Tektronix TYPE 109 User manual

I NSTRUCTION
MANUAL
TYPE 109
PULSE GENERATOR
------------------------------------------- S ’A ' - , f ' / 7
----------
Tektronix, Inc.
S.W . M illikan W a y • P. O . Box 500 • Beaverton, O regon • Phone Ml 4-0161 • Ca b le s: Tektronix
Tektronix International A.G .
Terrassenw eg 1A • Zu g , Sw itzerlan d • P H .0 4 2 -4 9 1 9 2 • Cab le : Tekin tag, Zu g Sw itze rland • Telex 53 .5 74
0 7 0 -2 9 9 563

WARRANTY
A ll Tektronix instruments are warranted
against defective materials and w orkm an
ship for one year. Tektronix transformers,
manufactured in our own plant, are w a r
ranted for the life of the instrument.
Any questions with respect to the w a r
ranty mentioned above should be taken up
with your Tektronix Field Engineer.
Tektronix repair and replacement-part
service is geared directly to the field, there
fore all requests for repairs and replace
ment parts should be directed to the Tek
tronix Field Office or Representative in your
are a. This procedure w ill assure you the
fastest possible service. Please include the
instrument Type and Serial number with all
requests for parts or service.
Specifications and price change priv
ileges reserved.
Copyright jjP 11963 by Tektronix, Inc.,
Beaverton, Oregon. Printed in the United
States of Am erica. All rights reserved.
Contents of this publication may not be re
produced in any form without permission
of the copyright owner.

CONTENTS
Section 1
W arranty
Characteristics
Section 2 Operating Instructions
Section 3 Applications
Section 4 Circuit Description
Section 5 Maintenance
Section 6 Calibration
Section 7 Parts List and Diagram

PULSE POLARITY
- ' T H *
WER flU R N OFF IO EXTEND
MERCURY SWITCH UFE
PULSE GENERATOR
{ RISETIME < 0,25 NANOSEC )
EXT. POWER OR MONITOR
AMPLITUDE
25
50 n
5G/V
CHG.
UNS
1
0,5 EXT.
PWR.
Type 109

SECTION 1
General Information
The Tektronix Type 109 Pulse Generator is a fast-risetime
pulse generator similar to the pulse generator section of the
Type 110. The Type 109 is capable of producing pulses of
different widths, calibrated-amplitudes, and polarities for
use in driving and testing the response of devices operating
in the nanosecond region.
An external network, supplied with the Type 109, provides
long duration pulses with an amplitude decay of only 10%
in 400 nanoseconds. This network is useful for testing ampli
fier linearity or tuning delay lines.
Pulse Amplitude
Three calibrated ranges: 0 to 0.5 v, 0 to 5 v, 0 to 50 v.
Accuracy is within 3% of the front-panel full-scale mark
ing. Each range is continuously variable.
± 5 0 volts is the maximum calibrated amplitude using the
internal power supply; ± 3 00 volts maximum allowed
using an external supply and either or both contacts.
Polarity
Positive or negative.
CHARACTERISTICS
Vertical Deflection Factor: 200 mv/cm
Equivalent Sweep Rate: 0.2 nsec/cm
Charge Line: One 10-nsec 50-ohm RG-8A/U
cable connected between 5012 CHG. LINE
1 and 50 ß CHG. LINE 2 connectors.
Risetime
Less than 0.25 nsec (nanosecond or 10“9 second). Pulse
risetime is illustrated in Figs. 1-1 and 1-2.
Fig. 1 -2 . W a v e form show ing the Typ e 1 0 9 pulse d isp lay e d on a
Te k tron ix Type 661 (S am plin g ) O scilloscop e . Combined risetim e
of the system , betw een 1 0 % and 9 0 % am p litude le ve ls , is less
than 0 .4 nanoseco nds.
RISETIM E LESS TH A N 0 .2 5 nsec
1 nsec
Fig. 1 -1. A d ouble e xp osure p h o tograp h o f the output pulse from
the Type 1 09 (no e xte rn a l ch arg e lin e ) an d a 1 g ig ac ycle /s e c
timing tra in . The w avefo rm s a re d isp lay ed on a T e ktro nix 0 .12 -ns e c
risetim e research -ty pe oscillo scope. This photo graph sh ow s the
risetim e to be w e ll u n de r 0 .2 5 nsec. The minimum pulse w idth is
a p p ro x im a te ly 0 5 nsec. Note the freedom from o vershoot.
Duration
From approximately 0.5 nsec measured at the 50% ampli
tude level (see Figs. 1-1 and 1-3), to a maximum of 100
nsec at a repetition rate between 550 to 720 cycles using
a single charge line to both contacts; 300 nsec at half
repetition rate by using either one open-ended charge
line with the unused contact grounded or by using two
separate open-ended charge lines.
An accessory, a special external charge network sup
plied with the Type 109, provides a fast-rise pulse with an
RC decay (see Fig. 1-4) and a pulse amplitude which is
about one-tenth that indicated by the front-panel mark
ing. Decay is only 10% in 400 nsec (see Fig. 1-5). Fig.
1-6 shows the pulse waveform at an equivalent sweep
rate of 1 nsec/cm.
®1-1

