0% found this document useful (0 votes)
117 views

22 Oscilloscope Tested

This document summarizes the results of testing 22 oscilloscopes. It provides details on the selection process for the scopes tested and lists the brands tested along with some basic specifications. It also notes that some brands are rebadged versions of Asian products sold under different names at various price points. The document aims to help readers select an appropriate oscilloscope by providing this overview of popular models and their performance.

Uploaded by

Sentral Avr
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
117 views

22 Oscilloscope Tested

This document summarizes the results of testing 22 oscilloscopes. It provides details on the selection process for the scopes tested and lists the brands tested along with some basic specifications. It also notes that some brands are rebadged versions of Asian products sold under different names at various price points. The document aims to help readers select an appropriate oscilloscope by providing this overview of popular models and their performance.

Uploaded by

Sentral Avr
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 13

Scope for ‘S

22 oscilloscopes tested

For electronics engineers


and technicians, the
benchtop oscilloscope is
an indispensable tool. If
you don’t already have a
‘scope’, or if you’re
looking to acquire a new
scope in the near future,
our survey will help you
select the proper one.

Test engineers:
Arjan Floris
Marcel Koenders
Henny Kuipers
Geert-Jan Laanstra
Stefan Lenk
Harald Profijt

Text and
project supervision:
David Daamen

14
copes
At first glance, the number of different makes and models
from 109 to more
than 22,000 pounds
If you examine the list of selected instruments, you will
of oscilloscopes currently available appears to be too probably notice that several other brands are missing.
large for a comprehensive survey. Even making a respon- For example, Philips is missing, but they no longer
sible choice when purchasing an oscilloscope appears to make oscilloscopes. Hewlett-Packard is also not repre-
be practically impossible. However, the product range of sented, since the ‘test and measurement’ division of
most manufacturers actually consists of only a few basic Hewlett-Packard is now called Agilent – and Agilent is
types, which are available in models having different indeed represented here. Other big names, particularly
specifications. If we assume that the members of such a in the digital segment (such as LeCroy and Tektronix),
product family are more or less the same with regard to are also included.
design, construction and basic functionality, the problem For some brands, such as Iwatsu, Protek and Kenwood,
is considerably simplified. we were unable to find a dealer who was prepared to
make a demo instrument available.
We had also originally planned to examine the Voltcraft
Selection 320 (GW Instek GDS830). However, the screen
That is thus more or less the method we used in selecting remained blank after the unit was switched on, so it was
the oscilloscopes for our survey. From manufacturers and not included in the results.
dealers that actively offer equipment, we requested demo We also made a few interesting discoveries. For
models that are representative for a particular family or instance, certain brands turn out to be ‘rebadged’ Asian
brand. However, not all product families of all brands products. This is not particularly surprising, since as you
are represented here, largely due to limitations related to doubtless already know, some vendors sell a lot of equip-
the scope of the test. Nevertheless, we are confident that ment under house brands, such as Conrad with Voltcraft.
the present selection gives a good impression of what is The same is true of Peaktech and Dynatek.
available in the area of oscilloscopes and how the vari- What is actually interesting is the following: we found
ous brands generally score. one instrument (EZ Digital OS-5030) that is offered under
these three brand names at three distinctly different
prices. Depending on the label on the front, you can pay
Preliminary remarks anywhere from £227 (€329) to £400 (€579). The dif-
Before introducing the instruments to you, we have sev- ference can thus be as much as £172 (€250). That’s
eral general comments. something we thought you’d like to know.
The firm Chauvin-Arnoux, which manufactures the Metrix Also, many Goodwill Instek oscilloscopes are also avail-
brand, was invited to participate in the test but declined. able under other brand names, for example Voltcraft.
As they will soon be introducing new models, they All of the selected participants are listed on the following
decided not to participate in this test. pages.

Technospeak
Bandwidth The frequency range within which the signal level (amplitude) has not dropped more than 3 dB.
Zin Input impedance, a combination of resistive and capacitive components.
Y range Range of the vertical scale. Indicates the amplitude of an externally applied signal.
X range Range of the horizontal scale. Indicates time. Normally controlled by timebase.
X-Y-mode Both the horizontal and the vertical scale is controlled by an externally applied signal.
Z-modulation Light intensity variation under the control of an externally applied signal.
Rise time Minimum time required to move the electron beam from 10 % to 90 % of screen height.
Trigger Signal indicating readiness to start writing to the screen.
auto Trigger automatically reset by next trigger.
normal As with auto, however electron beam quenched in the absence of a trigger.
single Trigger is no longer reset. With memory oscilloscopes, the triggered waveform remains on the screen.
edge Triggering takes place on signal edges.
pattern Triggering takes place on measured signal.
pulse width Triggering takes place on a certain pulsewidth found in the measured signal.
TV Triggering takes place on a video signal.
Delay range The range available for delaying write actions to the screen (post-trigger). If a delay line is available, neg-
ative times are also possible. A part of the waveform for triggering may then be displayed (pre-trigger).
Sample rate Rate (frequency) at which samples of the input signal are taken.

