Tektronix 492P, 492 Datasheet

We

go where

you

go with lab quality

spectrum analysis.

492/492P Spectrum Analyzers

Lab quality you can

Fully calibrated in

amplitude and frequency.

Crt readout of parameters.

Reference level, center fre­quency, vertical scale factor,
frequency span, frequency
range, resolution bandwidth,
and rf attenuation are dis-

played right on the screen
for easy reference and

photographic documenta­tion. Convenient bezel iden­tification of crt parameters.

Programmable control set­tings and signal display in­formation are available via

GPIBinthe492R

80 dB dynamic range.

Wide frequency

50

kHz

to

21 GHz internally
and to 60 GHz with Tektronix
calibrated external

waveguide mixers. Opera­tion to 220 GHz with com-

mercially available external

waveguide mixers.

Automatic modes.
Wide range of options.

Goes where you go.

With its portable form factor
and single-handle carry, the
492 moves easily around the

lab or systems test area, fits
under an airplane seat, or
goes out with you on a field
call.

Environmentalized per
MIL-T-28800B.

range.

get

a handle on.

Copyright

©1980,

Tektronix,

Inc. All rights reserved.

Full

programmability.

492P model provides IEEE

488 compatibility via

GPIB

interface bus. Allows full

programming of all signal-

affecting front-panel control

settings and processing of

stored spectrum displays.

Digital

storage and signal

processing.

SAVE

A, B MINUS

MAX HOLD, and AVERAGE

modes. Lets you compare,
subtract, save maximum
values, or noise average
(smooth) your spectrum

SAVE

A,

displays.

Microprocessor aided
ease of operation.

Simple three-knob operation
performs your measure­ments quickly and easily:

1-Set

center frequency,

2-Set span/div,

3-Set reference level.
Automatic control on most
often used functions.

Power-on initiates input

protection and normal

start-up settings.

Input protection.

Freedom from spurious
responses.

Internal calibrated

selector available with Op-

tion

01.

Eliminates harmonic

mixing products and im-

ages, and increases

dynamic range for harmonic

YIG

pre-

measurements in the micro-

wave range.

A spectrum analyzer with unmatched

convenience and capability.
In one compact package.

Simple to use.

Operation of the 492 is as

easy as 1 (set the input ref­erence level), 2
center frequency), and 3
(set the frequency span).

Most-used functions are

automatically controlled.

Setting the reference level

automatically selects the

proper

//

gain and rf attenu-

ation. Setting the span/div

automatically selects the

proper resolution and scan
time. These preprogrammed
controls of interrelated func­tions save you time and
simplify your measurement
task.

(set

the

The center frequency con-

trol with constant tuning rate

(CTR) provides smooth fre-

quency adjustment with just

one knob regardless of the

span being used. CTR
allows the operator to posi­tion a signal more quickly
and more precisely than a
conventional tuning system.

Digital storage and signal

processing.

Digital storage allows for
flicker-free, easy to interpret
displays that may be held in

memory as long as instru-

ment power is on.
Available as Option 02,

digital storage and signal

processing enhance the
492's performance and
ease of operation.

VIEW

A,

VIEWS

Contents of the selected

memory are displayed. All

SAVE A mode.

SAVE A mode — The spec­trum stored in memory A is
displayed. If

is selected at the same time,

memory A and B are both

displayed simultaneously for

comparison (data viewed in

memory B is updated con-

tinuously).

modes —

VIEWS

mode

stored displays are updated
continually (except with

SAVE A mode). A and B

memories can be combined
for high resolution (1000

point) storage.

B

MINUS

SAVE A is automatically ac-

tivated and the algebraic

difference of the continu-

ously updated contents of

memory B and the stored
contents of memory A is

displayed. Positive and

negative differences are

displayed above and below

an internally selectable zero

reference screen position.

SAVE

A mode —

MAX

HOLD

mode.

MAX

HOLD mode

—the

memory stores the highest
amplitude signal detected

for each frequency display,

allowing you to maintain and

monitor maximum

values.

This feature is especially

useful in measuring signal

drift and stability, in record-

ing peak amplitudes, and in
logging the presence of
random signals.

AVERAGE mode — Move-

able cursor sets the level of
signal peak detection or av­eraging. All signals above

the cursor are peak de-

tected and then digitally

stored; all signals below the

cursor are averaged before

storage.

Portable form factor.

Compact size and light

weight combine to offer un-

matched portability in a lab-

oratory quality analyzer. The

492 can easily be moved in
the design lab or systems

test area — or wherever else
you need it. It even fits under

an airplane seat.

Wide frequency range.

The 492 offers unequalled

frequency coverage. From

50

kHz

to

220

GHz, with

amplitude calibrated

waveguide mixers to

60 GHz available from Tek-

tronix. A broader range than

any laboratory analyzer and

with the performance you

need to handle tough labo-

ratory measurements.

Wide frequency range.

Programmability/IEEE

(GPIB)

compatibility.

The programmability and

interface bus capability

available with the 492P pro­vide added measurement
versatility.

Repetitive or large quantity
data collection with consis­tent and rapid results is

made easy, as is recording

of data in hard copy form.

Automated testing and

monitoring are also possi-

ble. The GPIB interface en-

ables full program control of

front panel settings and of

special modes like

12 dB/div and "smart" func-

tions. When an external

controller is used, auto-

mated data correction and

analysis can enhance re-

sults and make possible

complex measurements

such as total harmonic dis-

tortion and power spectral

density.

Amplitude resolution of

0.25

dB.

High amplitude accuracy is

possible through 0.25 dB

steps, an improvement over

the well-known if substitution
technique. Superfine 0.25
dB control increments pro-

vide 0.05 dB per 0.25 dB
step accuracy. This per-

formance increase is

achieved without a separate

vernier control.

?

!

FINE

492P makes spectrum analysis

automatic. And easy.

Two instruments in one.

The 492P is a fully program-

mable version of the 492

Spectrum Analyzer. It incor-

porates all of the 492's lab

quality performance and

ease of use features when

used as a manual instrument.
Push the "Reset to Local"

button and the 492P be-

comes a
from the front panel. But, most

important, the 492P opens

the way to automated spec-

trum analysis and documen-

tation via its IEEE-488

interface.

This versatility makes the
492P useful in many applica-

tions in the lab, factory or

field.

492—with

operation

(GPIB)

With or without a controller.