Characteristics— Type 109
Vertical Deflection Factor: 200 mv/cm
Equivalent Sweep Rate: 1 nsec/cm
Charge Line: None
1
f"
t
t
—
Vertical Deflection Factor: 200 mv/cm
Equivalent Sweep Rate: 50 nsec/cm
Fig . 1-3. M inim um p ulse -w idth w ave form a s d isp lay e d on a Type
661 (S a m p lin g ) O scillo sco pe.
Fig . 1 -5 . W avefo rm obtained using the sam e charge netw o rk as
described in F ig . 1-4 but the eq uiva le nt sweep ra te o f the Type
661 is incre ased to 5 0 n sec/cm . This p ho tograp h show s that the
am plitu d e (o r R C ) dec a y is less th an 1 0 % in 4 0 0 nsec.
\
_ \
T 1
i
i
i
i
i
-i—
i
4—1
H—
1
- 1 —
5RC
Vertical Deflection Factor: 200 mv/cm
Equivalent Sweep Rate: 5 /zsec/cm
(Sampling) Oscilloscope: Type 661
Fig 1 -6 . Sam e conditions as Figs. 1-4 an d 1-5 e xce p t that the
eq uiv a len t sw eep rate is 1 nsec/cm .
Fig. 1 -4 . W a v e form produced by using the e xte rn al cha rge n e t
w ork supp lie d as a sta n da rd ac c e ssory. The netw ork connects to
one of the 5 0 12 C H G . LIN E connectors on the T ype 1 09 and
grounds the unused connector.
Repetition Rate
Infernally adjustable for proper operation at a rate
between 550 to 720 pulses per second.
Alternate Pulses
Unequal charge lines produce alternate pulses of differ
ent time durations but of the same amplitude and polarity
using the Type 109 internal power supply as shown in
Fig. 1-7.
External charge voltages permit alternate pulses of
different amplitudes and/or polarity to be generated as
shown in Figs. 1-8 and 1-9.
1-2

Characteristics— Type 109
'
sTort Pulse : l<
>ng P j l s e
l 111
5(
Amp
>%
litude •»*
L<vel
Vertical Deflection Factor: 200 mv/cm
Equivalent Sweep Rate: 5 nsec/cm
Charge Lines: One open-ended 9-nsec
cable connected to 50 12 CHG. LINE 1
connector and one open-ended 20-
nsec cable connected to 5 0 12 CHG.
LINE 2 connector.
Fig. 1 -7. Sin g le-e x p o s u re p ho togra p h illu stratin g the a lte rn ate
pulse fe a tu re , u sing the in tern a l p o w er of the Type 1 0 9 .
V ertica l D eflectio n Fa c to r: 9 .5 v/cm
Sw e e p R ate : 5 nsec/cm
(R e a l-T im e ) O scillo sc op e : Typ e 5 1 9
Fig. 1 -8 . Sin gle -e xp osu re p hoto g raph sh o win g a lte rn a te recurrin g
pulses of d iffe re nt w id ths an d am p litu des produced b y using 5-nsec
and 8 .8 -n se c ch arg e lin e s, and e xte rn a l ch arg e vo ltag es of the
sam e p o larity.
V e rtical D eflection Facto r: 9 .5 v/cm
Sweep R ate: 5 nsec/cm
(R eal-T im e) O scillo sco pe: Typ e 5 19
Fig. 1 -9 . D o uble-exposure pho togra p h shows th at s im ila r co nd i
tions as those in F ig . 1-8 w ere used e xcept the e xte rn al ch arge
vo ltag e s a re of op p o site p o la ritie s .
Output Impedance
50 ohms.
Power Requirements
Line Voltage— 105 to 125 volts, or 210 to 250 volts, 50
to 800 cycles.
Power—Approximately 50 watts at 117 volts line voltage.
Mechanical
Construction— Light-weight, shock-resistant aluminum alloy.
Finish— Photo-etched, anodized front panel. Textured-
aluminum cabinet with vinyl-based blue finish.
Dimensions— 7% " high, 5” wide, and 11’A ” deep overall.
Net Weight— 8 lbs. 7 ounces (Type 109 only).
Accessories
1— Power Cord, (161-010).
1—Three-wire to two-wire power cord adapter, (103-013).
3— 5-nsec cables, 50-Ohm GR RG-8A/U, 40" long, (017-
502).
1— Charge Network, (017-067).
2— Instruction Manuals, (070-299).
1-3

