Tektronix S-4 User manual

INNSTRUCTION
MAN UAL Serial Number
Tektronix, Inc.
S.W.
Millikan
Way
@
P.
©.
Box
500
@
Beaverton,
Oregon
97005
@
Phone
644-0161
@
Cables:
Tektronix
070-0896-00
169

oa
D
a
or
3 ee
2
en. oe
cs
i
—
_—
. a
—
oo
a
Type S-4
®

CONTENTS
Section 1 Specification
Section 2 Operating Instructions
Section 3 Circuit Description
Section 4 Maintenance
Section 5 Performance Check/Calibration
Abbreviations and Symbols
Parts Ordering Information
Section 6 Electrical Parts List
Mechanical Parts List Information
Section 7 Mechanical Parts List
Section 8 Diagrams
Mechanical Parts List Illustrations
Accessories
Abbreviations and symbols |used in this manual
are based on or taken directly from IEEE Stand-
ard 260 ‘Standard Symbols for Units,’’ MIL-STD-
12B and_other standards of the electronics in-
dustry.
at the rear-ofthe-manuat
Cha
nge information,| if any, is located
£
i
Type S-4

oo
a
3 = 2 fal g at &
r)
ee
|
:
oe
:
oe
ee
Lae
oe
ee
:

Type 5S-4
SECTION I
TYPE
S-4
SPECIFICATION
Change
information,
if
any,
affecting
this
section
will
be
found
at
the
rear of the manual.
General Information
The Type S-4 Sampling Head is an input signal processing
unit for Tektronix sampling units such as the Type 352, 355,
and 386. Input characteristics of the sampling system are
determined by the Type S-4. The Type S-4 has an input im-
pedance of 500 and a risetime of 25 ps or less.
The Type S-4 can be installed directly into a sampling unit
or used remotely (with the Type 3S2 or Type 355) on an
optional 3 foot or 6 foot extender cable. When used with a
Type 356 Sampling Unit the Type S-4 is plugged into the
dual-input extender cable supplied with the Type 356.
A portion of the signal delivered to the sampling head
input is coaxially coupled to the sampling unit for use as an
internal trigger signal. The Type 382 allows selection of the
trigger pickoff signal from the Channel A or Channel B
sampling head. Type 385 and 356 Sampling Units permit
use of the trigger pickoff signal only from Channel A. Use
of this trigger pickoff signal is dependent upon the type of
sampling unit and sampling sweep unit used in the system.
See the instruction manual for your sampling unit and
sampling sweep unit for further information.
Vertical deflection factor of the sampling system is label-
ed at the top of the Type S-4 as mVOLTS/DIV; the label
refers to the sampling unit Units/Div switch of the corre-
sponding channel.
Digital Unit Programming Connections
The Type S-4 has two contacts at its rear connector that
program the decimal and units-of-measure lamps of a Tek-
tronix Type 6RI1A or Type 230 Digital Unit. These connections
are not used with a Type 382 Sampling Unit. When using a
Type 355 or Type 386, one of the sampling head contacts
notifies the digital readout unit, through the sampling unit
digital control circuits, that the Type S-4 is a voltage meas-
uring head. This causes the Volts lamp of the digital readout
unit to light. The other sampling head contact notifies the
digital readout unit, through the sampling unit digital control
circuits, that the Type $-4 sensitivity requires no decimal shift
of the numbers around the sampling unit Units/Div switch.
ELECTRICAL CHARACTERISTICS
Characteristics
The following characteristics apply over an ambient tem-
perature range of 0°C to +50°C and after a five minute
warmup providing the instrument was calibrated at a tem-
perature between +20°C and +30°C.
Characteristics listed below apply only after the Type S-4
has
been
properly
mated
to
the
sampling
unit
and
indicator
oscilloscope
and
after
these
units
have
been
given
sufficient
@
warmup time. To determine the particular system warmup
requirements, refer to the related amplifier and indicator
oscilloscope instruction manual. A procedure for mating the
Type
S-4
to
the
sampling
unit
can
be
found
in
the
Operating
Instructions section of this manual
ELECTRICAL CHARACTERISTICS
Performance Requirement
50Q within 5%
Characteristic
Input Resistance
Response to Step
Signals
Risetime (10% to 25 ps or less (Verify using step 9,
90%) Section 5.)
Pulse Flatness Devi-
ation (after 90%
point on display}
Ops to 400 ps —10% to +10% with a total of
20% or less. (Verify using step 10,
Section 5).
0% to +10% with a total of 10%
or less. (Verify using step 10, Sec-
tion 5).
—2% to +2% with a total of 4%
or less. (Verify using step 10, Sec-
tion 5).
400 ps to 25 ns
25 ns or more
Maximum Operating | 1V P-P
Signal Voltage
Safe
Overload
Signal
|
Not
to
exceed
+
or
—5V
(DC
Voltage
plus
AC
peak}
Adjustable in the sampling unit to
unity for signals up to 500 mV P-P;
when the loop gain is adjusted to
unity with positive input signals,
the loop gain will be within 5%
of unity with negative input sig-
nals and vice versa.
Loop Gain
Baseline
Shift
(with
10
mV
or
less,
30Hz
to
50kHz
change
in
repetition
rate)
5mV or fess, measured tangen-
tially
Displayed Noise
ENVIRONMENTAL CHARACTERISTICS
Storage Operating
Temperature—As stated
above Electrical Character-
istics table
Temperature—
—40°C to +65°C
Altitude—To
50,000
feet
Altitude—To
15,000
feet
1-1

Specification—Type 5-4
MECHANICAL CHARACTERISTICS
Dimensions— Height = 2 inches
Width = 1%, inches
Length = 4%, inches
Construction— Epoxy laminated circuit boards. Aluminum
Accessories—
wrap around cabinet with aluminum casting
at front and rear. Anodized aluminum front
panel.
An illustrated list of the accessories supplied
with the Type S-4 is at the end of the Me-
chanical Parts List pullout pages.