Switches on the rear panel
select the mode of operation

as a GPIB instrument. In the

normal TALKER/LISTENER

mode, the 492P listens to and
executes commands from a
GPIB controller. All important
front panel settings can be

LF

OR

TALK ONLY

LISTEN ONLY

EOI————- • '

1

operated remotely. Some
functions are controlled with
more detail through the bus
than possible from the front

panel. For example, SPAN

can be set with two digit pre-

cision anywhere within the

range of the instrument,

making possible special

spans such a 4.8 kHz per div.

Also, via the GPIB, the vertical
scale may be set for 1 to

dB per division in 1 dB incre-

ments. When requested, the
492P will report instrument
settings, internal status or

data

For
troller is not available, the

492P can be set to the TALK

ONLY

modes.

GPIB ADDRESS

EOI

from its display.

field

use where a con-

or LISTEN ONLY

In

the

TALK

ONLY

mode, it outputs waveforms

and front panel settings in a
fixed format for data logging
to a digital tape, such as the
Tektronix 4924. In the LISTEN

ONLY

mode,

the

492P

comes a display for waveform
data sent on the bus and will

respond to commands for
measurement setup.

15

be-

When used with the Tektronix 4052 Graphic Computing System

controller and 4631 Hard Copy Unit, the 492P can provide test
results in both graphic and numeric form for the evaluation of

microwave signal sources.

Easy to use.

The 492P is designed for
ease of operation via the

GPIB,

just

as

the

492

is

designed for front panel
operational ease. Most com­mands for program control

are simply abbreviations of

the front panel nomenclature.

For example, to set center

frequency, send the ASCII

characters
To query the center fre-

quency, send "FREQ?" The

response is "FREQ 5.2 E

+

09!'

The 492P's high level

command language and the
similarity of commands and

responses simplify pro-

gramming and make pro-

gram listings easily readable
for editing.

"FREQ

5.2 GHz."

Automated setup.

The 492P is not only an

obedient servant under pro­gram control, but smart
enough to do many things on

its own. For example, it will

provide all of its settings with

one command. Just send

"SET?" You'll get back a block

of characters listing all of the

front panel and internal set­tings. Store this response in
the controller or on a tape

To return

original measurement condi-

the

492P

to

file.

the

tion, send the same charac-

ters back to the analyzer. To

initialize

"power on" settings, send

"INir

the

492P

to

its

Internal processing
simplifies programming.

The 492P's power reduces

processing time and eases

the software development

task. For example, the 492P
can identify all signal peaks

above a specified amplitude

threshold and the maximum

and minimum points on the
display waveform.

Having identified a point of
interest in its digital storage
memory, the 492P can then
perform certain operations

automatically. It can tune to
center a signal on the dis-

play before narrowing the
span. Send the command,

"CENSIG"

automatically. Another com-

mand,

cally changes the sensitivity

so the signal peaks at the top

of the screen reference level.

The display now reads di-

rectly in dB below the signal

peak

screen display for precision

and it's done

"TOPSIG"

and assures an on

automati-

amplitude measurements.

Programmable 492P saves

time

and

simplifies

A UHF TV signal with

492P Programmable Spectrum Analyzer, sent
digital form and then

VERT DISPLAY RF ATTEN

I0DB/

The same spectrum displayed above is drawn by the 4052
using the Tektronix 4662 Interactive Digital Plotter for high

resolution documentation.

multiburst

displayed

20DB

0-1.8

your

test pattern is acquired by the

on the 4052 Graphic System.

FREQ

RANGE RES BANDWIDTH

100

via

FREO

the

SPAN/DIV

S3BKHZ/

KHZ

job

GPIB

in

Automated signal tracking.

Keeping a drifting signal on
screen in a narrow frequency
span per division is no longer
troublesome. With the internal
processing functions and the
ability to repetitively execute
instructions in its input buffer,
the 492P can track drifting
signals with virtually no help
from the controller.

Auto peaking.

Automatic peaking provides
greatest accuracy when
measuring at widely differing
frequencies using the op­tional internal preselector filter
(1.8-21 GHz). Send "PEAK
AUTO" from the controller to
the 492P and an internal
routine automatically adjusts
the preselector tracking for
maximum signal response.

Enhanced data
presentation via GPIB.

Displayed waveforms can be
outputted from digital storage
to a calculator, computer, or
data storage device such as
a tape or disc file. The data
may be graphed on a digital
plotter, lower left, or displayed
on the screen of a controller
such as the
programmable Tektronix 4052
Graphic System, top left.

Mathematical operations
such as calibration and cor­rection of the original data
can be performed in the con­troller. Data can be combined
from several different fre­quency spans to make a

BASIC-language

composite plot. Raw data can
be converted into different
units such as micro-volts or
dBc. The two-way communi­cation ability of the 492P
permits a real time compari­son of a controller-generated
spectrum, or set of limits, and
an incoming signal on the
492R The computational
power of the controller can
be used to solve complex
analytical measurements
such as total harmonic distor­tion, photos at top of page 9

(facing)

density.

Full programmability
completes the task.

Complete automation of the
entire measurement saves
time and eliminates many
operator errors or inconsis­tencies. Also, some pro­grammed measurements

may be made more precisely

and thoroughly than by

manual methods.

harmonic distortion meas­urement, for example, the
analyzer can be controlled to
tune each harmonic fre­quency separately, permitting
an enlarged high resolution
display of the signal peaks,

lower left, page 9. After the

peak amplitude is measured,
the span can be widened and
the analyzer tuned to find the

next harmonic. Internal pro­cessing greatly simplifies the
controlling program and

speeds the measurement.

or power spectral

For

a total

FREQUENCY

AMPLITUDE

DBM

RELATIVE

DB

A 100 MHz signal and its harmonics displayed on the 492P with
a full screen span of 500 MHz.

1.0E+8

2.0E+3

3.0E+8

4.0E+8

5.8E+8

6.8E+8

7.0E+8

8.0E+8

TOTAL

HARMONIC

DO YOU

MAHT

TO DO ANOTHER ANALYSIS

Automated total harmonic distortion test results computed by
the Tektronix 4052 and outputted on the Tektronix 4631 Hard

Copy Unit.

Put it to work.