NOTES

SECTION 2
OPERATING
INSTRUCTIONS
AM PLITUD E EXT. PO W ER OR M O N IT O R
Ad ju sts pulse am p litude from
zero to the fu ll-sca le vo ltag e
set b y the V O LT A G E RA N G E
co ntrol.
V O LT A G E RA N GE
Selects full-scale am p litud e of the
pulse or sw itches to e xte rn al
p ow e r.
PULSE PO LA R ITY
Contro ls p o larity of the output
pulse using in tern al vo lta g es.
PULSE P O LARITY
PO W E R
Line po w er sw itch.
For ap ply ing e xte rn al pow er to
ch arge the lin e , or fo r monitoring
the charging w avefo rm on an
oscilloscope.
50 ß C H G. LIN E 1
50 ß C H G . LIN E 2
Connect ch arge lin e s here to ob 
ta in pulses of desired du ratio n .
50 ß OUTPUT
Output p ulse from the T ype 1 09 .
TEKTRONIX, INC., PORTLAND, OREGON, U.S .A.
Fig . 2 -1 . Functions o f the fro nt-p ane l co ntrols.
General Information
The Type 109 Pulse Generator produces fast rising pulses
at a repetition rate between 550 and 720 cycles per second.
The risetime of the pulses is less than 0.25 nanosecond, the
polarity can be selected, and both the amplitude and dura
tion are variable.
The Type 109 is intended for use with fast rise sampling
(equivalent-time) systems or conventional (real-time) oscillo
scopes.
The Type 109 is fully transistorized, except for a VR tube,
and requires no warmup time before operating. As soon
as the POWER ON switch is turned on, the Type 109 is
ready to operate.
Power Requirements
The Type 109 can be operated from either 117 or 234
volts nominal. The only changes necessary to convert from
one operating voltage to another are in the wiring of the
power transformer primary. The power transformer, T601,
has two separate primary windings. The windings are con
nected in parallel for 117-volt operation and in series for
234-volt operation.
A small metal tag located near the power receptacle at
the rear of the instrument indicates the nominal line voltage
for which the Type 109 was wired at the factory. If wired
for 117 volts, the instrument will operate properly with
®2-1

Operating Instructions— Type 109
Fig. 2 -2 . Po wer transfo rm e r connections for 1 1 7- an d 2 34 -v o lt o peratio n .
line voltages between 105 to 125 volts. If wired for 234
volts, the instrument will operate properly from 210 to 250
volts.
To change the power transformer connections for opera
tion on another line voltage, change the location of the
bare wire straps at the primary terminals. Fig. 2-2 shows the
wire-strap connections required for each nominal operating
voltage. When changing the location of the straps, it is
not necessary to move any of the plastic insulated wires.
After the transformer connections are changed, the voltage
indicated on the metal tag should be covered with another
tag which conforms to the new operating voltage.
Cooling
Before operating the Type 109, choose a location that
will not block the ventilating holes in the cabinet. Allow
at least two inches of clearance at the top and sides of the
instrument. With sufficient clearance, air can freely circulate
through the instrument and keep it cool.
If the temperature inside the instrument should become
excessively high, a thermal cutout in the instrument will
open, shutting off the power and the pilot light will go
out. When the interior temperature of the instrument drops
to normal, however, the cutout contacts will close and the
pilot light will light to indicate that power is restored.
Normal Operating Position
The upright position is the normal operating position
for the Type 109. If the instrument is placed on its side
or turned upside down, the mercury in the mercury switch
will flow to the contact end of the reed switch and short
the contacts. Such a short circuit does not harm the instru
ment, but it does prevent the Type 109 from generating any
output pulses.
Cabling Considerations
The Type 109 is designed for use with 50-ohm cables.
The charge lines must be 50-ohms for proper impedance
match. The output pulse must be applied through high-
quality 50-ohm cables or suitable impedance-matching
devices to keep losses down and maintain the waveform.
Use RG-8A/U for signal connections. If a signal delay cable
is needed, use the Type 113 Delay Cable. The only excep
tions are the cables used to supply external power to the
EXT. POWER OR MONITOR connectors to charge the lines.
The impedance of these cables is not critical, and virtually
any value can be used.
If proper signal cables are not used, reflections and atten
uations will occur which will produce undesirable side ef
fects. The GR 50-ohm cable connectors are used because
they easily connect together and they have reasonably
constant impedance.
Selecting the Pulse Duration
1. Using External Ch a rge Lines
Pulses from the Type 109 are generated by discharging
charged coaxial lines into a load through a solenoid-
operated mercury switch. The Type 109 uses two such
charged lines with the mercury switch discharging them
alternately. The charge lines must be connected externally
to the 50Q CHG. LINE 1 and 50Q CHG. LINE 2 connectors
on the front panel.
The physical length of the charge lines directly deter
mines the duration of the output pulses. The output pulse
duration is equal to twice the transit time of the charge
line used, plus a small built-in charge time due to the lead
length from the GR panel connectors to the mercury switch
contact point.
The transit time of the cable is defined as the time
required for a signal to pass from one end of the line to
the other. For a 10-nsec charge line then, the duration of
the output pulse would be 20 nanoseconds, plus about 0.5
nanosecond (minimum) due to the switch leads. Since two
charge lines are alternately discharged into the load, it is
possible to have alternate pulses with different time dura
tions by using charge lines of different lengths. Fig. 1-7 in
the Characteristics section demonstrates this feature.
It is also possible to have pulses of exactly the same
duration by using the same charge line. For this application,