Type S-4
SECTION 2
OPERATING
INSTRUCTIONS
Change information, if any, affecting this section will be found at the
rear of the manual.
General Information
This section provides the basic information required for
operation of the Type S-4 Sampling Head, including installa-
tion and First Time Operation.
The Type S-4 can be used on an extender cable without
compromising the response of the measurement system. Sig-
nals are applied to the 50 Q input of the sampling head
through a 3 mm, 50 © coaxial connector located at the cen-
ter of the front panel.
Unity
loop gain can be obtained with input signals of up
to 0.5 volt (peak-to-peak). Unity loop gain ensures the most
accurate displays, of particular importance during fast
changes in input signal amplitude. See the Basic Sampling
Principals Section of your sampling unit manual for further
information.
The Type S-4 may be used with input signals up to 1 volt
(peak-to-peak). Specifications for the Type S-4 Sampling
Head are given in Section 1 of this manual.
NOTE
Attenuators, with threaded 3 mm connectors, are
available as optional accessories. These attenua-
tors are useful in reducing the amplitude of large
signals. Other optional accessories with 3mm
connectors include coaxial cables, a 50 termi-
nation, and adapters for interconnecting various
types of connectors. Refer to your Tektronix cata-
log or contact your local Tektronix Field Engineer
or Representative for further information about
optional accessories.
Installing the Type S-4 Sampling Head
Fig. 2-1 shows the Type S-4 partially installed in a Type
352 Sampling Unit with a Type 564 Oscilloscope and a
Type 3T2 Random Sampling Sweep unit. The sampling head
(or
heads)
can
be
plugged
into
the
sampling
unit
as
shown,
or used remotely on a special extender cable. Three and
six foot extender cables are available. Order the three foot
extender
cable
by
Tektronix
Part
No.
012-0124-00,
or
the
six
foot extender cable by Tektronix Part No. 012-0125-00. Con-
tact your local Tektronix Field Engineer or Representative
for price and availability of these optional accessories.
To insert the Type S-4 into a compartment of the sampling
unit, proceed as follows:
1. Pull the latch knob (Fig. 2-1) outward from the front
panel
(the
latch
knob
will
push
out
normally
when
the
unit
is inserted if the knob is left free to move).
2.
Insert
the
Type
S-4
slowly
into
the
compartment,
so
the
two
plastic
guides
in
the
compartment
engage
the
S-4.
®
Fig. 2-1. Inserting the Type S-4 Sampling Head in the sampling
unit.
3. Push the Type S-4 completely into the compartment.
4. Push the latch knob to lock the Type S-4 in place.
To remove the Type S-4 from the compartment, pull the
latch knob away from the front panel, then pull the unit
from the compartment.
To use the Type S-4 on an extender cable, install as
follows:
1. Pull the latch knob located on the head end of the ex-
tender cable outward from the panel (the latch knob will
push out normally when the extender is inserted if the knob
is free to move).
2. Insert the extender cable head end slowly into the de-
sired compartment in the sampling unit so the two plastic
guides in the compartment engage the unit.
3. Push the head completely into the compartment.
4. Push the latch knob to lock the extender cable head
end in place.
5. Connect the Type S-4 to the other end of the extender
cable
in
a
similar
manner,
and
set
the
latch
knob
to
hold
it
in place.
2-1

Operating Instructions—Type S-4
6. To remove the Type S-4 from the extender cable, pull
the latch knob on the front panel of the Type S-4, and re-
move the unit from the extender cable.
7. To remove the extender cable head from the sampling
unit compartment, pull the latch knob otuward from the front
of the panel, then pull the unit free.
Mating
The vertical deflection factor and the sampling system
loop balance (adjusted by the Gate Bal control) should be
adjusted when mating a Type S-4 to a sampling unit.
The mVOLTS/DIV label on the Type S-4 names the deflec-
tion factor units of the sampling unit Units/Div switch. For
example, with the Type S-4 installed in a Type 382 and the
Units/Div switch set at 100, each major division of CRT de-
flection corresponds to 100 millivolts of input signal (when
the Variable control is in the Cal position). The deflection
factor is mated to the indicator oscilloscope CRT by adjust-
ing the sampling unit Gain control. Refer to the sampling
unit manual for its Gain adjustment.
The Gate Balance adjustment of the Type S-4 is not great-
ly affected when the sampling head is changed from one
sampling unit to another, or is operated on an extender
cable. However, for precise offset measurements, the Gate
Balance control must be readjusted whenever such a relo-
cation is made. Location of the control and details of its
adjustment are given below under the heading First Time
Operation.
FIRST TIME OPERATION
In addition to the Type S-4, this procedure utilizes a Type
S-50 Pulse Generator, Type 3S2 Dual-Trace Sampling Unit,
Type 312 Random Sampling Sweep, and a Type 564 Storage
50 Trigger
input
2ns Flexible 3mm
coaxial cable Trigger cable
Fig. 2-2. Equipment setup used during First-Time Operation.
2-2
Oscilloscope. See Fig. 2-2. The Type 3T2 Random Sampling
Sweep permits viewing the signal in advance of the trigger-
ing event. Such sampling sweep trigger lead-time is required
when using the Type S-4 with the Type 3S2 for viewing fast
pulses without a pretrigger. The built-in pretrigger circuit of
the Type S-50 permits sequential sampling units, such as the
Type 375 or Type 3T6, to operate at their faster sweep rates.
Setup Information
1. With the Type 564 Power switch off, insert a Tektronix
Type 352 Sampling Unit into the vertical channel compart-
ment (left) and a Tektronix Type 3T2 Random Sampling
Sweep unit into the horizontal channel compartment.
2. Insert the Type S-4 into the Channel A compartment
(left) of the Type 352 leaving the latch knob free to move.
Once the S-4 is seated, push the latch to lock it in place.
3. Insert the Type S-50 into the Channel B compartment
(right) of the Type 3S2 leaving the latch knob free to move.
Once the S-50 is seated, push the latch to lock it in place.
4. Connect the Type S-50 output connector to the Type S-4
input connector using a 2 ns section of flexible coaxial line.
See cable bending limits discussed under Cable and acces-
sory Considerations later in this section.
NOTE
Connectors at both ends of the coaxial cable
should be firmly connected to mating connectors
or accessories. Tighten slightly more than finger
tight using a °/15 inch wrench. A good connection
is necessary to minimize reflections at the junction
of connectors.
5. Connect Pretrigger Out of the Type S-50 to the 50 2
Trigger Input of the Type 372.
6. Set the Intensity control on the Type 564 fully counter-
clockwise.
7. Connect the Type 564 to the power line and set the
Power switch to On.
8. Allow about 5 minutes warmup so the units reach oper-
ating temperature before proceeding.
9. Set instrument controls as follows:
TYPE 352
Display Mode CHA
Normal-Smooth Normal
Horiz Compatibility Sw. Sampling
Channel A controls
Position Midrange
DC Offset Midrange (5 turns from one
end)
Units /Div 200
Variable Cal
Invert Push in
Dot Response Midrange

Operating Instructions—Type S-4
TYPE 372
Time Position Midrange
Horiz Position Midrange
Samples/Div 9 o'clock position
Samples/Div Sw Variable
Display
Mode
Normal
t
Start
Point
Before
Trigger
t
Sweep
Rate
100
ns/Div
a
Range
1
ps
:
Dsiplay
Mag
x1
t
i
+Frys
Time
Magnifier x1
portion
jj
Variable
Cal
‘poe
| |
Trigger
Source
Ext
P|
Trig.
Sensitivity
Fully
clockwise
TYPE
564
Astigmatism
Midrange
Focus
Midrange
Intensity
Fully
counterclockwise
Upper
and
Lower
Screen
Non-Store
10. Advance the Type 564 Intensity control until the trace
brilliance (free running sweep) is at the desired viewing level.
Adjust Astigmatism and Focus controls.
11. Center the trace on the graticule with the Type 382 “A”
Position control and/or the DC Offset control.
12. Adjust the Type 312 Trigger Sensitivity and Time Posi-
tion
controls
for
a
stable
triggered
display
of
the
Type
S-50
:
|
output
pulse.
See
Fig.
2-3A
(displayed
signal
amplitude
has
Sages
been
adjusted
to
5
divisions
by
turning
the
Type
3S2
i
gor
foot
Channel A Variable control).
Display
Characteristics
of
Fastrise
Input
Pulses
In Fig. 2-3A the displayed pulse shows overshoot at the
end of the fastrise portion of the pulse. This overshoot is
normal when using the Type S-4 together with a fastrise
pulse generator such as the Type $-50. Since the 0% to
100% amplitude of the pulse is 5 divisions, each division
represents 20%. In Fig. 2-3A the overshoot appears to have
an amplitude of approximately 0.4 divisions or 8%.
The preshoot portion of the Type S-50 output pulse is the
relatively slow rise signal driving the Type S-50 output tunnel
diode.
The
fastrise
portion
occurs
as
the
output
tunnel
di-
ode switches. Thus, the region labeled as the preshoot por-
tion of the pulse in Fig. 2-3A is apparently much slower than
the fastrise portion. The presence of several dots indicates
samples taken during the preshoot, and the absence of
samples during the fastrise portion indicates the difference
in risetime. The appearance of these portions of the Type
S-50 output pulse at a faster sweep rate is shown in
Fig. 2-3B.
(C) Determining displayed risetime.
Set Type 3T2 Range switch at 100 ns and Time Magnifier
to X10. Set the Type 312 Start Point switch to With Trigger.
Turn the Type 312 Time Position control until a display simi-
lar to that of Fig. 2-3B is visible. This waveform shows that
after
a
rather
gradual
rise
requiring
about
5
ns
(5
divisions)
Fig.
2-3.
Fastrise
pulse
displays.