10/2004 - elektor electronics 15


Voltcraft Digimess
610-3 MO10

Brand Voltcraft / Goodwill Instek Brand Grundig Digimess


Model 610-3 / GOS310 Model HUC70-01 MO10
Display and type CRT, analogue Display and type CRT, analogue
Channels 1 Channels 1 (second cannel for x-y mode only)
Bandwidth 10 MHz (-3 dB, DC) Bandwidth 10 MHz (–3 dB, DC)
Zin 1 MΩ, 30 pF ±5 %) Zin approx. 1 MΩ, 30 pF
Y range 5 mV - 5 V / div (≤ 5 %) Y range 5 mV - 5 V / div (3 %)
Modes X-Y via additional input Modes X-Y via separate input
Max. input 400 V Max. input 400 V
Rise time unknown Rise time ≤ 35 ns
X range 0.1 µs - 0.1 s / div (≤ 5 %) X range 0.1 µs - 0.1 s / div (3 %)
Triggering auto, norm, TV Triggering auto, norm, TV
Trigger inputs internal, power, external Trigger inputs int, line, ext
Miscellaneous - Miscellaneous -
Dimensions 135 x 200 x 300 mm (W x H x L) Dimensions 220 x 90 x 270 mm (W x H x L)
Weight approx. 3 kg Weight approx. 3 kg
Including user manual, measurement cable with Including user manual, power cord, measurement
two croc clips, power cord probe
Price (rrp) £109 (€159) (Conrad) Price (rrp) £119 (€172.49) (Digimess), £132
(€191.32) (ex. VAT, Farnell)

Digimess GW Instek
MO20 GOS620

Brand Grundig Digimess Brand Goodwill Instek (also sold as Isotech and
Model HUC70-02 MO20 Voltcraft)
Display and type CRT, analogue Model GOS620
Channels 2 Display and type CRT, analogue
Bandwidth 20 MHz (–3 dB, DC) Channels 2
Zin approx. 1 MΩ (3 %), 25 ±5 pF Bandwidth 20 MHz (–3 dB, DC)
Y range 5 mV - 5 V / div (3 %) Zin approx. 1 MΩ, 25 pF
Modes ch1, ch2, add, alt, chop, invert ch2, x-y Y range 5 mV - 5 V / div (3 %)
Max. input 400 V Modes ch1, ch2, add, chop, invert ch2, x-y
Rise time ≤ 17.5 ns Max. input 300 V
X range 0.2 µs - 0.5 s / div (3 %) Rise time 17.5 ns
Triggering auto, norm, TV, auto PP X range 0.2 µs – 0.5 s / div (3 %)
Trigger inputs int, ch1, ch2, alt, line, ext Triggering auto, norm, TV-h, TV-v
Miscellaneous z-modulation Trigger inputs ch1, ch2, alt, line, ext
Dimensions 320 x 130 x 418 mm (W x H x L) Miscellaneous z-modulation, ch1 output
Weight approx. 6.5 kg Dimensions 310 x 150 x 455 mm (W x H x L)
Including user manual, power cord, two measure- Weight approx. 8 kg
ment probes Including user manual, power cord, two measure-
Price (rrp) £305 (€442.48) (Digimess), £332 ment probes
(€480.79) (ex. VAT, Farnell) Price (rrp) £350 (€508) (Havé-Digitap)
(from the same series: MO40, MO60; both 40 MHz) (from the same series: various 20-, 30- and
50-MHz instruments)

16 elektor electronics - 10/2004


B+K Hameg
Precision HM303-6
2120B
Brand B+K Precision Brand Hameg
Model 2120B Model HM303-6
Display and type CRT, analogue Display and type CRT, analogue
Channels 2 Channels 2
Bandwidth 30 MHz (–3 dB, DC) Bandwidth 35 MHz (–3 dB, DC)
Zin 1 MΩ (± 2 %), 22 ±10 pF Zin approx. 1 MΩ, 20 pF
Y range 5 mV - 5 V / div (3 %) Y range 5 mV - 20 V / div (3 %)
Modes ch1, ch2, add, alt, chop, invert ch2, x-y Modes ch1, ch2, add, chop, invert ch2, x-y
Max. input 400 V Max. input 400 V
Rise time 12 ns Rise time < 10 ns
X range 0.1 µs - 2 s / div (3 %) X range 0.1 µs – 0.2 s / div (3 %)
Triggering auto, norm, fix, single, TV-h, TV-v, vari- Triggering auto, norm, TV
able holdoff Trigger inputs ch1, ch2, line, ext
Trigger inputs ch1, ch2, alt, line, ext Miscellaneous component tester
Dimensions 370 x 180 x 440 mm (W x H x L) Dimensions 285 x 125 x 380 mm (W x H x L)
Weight 7.8 kg Weight approx. 5.6 kg
Including user manual, power cord, spare fuse, Including user manual, power cord, two measure-
two measurement probes ment probes
Options various probes, carrying case Price (rrp) £455 (€659.26) (DDS Electronics),
Price (rrp) £322 (€466.65) (Printtec) £485 (€702.91) (Conrad)
(from the same series: various 30-, 60- and 100-MHz instruments, (from the same series: a 50-MHz model)
some including an internal frequency counter and component tester)

EZ-digital Hameg
OS-5030 HM1004-3

Brand EZ-digital (formerly Goldstar / LG Preci- Brand Hameg


sion) / Dynatek / Voltcraft / Peaktech Model HM1004-3
Model OS-5030 / 8300 / 630-2 / 2035 Display and type CRT, analogue (microprocessor con-
Display and type CRT, analogue trolled)
Channels 2 Channels 2
Bandwidth 30 MHz (–3 dB, DC) Bandwidth 100 MHz (–3 dB, DC)
Zin approx. 1 MΩ, 30 pF Zin approx. 1 MΩ, 15 pF
Y range 5 mV - 20 V / div (±3 %) Y range 1 mV - 2 mV (5 %), 5 mV - 20 V / div
Modes ch1, ch2, add, dual, chop, alt, x-y (3 %)
Max. input 400 V Modes ch1, ch2, add, chop, invert ch2, x-y
Rise time approx. 12 ns Max. input 400 V
X range 0.2 µs - 0.2 s / div (±3 %) Rise time < 3.5 ns
Triggering auto, norm, TV-v, TV-h X range 50 ns – 0.5 s / div (3 %)
Trigger inputs vert. (dual/alt), ch1, line, ext Triggering auto, norm, TV
Miscellaneous z-modulation, ch1 output, blanking input Trigger inputs ch1, ch2, line, ext
Dimensions 316 x 132 x 410 mm (W x H x L) Miscellaneous second timebase, component tester, con-
Weight approx. 7.8 kg figuration memory, cursor measure-
Including user manual, spare fuse, power cord ments, autoset, RS232 interface
Price (rrp) £227 (€329) (Conrad), £399 (€579) Dimensions 285 x 125 x 380 mm (W x H x L)
(DDS Electronics) Weight approx. 5.9 kg
(from the same series: various 20-, 40-, 60- and 100-MHz instruments. Including user manual, power cord, two measure-
Also digital up to 250 MHz/200 MSa/s) ment probes, PC software
Price (rrp) £1008 (€1463.70) (DDS Electronics),
£1032 (€1497.02) (Conrad)