Operating Instructions— Type 109
one end of the cable is connected to the 500 CHG. LINE
1 connector and the other end of the cable is connected to
the 500 CH G. LINE 2 connector. Since the same cable is
used to generate both pulses of a pair, all pulses have
exactly the same duration and amplitude. This mode of
operation results in an opposite polarity pip at the center
of the pulse, due to capacitive coupling between the switch
contacts of SW750. The pulse Ifength will be twice the
delay time of the charge line plus one nanosecond.
When using the same charge line between connectors,
maximum pulse duration at full amplitude is limited by
the amount of time that the reed of the mercury switch
remains between contacts. This is the open-contact time.
To assure that the charge line has sufficient time to fully
charge during this time, the length of the charge line should
not exceed 50 nanoseconds. Therefore, to generate pulses at
their full amplitude, maximum pulse duration is limited to
about 100 nanoseconds. If longer duration pulses are gener
ated by using a longer charge line, the pulses may not
reach full amplitude. In addition, each pair of pulses gen
erated may not equal each other in amplitude due to
variations in open-contact times.
To generate longer pulses at full repetition rate, two
separate charge lines can be used. As an alternate method,
a single unterminated charge line can be used if you
disable the other line-charging network. To disable the
network, ground the unused 50Q CHG. LINE connector or
disconnect either R751 or R756. If the alternate method is
used, the repetition rate of the pulses will be one-half the
original rate. By using either of these methods, pulse dura
tions up to 300 nanoseconds can be satisfactorily obtained.
2. Using the C h arge Netw ork Accessory
Included with your Type 109 is a special Charge Network
designed to produce a fast-rise pulse that decays one RC
in about 5 microseconds. The amplitude decay is only 10%
in 400 nanoseconds (see Fig. 1-5, Characteristics section).
The Charge Network plugs into the 50Q CHG. LINE 2 con
nector and the banana plug goes into the center conductor
of the 50Q CHG. LINE 1 connector. The banana plug dis
ables or grounds the unused contact of the mercury switch.
Since one set of contacts are used to generate the pulse,
the repetition rate is one-half the normal rate. The output
amplitude of the pulse is about one-tenth that indicated by
the VO LTAGE RANGE and AMPLITUDE controls.
Maximum allowable external voltage that can be applied
to the Charge Network is 50 volts.
Selecting the Pulse Amplitude
1. Using the Type 109 Pow er Sup ply
In most applications the coaxial lines which are used to
generate the output pulses are charged by the internal
100-volt power supply of the Type 109. In these applica
tions, the pulse amplitude is controlled by the VOLTAGE
RANGE and AMPLITUDE controls. The VOLTAGE RANGE
control determines the range of adjustment of the AMPLI
TUDE control. The scale of the AMPLITUDE control, when
used with the setting of the VOLTAGE RANGE control,
indicates the approximate pulse amplitude. Using the
internal power supply, pulses with amplitudes between zero
and 50 volts can be produced.
Alternate pulses of different amplitudes can be produced
using the Type 109 internal voltage supply and separate
charge lines. To do this, connect a rheostat across either
of the EXT. POWER OR MONITOR connectors. If this is
done, however, the front-panel amplitude settings will not
be correct, due to the external loading.
2. Using an External Power Source
Pulses with amplitudes higher than 50 volts can be gen
erated if an external power source is used to charge the
coaxial lines. To use an external power source, first place
the VOLTAGE RANGE control in the EXT. PWR. position.
Then connect the external power source or sources to the
EXT. POWER or MONITOR connectors on the front panel.
The pulse amplitude obtained will be approximately one-
half the power source voltage, up to approximately 100
volts. At some higher voltage, the relay will suddenly
develop a higher arc drop resulting in a reflection at the
relay. This reflection will cause the output to be less than
one-half the power source. The voltage applied to the Type
109 to charge the lines should be limited to approximately
600 volts using one or both contacts to prevent damage to
the 47-k 2-watt limiting resistors, R752 and R757.
An additional advantage in using external power to
charge the coaxial lines is that alternate pulses of differ
ent amplitudes and/or polarity can be generated by using
two different power sources. This can be combined with dif
ferent length charge lines to produce not only different
amplitudes but different pulse widths as well.
Selecting the Pulse Polarity
The PULSE POLARITY switch controls the output polarity
of the pulses when internal power is used to charge the
coaxial lines. The polarity of the output pulses is the same
as the polarity of the charge voltage.
When external charge power is used, alternate positive
and negative pulses can be obtained by charging one
line with a positive source and the other line with a nega
tive source. If identical pulse widths are required for both
the positive and negative pulses, it will be necessary to
select identical charge cables. (A single charge cable
connected between the 50Q CHG. LINE 1 and 50L2 CHG.
LINE 2 connectors should not be used for this mode of
operation because the cable cannot be charged simultan
eously by both a positive and a negative voltage.)
2-3

NOTES

SECTION 3
Some fundamental factors to consider when preparing the
Type 109 Pulse Generator for use with other equipment
will be covered in this section of the manual. Several repre
sentative test systems including a few specialized examples
will also be illustrated and discussed. These systems pro
vide a basis for the development of other specialized
systems as required by specific applications.
Pulse Definitions
The following terms are commonly used in describing
pulse characteristics and are defined here for convenience.
The terms are illustrated and applied in Fig. 3-1. The input
pulse represents an ideal input waveform for comparison
purposes. The other waveforms represent typical output
waveforms in order to show the relationships. The terms
are defined as follows:
Risetime tr: the time interval during which the amplitude
of the output voltage changes from 10% to 90% of the
rising portion of the pulse.
APPLICATIONS
Falltime tf: the time interval during which the amplitude
of the output voltage changes from 90% to 10% of the
falling portion of the waveform.
Pulse Width (or Duration) tw: the time duration of the
pulse measured between the 50% amplitude levels of the
rising and falling portions of the waveform.
Time Delay td: the time interval between the beginning
of the input pulse (t = 0), and the time when the rising
portion of the output pulse attains an arbitrary amplitude
of 10% above the base line.
Tilt: a measure of the tilt of the full amplitude, flat-top
portion of a pulse. The tilt measurement is usually expressed
as a percentage of the amplitude of the rising portion of
the pulse.
Overshoot: a measure of the overshoot occurring gener
ally above the 100% amplitude level. This measurement is
also expressed as a percentage of the pulse rise.
Bear in mind that these definitions are for guide pur
poses only. When the pulses are very irregular (such as
excessive tilt, overshoot, etc.) the definitions may become
ambiguous. In such cases, a more complete description of
the pulse will probably be necessary.
o
<
o
>
INPUT
PULSE
TIM E
tr= R IS ETIM E
tf = FA LL TIM E
t-= PULSE W ID TH
td = TIM E DELAY
% T I L T = - X 1 0 0 %
D
% O V E R S H O O T —— X 1 0 0 %
D
■
-- »-j td J-* ---
Fig. 3 -1 . Term s used in d escribing pulse ch a racte ristics.
3-1