Operating Instructions—Type S-4
the signal amplitude suddenly increases by approximately
500 mV. The displayed pulse of Fig. 2-3B is typical of the
Type S-50 and Type S-4 combination.
In Fig. 2-3B the aberration occurring approximately 4.5 ns
after the fastrise portion of the pulse is a result of a reflec-
tion at the input gate of the Type S-4. A portion of the fast-
rise part of the pulse traveling from the Type S-50, through
the 2 ns length of coax and through the Type S-4 input con-
nector to the input gate, is reflected back to the Type S-50.
This round trip requires approximately 4.5 ns when the Type
S-50 and Type S-4 are interconnected with a 2 ns section of
coaxial line. Adding a second 2 ns section of line in series
with the first would result in the reflection being displayed
4ns (2 X 2ns) later, or 8.5ns after the fastrise portion of the
pulse.
Determining Displayed Risetime
1. Set the fastrise portion of the displayed pulse to grati-
cule center, as shown in Fig. 2-3B, using the Time Position
control of the Type 372.
2. Change the Type 3S2 deflection factor to 100 mV/Div
(Variable at CAL).
3. Vertically center the fastrise portion of the displayed
pulse using the Type 3S2 Channel A Offset control.
4. Set the Type 312 Time Magnifier at X50 and readjust
the triggering for minimum jitter using the Type 312 Sensiti-
vity control.
5. Set the Type 3T2 Display Mag at X10 and carefully
adjust the Time Position control for a display similar to that
of Fig. 2-3C.
In order to accurately determine the displayed risetime
the display should be photographed or stored (on a Storage
Oscilloscope CRT) to permit careful analysis. To store the
display, set the Type 3T2 Display Mode to Single Swp., the
Type 554 upper and lower screen for storage, and push the
Type 3T2 Single Sweep Start pushbutton.
The pulse risetime is found by determining the horizontal
distance between the 10% and 90% amplitude points and
multiplying this distance by the Time/Div setting of the
sampling sweep unit. It is recommended that the 0% to
100% amplitude be adjusted to exactly 5 divisions using the
Type 382 Variable control, and the display vertically center-
ed using the Type 382 DC Offset control. This will position
the 0% point 2.5 divisions below graticule center and the
100% point 2.5 divisions above graticule center. The 10%
amplitude will then appear 2 divisions below graticule cen-
ter and the 90% point 2 divisions above graticule center.
More information on locating the 0% and the 100% points
of the displayed waveform is given later in this section.
Gate Balance Adjustment
The Type S-4 Gate Balance may occasionally require re-
adjustment. The location of the control is shown in Fig. 2-4.
Channel A sampling head can be adjusted while on an ex-
tender cable or when inserted in the sampling unit. In order
to adjust the Gate Balance of a Type S-4 located in the
Channel B compartment of the sampling unit, it is necessary
to remove the oscilloscope side panel and any sampling
head plugged into the Channel A compartment.
2-4
—
Type S-4
Sampling Head
Fig. 2-4. Gate Balance location.
1. Connect a bench multimeter set for approximately 30 V
full scale between ground and the Type 3S2 Channel A
Offset jack.
2. Adjust the Type 382 DC Offset control until the multi-
meter reads 0 volts.
3. Increase the meter sensitivity and repeat the adjustment
for greater accuracy.
4. Disconnect the semi-rigid cable from the Type S-50,
leaving the other end of the cable connected to the Type
S-4 input connector.
5. Turn the Type 312 Trigger Sensitivity control clockwise
to free-run the trace.
NOTE
Operating the sampling head without the input
connector terminated by a 50 © resistor or coaxial
cable may cause a vertical shift of the zero signal
baseline by a few millivolts. This is caused by the
strobe kickout signal being reflected from the open
input connector, and arriving back at the sampling
gate while it is conducting. Because of this, termin-
ate the input circuit when setting the display zero
reference point. Also use a 15cm airline, or a
cable providing at least 500 ps of delay, between
the Type S-4 input and any tunnel-diode genera-
tor or circuit that is sensitive to signals entering
its output circuit.
6. Switch the Type 382 Units/Div control throughout its
range, adjusting the Type S-4 Gate Balance control for a
trace shift of not over one division as the Units/Div control
is switched from one end to the other.

5 ns/div_
100 0 mV /<
(B) Dot response set for unity loop gain.
Fig. 2-5. Typical display changes with dot response adjustment
when using random sampling operation.
OPERATING INFORMATION
Dot Response
Dot response of the sampling unit must be adjusted for
unity loop gain to minimize waveform distortion and dot
splattering. With unity loop gain the system will track the
input signal closely.
An example of the appearance of a displayed waveform
with the dot response adjusted for a gain greater than unity
is shown in Fig. 2-5A. In Fig. 2-5A the Type 312 is set for
random sampling (Before Trigger) and the Time Magnifier is
set at 20. The output signal of the Type S-50 Pulse Genera-
tor is being applied to the Type S-4 input through a 2X
attenuator. Fig 2-5B shows the effect on the displayed wave-
form of setting the dot response for a loop gain of unity.
Although proper adjustment of the dot response control is
particularly necessary during random sampling, improper
®
Operating Instructions—Type S-4
adjustment of dot response (loop gain) will also result in a
distorted waveform display during sequential sampling.
Consult the Basic Sampling Principals and the Operating
Instructions Section of your sampling unit instruction manual
for further information about dot response and loop gain.
Smoothing
Setting the Normal-Smooth switch of your sampling unit
to Smooth will result in a reduction of displayed noise. Since
switching to Smooth reduces loop gain to approximately
0.3, transient response problems caused by low loop gain
may show up. Smoothing should not be used when the Type
372 is used as a random sampler.
A loop gain of less than unity can be useful, if the result-
ing compromise is understood and the system is operated
properly. Random noise in the display is reduced when
loop gain is less than unity, since several consecutive
samples are averaged. The averaging may also slow down
the fastest risetime capability, depending upon the number
of dots contained in the step transition. By increasing the
number of dots (Samples/Sweep or Samples/Div) in a step
transition, the display will follow the actual step transition
more closely. If the number of samples (dots) in the step
transition is reduced below 12 the true waveshape may not
be displayed in Smooth.
Effect of Excessive Input Voltage
A maximum operating signal voltage of 1 volt (peak-to-
peak) is specified for the Type S-4 in Section 1 of this
manual. Signals having an amplitude of more than 1 volt
(P-P) at the Type S-4 input may be distorted by the sampling
head. Large signals must be reduced in amplitude to not
more than 1 volt (P-P) at the Type S-4 input.
A 15 volt (P-P) signal can be reduced to a suitable ampli-
tude at the Type S-4 input by inserting a series combination
of 10X and 2X attenuators in the coaxial line delivering the
signal. A series combination of 10X and 5X attenuators
would also be satisfactory.
Displayed waveforms with several values of signal volt-
age applied to the Type S-4 input are shown in Fig. 2-6.
Fig. 2-6A shows the displayed waveform with an input sig-
nal of 0.5 volt (P-P). In Fig. 2-6B the input signal has been
increased to 0.9 volt. Fig. 2-6C shows the distorted display
resulting from applying a signal of 1.1 volts (P-P) to the Type
S-4 input.
See the discussion on Sine Wave Signal Measurements
later in this section for information on signal amplitude
affect upon display amplitude accuracy.
Risetime Measurements
Accurately determining risetime from an observed display
requires that the position of the 10% and 90% points be
accurately known and that the horizontal distance between
these points be accurately measured. The timing accuracy
of the sampling sweep unit must also be known. Switching
the sampling sweep unit from one oscilloscope to another
requires readjustment of the horizontal gain control on the
sampling sweep front panel. See your sampling sweep unit
2-5