With the programmable 492P
on your measurement team,

repetitive measurements can

be done the same way every
time. Your throughput will

increase—and

dence in results. And, the

internal processing and high
level programming language

-2B.8

-25.64

-34.96

-33.16

-28.32

-37.64

-39.B8

-33

-43.56

DISTORTION EQUALS 80 PERCENT

(YES

-4.84

-14.16

-17.36

-7.52

-16.84

-18.28

-14.2

-22.76

OR

When you look at the total

performance capability of the

492R you'll recognize its

value: Ease of operation both

as a programmable and
manual instrument. Wide fre-

your confi-

quency range. The versatility

to go where you go. Into the

lab for automated testing; into

the field for data collection.

H0>?

make software development
faster. You get high power re­sults with easy programming.

Full

programmability

and measuring the signal peak with high amplitude resolution.

Internal processing commands

allows tuning to each harmonic frequency

in

the

492P

greatly simplify this

measurement.

Check the specs and see.

Frequency Related

Frequency Range — 50 kHz to

21 GHz with internal mixer, to
220 GHz with external mixers. Op-

tion 08 deletes coverage above

21 GHz (calibrated mixers to
60 GHz available from Tektronix).

Frequency Accuracy — ±0.2% or
5 MHz, whichever is greater, +20%
of span/div.

Frequency Readout Resolution

— Within 1 MHz.

Frequency Span per Division —

10 kHz to 200 MHz plus zero and
full band max span, down to 500 Hz
with Option 03 in 1-2-5
sequence.

Frequency Span Accuracy —

±5% of span/div, measured over

center eight divisions.

Resolution Bandwidth @ 6 dB
Points — 1 MHz to 1 kHz (100 Hz

for Option 03) in decade steps
within ±20%, manually or auto-

matically selected.

Resolution Shape Factor
(60/6 dB) — 7.5; maximum.

10

Spurious Responses

Residual (no input signal) —

-100 dBm or less referenced to
input mixer for fundamental
conversion.

Harmonics — At least -60 dBc for

full screen signal in the Min Distor-

tion mode to 21 GHz. At least -100

dBc for preselected Option

to

21

GHz.

01.1.7

Intel-modulation — 3rd order

products at least -70 dB down
from two full screen signals within

any

frequency span in the Min Dis-

tortion mode. At least -100 dB

down for two signals spaced more

than

100

MHz apart from

GHz for preselected Option

L.O. Emissions (referenced to
input

mixer)—

maximum; -70 dBm maximum for
Option

01.

10

dBm

1.7

to 21

01.

Stability
(after 2 hour warm-up)

Residual

for 2 ms
(50 Hz) for 20 ms with phaselock
Option 03.

Long Term Drift: 200 kHz/hour un-

phaselocked, 25 kHz/hour phase­locked for fundamental mixing.

Noise Sidebands — At least
75 dBc @ 30X resolution offset
(70 dBc for 100 Hz resolution) for

fundamental mixing.

FM—1

time

kHz peak-to-peak

duration, improves to

Amplitude Related

Reference Level

-123

dBm

maximum safe input) for 10 dB/div
and 2 dB/div log modes. 20 nV/div
to 2 V/div (1 W maximum safe input)

in the linear mode.

Reference Level

dB, and 0.25 dB for relative level (A)
measurements in log mode. 1-2-5
sequence and 1 dB equivalent in­crements in

Reference Level Accuracy —

Amplitude change of 0.25 dB

±0.05

dB, 1 dB

Range

to +40 dBm

Steps

LIN

mode.

±0.2dB,10dB

—

(+30

—10

dBm

dB, 1

±0.5 dB; to a maximum of

for 60 dB and ±2 dB for 90 dB

erence level change when gain
change and attenuation do not
offset each other.

Display Dynamic Range — 80 dB

@ 10 dB/div, 16 dB @ 2 dB/div and

±1.4

dB

ref-

8 divisions linear.

Display Amplitude Accuracy —

±1

dB/10

±2 dB/80 dB; ±0.4 dB/2 dB to
maximum of ±1
full screen in LIN

Resolution Bandwidth Gain

Variation— ±0

dB to maximum of

dB/16

5dB

SENSITIVITY AND FREQUENCY RESPONSE

Average Noise Level

to

18

01,

10%/div,

Preselection

-115dBm

-115dBm

-115

-115dBm

-100

-95

-100dBm

GHz,

01,

dB,

for 1 kHz Resolution

No

Preselected

Option 01

-110dBm

-110dBm

-110dBm

dBm

-110dBm

dBm

-95 dBm

(12GHz)

-90 dBm

per

(18

-85 dBm

1

formance

dBm

-95 dBm

-95 dBm

Pen Lift

IF

Out—

screen, -30 dBm display; 10

50ft.

GPIB

output control for 492 P

Miscellaneous

Sweep Time — 20

(10 s/div in auto) in 1-2-5 sequence.

Crt

center frequency, frequency range,
vertical display mode, frequency
span/div, resolution bandwidth, and

rf

attenuation.

Crt

Power — 90 to

250

with all options.

Environmental Characteristics —

Per MIL-T-28800B type III, class 3,

style C.

Configuration—Portable, 20 kg

(44

49.9 cm (6.9 x 12.9 x 19.7 in) without
handle or cover.

"Flatness

not apply to the

attenuator positions between 19 and

20

GHz.

1

GHz)

Readout — Reference level,

—8x10

VAC,

Ib)

Mixing

Frequency Range

50kHz-1.8GHz'
50kHz-4.2GHz'

1.7-5. 5 GHz

3.0-7.1 GHz

5.4-18GHZ

15-21

GHz

100MHz-18GHz"'

18-26

GHz

26-40 GHz
40-60 GHz

"Low

frequency end performance does not include effects due to zero Hz feedthrough.

"Flatness

attenuator positions between 19 and 20 GHz.

**

Includes frequency band switching error of 1 dB maximum.

Number (n)

1
1
1
1

3

3

With Tektronix optional high

6

10
10

and accuracy specifications do not apply to the

Input Characteristics

Internal

Mixer — Type N female

connector, VSWR
and 3.5 to 21 GHz, with 10 dB or

1.45

more attenuation.

Optimum Level for
Linear Operation —

-30 dBm referenced to mixer.

1 dB Compression Point —

-28 dBm from 1.7 to 2 GHz for Op-

tion 01; otherwise -10 dBm.

Maximum Safe Input Level —

+13 dBm without Option
+30 dBm (1 W) with Option

zero

rf attenuation.

Attenuator Power Limit —

+30 dBm (1 W) continuous, 75 W

peak for 1

/us

or less pulse width

and 0.001 maximum duty factor.