Applications— Type 109
Risetimes
The risetime of any particular assembly of the Type 109,
an oscilloscope (conventional or sampling), and accessory
pieces such as coax cables is a variable depending upon
the cable characteristics as well as individual risetimes. The
“ root of the sum of the squares” method can generally be
applied as an approximation method only, as skin effect
losses of the cables do not add properly using this method.
(The root-sum-squares method applies accurately to gaus-
sian systems only.)
As a general rule, if the equipment or signal being
measured has a risetime 10 times slower than the Type 109
and other related measuring equipment, the error is 1%.
This amount is small and can be considered to be neg
ligible. If the equipment being measured has a risetime
three times slower than the related measuring equipment,
the error is slightly less than 6% . By keeping these rela
tionships in mind, the results can be interpreted intelli
gently.
Basic Precautions
For faithful reproduction of the pulse certain precautions
should be followed. These can be summarized as follows:
(a) Use proper types of cables, terminations, attenuators,
and impedance matching networks. Low-impedance coaxial
cables are used with the Type 109 as signal conductors.
It is important that these cables be terminated in their
characteristic impedance (50Q) to prevent reflections and
standing waves unless you deliberately wish to improperly
terminate the cables. One application for improper termina
tion would be to boost the signal to an amplifier input by
leaving the end of a transmission line unterminated.
(b) Keep unshielded wires of uncertain impedance short so
that reflection and/or cross-coupling effects are not intro
duced. Keep ground-return paths short and direct.
(c) Shield measuring equipment leads to prevent unde
sired coupling to other parts of the circuit. Shielding is
especially required where radiation is a problem and where
high-impedance dividers or circuits are involved.
(d) Choose components which function properly at fre
quencies and risetimes encountered.
(e) Keep in mind inherent parameters in circuit com
ponents such as inductance present in capacitors or resistors.
(f) Consider the possible nonlinear behavior of circuit
components due to changes in voltage or temperature
coefficients.
(g) Consider the input impedance of measuring equip
ment. The impedance may be enough to cause loading
effects, detuning or undesirable reflections.
Connecting the Type 109 to the Device Under
Test
When connecting the Type 109 Pulse Generator output
to the device under test, observe the following precautions:
1. A complete dc-return path must be provided between
the device under test and Type 109 Pulse Generator 50fi
OUTPUT connector.
2. If the pulse is applied to a 50Cl load which has a
dc potential across it, the actual amplitude of the pulse is
the voltage set by the AMPLITUDE control less one-half the
dc voltage across the load. Do not allow more than 200
volts dc to be applied to the Type 109 Pulse Generator 500
OUTPUT connector. This limit will keep the internal com
ponents of the Type 109 from being damaged.
As an example, assume that the Type 109 Pulse Gener
ator output is connected to a load which has +10 volts
across it and that the AMPLITUDE control is set to +1
volt. The actual amplitude is found by substituting these
values in the following equation:
VA = Vs - VL = (+1) - (+10) = - 4 volts
2 2
where VA is the actual pulse amplitude, Vs is the voltage
setting of the AMPLITUDE control, and VL is the dc voltage
applied across the load.
3. If the load will not terminate the 50 Q output of the
Type 109 Pulse Generator (because it is not practical or
possible), then it will be desirable to use a 50-ohm coaxial
lead (between the Type 109 and the load) which is long
enough to delay the load’s reflection until after the time
of interest. The reflection will appear at a time equal to
twice the output lead delay plus the pulse length.
Some representative test systems involving the Type 109
and other related equipment are described and illustrated
in this portion of the manual. The systems to be described,
as mentioned earlier, may be used as a basis for the de
velopment of other more specialized systems required by
specific applications.
Using the Type 109 With Sampling Oscillo
scopes
One of the primary applications of the Type 109 Pulse
Generator is to use it for checking and calibrating sampling
oscilloscopes which have internal triggering capabilities.
Since this application is adequately covered in the instruc
tion manual for the sampling oscilloscope involved, no de
tailed explanation will be provided here.
In the usual application the Type 109 is used to drive
a test device so the output from the device can be ob
served and measured on the crt screen of the sampling
oscilloscope. Fig. 3-2 shows how the connections for this
application are made.
In other similar applications, using this setup, the test
device could be a test fixture. An example of a test
fixture that can be used is the Type 290 Transistor Switch
ing Time Tester, available through your Tektronix Field
Office.
Fig. 3-3 illustrates a sampling test setup where the de
vice under test is inserted in series with the charge line.
A clear picture of transmission-line characteristics can be
made using this setup. The presence of discontinuities
along a transmission line can be determined while the line
is under study by means of the oscilloscope display.
The Type 109 when used in conjunction with a sampling
oscilloscope provides an excellent means for measuring the
impedances of certain devices and cables. In an application
3-2

