary) measurement. azimuth will shift the phase of the
- Set the input coupling switch to signals which you can read directly
AC A more accurate method of phase off the display (remember that one
- Connect circuit's input signal to measurement uses the time-delay division equals 45 degrees). For
CHAN A principle. This is the same type greater resolution use the x10 mag-
- Connect circuit's output signal to measurement discussed previously nifier. Now each division represents
CHAN B under the heading, "How To Meas- 4.5 degrees.
- Alternate display ure Propagation Delay." It involves
- Internal trigger on CHAN A in looking at two signals simultane- This concludes the Basic Oscillo-
AUTO mode ously and observing any phase dif- scope articles. For more information
- Adjust vertical controls so sig- ference between the two. on specific oscilloscope applications,
nals are centered and approxi- Hewlett-Packard offfers many free
mately six divisions high One example of using the time-delay application notes. Several examples
- Adjust the sweep control so the method to make accurate phase are: AN152 - Probing in Perspec-
pulses look like those used for measurements is checking the out- tive, AN223 - Oscilloscope Meas-
making rise time measurements put of a stereo tape player. The head urements in Digital Systems,
alignment, or azimuth, must be pre- AN185-2 - Transmission Line
You should see the leading edges of cisely set for best high frequency and Matching and Length Measurings
two pulses separated by a measura- zero phase response. The necessary Using Dual-Delayed Sweep, and
ble distance. Measure the propaga- scope control settings are as follows: AN262 - Eliminating Time Base
tion delay at the 50% points (center Errors from Oscilloscope Meas-
horizontal graticule line) by count-
ing the number of divisions between
- Always use identical probes
- Set the input coupling switch to
urements. Many HP engineers and
customers have collaborated on
7
the two pulses and multiplying that AC these notes to pass their applications
times the setting of the sweep speed - Connect stereo tape unit's left research and experience on to you.
control. If you measure two divisions output to CHAN A Some notes are tutorial in nature,
and the TIMEDIV dial is set at 5 ns, - Connect stereo tape unit's right while others describe very specific
the propagation delay is 10 ns. For output to CHAN B "how to" procedures. All HP applica-
greater resolution, use the x10 mag- - Select Chop display (for low fre- tion notes are designed to help you
nifier or delayed sweep. quency test signal) obtain maximum use from your
- Internal trigger on CHAN A in Hewlett-Packard equipment. Please
AUTO mode contact your local HP office for more
- Adjust vertical controls so sig- information.
Using the Dual Trace Scope nals are centered and approxi-
to Measure Phase mately six divisions high Editor's N