Output Characteristics

Calibrator——20 dBm ±0.3

100

MHz ±0.01%.

1st

L.O.—+7.5

nominal.

2nd

L.O.——16

nominal.

Vertical

nominal.

Horizontal

kft

nominal.

dBm @ 50ft

dBm @ 50ft

Out—0.5 V ±5%/div, 1 kft

Out—0.5

V ±

Frequency Response

With

No

Preselection

±2,5dB

±1.5dB
±1.5dB

±2.5dB

±3.5dB
±3.5dB

wavei

guide

mixers

±3dB

±3dB
±3dB

30,

40,

50, and

—TTL,

5V

-15 dBm nominal

Control — IEEE 488

cm, P31 Phosphor.

132

48 to 440 Hz, 210 W max

(all options), 17.5 x 32.7 x

and accuracy specifications do

30,

40,

dB; ±5% of

mode.*

10dB

60dBr/

nominal.

/xs

to 5 s/div

VAC,

50, and 60

Attenuation

Preselected

Option 01

±1.5dB

±2.5dB
±2.5dB
±3.5dB

±5.0dB*"

±4 5dB

for

MHz,

input/

180

to

dBrf

full

Interpreting

the specifications

Low-End Frequency Performance.

Frequency Drift.

Low-End Frequency

Performance.

Typical low-end frequency

performance for the 492 is
shown opposite in figure

Unlike many spectrum

analyzers with microwave

measurement capability, the

492 is specified for mea-

surements down to 50 kHz.

Resolution filter shape,
phase noise sideband per-

formance, input mixer cou-

pling, and zero hertz local

oscillator feedthrough can

all contribute to limiting
low-end frequency per-

formance.

The 492 low-end frequency
specification is 50 kHz. This

is based on the fact that, for

a basic 492, 50 kHz corre-

sponds to center screen at

the narrowest available fre-

quency span of 10 kHz/div.

By comparison, a

phaselocked 492 (with Op-

tion 03), set at 100 Hz reso-

lution; has a center fre-

quency capability of 2.5 kHz

(500 Hz/div).

Performance below 50 kHz is

affected by local oscillator
phase noise, input coupling
and the zero hertz feed-

through. Practical per-

formance is limited to

approximately 5 kHz.

1.

Note that,

at

50

kHz, the

noise level is degraded to

approximately
a 1 kHz resolution bandwidth
due to noise sidebands

around the zero Hz feed-

through.
The zero Hz signal also lim-

its the ability to set a small

signal (e.g., -80 dBm)
screen in 2 dB/div. These

effects are not significant at
frequencies above 1 MHz,
and they can be circum­vented by using the A
amplitude mode.

Frequency Drift.

The relationship between
long-term, medium-term,

and short-term frequency

drift for the 492 is illustrated

in figure 2.

Spectrum analyzer fre-

quency drift is primarily

caused by changes in tem-

perature and manifests itself

as a shift in signal display

frequency even though input
signal frequency remains

constant. The apparent
change in frequency de-

pends on several factors,

including instrument tem-

perature at turn-on, en-

vironmental temperature,

time since turn-on, whether
or

not

the 492 is phase-

locked, the mixing harmonic

number, and the period of

time over which the meas-

urement is made.

-90dBmfor

to

full

-40

-60

o

•o

-80

-100

-120

1 10

Typicaf low end frequency performance for the 492 with Option 01

(see text for explanation.)

Figure 1

/

turr

warr

Representative drift characteristics of modern spectrum analyzers.

Figure 2

/

-on

-i-up
——

rift

) drift^vline

*•—

1000

.

100

Frequency (kHz)

Dngt

3rm

••

ta-^-

;

V

JVW

>s

Time

V-

/v

^

+*

1

^

v

shc

resi

nedk

>rm

rtter

dual

^

s

m

c

rift

^

m

-M

c

zer<

S

\

11

Interpreting the

specifications

Q

u

c

10,000

nul;

live

1,000

drift

100

Ion 3 teim d

10

11

40

20

Typical long term turn-on drift characteristics for an unphase-

locked 492 starting at room temperature (fundamental mixing.

Multiply by mixing N number for higher mixing products).

Figure 3

60 80

Time (Minutes)

100

specified two-hour warm-up
period.

Total cumulative drift after

the two-hour warm-up is

about 5 MHz. At this point,
the 492 has reached the
long-term drift specification

of 200 kHz/hr. The instru-

ment should be recalibrated

after warm-up to maintain
accuracy.

Frequency Drift During

Warm-Up

492 (Option 03).

Adding

r

rr

inut

03) to a 492 helps minimize

fora

Phaselocked

phaselock

(Option

drift. The first local oscillator

is stabilized (automatic at 50

kHz/div at fundamental

conversion unless manually
overridden), leaving only

120

140

1,000

drift caused by the second
local oscillator. The second

oscillator is not affected by

higher mixing product mul-

tiplication, and so drift is in-

dependent of input fre-

quency.

Furthermore, 492 phaselock

stabilization offers the

equivalent of infinite hold-in

range, so that the first local

oscillator will not break lock

as it drifts. When phase error
voltage reaches a prede­termined maximum, the tune
voltage is reset to maintain

lock.

A phaselocked 492 (Option

03) starting at room temper-

ature changes frequency by

no more than 500 kHz in the

ulat

ve c

rift

Frequency Drift During
Specified Warm-Up Time.

Because of the thermal na-

ture of frequency drift, the
492's behavior depends to a
great extent on external

temperature conditions as

they affect the internal tem-

perature of the instrument. A
492 that has been out in the

cold before warm-up may

drift more than one that has

been in the laboratory.

Frequency Drift During

Warm-Up For An
Unphaselocked 492.

Turn-on drift as a function of
time is caused by environ-

12

mental changes. Because

the 492 is designed to oper-

ate in a wide variety of envi­ronments, drift is a more im-

portant consideration

for

the
492 than for a spectrum
analyzer designed for lab

use only.

For a typical unphaselocked
492 starting out at room

temperature, frequency drift

during the first half hour of

warm-up is about 4 MHz, as
shown in figure 3.

The cumulative drift speci­fications shown in figure 3

represent a slope of several
hundred kHz/min at turn-on,

down to a few kHz after the

:rrr

100

nutf:

drift

120

140

Icngt

40

Typical long term turn-on drift characteristics for a phaselocked
492 (with Option 03) starting at room temperature.