Applications— Type 109
Fig. 3 -2 . Sam p lin g test setup w h ere the Typ e 1 0 9 is used to d rive a test dev ice or fixtu re .
of this sort, the device is connected as part of the charge
line while the output from the Type 109 is applied to the
oscilloscope input. If the impedance of the inserted device
is exactly 50 ohms, it will merely increase the time that
the amplitude of the Type 109 waveform remains constant.
The displayed waveform will also indicate double the de
lay time for the inserted device.
The test device, such as a piece of coaxial cable, a
connector assembly, or a delay line, can be connected into
the charge line of the Type 109 in the manner shown in
Fig. 3-3. In Fig. 3-4 the Type 109 waveform is shown when
a length of 125-ohm cable is connected into the charge line
in series with two lengths of 50-ohm cable. The portion of
the waveform due to the 125-ohm section is about 60%
as high as the first portion of the waveform due to the
50-ohm cable. The duration of the Type 109 waveform due
to the 125-ohm cable is twice the delay time of the cable
so it is evident from the picture that the true delay time
of the 125-ohm cable used is actually 5 nsec.
In Fig. 3-4 the relative amplitude of the portions of the
waveform bear a definite relationship to the impedance
of the device that generated that portion. The impedance
of an unknown device can thus be measured by compar
ing the amplitude of the portion of the Type 109 wave
form produced by it against the amplitude of the initial
portion due to the 50-ohm system. The method is generally
limited to the first reflection, unless the deviations are
small, due to multiple reflections and reflection losses.
If we call the amplitude produced by the 50-ohm system
V0 and the amplitude produced by the inserted device Vx,
then the impedance of the inserted device is given by the
formula:
Z = 50 (2 — — 1)
Vx
3-3

Applications— Type 109
_________________ P O R I IO N C UE TO 125 O JJN IE _
—H
V *-1 V v , V u
Vo— w " n r
iii r
mJ
1 J 1 1
AWA
V: se c
: p f f
OND i
p n i f
\R Y
"INK
V
;— V
Equivalent Sweep Rate: 5 nsec/cm
(Sampling) Oscilloscope: Type 661
Fig. 3 -4 . W a v efo rm o btained when a section o f no m in al 125-ohm ca b le is connected as p a rt of the charge lin e fo r the T yp e 1 09 .
In Fig. 3-4 the ratio of V0 to V, is approximately 1.75.
Using this in the above formula gives the correct impedance
of 125 ohms used to produce the waveform.
It is essential in applications of the type described here
that no shorts, terminations, terminated adapters, or attenu
ators having low shunt resistance to ground are used in
the charge line of the Type 109. If devices such as these
are used, they will prevent the charge line from charging to
the correct voltage and will thereby prevent the Type 109
from producing an output waveform. Where it is necessary
to match one type of connector to another, low-loss un
terminated adapters should be used.
the 50 n cable (used as the reference) and determine their
characteristics. If this pad is constructed and used, the
formula for determining the impedance is as follows:
where: Z = the unknown impedance.
V0 = the peak amplitude produced by the
50 Q reference impedance.
Vx = the peak amplitude at the time of
the reflection.
(If component under testN I '
is connected here
place of cable under
test, ground other end p
of component to 50 ß i
cable shield.) I
est, I '
i Y
X A
?-T
From
Type 109 (A
50 U OUTPUT V7
Connector
,
__ i66.7 Q
AAA
----
100 Q, 1%
Vt w
(Cable-under-test cen
ter conductor can be left
open or shorted to coax
shield.)
Cable or Component
under test
50 S3 cable (Used as a
reference impedance)
66 .7 Q
---- W v
lOOfi, 1%
V i w
Dividing Pad ^
, \ To 50
Input of
Oscilloscope
All Resistors: 1 % , V i watt
Fig. 3 -5 . T h re e -w a y d iv id in g pad test circuit fo r m e asuring im
ped an ce b y reflection.
As an alternate method, you can construct a dividing
pad such as the one shown in Fig. 3-5. This pad overcomes
a disadvantage of the previous method. With this pad you
can connect low-shunt resistance components or devices to
Using the Type 109 With a Conventional Oscil
loscope
The Type 109 can be used with conventional oscilloscopes
in much the same manner as with sampling oscilloscopes
that have internal triggering capabilities. Fig. 3-6 shows the
connections required. The oscilloscope used in this setup
must have an internal delay line if the signal from the de
vice is fast-rise non-repetitive or has a low duty cycle.
This setup is useful where the frequency response of the
device falls within the bandwidth limitations of the vertical
amplifier system of the oscilloscope. Internal triggering on
the applied signal is the method that must be used to trig
ger the sweep. This method of triggering is convenient
since no external triggering connections are required.
A second system is shown in Fig. 3-7. Here the device
under test, besides being pulsed by the Type 109, is able
to provide external triggers to the oscilloscope. It is there
by possible to observe the shaping and amplification of a
signal in the circuits of the device without resetting the
oscilloscope triggering controls for each observation. If the
external triggering signal is derived from the waveform
at the input circuit of the device under test, the time re
lationship and phase between the output and input wave-
O
3-4