Operating Instructions—Type S-4
mi Gis casa
BEG ai
“086 fhe
(B) —900 mV input signal
(C) —1.1V input signal
Fig. 2-6. Effect of input signal amplitude on display amplitude.
2-6
instruction manual for timing accuracy specifications and
complete instructions for checking timing.
During First Time Operation earlier in this section a stor-
age oscilloscope was used to display risetime. Ability to
store a display is an advantage when determining risetime
from the CRT display, especially when using a high dot den-
sity and a low signal repetition rate. An alternate method
of determining risetime when a storage oscilloscope is not
available and the display rate is slow is to photograph
the display.
0% and 100% Levels with the Type S-4/S-50
Combination
When using the Type S-4/S-50 combination a special tech-
nique is used in locating the 0% and 100% amplitude levels
of the displayed waveform. In Fig. 2-7 two reference position
are shown. Reference point 1 is at the corner preceeding the
fastrise portion of the display. Reference point 2 is at the
corner following the fastrise portion of the display. Both
reference points are located where the rate of change of
slope is maximum (where the radius of curvature is least). The
0% zone is centered 35 ps before the first reference point
and the 100% zone is centered 35 ps after the second refer-
ence point. Each zone is 35 ps in width. The 0% and 100%
levels are the average amplitude of the zones.
Follow the procedure below in determining the displayed
risetime:
1. Locate reference point 1 and 0% zone.
2. Locate reference point 2 and the 100% zone.
3. Using the sampling unit DC Offset and Variable con-
trols, set the 0% level 2.5 divisions below graticule center
and the 100% level 2.5 divisions above graticule center. Re-
member the level is the average value within the zone.
4. Locate the-10% point on the fastrise portion of the dis-
played waveform (the point where the trace crosses the hori-
zontal graticule line located two divisions below the graticule
center).
5. Locate the 90% point on the fastrise portion of the dis-
played waveform (the point where the trace crosses the
the
horizontal gratucule located two divisions above grati-
cule center).
6. Multiply the horizontal separation, in divisions, be-
tween the 10% and 90% points by the Time/Div setting of
the sampling sweep unit to determine the displayed risetime.
The specified combined risetime of the Type S-4/S-50 combi-
nation is 35 ps.
Displayed Risetime and Actual Risetime
Differences in displayed and actual signal risetime will
result unless the risetime of the sampling system is about
four times faster than the risetime of the signal to be
measured. As used in this discussion, the sampling system
is considered to be made up of the sampling head and any
coaxial line, connectors or accessories between the signal
source (the Type S-50) and the sampling head. Displayed
risetime is dependent upon the risetime of the sampling
head, the risetime of the pulse generator, and the risetime
of the cable and connecting components between the gener-
®

Reference L ane
point 2 4 [ tl 35 ps
h0O%
Level
~--}--4-San5-
[|__|
90% a }. mm
oT At 35 pse} | »
HI P a Zone
HD + | center
I t/ I
Le Trt
T tht ptt
Zone { an H
center ! an |
iy L
[+35 Pst Vv tT
t t gm 10 Yl
|
AI
ecto
panto
n
ap
==
0%
Level
| a, !
y_100% 4 <x, Reference
“zone point 1
| 35 ps | {| 4
Fig.
2-7.
Special
$-4/S-50
system
risetime
measurement
technique.
ator and sampling head. With signals as fast as about
100 ps arriving at the Type S-4 input, the displayed risetime
will be very close to the actual signal risetime.
Since the risetime of the Type S-50 Pulse Generator output
signal is much faster than 100 ps, the displayed risetime
(with this pulse applied to the Type S-4 input} will not be
the actual signal risetime.
The risetime of signals that are faster than 100 ps can
still
be
determined
using
the
Tektronix
Type
S-4
Sampling
Head. However, the signal risetime is calculated using for-
mula
(2-2)
below.
Signal
risetime
calculated
using
formula
(2-2)
will
provide
a
close
approximation
of
the
actual
signal
risetime.
The relationship between displayed risetime, signal rise-
time, signal path risetime, and sampling head risetime is
shown
by
formula
(2-1)
below.
What
was
referred
to
as
system
risetime
has
been
broken
down
into
T,
of
signal
path,
and
T,
of
Type
S-4
in
formula
(2-1).
T,
of
signal
path
includes the combined risetime of all coaxial cables, con-
nectors,
or
accessories
between
the
signal
source
and
the
Type
S-4
input.
If
the
signal
path
consists
of
a
coaxial
cable
having
a
risetime
of
8
ps
and
connectors
and
acces-
sories
having
a
risetime
of
6ps,
the
combined
signal
path
risetime is
(8)? + (6) = 10 ps.
T, (displayed) ~
VT,
signal)?
+
(T,
of
sig.
path}?
+
(T,
of
Type
S-4)?
=
(2-1)
For
example,
assuming
an
input
signal
having
a
risetime
of
25
ps;
connecting
cable,
accessories,
an
dconnectors
hav-
ing
a
risetime
of
10
ps;
and
a
sampling
head
with
a
rise-
time
of
25
ps
gives:
T, (displayed) = J (25)? + (10)? + (25)?
T, (displayed) = 625 + 100 + 625
T, (displayed) = 37 ps
Operating Instructions—Type S-4
Transposing
formula
2-1
to
permit
calculation
of
signal
risetime
when
the
dispalyed
risetime,
signal
path
risetime,
and
sampling
head
risetime
is
known
gives:
T, (signal) =
VT,
displayed}?
—
(T,
of
Type
S-4)?
—
(T,
of
sig.
path)?
(2-2)
Additional
information
concerning
the
effect
of
connect-
ing
cables
on
the
displayed
signal
risetime
may
be
found
under
Cable
and
Accessory
Considerations
later
in
this
section.
Sine Wave Signal Measurements
The
parameters
of
the
Type
$-4
are
specified
using
pulse
or
time
domain
parameters
and
nomenclature.
The
fre-
quency
response
characteristics
of
the
unit,
while
not
speci-
fied,
are
such
that
it
may
be
used
for
relative
sine
wave
measurements
through
its
calculated
upper
3
dB
rolloff
point
of
14
GHz.
Because
of
the
traveling
wave
gate
used
in
the
Type
S-4,
it
is
capable
of
handling
very large
signals
as
compared
to
other
microwave
samplers.
Even
so,
the
peak
to
peak
sig-
nal
amplitude
should
be
held
to
as
low
a
level
as
possible
consistent
with
good
signal
to
noise
ratio
when
making
checks
in
the
5
to
15
GHz
region
to
avoid
compression
effects.
Cable
and
Accessory
Considerations
When
connecting
a
signal
to
the
Type
S-4
it
is
important
that
the
signal
coupling
cable,
connectors,
attenuators
and
any
other
devices
between
the
signal
source
and
sampling
head
have
the
correct
characteristic
impedance,
be
capable
of
handling
the
signal
frequency
without
causing
excessive
loss
or
distortion,
and
be
of
high
quality.
Care
must
be
taken
to
firmly
tighten
all
connectors,
or
unwanted
reflec-
tions
may
result.
Accurate
displays
of
signals
containing
very
fast
transi-
tions
or
very
high
frequencies
are
possible
only
if
the
cables
carrying
the
signal
to
the
sampling
head
do
not
distort
the
signal.
The
Type
S-4
input
resistance
is
502
and
requires
the
use
of
high
quality,
low
loss,
50
coaxial
cables
and
attenuators
between
the
signal
source
and
the
sampling
head
input
connector.
Using
an
adapter
to
change
from
3
mm
to
a
larger
diamter
coaxial
cable
will
resutl
in
a
loss
of
system
bandwidth.
The
Type
S-4
sampling
head
is
useful
from
DC
up
to
at
least
14GHz.
Standard
and
optional
3mm
_
accessories
available
from
Tektronix
are
intended
for
use
at
frequencies
to
18
GHz
and
are
generally
useful
at
frequencies
to
25
GHz.
Since
any
accessories
in
the
path
between
the
signal
source
and
the
Type
$-4
input
will
increase
the
displayed
signal
risetime,
use
only
high
quality
accessories
of
ade-
quate
high
frequency
capability.
Accessories
(3mm)
supplied
with
the
Type
$-4,
or
available
as
optional
equipment,
in-
clude
flexible
and
semi-rigid
coaxial
cable,
atfenuators,
adapters
to
change
from
3
mm
to
larger
size
coaxial
cable,
and
a
power
divider.
See
your
Tektronix
Field
Engi-
neer
or
Representative
for
price
and
availability
of
optional
accessories.
The
physical
and
electrical
characteristics
of
the
cable
determine
the
characteristic
impedance,
velocity
of
propa-
gation
and
nature
of
signal
losses.
2-7