Figure 4

60 80

Time (Minutes)

Resolution Bandwidth.

Harmonics and Intermodulation.

Figure 5

first two hours. After the

two-hour warm-up, the in-

strument drifts a maximum

of 25 kHz/hr — a specifica-

tion that is almost ten times

better than without phase­lock. These specifications

represent an approximate

drift of 10 kHz/min at turn-on

and 400

Hz/min

after two

hours, as shown in figure 4.

Medium-Term Drift. A

phaselocked 492 (with Op-

tion 03) exhibits excellent

medium-term drift at 100
Hz/sec. An unphaselocked

492 will exhibit medium-term
drift of about 2 kHz/sec.
Medium-term drift is random
back and forth movement of

the displayed signal after

the instrument has warmed

up.

Incidental FM

(Short-Term

Drift), A phaselocked 492

exhibits 50 Hz incidental

fm,

and a basic 492 exhibits
1 kHz peak-to-peak. Inci-

-60

+-H-

dental fm (caused by

short-term drift) takes place
in a very short time — often

in milliseconds,

at

the most

in seconds — and it must be

measured in a very short

time as well.

Resolution Bandwidth.

492 resolution bandwidths

are specified at the 6 dB

down points. This specifica-

tion represents the spacing

of two equal amplitude sig-

nals that can be well re-

solved, as illustrated in

figure 5.

Depending on the applica-

tion, other bandwidths may

be of interest: for example,

3 dB, random noise, and

impulse bandwidths. For
most measurements, 492

random noise and impulse

bandwidths are approxi­mately equal to the 6 dB

bandwidth, and the 3 dB

bandwidth is approximately

0.75

of

the 6 dB

bandwidth.

Typical approximations are

given in table

1.

I +n

Typical

Random

Noise

Bandwidth

100

Hz

1 kHz

9.5kHz
100kHz

1 MHz*

Specified 6 dB

Resolution
Bandwidth

±20%

100 Hz
1 kHz

10kHz

100kHz
1 MHz

*The

pulse stretcher must be activated to achieve full impulse bandwidth at 1 MHz resolution setting.

Table 1.

Typical

3dB

Bandwidth

75

Hz

750

Hz

7.5kHz
75kHz

750 kHz

Harmonics and

Intermodulation.

The 492 specifications for

harmonics and intermodula­tion are based on a

-30

dBm

rf

signal level at

the first mixer. The following

worst-case specifications

relate to the level referenced

to the rf signal (in dBc):

Harmonic

Products

—

-60dBc

Third-Order Intermodu-

lation —

-70dBc

The distortion-tree dynamic

Typical

Impulse

Bandwidth

100

Hz

1 kHz

9.5kHz

100kHz

800kHz"-

Optimal distortion-free

dynamic range occurs when

the distortion products

equal the sensitivity noise

level (n) in dBm.

For the 492 with preselector

at 100 Hz resolution, n =

-118

dBm for fundamental

conversion. Thus,

S0 =

(N-1)

N

=

(2)5

+(-118) = -36 dBm,

and

range for any input level can

be computed from the inter-

cept point equation:

I =

_A_

+ S, where

N-1

I = Intercept point in dBm.

A = Distortion level relative

to input carriers in dB.

N = Distortion order numbers.

S = Input signal level to the

mixer in dBm*
For second-harmonic distortion,

I =

60_ + (-30) = +30

dBm.

2-1

For third-order distortion,

I =

70_ + (-30) = +5

dBm.

=

(2)

Best distortion-free dynamic
range for closely spaced

signals is 82 dB. Inter-

modulation dynamic range

reaches 100 dB for signal

spacings greater than

100 MHz

21 GHz frequency range.

"Input level to the mixer is spectrum analyzer

input level

+10 dBm input with 30 dB of rf attenuation is

a mixer input level of +10-30 = -20 dBm.

N

(123)

=82

3

for

the

lessrf

attenuator level. Thus, a

dBc.

1.7

3-1

to

13

Interpreting the

specifications

x

ir

1 M

10k

100

coa

HzE

Hzd

Hzb

W

W

W

-30

-40

-50

-60

-70

m

pn

•o

-90

-110

-120

1 10

Sensitivity showing average noise level

492 (Option 01).

Figure 6

Sensitivity.

Sensitivity for a preselected

492 (Option 01) is shown in

figure 6.
Note that, although a non-

preselected 492 has 5 dB
lower input loss, and there­fore 5 dB better sensitivity,
the spurious-free dynamic
range of the preselected in-

strument is superior.

Sensitivity is specified in
terms of the smallest ob­servable signal, and is
therefore determined by
spectrum analyzer internal

noise level. Noise level de-

pends on resolution band­width and local oscillator
mixing multiplication
number. (Refer to figure 6.)

pu

t

i^J-

r

J^k

r

_r

—

1~

^

Input Frequency (GHz)

fora

preselected

Dynamic Range.

On-screen dynamic range of
the 492 is 80 dB at a vertical
display scale factor of

10dB/div.

Intermodulation distortion
dynamic range for signals
spaced closer than 100 MHz
is consistent with on-screen
dynamic range at 80 dB.

For small signals or modula-

tion components next to a

large carrier, the dynamic
range can be considerably
greater. Limiting factors are
resolution bandwidth filter

shape, phase noise side-

band characteristics, and

input overload capability.

For resolution bandwidths

wider than 10 kHz, filter

shape is the limiting factor.
Figure 7 shows a 100 kHz

vavf

\

ipttt

uid

i

; -

-»-

resolution bandwidth filter
that exhibits 1.1 MHz
bandwidth at 80 dB down.
Phase noise sidebands are
the limiting factor closer in.
Figure 8 shows the same
signal with a 1 kHz resolution

bandwidth filter.

100

.

Note that the phase noise
sideband pedestal breaks

out at 70 dB down, dropping

to

75

dB

down

at

20

kHz

offset from the carrier. This

allows easy observation of
close-in sidebands 10 kHz
out and 70 dB down.

IEF

-10

;

100

76

50

75

50

25

I

I

1

__

Figure 7

ill.

:

MHz

t

,

,V,

, , , ,

..L..I._I •

-...<--! .1 1

80 d

Be

-20

-30

-40

-20

-30

-<0

i

^50

-60

-70

I

D
G

dB

i i

14

Sensitivity.

Dynamic Range.

Amplitude Accuracy.

Reference Level.