Applications— Type 109
Fig . 3 -6 . Test setup w here a con ven tional oscillosco p e is used in conjunction w ith the Type 10 9 .
forms may be seen and compared on the oscilloscope
screen. Bandwidth limitations of the oscilloscope must be
considered when using the setup. In addition, consider the
fact that the oscilloscope should have an internal delay line
so the leading edge of a single waveform, or fast-rise low-
duty-cycle pulse can be observed.
Direct Connection to Deflection Plates
In some cases the output signal from a 50-ohm device
under test can be observed by direct connection through
coupling capacitors to the vertical deflection plates of a
conventional oscilloscope. Thus, the limited bandwidth of
the oscilloscope vertical amplifier can be bypassed. The
discussion that follows first describes a method for coupling
the Type 109 pulses to the vertical deflection plates. Next,
a specific setup is used as an example to show the results
that can be expected. When the setup is working properly,
the device under test can then be inserted into the signal
line and the output from the test device observed.
The following factors pertaining to the vertical deflection
plate system will be considered: dc operating potential of
the plates, lead inductance, deflection plate capacitance,
transit-time limitations, delay lines, and deflection factor.
(If the last factor is a prohibitive limit, then a Tektronix
Sampling Oscilloscope should be considered as another way
to avoid the bandwidth limitations of the vertical amplifier
and at the same time obtain excellent sensitivity.)
A typical circuit for ac-coupling directly to the vertical
deflection plates is shown in Fig. 3-8. This circuit permits
the internal vertical amplifier of the oscilloscope to be by
passed, but still allows the normal dc operating and posi
tioning voltages to be applied to the deflection plates from
the internal vertical amplifier. However, when using this
circuit, you must use a high-quality external delay line
Fig. 3 -7 . Test setup w h e re the o scilloscope is e x te rn ally trig gered by the device .

Applications— Type 109
Fig. 3 -8 . D e tailed method o f ac-coupling the sig na l d irectly to the vertical deflection p late s.
which will delay the pulse about 120 nsec to get the pulse
on the crt screen.
Only approximate values for two of the parts are given.
The values of these and all other parts depend upon the
crt, cable impedance, and lead lengths. Later on, some set
ups employing a 540-Series Oscilloscope are illustrated and
described to show how optimum results were obtained by
modifying the circuit shown in Fig. 3-8 to suit the specific
application.
The coupling capacitors, Cl and C2, provide the means
for ac-coupling to the plates. When selecting these parts
for use in the circuit, keep in mind that the physical size
should be small to reduce the lead-length inductance.
The cable which connects to the terminating resistor
should be long enough so that if a double-transit time re
flection appears, it can be easily identified from the input
signal. Then, the undesired termination error causing the
reflection can often be corrected by physical or electrical
adjustments at the termination.
To find the resonant frequency (f0) of the lead induc
tance and the deflection plate capacitance (C) for use in
the equation shown in Fig. 3-8, use the method that follows.
Turn off the oscilloscope power and disconnect the vertical
amplifier output leads to the crt. Cut a wire loop which
equals the total length of C l, C2, R l, R2, R3, R4. Substi
tute the wire loop for these components between the verti
cal deflection plate pins. Bring a grid-dip meter near the
loop and measure the resonant frequency.
A convenient method for making connections to the crt
deflection plates is to use clips removed from a standard
miniature tube socket.
After removing the wire lead, measure the total capaci
tance between the plates with a Tektronix Type 130 LC
Meter, or equivalent, at the deflection plate pins. Capaci
tance between the plates can also be found by referring
to a list of crt specifications.
The value of Rl is found by solving the equation in Fig.
3-8. Make R2 the same value as R l.
Since the deflection plates are placed close to the path
of the electron beam, a small amount of current will flow
in the deflection plate circuits. This current flow varies
nonlinearly with the beam position. The values of the re
sistors R3 and R4 must be selected to keep the current
flow from producing' a large voltage drop at the deflection
plates. If the resistances are too great, the voltage drops
may become large enough to cause serious positioning
difficulties, defocusing, or distortion. These effects are most
noticeable when the beam is positioned near one side of the
crt. On the other hand, if the resistances are too small
the short r-c time constant of the coupling circuit may cause
the low-frequency response to be limited.
The risetime limitation is a combination of the limitation
imposed by the resonant frequency (fc) (which limits the
risetime arriving at the deflection plates), and the transit
time (tr) of the electron beam through the deflection plate
system (which limits the deflection plates' ability to change
the beam position rapidly).*
The deflection factor can be found from the reference
chart, or it can be measured as follows: Connect a dc volt
meter between the vertical plates when the internal vertical
amplifier is connected to the deflection plate pins. Measure
the voltage change when the beam is positioned vertically
over the full height of the graticule. Divide this voltage
excursion by the graticule height in centimeters to obtain
the deflection factor in volts per centimeter.
If the output leads from the internal vertical amplifier
of the oscilloscope are disconnected and the power is on,
do not allow the leads to come in contact with the chassis
or tube shield. A short circuit of this type can damage the
amplifier circuits.
Typical risetime figures for Tektronix cathode-ray tubes
are listed in Table 3-1.
As a specific example, showing before and after results,
a Type 540-Series Oscilloscope with a Type K Plug-In Unit
was used in the setup shown in Fig. 3-9. In this example
* Se e I. A . D. Lew is and F. H. W e ll, M illim icro second Pulse Tech
niq u es, Second edition — 1 9 5 9 , C h ap ter 6 , Pergam on Press, Lon
don and N ew Y ork .
3-6