Operating Instructions—Type S-4
It is important to use cables or airlines, in the signal-han-
dling side of the system, that are as short as possible. Use
of the optional 3 foot or 6 foot head extender cable per-
its locating the Type S-4 near the signal source and thereby
keeps the input coaxial cable short and signal losses to a
minimum. Signal losses along a coaxial line, due to skin
effect or dielectric loss, result in a loss of high-frequency
signal amplitude as the signal travels along the line. Dielec-
tric loss along the coaxial line also results in an increase
in the displayed risetime of fast step pulses. Dielectric and
skin effect losses increase with signal frequency and the
length of the coaxial cable. Fig. 3-8A shows the displayed
waveform when the 25 ps output pulse of the Type S-50 is
fed to the Type S-4 input through a semi-rigid 3mm coaxial
line available as an optional accessory. In Fig. 2-8B the
25 ps pulse is fed to the S-4 input through a 5ns length of
flexible 3mm coaxial cable, Tektronix Part No. 015-1005-00.
Fig. 2-8C shows the risetime of the displayed signal when the
same 25 ps pulse is sent through a 5 ns length of good quality
coaxial cable similar to Type RG-58/U. The importance of
using a short length of the proper type coaxial between the
signal source and the Type S-4 input is apparent from com-
parison of the waveforms of Fig. 2-8.
Another factor that can result in loss of signal in a coax-
ial cable is use of coaxial cable having a diameter that is
large compared to the wavelength of the highest frequency
components of the signal. The Type S-4 input connector is
intended for use with Teflon dielectric coaxial cable having
an outer conductor with an outside diameter of 0.141 inches.
Resistive losses in large diameter coaxial cable is less
than in smaller diameter cable for frequencies below the
critical value.
CAUTION
The two lines listed under part numbers 015-1015-
00 and 015-1005-00 must be treated with care
to obtain maximum life and least signal distortion.
The semirigid line may have its shape changed just
twice in its lifetime, providing the minimum bend
radius is no less than 10 times its OD, or approxi-
mately 1% inches. The flexible (2ns delay)
cable must not be given a bend radius less than
10 times its OD, or approximately 24% inches.
(The same applies to a 5ns signal delay flexible
cable, Tektronix Part No. 015-1006-00.) The two
flexible cables may have their outer conductor
damaged by tighter bending. Few flexures assure
longer life.
Use of Probes
A Tektronix P6034 or P6035 probe may be used with
the Type S-4 Sampling Head. Connecting either of these
probes to the Type S-4 input requires using a GR to 3 mm
adapter. It should be kept in mind that the high frequency
capability of the Type S-4 will be limited to that of the
probe.
2-8
PEE
|
| | |
HEHEHE HEH
(C) 5ns length of cable similar to RG 58/U.
Fig. 2-8. Effect of cable length, diameter, and quality on dis-
played risetime.
®