-10

-20

-30

-40

-50

-60

-70

-80

-90

-100

-110

-120

-130

-140

Typical filter shape, phase noise sidebands, and sensitivity factors

for fundamental mixing on a preselected 492 (Option 01).

Figure 9

1

10 100

Offset Frequency (kHz)

1,000

10,000

Figure 8

Figure 9 shows typical filter
shape, phase noise side-

bands, and sensitivity fac-

tors for fundamental mixing

on a preselected 492 (Op-

tion 01).

Amplitude Accuracy.

The amplitude reference for

the

492

is

a -20dBm,
100 MHz calibrator. The
amplitude accuracy of this
signal is

±0.3dB.

Factors

affecting measurement ac-

curacy depend on input sig-

nal frequency and amplitude

as follows.

"Flatness and accuracy specifications

not apply to the

attenuator positions between 19 and

20

GHz.

30,

40, 50, and 60 dB

dc
rf

Frequency response per

band.

The frequency response for

the first band is referenced

to

the

100

MHz calibration

point.

A band-to-band reference

error of 1 dB must be added

to the per-band frequency
response specifications

when the instrument is oper­ating outside the first band.

Amplitude measurement

error resulting from fre­quency response depends
on the frequency of the in-

coming

signal.*

Approxi-

mately ±1 dB is contributed

by the preselector (Option

01), and the remainder is
contributed by other input
circuits.

A nonpreselected 492 has

almost 1 dB better amplitude
accuracy.

Amplitude Display.

An amplitude display error

of0.4dB/2dB,
and 2 dB/80dB

1 dB/10dB,

occurs

when signal display ampli­tude is not at the full screen

reference level.

Reference Level.

Gain/attenuation errors are

introduced when the refer­ence level is changed from
the

-20

dBm

calibrator

level.
Absolute reference level in

dBm can be changed in

1

dB

and

10

dB

steps.

Reference level change er-

rors are

±0.5

±0.2dB/1

dB/10

dB,

dB accumulat-

ing to ±1.4dB/60dB, and

±2

dB/90dB.

This error

analysis is worst case, pro-

viding gain and attenuation

are not switched in simul-

taneously. This is normally

prevented by the 492's au-

tomatic reference level
mode.

A -20 dBm calibration level

should preclude having to

switch gain and attenuation

together.

A10

dB

minimum

rf input

attenuation setting during
calibration allows for small
signals that require some
attenuation to provide good
impedance match. The user
also has the option of cali-

brating at -20 dB and zero

dB attenuation by activating

the

MIN

NOISE control.

15

Interpreting

the specifications

Relative Reference Level Steps.

0

-2n

„

2

Q

ut

4

=1

to

•a

+l 6

-80

8

10

02

Cumulative maximum absolute amplitude measurement error as a
function of signal amplitude and frequency for a preselected 492
(Option 01) assuming calibration using the internal -20 dBm 100 MHz
reference.

Figure 10

Frequency Range

50kHz-1.8GHz

1.7GHz-5.5GHz

3.0GHz-7.1 GHz

5.4GHz-18.0GHz

15.0 GHz-21.0 GHz

18.0GHz-26.5

26.5GHz-40.0GHz

40.0 GHz-60.0 GHz

"10

dB steps over a 70 dB

"Using

optional

Achievable amplitude accuracy for a preselected 492 is shown above.

Table 2.

rlBm

ff

4

dBm

sign;

4 6 8 10 12 14 16 18 20 22 24 26

±dB

GHz**

accessory

ji

sinn

1-

-4­i

Measurement Frequency (GHz)

Frequency

Response

range

high

performance

Max.

1.5

2.5

2.5

3.5

5.0
3
3
3

IF Gain*
Variation

±dB

mixers

Max.

0.5

0.5

0.5

0.5

0.5

0.5

0.5

0.5

Calibrator

Output

±dBMax.

0.3

0.3

0.3

0.3

0.3

0.3

0.3

0.3

±dB

Total

2.3

3.3

3.3

4.3

5.8

3.8

3.8

3.8

Max.

Relative Reference Level
Steps.

The 0.25 dB steps are ena-

bled in the 2 dB/div vertical
mode when "Fine" steps are
selected. These steps are
used for accurate relative

Typical accuracy limits in
each band for a preselected

492 are summarized in table

2. Worst case accuracy
limits including band-to­band error of 1 dB are illus-

trated in figures 10 and

amplitude difference mea­surements. The total mea­surement range is 50 dB,
with an accuracy of

0.05 dB/0.25dB,

0.4dB/2dB,1

and

2 dB/50 dB.

dB/10dB,

OdB'ref

50 d 3 ref

level

evel

ciang

;han

je

10

24

6 8 10 12 14 16 18 20 22 24 26

Cumulative maximum error in measurement of relative amplitude of

two signals as a function of reference

separation,

Figure 11

fora

Frequency Difference (GHz)

pre-selected

level

492 (Option

change and frequency

01).

11.

16

Option

Configurations

Two basic models of spec­trum analyzer are available,

the 492 Spectrum Analyzer

and the 492P Programmable

Spectrum Analyzer. You can

order either model with any

combination of the available

performance options and

optional accessories to con-

figure the instrument for your

requirements.

The 492P is fully compatible
with the IEEE-488 interface

bus (GPIB) and can be op­erated through any control-

ler capable of interfacing

through the bus, especially

the Tektronix 4050 Series

systems. All important front

panel controls function

digitally and can be oper-

ated by the controller. Thus,
the 492P not only transmits

data to the controller, but
may be programmed by that
controller. A remote/local

switch

on

the

492P

front

panel provides for local,

manual control of the
analyzer, overriding pro­gram control.

The front panel may also be

locked out by the controller

for testing situations in

which the possibility of local
control might invalidate

measurement results.
For general use spectrum

analysis, optimum perform­ance may be obtained with

the inclusion of Options

02, and 03.

01,

Option 01 — Internal
Preselection.

With this option, internally

generated image and har-

monic mixing spurious re-

sponses are effectively

eliminated. This results in a
display that is much easier

to interpret. In the frequency

range

of

50

kHz

to

1.7

GHz,

a low-pass filter is used to

limit spurious responses. In
the range of
21 GHz, a

selector is used. Internal

calibrated preselection re-

duces the requirement to

examine each signal to ver-

ify authenticity.