Applications— Type 109
Fig . 3 -9 . A p p ly in g a fa st-rise pulse from the Type 1 09 to the inp ut of a Type 54 0-S erie s O sc illo 
sco pe. The illu stra tio n show s the setup th at w as used and the resu ltan t w av e form .
TABLE 3-1
CRT REFERENCE CHART
TEKTRONIX OSCILLOSCOPE
TYPE (includes rackmounts)*
CRT
(Type No.)
NOMINAL VERT.
DEFLECTION
FACTOR-
volts/cm
NOMINAL
RISETIME-
nsec
COMPARATIVE
WRITING
RATE**
531, 535 T51 11.0 to 14.0 2.0 V s t o %
531 A, 533, 535A T533 8.5 to 10.5 2.0 V s t o y 4
541, 545 T54 6.0 to 7.0 2.5 V s to y4
541 A, 543, 545A T543 6.0 to 7.0 2.5 V s to y4
551 T551 6.0 to 6.5 3.0 V s to y4
555 T555 6.4 to 7.3 3.0 V s t o y 4
517A (12 kv) T517 (T54-H) 7.0 to 8.0 2.0 V s
517A (24 kv) T517 (T54-H) 14.0 to 16.0 1.5 l
517 (12 kv) 5XP 17.0 to 30.0 1.5 V s
517 (24 kv) 5XP 34.0 to 40.0 1.0 1
* Refer to the Instruclio n M an ual for the o scilloscope you are using to d ete rm in e sw eep -ra te lim ita tion s.
* * W ritin g rate com pared to Type 5 1 7 crt, 24 kv, P I 1 p ho sphor. W ritin g rate of T5 17P 1 1 at 24 kv is 1 00 0 to 1 20 0 cm/fisec recorded
on TRI-X film at f l. 9 with 4 .2 to 1 reductio n. W ritin g rate is in creased by a facto r of 2 to 3 using PO LAR O ID 3 0 0 0 speed film ,
even fa ste r for p refogged film . W ritin g rate is a function of film sp ee d , lens f-stop , op tical reduction fa cto r, type of pho sphor, sen si
tiv ity of film to color o f p ho sp hor, accele rating p o ten tia l, an d beam current. Consult yo u r local Tektro n ix Field En gineer or O ve r
seas Repre sentative for more info rm ation on w riting rates and p ho spho rs.
the pulse from the Type 109 was applied to the Type K
Unit to show how the pulse rise and fall times are limited
by the bandwidth of the vertical amplifier.
Fig. 3-10 shows the setup which was used to apply the
Type 109 pulse to the vertical deflection plates through a
Deflection Plate Connector (Part No. 013-017). The result
ant waveform is shown in the same illustration. The deflec
tion plate connector circuit that was used is similar to the
one shown in Fig. 3-11 except C3 and R5 were not added.
The rolloff and gradual slope of the fop portion of the
waveform is caused by the use of the 120-nsec RG-8A/U
cable. A long cable is necessary to obtain sufficient delay
to display the leading edge of the waveform on the crt
screen.
To obtain the best results, C3 and R5 were added (See
Fig. 3-11). The setup and the resultant waveform are shown
in Fig. 3-12. Even by using two Type 113’s that contain
high-quality cable, some rolloff occurred and components
C3 and R5 had to be added to sharpen the leading corner
and make the top of the waveform as horizontal as possi
ble. If cw signals are applied to the deflection plate con
nector instead of pulses from the Type 109, remove C3, R5,
and the Type 113’s. Use a short 50-ohm GR-8A/U cable
in place of the Type 113's.
By inserting a test device in series with the signal delay
cable using the same setup shown in Fig. 3-12, you can
observe and measure not only the output of the device but
also the time delay introduced by the device. If the device
is linear, it is unimportant where the device is inserted into
the signal delay cable. If the device is nonlinear, if may be
advantageous to place the device between the delay cable
and the deflection plate connector, especially if the delay
cable risetime is significant, as when RG-8A/U cable is
used in place of the two Type 113’s.
3-7

Applications— Type 109
Open end.
Fig . 3 -1 0 . A p p lyin g the Typ e 109 p ulse through a d eflectio n p late connector netw o rk to the crt
ve rtical d eflection pla tes. Note that the use of R G - 8 A /U ca ble as a sig n al d e la y line causes ro lloff
and a g rad u al upw ard slope o f the w a v e form .
Fig. 3 -1 1 . C ircuit dia g ra m o f actual circuit used to ob tain the w avefo rm show n in Fig . 3 -1 2 . Com
po nents C3 a nd R5 aid in sh arp en in g the le ad ing co rner and fla tten in g the top of the w aveform to
offse t the slig h t d eterioratio n cause d by using a long sig n a l de la y line.
3-8