17
10/2004 - elektor electronics
Leader GW Instek
LS8105A GOS6103

Brand Leader Brand Goodwill Instek (also sold as Isotech and


Model LS8105A Voltcraft)
Display and type CRT, analogue Model GOS6103
Channels 3 Display and type CRT, analogue (microprocessor con-
Bandwidth ch1, ch2: 20 MHz @ 1mV/div, 100 trolled)
MHz @ > 1 mV/div; ch3: 100 MHz (–- Channels 2
3 dB, DC) Bandwidth 100 MHz, 20 MHz @ 2 mV/div (–3 dB,
Zin 1 MΩ (±1 %), 20 pF DC)
Y range 1 mV - 2mV / div (5 %), 5 mV - 5 V/div Zin 1 MΩ (±2 %), approx. 25 pF
(2 %) Y range 2 mV - 5 V / div (≤ 3 %)
Max. input 400 V (ch3 50 V) Max. input 400 V
Rise time 3.5 ns (5 mV - 5 V/div), 17.5 ns (1 mV Rise time 3.5 ns (17.5 ns @ 2 mV/div)
- 2 mV/div) X range 50 ns – 0.5 s / div (3 %)
X range 50 ns – 0.5 s / div (3 %) Triggering auto, norm, TV, holdoff (1 µs - 5 s)
Triggering auto, norm, fix, single, TV-h, TV-v, vari- Trigger inputs ch1, ch2, line, ext
able holdoff Miscellaneous z-modulation, trigger output, auto-range
Trigger inputs ch1, ch2, ch3, alt, line, ext (time/div)
Miscellaneous z-modulation Cursor measurements frequency, period, pulsewidth, duty-cycle
Dimensions 300 x 150 x 400 mm (W x H x L) Dimensions 310 x 150 x 455 mm (W x H x L)
Weight 8.8 kg Weight approx. 9 kg
Including user manual, power cord, spare fuse, Including user manual, power cord, two measure-
two measurement probes ment probes
Options rackmount adapter, probe case, front Price (rrp) £948 (€1376) (Havé-Digitap)
cover, various probes (from the same series: various 50- and 200-MHz instruments)
Price (rrp) £1148 (€1665) (elQuip)
(From the same series: 20- and 50-MHz instruments and 100-MHz
with cursors)

Voltcraft Hameg
650AD 1507-3

Brand Voltcraft / Goodwill Instek Brand Hameg


Model 320 / GRS6052A Model 1507-3
Display and type CRT, analogue & digital Display and type CRT, analogue & digital
Channels 2 Channels 2
Bandwidth 50 MHz (–3 dB, DC) Bandwidth 150 MHz (–3 dB, DC)
Zin 1 MΩ (±2%), approx. 25 pF Zin approx. 1 MΩ, 15 pF
Y range 1 mV - 20 V / div (±3 %) Y range 1 mV - 2 mV/div (5 %), 5 mV - 20 V /
Max. input 400 V div (±3 %)
Rise time approx. 7 ns (5 mV - 20 V/div) Max. input 400 V
X range analogue: 0.2 µs – 0.5 s / div (±3 %), Rise time < 2.3 ns
digital: 0.2 µs - 100 s X range analogue: 50 ns – 0.5 s / div (3 %),
Triggering auto, normal, TV, variable holdoff digital: 0.1 µs - 100 s / div (3 %)
Delay range unknown Triggering auto, normal, single, TV, time
Trigger inputs ch1, ch2, line, ext delay/event delay, edge, pulse width
Sample rate 8 bits @ 100 MSa/s (real-time) Delay range approx. 70 ns
Memory 16 kB (per channel) Trigger inputs ch1, ch2, line, ext
Acquisition sample, peak detect (> 25 ns), average, Sample rate 8 bits @ 200 MSa/s (real-time)
envelope, persistence Memory 2 kB (per channel)
Cursor measurements ∆V, ∆t, 1/∆t Acquisition sample, peak detect (> 5 ns), average,
Miscellaneous autoset, waveform-save/recall, configu- single, roll, envelope
ration memory, RS232 interface, z-mod- Cursor measurements ∆V, ∆t, 1/∆t
ulation, ch1 output Miscellaneous second timebase, autoset, component
Dimensions 275 x 130 x 370 mm (W x H x L) tester, RS232 interface
Weight 8.5 kg Dimensions 285 x 125 x 380 mm (W x H x L)
Including user manual, power cord Weight approx. 6.5 kg
Price (rrp) £699 (€999) (Conrad) Including user manual, software, two probes
(from the same series: a 30-MHz version from GW-Instek) Options multifunction interface, optical interface
Price (rrp) £1413 (€2049) (Conrad), £1446
(€2098)) (DDS Electronics)
(From the same series: 200-MHz variant)