Measurement capability is

enhanced with Option 01 by

an increase in dynamic

range from 80 dB in the

basic analyzer to

signals separated by

MHz). This is because the

automatic tracking pre-

selector rejects signals out-

side its bandwidth by 70 dB

or more.

Option 01 also includes

a

limiterto

dBm

input protection to the
first mixer
Above
mixer is protected by the

preselector.

1.7

GHz

tracking

provide +30

up

to

1.7

1.7

GHz the input

YIG

100

GHz.

to

pre-

dB (for

100

Option 02

—Digital

Storage.

Spectra may be digitized with
500 point resolution and held
in one of two memories, A or

B—or

in a

1000

point mem­ory created by combining A
and B. Once in memory
spectra may be displayed

with a bright flicker free trace,

making prolonged viewing or

photography

cially for single sweep or slow
sweep speeds. Digital stor­age also adds several internal
data processing features:

Digital averaging — Data at

each frequency point in

memory is summed and di­vided by the number of

samples at that frequency.

Peak Detection — Data at
each frequency point is

continuously updated with

peak-detected values.

Digital
able cursor allows the

operator to obtain a com-

bined peak/average display

— data above the cursor is

shown peak-detected, data
below the cursor is aver­aged.

With MAX HOLD mode, the

highest amplitude attained at

each of the

easy—espe-

cursor — An adjust-

1000

points dur-

ing successive sweeps

stored and displayed. This
mode is useful for measuring
peak-to-peak drift over a time
interval or in making swept

response measurements of

filters without a tracking
source.

With SAVE A mode, one sig-

nal is stored in the A memory

for later examination. This in-

formation is not updated and

is useful in instances of later
comparison with other signal

information.

SAVE A

In

mode,

the B MINUS

the A signal is

is

stored and not updated, then

arithmetically subtracted from

the B signal, which may be

continually updated. This
mode is most useful for com­paring signals such as in pro­duction test comparison of a
signal with a standard, or for

calibrating frequency re-

sponse uncertainty out of a
measurement.

17

Ordering Information

With

the AVERAGE mode,

the display is divided by a

horizontal cursor. Above the

moveable cursor, signals

are peak detected and dis-

played. Below the cursor,

signals are averaged. Av-

eraging is useful for appli­cations in which signals

must be

presence of high noise
levels. The trace smoothing

that occurs through av-

eraging simplifies the

process.
Digital averaging and video

filtering may be used jointly

or independently depending
on the nature of the signal.

In
age, slowly swept signals are
easy to observe and photo-

graph, and do not require

intensity or other display re­adjustments.

Option 03 — Frequency

Stabilization/100 Hz

analyzed

addition,

in the

with digital stor-

Resolution.

With this option, phase-

locked local oscillator

stabilization provides ex-

ceptional display stability

and low noise sidebands,

18

and results in less frequency

drift and less residual

Thus, the 492 user can ob-

serve and measure charac­teristics of lower modulation

frequencies. As part of Op-

tion 03, improved resolution

(100 Hz) and narrow span of
500 Hz/Div. provide in­creased measurement

fm.

capability for close in
sideband analysis. The

special purity of clean

oscillators may thus be

measured directly at micro-

wave frequencies. The 492

retains its one knob center
frequency control with Con­stant Tuning
even with

Option 03 is recommended

when the 492 will be used at

spans less than 50 kHz per

division, and is required for

spans of less than 10 kHz

per division. Phaselock oc­curs automatically and is a
function of the setting of the
span/div control. For con-

venience in operating the

analyzer in fixed tuned re­ceiver (zero span) mode,

phaselock may be deacti-

vated by a front panel
control.

Option 08 — Delete

Rate(CTR)

phaselock.

External Mixer Capability.

Option 08 deletes external
mixer capability for extend­ing frequency range above

21

GHz.

Option 20 — General Pur-

pose 12.5 GHz to 40 GHz

Waveguide Mixer Set.

This option extends the op-

erational upper frequency of

the

492

or

492P

to

40

GHz.

The actual waveguide range

is

12.5

to

40

GHz. This op-

tion package, designed to

provide economical use of
the analyzer at frequencies
above 21 GHz, consists of

three waveguide mixers and
a connecting coaxial cable.

Option 21 — High Per-

formance 18 GHz to
40 GHz Waveguide

Mixer Set.

This option consists of two

waveguide mixers and a

coaxial cable, and extends

the operational upper fre-

quency of the 492 or 492P to

40 GHz. The actual wave-

guide range is 18 to 40 GHz,

and the mixers are designed

so that, in operation with the
analyzer, the system is cali­brated in amplitude, flat-

ness, and sensitivity.

Option 22 — High Per-

formance 18 GHz to
60 GHz Waveguide

Mixer Set.

This option consists of three

waveguide mixers and a

connecting coaxial cable,

and extends the operational
upper frequency of the 492

or 492P to 60 GHz. The ac-

tual waveguide range is 18
to 60 GHz, and the mixers

are designed so that, in op-

eration with the analyzer, the

system is calibrated in

amplitude, flatness, and

sensitivity.

Combined

options.

The basic 492 Spectrum

Analyzer or the basic 492P

Programmable Spectrum

Analyzer may be ordered with

many combinations of op-

tions and optional acces-

sories. The prices are additive

(except for one delete option,
Option 08, which deletes ex-

ternal mixer capability). Note

that the standard accessories

are listed only once (with the

basic 492), but that one set is

included with either spectrum

analyzer.
Also listed with the 492P are

the companion 4052 Graphic
Computing System Controller,

4631 Hard Copy Unit, 4662

Interactive Digital Plotter and

4924 Digital Cartridge Tape

Drive.

492 Spectrum

Including the following listed stand­ard accessories: Diplexer Assem-

bly

(015-0385-00);

Cable, N to N Connectors, 6 foot

(012-0114-00);

BNC to BNC

(1)

(012-0076-00);

to BNC Female (1) (103-0045-00);

CRT

Mesh Filter

Fuse 2 A, Fast Blow (1)

0021-00);
(159-0017-00); Power Cord,

(1)

(161-0118-00);
(343-0170-00);
Blue

(1)

Filter, Amber

CRT Light Filter,

0115-02);

0653-00); Operator's Manual

Operator's Handbook

Service Manual, Volume 1
Service

492P Programmable Spectrum
Analyzer

Including Programmers Manual

and the listed standard 492

accessories.

Analyzer

50

50 n Coaxial Cable,

Connectors',

Adapter, N Male

(1)

(378-0726-01);

Fuse 4 A

(378-0115-00);

CRT

Manual,

.............