Type S-4
SECTION 3
CIRCUIT
DESCRIPTION
Change
information,
if
any,
affecting
this
section
will
be
found
at
the
rear of the manual.
General
This section of the manual contains a block diagram
description
of
the
Type
S-4
Sampling
Head
followed
by
a
detailed
circuit
description.
Read
the
section
on
Basic
Tek-
tronix
Sampling
Principles,
located
in
the
associated
sam-
pling
unit
manual,
if
the
function
of
a
particular
Type
S-4
circuit
is
not
clear.
Both
the
block
and
schematic
diagrams
are located at the rear of this manual.
The
Type
S-4
is
the
signal
input
section
of
a
sampling
system,
and
determines
the
vertical
channel
input
character-
istics.
The
special
traveling-wave
gate
at
the
input
differs
from
the
sampling
bridge
described
in
the
Basic
Sampling
Principles
section
of
the
sampling
unit
manual.
BLOCK DIAGRAM
Both
the
Block
Diagram
and
the
circuit
schematic
show
the
Type
S-4
circuits
outlined
by
dashes,
indicating
the
indi-
vidual
circuit
boards
within
the
unit.
One
circuit
board
is
devoted
to
the
generation
of
the
sampling
gate
strobe
pulses,
one
to
the
sampling
gate,
one
to
the
blow-by
and
trigger
pickoff
amplifier
and
one
to
the
preamplifier.
Strobe Generator
The
Strobe
Generator
develops
heavy-current
(several
mA),
short-duration,
push-pull
pulses
that
drive
the
Sampling
Gate
into
balanced
conduction.
Output
occurs
at
the
time
of
each
sample
when
a
command
pulse
arrives
from
the
Strobe
Driver
circuit
of
the
associated
sampling
unit.
The
pulse
lasts
for
a
period
of
about
125
ps,
turning
on
the
Sampling
Gate
diodes.
At
the
end
of
the
pulse,
the
diodes
turn
off
very
quickly,
retaining
a
portion
of
the
input
signal.
Strobe
pulse
duration
and
fall
time
are
fixed
by
the
Snap-Off
diode
Dé61,
and
the
two
clipping
lines.
The
Strobe
Genera-
tor
parts
layout
and
circuit
board
construction
are
carefully
controlled
during
manufacture
to
ensure
proper
strobe
pulses
being
sent
to
the
Sampling
Gate.
Sampling Gate
The
Sampling
Gate
(called
the
Sampling
Bridge
in
other
Tektronix
sampling
circuitry)
connects
the
input
signal
to
the
Preamplifier
only
during
the
short
time
when
each
sample
is
taken.
The
six
diodes,
D2A
through
D2F,
form
what
is
technically
termed
a
traveling
wave
gate.
It is
the
way
the
end
of
the
Strobe
Drive
pulse
travels
through
the
gate
that
determines
the
step
response
of
the
sampler.
In
other
Tek-
tronix
sampling
bridges,
the
step
response
risetime
is
con-
trolled
by
the
strobe
pulse
duration.
In
the
Type
S-4,
the
step
response
risetime
is
controlled
instead
by
the
length
of
time
@®
it
takes
the
strobe
pulse
end
to
travel
through
part
of
the
Sampling
Gate.
At
the
end
of
each
Strobe
pulse,
part
of
the
input
signal
is
stored
temporarily
between
Sampling
Gate
diodes
and
is
then
fed
to
the
Preamplifier
input
at
a
rate
much
slower
than
the
step
response
risetime.
It
is
only
the
special
travel-
ing
wave
gate
that
has
its
electrical
environment
controlled
for
minimum
reflections
of
fast
pulse
signals.
All
circuit
parts
that
pass
the
sampled
error
signal
(see
the
sampling
unit
manual
section
on
Basic
Sampling
Principles
for
defini-
tion
of
error
signal)
to
the
Preamplifier
handle
only
moderate
rate-of-rise
signals.
A
special
10
X
(20db)
500
attenuator
helps serve
as
a
high-quality,
high-frequency
termination
to
the
input
connector
and
the
50
Q
environment
of
the
travel-
ing
wave
gate.
Part
of
the
input
signal
is
continuously
passed
to
the
Trig-
ger
Amplifier
Board
and
the
Blow-by
compensating
and
Trigger
Pickoff
circuit.
The
signal
passes
through
the
10
X
attenuator
directly
to
the
Trigger
Amplifier
Board
and
is
terminated
in
50
ohms
there.
Gate Bias
Reverse
bias
is
applied
to
the
Sampling
Gate
diodes
by
the
Gate
Bias
circuit.
The
average
voltage
of
the
gate
bias
is
controlled
by
the
Gate
Balance
circuit
and
the
associated
sampling
unit
DC
Offset
and
Feedback
signals.
The
Sam-
pling
Gate
output
is
DC
coupled
to
the
Preamplifier
input
through
a
portion
of
the
Gate
Bias
circuit.
Blow-by and Trigger Pickoff
The
primary
function
of
the
Blow-by
and
Trigger
Pickoff
circuit
is
to
cancel
capacitively-coupled
unwanted
signals
that
normally
bypass
the
Sampling
Gate.
These
unwanted
signals
are
called
“blow-by’’
signals.
The
Type
S-4
special
traveling
wave
Sampling
Gate
con-
tains
two
blow-by
reducing
diodes
that
minimize
the
ca-
pacitively-coupled
unwanted
signals.
The
special
Sampling
Gate
construction
and
the
blow-by
circuit
cancel
essentially
all
of
the
unwanted
signals
that
would
otherwise
distort
the
oscilloscope
sampling
display.
The
Blow-by
and
Trigger
Pickoff
circuit
receives
an
atten-
vated
portion
of
the
input
signal
and
terminates
the
10
X
attenuator
that
is
located
between
the
Sampling
Gate
and
the
Trigger
Amplifier
Board.
The
circuit
amplifies
and
inverts
the
signal
and
applies
it,
as
a
blow-by
correction
signal,
through
a
small
capacitor
to
the
output
side
of
the
Sam-
pling
Gate
blow-by
limiting
diodes.
Magnitude
of
the
blow-
by
correction
signal
is
adjusted
during
calibration
by
the
Transient
Response
control.
3-1

Circuit Description—Type S-4
The trigger pickoff function of the Blow-by and Trigger
circuit provides an in-phase signal source for internally trig-
gering the sampling sweep unit. Not all channels of all
sampling units [in which the S-4 operates) couple the trigger
pickoff signal to the sampling sweep unit. If the signal is
not used, the sampling unit terminates the output in 50 ©,
allowing the circuit to remain functional for blow-by correc-
tion.
Preamplifier
The Preamplifier circuit both amplifies and time-stretches
the signal it receives from the Sampling Gate. The signal
received is a portion of the difference between the Feedback
combined with the DC Offset voltage and the input signal.
This “error signal” is amplified and AC coupled to the Post
Amplifier in the sampling unit. The Preamplifier gain is ad-
justable to aid in setting the overall sampling head and
sampling unit “loop” gain to unity for proper dot response.
CIRCUIT DESCRIPTION
Refer to the main schematic diagram during the following
descriptions. The Type S-4 Sampling Head uses the power
supplies of the indicator oscilloscope and associated sam-
pling unit. Interconnections to the sampling unit circuits are
by two connectors at the sampling head rear. The following
description includes references to circuits in the sampling
unit and sampling sweep unit, ail units forming one sam-
pling system. Reference to diagrams and circuit descriptions
in the instruction manuals for the other sampling units will
help in gaining a full understanding of the circuit relation-
ships.
Strobe Generator
The Strobe Generator circuits are all located on the (right
side) Strobe circuit board. The generator contains two basic
circuits, an Avalanche circuit that delivers fast push-pull
pulses, and the Snap-off diode and clipping lines circuit.
Both circuits work together to produce the push-pull strobe
pulses that drive the special traveling-wave Sampling Gate
through two equal transmission lines. Both sets of circuitry
are physically arranged over separate ground planes that
are identified on the main schematic diagram. The two
ground planes are connected together by R52 and R54, to
damp out ringing between the two circuit areas at sample
time.
The Avalanche Circuit converts the Strobe Drive
pulses from the sampling unit to very fast push-pull pulses
that drive the Snap-off diode to non-conduction.
The Strobe Drive pulse is transformer-coupled by T75 to
the base and emitter of Avalanche transistor Q69. Two out-
puts are AC-coupled from Qé69, one from the collector and
the other from the emitter. The Avalanche Volts control ad-
justs the collector voltage of avalanche transistor Q69. The
typical quiescent voltage at Qé69 collector is about +15
volts. This voltage sets the amplitude of the signals that
drive the Snap-off diode circuit and assures the normal ava-
lanche action of Q69 when driven by the Strobe Drive sig-
nal. Before Avalanche conduction, there is a potential of
about 60 volts between Q69 collector and emitter.
3-2
The negative Strobe Drive pulse is transformed into a
hard forward bias signal to Q69. Normal avalanche action
follows with the collector going negative and the emitter
going positive. This fast-rise, push-pull signal is capacitively
coupled to the Snap-off circuit.
The Snap-off Circuit operates as a current switching
circuit to apply some of the push-pull Avalanche current sig-
nal at snap-off time to the Sampling Gate. The circuit con-
sists of a Snap-off Current control R57, Q55, Snap-off diode
Dé1, two clipping lines and associated components. Be-
tween drive pulses from the Avalanche circuit, the Snap-off
diode D61 is forward-biased by the current in @Q55. The
current value is set by the Snap-off Current control R57. The
current in Dé1 is typically 20mA. This forward current
assures that D6] has many carriers within its junction region.
The push-pull signals from the Avalanche circuit cause D6]
junction carriers to reverse direction as a heavy reverse cur-
rent. This heavy reverse current stops suddenly as all the
carriers clean out of the junction. As the reverse current
“snaps to a stop, the push-pull avalanche signals are
suddenly coupled into the clipping lines by C52 and C54
and toward the Sampling Gate by R51-C51 and R53-C53.
The fast-rise step which appears at each clipping line in-
put is propagated down the line. About 50 ps later the
steps reach the short circuited ends of each clipping line.
The step is then reflected, equal in amplitude and opposite
in polarity, back to the input end of each clipping line. This
cancels the signals moving toward the Sampling Gate. This
action results in a positive Strobe pulse being delivered to
P53, and a negative Strobe pulse being delivered to P51.
The clipping lines have a surge impedance of approxt-
mately 10 Q with a very good short circuit at the shorted
ends. The high quality of the short, and the line low imped-
ance causes the reflected signal to stop the Strobe pulses
very rapidly. Such rapid ending of the Strobe pulse and the
characteristics of the traveling wave Sampling Gate causes
each sampling diode to turn off in approximately 10 ps.
The fast diode turn-off is discussed below in the Sampling
Gate description.
Ringing of the Snap-off diode and the clipping Sines is
very rapidly. Such rapid ending of the Strobe pulse and the
the lines ends. Ringing is also limited to some degree by
the two equal resistors R60 and R62 that are in DC series
connection with the Snap-off diode Dé1. The primary func-
tion of R60-R62, however, is to provide a DC path for Dél
forward conduction while at the same time allowing the
Avalanche signals to pass on into the clipping lines and
Sampling Gate. As the Avalanche pulses end and as both
the Avalanche and Snap-off circuits recover for another
sample, R60-R62 re-establish the correct quiescent charges
on C63 and Cé4. C59 assures a very low impedance AC
path for the ground side of R60 so that R60 and R62 will
both have the same affect upon the circuit.
Another technique for minimizing Strobe ringing is that of
R51 and R53 being in series with the output lines to the Sam-
pling Gate. The output connectors that connect the Strobe
board to the Sampler board have a 50 Q characteristic
impedance. R51 and R53 raise the Clipping lines 10 Q im-
pedance drive to 50 © to properly match the connections.
Thus, at the end of the Strobe pulse, reflections back from
the Sampling Gate are reverse terminated. These resistors
and the resistors discussed in the preceeding paragraph all
®