18 elektor electronics - 10/2004


Wittig Voltcraft
22-300 6150C

Brand Wittig Technologies Brand Voltcraft / Goodwill Instek


Model 22-300 (Bench Scope) Model 6150C / GDS820C
Display and type LCD (monochrome, 320 x 240 pixels), Display and type LCD (colour, 320 x 240), digital
digital Channels 2
Channels 2 Bandwidth 150 MHz (–3 dB, DC)
Bandwidth 20 MHz (–3 dB, DC) Zin 1 MΩ (±2 %), approx. 22 pF
Zin 1 MΩ, 20 pF Y range 2 mV - 5 V / div (±3 %)
Y range 1 mV - 4 V / div (±2 %) Max. input 300 V
Max. input 40 V Rise time approx. 7 ns (5 mV - 20 V/div)
Rise time approx. 3.5 ns X range 1 ns - 10 s / div (±0.01 %)
X range 200 ns - 200 ms / div (±0.5 %) Triggering auto, normal, single, TV, time
Triggering auto, int., line, TV delay/event delay, edge, pulse width
Trigger inputs ch1, ch2, ext Delay range 100 ns – 1.3 ms / 2 - 65000 events
Sample rate 8 bits @ 100 MSa/s (real-time) Trigger inputs ch1, ch2, line, ext
Memory 16 kB (per channel) Sample rate 8 bits @ 100 MSa/s (real-time)
Miscellaneous second timebase Memory 125 k (per channel)
Dimensions 300 x 138 x 70 mm (W x H x L) Acquisition sample, peak detect (> 10 ns), average,
Weight approx. 1.35 kg FFT
Including user manual, power cord, probe, PC Cursor measurements ∆V, ∆t, 1/∆t
software (via RS232 or USB; incl. FFT) Miscellaneous autoset, save/recall, waveform trace
Options - save/recall, RS232, probe-calibration
Price (rrp) £482 (€699) (Conrad) Dimensions 254 x 142 x 310 mm (W x H x L)
(from the same series: a 4-channel version) Weight 4.1 kg
Including user manual, power cord, 2 measure-
ment probes
Price (rrp) £896 (€1299) (Conrad)
(From the same series by GW-Instek: various 100-, 200-MHz instru-
ments, also with colour display)

LDS Tektronix
Sigma TDS2024
60-4
Brand LDS Test and Measurement Brand Tektronix
Model Sigma 60-4 Model TDS2024
Display and type LCD (colour touchscreen, 800 x 600), Display and type LCD (colour, 320 x 240), digital
digital Channels 4
Channels 4 Bandwidth 200 MHz (–3 dB, DC)
Bandwidth 200 MHz (–3 dB, DC) Zin 1 MΩ (±2 %), 20 ±3 pF
Zin approx. 1 MΩ, 9 pF Y range 2 mV - 5 mV / div (±4 %), 10 mV - 5 V
Y range 2 mV - 5 V / div (±1 % ±2 LSB) / div (±3 %)
Max. input 400 V Max. input 450 V (< 100 ms), 300 Vrms
Rise time unknown Rise time < 2.1ns
X range 50 ns - 100 s / div (±25 ppm) X range 2.5 ns - 10 s / div (±0.001 %)
Triggering auto, pre- and post trigger, edge, pulse Triggering auto, edge, video, pulse width (glitch),
width and misc. adv. trigger options holdoff (500 ns - 10 s)
Delay range 399 s Delay range - 4 div x s/div - 20 ms
Trigger inputs ch1, ch2, ch3, ch4, ext Trigger inputs ch1, ch2, ch3, ch4, ext
Sample rate 8/10 bits @ 200 MSa/s (real-time) Sample rate 8 bits @ 2 GSa/s (real-time)
Memory 200 k (per channel, cascade with fewer Memory 2500 samples per channel
channels) Acquisition normal, average, peak detect (> 12 ns)
Acquisition +, –, *, ÷, invert filter, FFT, averaging Cursor measurements ∆V, ∆t, 1/∆t
Cursor measurements 2 vertical cursors, 2 horizontal cursors Miscellaneous set-up and waveform save/recall, trig-
Miscellaneous VGA, RS232, Centronics, Ethernet, ger output
USB, PS2, soundcard, HD recording Dimensions 323.8 x 151.4 x 124.5 mm (W x H x L)
Dimensions 378 x 254 x 302 mm (W x H xL) Weight 3.6 kg
Weight approx. 8.2 kg Including user manual, two measurement probes
Including user manual, power cord Options GPIB and RS232 -interface, rackmount
Options Enhanced Aanalysis, Synchroscope, kit, programmer and service manual,
Direct to disk various carrying cases
Price (rrp) £5045 (€7316) (Abtronix) Price (rrp) £2020 (€2930), TDS2022: 2 channels,
(From the same series: 5 MHz/10 MSa/s and 25 MHz/100 MSa/s £1731 / €2510 (ex. VAT, CN Rood)
models) (From the same series: various 60-,100- and 200-MHz models)

10/2004 - elektor electronics 19


Yokogawa Agilent
DL1740EL 54642D

Brand and Model Yokogawa DL1740EL Brand and Model Agilent 54642D
Display and type LCD (colour), digital Display and type CRT (monochrome), digital
Channels 4 Bandwidth 500 MHz (–3 dB, DC)
Bandwidth 500 MHz @ 10 mV - 10 V/div, 400 MHz Channels 2
@ 2 mV - 5 mV/div (@50 Ω, –3 dB, DC) Zin 1 MΩ (±1 %) / 50Ω (±1 %)
Zin 1 MΩ (±1 %) / 50Ω (±1 %) Y range 2 mV - 5 V / div (±2 % full scale)
Y range 2 mV - 10 V / div (±1.5 % • 8 div) Max. input 400 V @ 1 MΩ / 5 Vrms @ 50 Ω
Max. input 400 V @ 1MΩ / 10 V @ 50 Ω Rise time approx. 700 ps (=0.35 / Bandwidth)
Rise time 700 ps (=0.35 / Bandwidth) X range 1 ns - 50 s / div (±0.005 % value /
X range 1 ns - 50 s / div ±0.1 % screenwidth / ±20 ps)
Triggering auto, normal, single, event/time-delay, Triggering auto, normal, single, edge, pattern,
edge, pattern, pulse width, (HD)TV, I2C, SPI pulse width, TV, duration, sequence,
Delay range 0-4s I2C, SPI, CAN, LIN, USB
Trigger inputs ch1- ch4, line, ext Delay range 60 ns - 10 s
Sample rate 8 bits @ 1 GSa/s (real-time) Trigger inputs ch1, ch2, ext
Memory 4 MW (per channel) 8 MW (interleaved) Sample rate 8 bits @ 1 GSa/s (real-time) (2 GSa/s
Acquisition normal, envelope, average, +, –, *, real time interleaved)
binary operations, inverted, differenti- Memory 4 MB per channel
ate, integrate, FFT Acquisition normal, envelope, average, peak detect
Miscellaneous I2C-, CAN-, SPI-analysis, USB, VGA, GPIB (1 ns), math, binary operations, FFT
Dimensions 220 x 265.8 x 264.1 mm (W x H x L) Miscellaneous 16 dig. inputs, set-up and waveform
Weight approx. 5.4 kg save/recall, RS232, Centronics, trigger out
Including user manual, one probe per channel, Dimensions 322.6 x 172.7 x 317.5 mm (W x H x L)
floppy or card interface Weight 6.82 kg
Options RS232 and Ethernet, inbuilt printer, Including user manual, various probes, software &
probe power supply, various probes RS232 cable
Price (rrp) approx. £5862 (€ 8500) (Yokogawa Options screening, rackmount kit, cover, printer,
Europe) carrying case, GPIB interface
(From the same series: comparable models with fewer Memory/other Price (rrp) £6436 (€9332) (£4614 / 6691 with-
options) out digital inputs, prices ex. VAT, Agilent)
(from the same series: various 60-, 100- and 350-MHz instruments.
Also 600-MHz and 1-GHz models)