Fast

Cord

CRT Light Filter,

(1)

(378-0115-00);

Gray

Visor

(1)

Volume 2 (2).

$18,500.00

fi

Coaxial

18

inch

(159-

Blow

115-V

Clamp

CRT Light

(1)

(378-

(016-

(1);

(1);

(1);

$22,500.00

(1)

(2)

(1)

and

4052 Graphic Computing System

Controller
4631 Hard Copy Unit

4662 Interactive Digital

Plotter

4924 Digital Cartridge Tape

Drive

............

..

...............$

................$

$10,950.00

$ 4,850.00

4,495.00

2,695.00

Option Ordering

Information.

Option

01—Internal

Preselection

Provides calibrated preselected

filtering of input to first mixer for
each frequency band.

Option

Storage Add

Provides multiple memory display

storage with SAVE A, MAX HOLD, B

MINUS SAVE A, display average,

and storage bypass.

Option

Stabilization/100 Hz

Resolution
Provides first local oscillator stabili-

zation by phaselocking the oscil-

lator to an internal reference. Also

provides 100 Hz resolution.

Option

Capability

Deletes internal switching front
panel connector and external di-

plexer to connect and use external

waveguide mixers.

Option

to 40 GHz Waveguide Mixer

Set Add $520.00

Includes three mixers (12.4 to 18

GHz,

18

40 GHz) and attaching hardware to

extend the 492 upper frequency.

Option

to 40 GHz Waveguide Mixer

Set Add 1,970.00

Includes two mixers (18 to 26.5 GHz

and 26.5 to 40 GHz) and attaching
hardware to extend the 492 upper
frequency.

Option

to 60 GHz Waveguide Mixer

Set

Add $3,220.00

Includes three mixers
GHz, 26.5

GHz) and attaching hardware to

extend the 492 upper frequency.

.......

Add

$3,800.00

02—Digital

$1,700.00

03—Frequency

........

08—Delete

....

20—General-Purpose

to

26.5

21—High

22—High

to

GHz,

Add

$3,250.00

External Mixer

Subtract

GHz,

and 40 to 60

$1,250.00

and

26.5

Performance

Performance

(18

to 26.5

12.5

to

18

18

Optional Accessories

The following listed accessories are
optional with all models and config-

urations of the 492 system, and may
be ordered in any combination.

General Purpose

Waveguide Set

(016-0656-00)

12.4

to

18

(119-0097-00)

18.0

to

26.5

(119-0098-00)

26.5

to

40

(119-0099-00)

Cable

(012-0748-00)

Case

(004-1651-00)

High Performance

Waveguide Mixer Set

(016-0662-00)
18

to

26.5

(016-0631-01)

26.5

to

40

(016-0632-01)

Cable

(012-0649-00)

Case

(004-1651-00)

High Performance
Waveguide Mixer Set

(016-0657-00)

18

to

26.5

(016-0631-00)

26.5

to

40

(016-0632-00)

40

to 60 GHz Mixer

(016-0634-01)

Cable

(012-0649-00)

Case

(004-1651-00)

Microwave Comb Generator

(067-0885-00)

C-5C

Camera

TV Trigger Synchronizer

(015-0261-00)

Hard Case (transit)

(016-0658-00)

Soft

Case

12.5

to

.........$

GHz Mixer

.........

GHz Mixer

.........

GHz Mixer

.........

... 20.00

....

18

to

.........

GHz Mixer

.........

GHz Mixer

.........

... 25.00

....

18

to 60 GHz

.........

GHz Mixer

.........

GHz Mixer

.........

.........

... 25.00

....

.........

.........

.........

.........

(016-0659-00)

40

GHz

550.00

150.00

190.00

220.00

35.00

40 GHz

$2,005.00

975.00

975.00

35.00

$3,255.00

925.00

925.00

1,250.00

35.00

$1,800.00

425.00

360.00

495.00

125.00

Note: The 492 Spectrum Analyzer

system is compatible with all

Tektronix C50 Series

cameras.

19

For the address of your nearest

Tektronix Field Office, contact:

U.S.A., Asia, Australia, Central & South

America, Japan

Tektronix,

Inc

P.O.

Box

1700

Beaverton, OR

Phone: [phone]

Oregon only

503/644-0161

Telex: 910-467-8708

Cable: TEKTRONIX

Europe,

Middle East

Tektronix International, Inc.

European Marketing Centre

Postbox827

1180 AV Amstelveen
The Netherlands
Telex: 18312

Canada

Tektronix Canada Inc.

P.O.

Barrie, Ontario

Phone: [phone]

Tektronix Distributors to serve you

around the world:

Argentina, Australia, Austria, Belgium,

Bolivia,

Costa Rica, Denmark, East Africa,

Egypt, El Salvador, Federal Republic of
Germany, Finland, France, Greece, Hong
Kong, Iceland, India, Indonesia, Iraq, Israel,
Italy, Ivory Coast, Japan, Jordan, Korea,
Kuwait, Lebanon, Malaysia, Mexico,
Morocco, The Netherlands, New Zealand,
Norway, Pakistan, Panama, Peru,
Philippines, Portugal, Republic of South
Africa, Saudi Arabia, Singapore, Spain,
Sri Lanka, Sudan, Surinam, Sweden,

Switzerland,

Turkey, Tunisia, United Kingdom, Uruguay,

Venezuela, Zambia.

97075

800/644-9051

Africa,

Box

6500

L4M

4V3

Brazil, Canada, Chile, Colombia,

Syria, Taiwan, Thailand,

Ecuador,

Copyright © 1980, Tektronix, Inc. All rights
reserved. Printed in U.S.A. Tektronix prod­ucts are covered by U.S. and foreign pa-

tents,

issued and pending. Information in
this publication supersedes that in all pre­viously published material. Specification
and price change privileges reserved.

TEKTRONIX, TEK, SCOPE-MOBILE, and

&jl

are registered trademarks of Tektronix,

Inc.

TELEQUIPMENT is a registered

trademark of Tektronix U.K. Limited. For
further information, contact: Tektronix, Inc.,

P.O. Box 500, Beaverton, OR 97077.

Phone:

503-644-0161;

Cable:

Tektronix. Subsidiaries and distrib-

utors worldwide.

TWX 910-467-8708;

Performance

worth the

___name

Tektronix

COMMITTED TO EXCELLENCE

5/80