assure that there is no possibility of double strobing of the
Sampling Gate due to ringing of the Strobe Generator
circuits.
Two separate ground planes are identified on the main
schematic diagram between the Avalanche and Snap-off
circuits. Two resistors connect the two ground plane ends
together to damp any natural ringing when the case is off.
Ringing can occur without R52 and R54 because the two
ground planes are connected together at the instrument
front and rear through the Sampler board. Rake-type
ground symbols (USA Standard ASA Y 32.2, 1967) on the
diagram identify chassis ground return circuits connected to
the front panel or case. Triangle-type ground symbols are
used throughout the Sampling Gate area, Strobe Generator
and Trigger board areas to indicate these circuits all share
a common grounding system with the signal input connector,
J1. Component layout is organized so the strobe signal
ground currents balance to zero at the center of the Sam-
pling Gate, and thus do not contribute unwanted currents
to the input signal. Q69 circuit is capacitively isolated from
both ground planes as much as possible, with its primary
signal output path being through Cé63-Cé4 and C52-C54.
There is a small third terminal capacitance to ground be-
twean the two windings of T75, but it is not large enough
to disturb the balanced drive from Q69 to Dé].
Sampling Gate and Gate Bias Circuits
The Sampling Gate consists of six series diodes, D2 A
through F, and four resistors all mounted on a ceramic sub-
strate inside the input circuit holder. See Fig. 3-1 during the
following dicussion.
The signal that arrives at J] travels right on through the
Sampling Gate area fo a 3 mm 50 10X (20 db} attenuator.
Ouput of the attenuator feeds the Trigger and Blow-by Com-
pensation circuit where it is terminated in 50 2. When not
being strobed, D2C and D2D present a very small value of
capacitance to the input signal. When the diodes conduct
due to Strobe current, the input line is coupled to the travel-
ing wave gate.
Gate Biasing
Quiescent condition of the Sampling Gate diodes is con-
trolled by the Gate Bias circuit. D2A, D2B, D2E and D2F are
each reverse biased about 0.5 volts by D20, D23, D25 and
D28. D2C and D2D are reverse biased about 1.2 volts each,
a total of about 2.45V is developed across the parallel
combination of R24-C25 of the Gate Bias circuit. The Gate
Bias circuit receives its current from the two 50-V supplies
and R30-R31-R32-R33 and R34 of the Gate Balance control
circuit. The diodes of the Gate Bias circuit are used as 0.5-
volt Zener diodes. R2A-R2F with C24 and R2B-R2E with C25
isolate the traveling wave gate segments from the Preamp
and Bias circuits R21-R26 conduct DC bias potentials to the
gate but allow a sampled error signal to drive C24 and the
Preamplifier input.
Except for the very short strobed conduction time (about
125 ps), the Sampling Gate diodes do not conduct. A
portion of the input signal is coupled through stray capaci-
®
Circuit Description—Type 5-4
tances to the Preamplifier input, and disturbs the normal
sampling display. This effect is known as blow-by”.
Blow-by Correction
The basic risetime of the traveling wave Sampling Gate
is 25 ps. In this extremely short interval of time, only a very
small amount of charge can be collected in one sample.
What small charge is collected, is called the error signal
and is amplified by the Preamplifier and the sampling unit
circuits. The Preamplifier has a shaped frequency response
best suited to amplifying each samples small charge and
delivering a proper error signal to the sampling unit.
It is impossible to build a sampling gate without a path
for blow-by displacement currents that cause the Preampli-
fier to respond to a part of some input signals and thus
distort the display. The Type S-4 Preamplifier responds to
signals in the 0.5 to 3 MHz bandpass region. Signals much
higher or lower in frequency are not amplified. Those blow-
by signals in the Preamplifier bandpass region that do get
through the Sampling Gate are virtually eliminated by the
Blow-By correction amplifier Q10. The traveling wave gate
is constructed by double differentiate signals in the Pre-
amplifier response domain while not affecting the small
sampled charge quantities obtained during each sample
interval.
Blow-by rejection in the Sampling Gate is assured by the
double differentiation networks that have very fast RC times.
They are located between the signal input and the Preampli-
fier, The first differentiation network is formed by reverse
biased diodes D2C and D2D and their respective resistors
to ground (see Fig. 3-1}, R2B and R2E. The second network is
formed by diodes and resistors D2B, D2E and R21, R26,
respectively. All the Sampling Gate diodes are normally
non-conducting and are therefore very small capacitors. The
differentiating networks eliminate all but about 5% of the
possible blow-by effects. The Blow-By correction amplifier
feeds an inverted replica of the input signal into the Sam-
pling Gate output, a small quantity of charge passes
through C19 to cancel most of the remaining blow-by signal.
Traveling Wave Sampling Gate
Push-pull strobe pulses from the Strobe Generator cause
the six Sampling Gate diodes to conduct for about 125 ps.
While the diodes are conducting, the signal at the input con-
nector propagates down the diode transmission paths as
well as toward the 10X attenuator, R3. As the fast falling
strobe pulse end begins to propagate into the diode trans-
mission paths, the diodes are quickly switched off (into
reverse bias}, one set after another. First, diodes D2A and
D2F turn off, then D2B and D2E, and finally D2C and D2D.
Turn off time of the diode sets is controlled by the mechan-
ical spacing and thereby the propagation time between
sets. The Sampling Gate risetime is determined by these
propagation times, and is thus not heavily influenced by
the strobe pulse amplitude or duration. The sampling pro-
cess is thus one of trapping signal charge between sets of
diodes, in particular, between set D2A-D2F and the set D2B
and D2E. The risetime is very close to the double transit time
between sets.
3-3