LeCroy Tektronix
Wavesurfer TDS7154B
424
Brand and Model LeCroy Wavesurfer 424 Brand and Model Tektronix TDS7154B
Display and type LCD (colour, 800 x 600), digital Display and type LCD (col. touchscreen, 1024 x 768),
Bandwidth 200 MHz (–3 dB, DC) dig.
Channels 4 Bandwidth 1.5 GHz (–3 dB, DC)
Zin 1 MΩ / 16 pF, 50Ω (±1 %) Channels 4
Y range 1 mV - 10 V / div (±1.5 % value + 0.5 Zin 50 Ω (±2.5 % @ 25 °C)
% full scale) Y range 2 mV - 1 V / div (±2.5 % - 3 %)
Max. input 400 V @ 1 MΩ / 5 Vrms @ 50 Ω Max. input 1 VRMS @ < 100 mV/div,
Rise time 175 ns 5 VRMS @ ≥ 100 mV/div
X range 1 ns - 1000 s / div Rise time approx. 200 ps
Triggering auto, normal, single, stop, edge, glitch, X range 50 ps - 10 s/div
pattern, pulse width, TV, line Triggering auto, edge, glitch, div. adv. triggers
Delay range 2 ns - 20 s Delay range 5 ns - 250 s
Trigger inputs ch1, ch2, ch3, ch4, ext Trigger inputs ch1 - ch4, line, ext
Sample rate 8 bits @ 1 GSa/s (real-time) ) (2 GSa/s Sample rate 8 bits @ 1 ch @ 20GSa/s, 2 ch @
real-time interleaved) 10GSa/s, 3-4 ch @ 5 GSa/s (real-time)
Memory 250 kpts @ 3-4 ch, 500 kpts @ 1-2 ch Memory 4M @ 1 ch, 2M @ 2ch, 1M @ 3-4 ch
Acquisition normal, envelope, average, FFT, math Acquisition normal, envelope, average, peak detect
Miscellaneous set-up and waveform save/recall, auxil- (> 400 ps), math, spectrum analysis
iary- and DC output, e-mail, Centronics, Miscellaneous set-up and waveform save/recall, trigger-
Ethernet, USB, VGA, RS232 reference- and ch3 output, Centronics, Eth-
Dimensions 322.6 x 172.7 x 317.5 mm (W x H x L) ernet, GPIB, USB, 2 x VGA, RS232
Weight 6.82 kg Dimensions 455 x 287 x 435 mm (Wx H x L)
Including user manual, measurement probes, soft- Weight 17.7 kg
ware Including incl. manual, two probes, software,
Options additional memory, advanced trigger- SMA- and BNC input modules
ing, extended math Options incl. memory, probes, trigger functional-
Price (rrp) £3070 (€4450) (ex. VAT, EMV Benelux) ity, software modules
(From the same series: various 200-, 350- and 500-MHz instruments) Price (rrp) £18034 (€26150) (ex. VAT, CN Rood)
(from the same series: various 500-MHz, 1-, 2,5-, 4- and 7-GHz instru-
ments)

20 elektor electronics - 10/2004


LeCroy Agilent
WavePro 54843A
7200
Brand and Model LeCroy WavePro 7100 Brand Agilent
Display and type LCD (colour touchscreen, 800 x 600), Model 54843A Infiniium
digital Display and type LCD (colour, 640 x 480), digital
Bandwidth 1 GHz (–3 dB, DC) Bandwidth 2.5 GHz (–3 dB, DC)
Channels 4 Channels 4
Zin 1 MΩ / 11 pF, 50 Ω (±1 %) Zin 50 Ω (±1 %)
Y range 2 mV - 1 V / div (±1 % typical) Y range 1 mV - 1 V / div (±2 % full scale)
Max. input 100 V @ 1 MΩ / 5 Vrms @ 50 Ω Max. input 5 Vrms @ 50 Ω
Rise time approx. 225 ps Rise time approx. 168 ps
X range 20 ps - 10 s / div X range 10 ps - 20 s / div (timebase accuracy:
Triggering edge, slope, line, glitch, pattern, width, 1 ppm)
state, logic, hold-off (20 s) Triggering auto, normal, single, edge, glitch, line,
Delay range -hor. timescale - 10000 divisions state, delay, pulse width, transition
Trigger inputs ch1 - ch4, line, ext. Delay range 80 ns - 320 ms
Sample rate 8 bits, 1-2 ch @ 20 GSa/s, 3-4 ch @ Trigger inputs ch1 - ch4, line, ext
10 GSa/s (real-time) Sample rate 8 bits @ 20 GSa/s (real-time per chan-
Memory 1Mpts @ 3-4 ch, 2 Mpts @ 1-2 ch nel)
Acquisition normal, envelope, floor, averaging Memory 262 k per channel
Miscellaneous auto set-up, auxiliary input and output, Acquisition normal, average, peak detect, +, –, *,
floppy drive, Centronics, Ethernet, USB, /, invert, differentiate, integrate, FFT
VGA, RS232 Miscellaneous set-up and waveform save/recall,
Dimensions 397 x 264 x 491 mm (W x H x L) RS232, Centronics, GPIB, trigger- and
Weight 18 kg timebase output, VGA, VGA (waveform)
Including user manual, cover, power cord Dimensions 437 x 216 x 440 mm (W x H x L)
Options incl. additional memory, probes, Weight 13 kg
advanced triggering, extended math, fix- Including user manual, power cord, software
ing material, GPIB Options incl. rackmount kit, various software
Price (rrp) approx. £15835 (€22950) (ex. VAT, extensions
EMV Benelux) Price (rrp) £19488 (€28259) (ex. VAT, Agilent)
(from the same series: 1- and 3-GHz models) (From the same series: 2-, 4- and 6-GHz instruments)

Gameboy oscilloscope
Did you know Elektor Electronics supplies a homebrew
digital memory oscilloscope? A brilliant design by Steve
Willis, the GBDSO (Gameboy Digital Sampling Oscillo-
scope) was published in the October and November
2000 issue. It is an extension card that turns the Game-
boy Classic, Pocket, Colour or Advanced into a digital
‘scope. The GBDSO rocked the electronics hobby world
and that’s why we continue offering ready-built GBDSO
modules for just £90 plus P&P. Of course, the GBDSO is
no match for any of the digital oscilloscopes discussed
in this article, but it is and remains a useful little instru-
ment with an excellent price/performance ratio. For
those who do not remember the above articles, the
main GBDSO specs are:

– Dual trace display


– Bandwidth 100 kHz
– Sample-rate max. 1 MSa/s
– Timebase 5 µs - 100 s/div
– X-Y recorder
– Real-time FFT-mode with dB scale
– Adjustable persistence in X-Y mode
– PC link for image and or date transfer
– 5-hour operation from NiMH battery pack
– Averaging and auto-trigger functions
– Reference trace storage
– Available ready-built and tested
from Elektor Electronics

10/2004 - elektor electronics 21


About the selection
Elektor Electronics has attempted to make a selection of relevant products that is representative for the current market situation while
including the availability factor. We realise this overview is not complete — space restrictions for printed articles had to be
observed. For completeness’ sake the table shows an overview of brands and websites of associated manufacturers, importers or
national representatives. Not all brands and models are available in the UK or other countries where Elektor Electronics is pub-
lished. A list showing a complete overview of all current models supplied by the manufacturers participating in this test may be
found on our website.

Manufacturer Website Representative


Agilent Technologies www.agilent.com www.agilent.co.uk
B&K Precision www.bkprecision.com www.metrix-electronics.com
Chauvin Arnoux (Metrix) www.chauvin-arnoux.com www.chauvin-arnoux.co.uk
EZ Digital (Goldstar, LG Precision) www.ezdgt.com/eng www.sjelectronics.co.uk*
www.conrad.com*
Grundig-Digimess www.digimess.co.uk www.digimess.co.uk, www.telonic.co.uk
GW Instek www.goodwill.com.tw www.sjelectronics.co.uk
www.conrad.com*
Hameg www.hameg.com www.sjelectronics.co.uk
Hitachi Denshi www.hitachi.com www.sjelectronics.co.uk
Iwatsu Test Instrumenst www.iti.iwatsu.co.jp/e/
Kenwood (Tecstar) www.tecstar.co.uk www.tecstar.co.uk
Leader Instruments www.leaderusa.com www.elquip.nl
LDS Nicolet www.lds-group.com www.gould-nicolet.co.uk
LeCroy www.lecroy.com www.lecroy.co.uk
PeakTech www.peaktech.de www.peaktech.de
Protek (Hung Chang) www.protektest.com
Tektronix www.tektronix.com www.tektronix.co.uk
Voltcraft www.conrad.nl www.conrad.com
Wittig Technologies www.wittigtechnologies.com
Yokogawa www.yokogawa.com/tm/ www.yokogawa.co.uk
* some models under different brand names

Test method that. In this regard, their prices are an excellent match to
their performance.
Due to the nature of the selection, in this comparison the
emphasis by definition cannot lie on the ‘standard’ speci-
fications. We thus didn’t verify any input sensitivities or No-nonsense two-channel scopes
bandwidths. However, that’s not strictly true — we were Next we have a set of five 20- and 30-MHz oscillo-
indeed curious with regard to a few instruments, but scopes. The GW Instek GOS620 has a convenient
more about that later. layout and is very pleasant to use. The image of the test
In the context of a test such as this, there’s also not much signal is especially stable and sharp, but the display is
point in comparing specs. That’s because manufacturers quite a bit less bright than that of the Digimess MO20,
in this segment think twice before stating a specification for example. The latter also has an adjustable trigger
they can’t prove. delay (‘holdoff’ function). This can be handy for display-
So how did we actually do the survey? We connected a ing complex signals, since it allows an event that would
PAL video signal (from the Test Pattern Generator in the otherwise generate a trigger to be skipped. That’s a nice
September 2003 issue of Elektor Electronics) to each of bonus for a scope in this price class. However, we must
the oscilloscopes. This is a relatively complex signal, so note that this instrument makes a somewhat less sturdy
we had to spend some time with each instrument before impression than its direct competitor, and the display of
obtaining an image on the screen. That proved to be a the test signal is somewhat less steady than with compa-
good way to quickly obtain an impression of the general rable scopes.
quality and usability of the instrument in question. In the The Hameg 303-6 is also very easy to use, although it
following descriptions, we give our opinions regarding is annoying that the focus setting also changes (consider-
image quality, ease of use and triggering performance, ably) when the intensity is changed. Most of the other
as well as our general impression of the instrument’s relia- instruments suffer far less from this problem. Other than
bility. that, this model is especially sturdy; it gives the impres-
sion that it will never break. It also includes the holdoff
function.
The pipsqueaks The EZ Digital OS-5030 does not have trigger holdoff,
Let’s start at the bottom end of the selection with two sin- but it is still an excellent instrument. Ease of use is right
gle-channel 10-MHz oscilloscopes, the Voltcraft 610-3 on, and the signal display is bright and clear. As already
and the Digimess MO10. In all honesty, we must admit mentioned, this scope is also available from Conrad,
that we had certain preconceptions here. We assumed Dynatek and Peaktech under their own brand names and
these instruments wouldn’t have all that much to offer. model numbers.
That turned out to be only partly justified. Although their Finally there is the B+K Precision 2120B. This scope
miniature screens are far too small for making serious looks quite robust, and the display is basically good.
measurements, they accepted the video signal without However, the sample we received for examination had
any complaints. They are thus suitable for quickly obtain- an interference signal that ran across the display. Unfor-
ing an impression of a signal, but not anything more than tunately, we were unable to discover the cause, but we