Circuit Description—Type S-4
+50V
SAMPLING
GATE
GATE
BIAS
INPUT L
CIRCUIT
PARTIAL
SAMPLER
BOARD
PREAMP
BOARD
T
C19
R30
CHIP &
HOLDER
R31
dv
l
PREAMP —<
~ | ~5 pF
| Strays
I
rh,
Feedback
R32 3 <_<
GATE
BAL
| STROBE
R33
R34
LA
l 2
—50V
GENERATOR
Fig.
3-1.
Traveling
wave
Sampling
Gate
and
Gate
Bias
circuit
components,
with
input
signal
blow-by
differentiating
paths
shown
by
dotted
lines.
Once the signal charge trapping process has taken place
{at
the
strobe
pulse
end),
the
Preamplifier
begins
to
receive
the
trapped
charge
by
conduction
through
resistors
R2A
and
R2F
(See
Fig.
3-2).
The
Preamplifier
input
capacitance
is
much
larger
than
the
capacitance
of
the
transmission
line
segments
between
the
diode
sets,
causing
practically
all
of
the
trapped
charge
to
finally
transfer
to
the
Preamplifier
input.
The
transfer
time
constant
is
approximately
10
nano-
seconds.
Because
of
the
carefully
adjusted
balance
of
the
plus
and
3-4
minus
strobe
signals
(set
by
the
physical
position
of
R51
and
R53
in
the
Strobe
Generator},
the
Preamplifier
receives
only
the
trapped
signal
charge
and
nothing
from
the
Strobe
Gen-
erator.
Preamplifier
The
Preamplifier
circuit
amplifies
and
time-stretches
the
error
signal
pulse
from
the
Sampling
Gate,
and
AC
couples
it
to
the
Post
Amplifier
in
the
associated
plug-in
unit.

502
— Strobe
source
\ Daa
N
—_—
502
Zo
—
/\- — — ——
502
+ Strobe
source
Fig.
3-2.
Error-signal
source
and
path
to
Preamp
shown
by
dotted
lines.
Input
transistor
Q36
operates
as
a
very high
input
imped-
ance,
high-gain
inverting
amplifier.
Temperature
compensa-
tion
for
Q36
is
accomplished
by
thermistor
R33
in
parallel
with
R39.
A
total
of
4.4mA
channel
current
passes
from
the
+50
volt
supply
through
R36,
Q36,
the
parallel
combina-
tion
of
R33-R39
and
R40
to
the
—50
volt
supply.
C39
assures
that
Q36
AC
gain
is
high,
while
its
DC
gain
is
less
than
1.
Q43
and
Q46
are
connected
as
an
operational
amplifier
with
a
very
low
output
impedance
at
Q46
emitter.
The
out-
put
is
coupled
by
C50
to
the
90
Q
input
resistance
Post
Ampli-
fier
in
the
sampling
unit.
D41
protects
Q43
base
from
high
negative
voltage
if
Q46
is
removed
from
its
socket.
Current
in
R40
with
DC
negative
feedback
by
R48
sets
the
output
DC
voltage
level
of
the
amplifier
at
about
7.3
volts.
Negative
AC
feedback
from
the
emitter
of
Q46
through
C49
and
Gain
control
R49
to
Q43
base
controls
the
AC
gain
of
the
Preamplifier.
Gain
control R49
allows
the
AC
feedback
to
be
adjusted,
thereby
adjusting
the
gain.
The
AC
gain
is
adjusted
during
calibration
so
that
the
sampling
head
has
the
correct
amplitude
output
signal
to
the
sampling
unit
Post
Amplifier.
The
Type
S-4
sampling
efficiency
is
less
than
2%.
(See
the
sampling
unit
instruction
manual
Basic
Sampling
Principles
for
sampling
efficiency
definition.)
With
a
low
samplng
efficiency,
the
Preamplifier
positive
and
negative
signal
out-
Circuit Description—Type 5-4
put
impedance
must
both
be
low.
@46
assures
a
low
output
impedance
for
positive-going
output
pulses,
and
D43
assures
a
low
output
impedance
for
negative-going
output
pulses.
D43
conducts
only
for
large
negative
output
signals
that
fall
at
a
rate
faster
than
Q46
emitter
can
follow.
Q46
emit-
ter
remains
a
low
output
impedance
for
small
negative
out-
put
signals.
Components
not
already
mentioned
above
include:
C47
and
R46
that
both
prevent
parasitic
oscillations
of
Q46,
and
C45-C46
that
decouple
the
+15-volt
supply
for
both
low
and
high
frequency
current
pulses
of
Q46.
Blow-by and Trigger Pickoff
The
Blow-by
and Trigger
Pickoff
circuits
consist
of
QO,
Q14
and
associated
components
connected
as
a
common-
emitter
paraphase
amplifier.
The
primary
purpose
of
the
circuit
is
to
cancel
unwanted
high
frequency
capacitively-
coupled
signals
that
bypass
the
Sampling
Gate.
The
secon-
dary
purpose
is
to
provide
a
trigger
pickoff
signal
to
the
sampling
unit
for
internal
triggering
of
the
sampling
sweep
unit.
Input
signal
to
the
Blow-by
and
Trigger
Pickoff
circuit
is
a
portion
of
the
Type
$-4
input
signal.
The
signal
is
fed
to
the
base
of
Q10
through
R3.
R10
terminates
the
509
im-
pedance
of
the
input
and
assures
no
reflections
back
into
the
input
circuit.
The
output
from
QI0
collector
feeds
an
AC
coupled
signal
to
the
output
side
of
the
Sampling
Gate,
cancelling
the
blow-by
signal.
The
output
from
Ql4
collec-
tor
feeds
a
DC
coupled
trigger
pickoff
signal
to
the
sam-
pling
unit
for
use
by
the
sampling
sweep
unit.
Feedback Limiting
The
Preamplifier
input
is
the
input
terminal
of
an
integrat-
ing
pulse
amplifier
that
includes
the
sampling
unit
Post
Amplifier,
AC
Amplifier,
Memory
Gate
and
Memory
Ampli-
fier.
Feedback
from
the
Memory
Amplifier
output
is
DC
coupled
to
the
Type
S-4
Preamplifier
input
through
the
sampling
unit
Feedback
Attenuator
and
the
Type
S-4
Gate
Bal
control.
As
a
sample
is
taken,
the
error
signal
is
amplified
and
converted
to
a
DC
signal
for
both
the
CRT
and
the
feed-
back
that
arrives
through
pins
2
and
B
of
the
sampler
board.
The
maximum
feedback
is
limited
to
approximately
+1.2
volts
by
two
resistive
dividers
and
D80
and
D82.
The
limiting
is
to
prevent
excessive
feedback
voltage
from
reaching
the
Sampling
Gate
when
the
sampling
unit
Units/
Div
switch
is
changed
between
positions.
In
certain
condi-
tions
of
operation,
the
excessive
feedback
can
cause
the
Sampling
Gate
to
go
into
a
mode
of
conduction
that
holds
the
CRT
beam
off
screen.
By
limiting
the
feedback
ampli-
tude,
overdriven
displays
will
leave
the
CRT
screen
but
will
return
just
as
soon
as
the
overdrive
is
removed,
Without
the
limiting,
an
overdrive
signal
could
cause
the
display
to
disappear.
3-5

NOTES