22 elektor electronics - 10/2004


but a 100-MHz scope with cursors is
also available under this brand name.
With regard to the LS8105A, we
noted that it is pleasantly easy to use,
solidly built and has outstanding
image quality. The accompanying
manual is not brilliant, but that’s actu-
ally the only criticism.

Digital or analogue?
The Voltcraft 650AD and Hameg
1507-3 are combined instruments.
They are actually ‘normal’ 50- or 100-
MHz scopes (respectively) with addi-
tional digital storage scope capability.
This directly overcomes one of the
most significant drawbacks of digital
scopes, which is that a purely digital
oscilloscope can never display the
actual waveform. What it displays is
always a reconstruction of a sampled
signal. As a result, it can happen that
the image on the screen has no
resemblance at all to the measured
signal. With a combined scope, you
can always quickly check whether the
digital result is accurate.
In this regard, we must note here that
the digital capability of the combina-
tion models included in this test is only
limited. They have several cursor func-
tions for making measurements and
assume it was something particular to this specific unit. In can be operated via a PC, but the sampling rate is actu-
other respects, this instrument is just as good as the others ally too low for good reproduction of a signal such as
with regard to features and operation. our test signal. In practice, the specified bandwidth is
thus only usable in the analogue mode. Another factor is
that the noise contribution from the A/D converters is
More megahertz generally rather large. This causes the traces of digital
Next we have three analogue oscilloscopes with a band- oscilloscopes to be ‘fatter’, so details more quickly
width of 100 MHz. As expected, all three perform a become literally lost in the noise.
good deal better than the less expensive instruments. This Other than that, there are no remarkable differences or
shows once again that it’s too simple to just compare details to be reported with regard to these combined
specifications. We’d like to reiterate that it’s quite proba- instruments, and the performance of the analogue portion
ble that the models of the brands in the previous group is good.
with bandwidths of 100 MHz (or more) would also per-
form much better.
The larger bandwidth increases the quality of the dis- Digital storage oscilloscopes
played test signal. Not only is the image sharper, which The Wittig 22-300, the simplest instrument in the ‘pure
makes it easier to distinguish rapid changes, but the writ- digital’ category, is limited with regard to more than just
ing speed of these instruments is also quite a bit higher. its bandwidth and sampling rate. Its operation, measure-
The display is refreshed more often, making the image ment capabilities and display fall short of qualifying it for
considerably more steady on the screen. serious use. This is offset by its modest price. However,
The Hameg 1004-3 and GW Instek 6103 are you would be wise to consider purchasing a good ana-
microprocessor-controlled. The advantage of this is that logue instrument for the same money. If you specifically
certain settings can be stored in memory. In situations need storage capability, it’s worth giving thought to sav-
where a number of similar signals must be regularly mea- ing up for an instrument with greater capability.
sured, that can save a considerable amount of time. The same feeling prevails with the Voltcraft 6150,
However, it also has drawbacks. For instance, it some- although it has significantly more extensive features. It
times takes a bit of searching to find out exactly how the has a variety of measurement functions, including spec-
instrument is configured, and displaying the various set- tral analysis using fast Fourier transform (FFT), that can
tings on the screen affects the displayed waveform. The be used via a clear, well-organised menu. Nevertheless,
latter factor is sometimes annoying with the 6103. we could not obtain a good display of the test signal, but
The Hameg and GW Instek instruments also allow mea- as already noted, that can’t be expected from an instru-
surements to be made on the displayed waveform using ment with such a limited sampling rate.
cursors. We think that once you’ve worked with cursors, The next instrument, the LDS Sigma 60-4, is some-
you’ll miss them if your scope doesn’t have this feature. If what unique. Although the sampling rate of this instru-
your budget permits it, this function is thus to be recom- ment is 200 MSa/s, which cannot be considered to
mended. especially high, it does have several unusually useful
The Leader LS8105A does not have cursor capability, features. The Sigma is particularly suitable for long-term

10/2004 - elektor electronics 23


signal analysis. This is demonstrated by features such as instrument is perfectly in order.
hard-disk recording, extensive triggering options, analy- Actually, that’s true of all of the brands at this level and
sis functions and report-generation options. This is also above — the technical aspects are not the issue. The
the first instrument in this test that is built on a PC plat- most significant differences among the various instru-
form. ‘Standard’ MS Windows is used as the operating ments lie in their housing, operation, features and display
system, so data from the measurement programs can be quality, among other things.
easily transferred to other programs (including user-writ- The Agilent 54642D has remarkably good display
ten programs). quality for a digital scope, despite the fact that it uses a
Despite the fact that the acquisition software runs under relatively small monochrome screen. However, the on-
Windows, a non-standard menu structure has been cho- screen menu could have been made a bit smaller; we
sen. We don’t consider this to be a logical choice, since found it rather obtrusive. In contrast to the -A version of
it costs significant extra effort to become adept at oper- this Agilent model, the -D version has a full sixteen digital
ating the scope. This is further complicated by the fact inputs in addition to two analogue inputs. Besides stan-
that only a few real knobs are present for changing set- dard triggering, this scope can also work with I2C, SPI,
tings. These knobs are constantly assigned different func- CAN, LIN and USB signals.
tions, depending on the menu selected using the touch The final member of this group is the LeCroy
screen (or the keyboard and mouse). We don’t consider Wavesurfer 424. We are enthusiastic about this instru-
this convenient. ment, although it also has its own peculiarities. For
The Tektronix 2024 is the first digital scope in this test instance, the automatic calibration feature often ‘gets in
with a serious sampling rate — no less that two gigasam- the way’, although the frequency of automatic calibration
ples per second. This is where the concept ‘digital’ starts does decrease after the instrument reaches its operating
to become attractive. All of its functions — triggering, temperature. In addition, it can be switched off for mea-
measurement options, etc. — are just right. This full- surements where accuracy is not important. We were
grown digital scope is also packaged in an especially pleased that we could quickly figure out how to do this,
small housing. However, you should bear in mind that without using the manual. That’s probably in part due to
with regard to screen resolution, this instrument is far the fact that a ‘Windows-like’ user interface was chosen
from being the equal of a good analogue scope. for this instrument. As a result, using the Wavesurfer is
Next we have the Yokogawa DL1740EL. This scope generally reasonably intuitive. The display is excellent and
is also packaged in an unusual enclosure with an attrac- generously sized. Using the extra functions, such as zoom-
tive colour screen. In addition, the sample we received ing in on a portion of the signal, is thus a real pleasure.
was equipped with a built-in thermal printer. Like the The instrument could handle our test signal without any
Sigma 60-4, this instrument is based on a PC platform problems, although when we experimented with the
and has its own operating system. Although some of the mathematical functions we noticed that everything slowed
knobs are also assigned different functions depending on down. That’s hardly surprising when you consider how
the menu selection, it has dedicated knobs for the basic much data has to be processed every second.
settings (unlike the Sigma scope). It is thus possible to Ultimately, all of the instruments in this group suffer more
quickly learn how to use the instrument, despite its exten- or less from this problem. If you’re considering purchas-
sive features. ing an oscilloscope in this price class, we recommend
This instrument is also distinguished by extensive trigger- making comparative tests of several different instruments
ing options. For instance, it can be triggered on a spe- using your own specific measurement setups.
cific I2C pattern, which allows the values of the bits to be This advice is certainly applicable to the final three instru-
viewed directly. Triggering on the test signal did not pre- ments in this test: the Agilent Infiniium 54843A,
sent any problems. With regard to technical aspects, this LeCroy Wavepro 7100 and Tektronix TDS7154B.

24 elektor electronics - 10/2004


More than 35 gigasamples per second
The three top-end models in the test collectively achieve a sampling rate of no less than 35 GSa/s. The Agilent Infiniium 54843A,
LeCroy Wavepro 7100 and Tektronix TDS7154B are all high-quality instruments that are typically used in research and develop-
ment environments, such as in universities and the high-tech industry. The fact that these instruments are intended to be used for spe-
cific measurements on very broadband signals can for example be seen from the fact that the standard versions of the Agilent and
Tektronix instruments have only 50-Ω inputs. This means that normal probes cannot be connected to them. You should pay attention
this when purchasing an oscilloscope, as good probes can take up a significant portion of the budget – as much as 20 per cent or
more. In this area, DIY pays off, see Poor Man’s 1-GHz Active Probe design elsewhere in this issue.
With the rest of the scopes we did not check any of their specifications, but with these instruments we couldn’t restrain our curiosity.
After all, bandwidths on the order of several gigahertz are not a matter of course, even with modern technology. However, our
measurements showed that all three of these instruments easily meet (or even surpass) their specifications.

Agilent Infiniium 54843A


This instrument has the smallest housing of them all, its design is relatively simple, and the front
panel has a good, well-organised layout. It’s a pity that the screen is too small for the opening in
the front panel, so the metal surround is visible. This may not matter to its performance, but it’s still
a bit untidy for an instrument in this price class. The screen has the lowest resolution of the three,
but it is the brightest. The user interface, in which a lot of the ‘Windows look’ can be seen, does
not give the impression of being particularly well thought through. The response time is short, and
the most important settings are readily accessible. However, it is often necessary to use the mouse
to adjust the settings.
The triggering options are not especially extensive, but they should be adequate for most applications. Several standard auto-
matic measurement functions are present, such as measuring and analysing amplitude, time, frequency, eye patterns and jitter.
Besides the standard mathematical functions that can be used for signal processing, several analysis options are available for
this instruments.
Signals can be displayed in up to four separate grids (multi-grid), or in a maximum of two grids using the megazoom function.

Tektronix TDS7154B
The housing of ‘the Tek’ is significantly larger, and the layout of the front panel is rather crowded.
The knobs appear sturdy, but they are somewhat difficult to operate. The display is large, but unfor-
tunately the reproduction of the waveform does not match the high resolution of the screen. All func-
tions can be controlled using the touchscreen. The response time is reasonably good, but it some-
times takes a rather large effort to adjust even simple settings. The instrument can also be operated
using an optional mouse and keyboard. Besides the usual connectors, this instrument has a display
output for the scope image, in addition to the normal VGA output. The triggering options are quite
extensive, as are the automatic measurement functions. Besides the standard mathematical functions,
four functions can be entered as formulas for manipulating signals, functions, measurements or data in the memory.
The signals are shown together in a single grid, and the zoom functions appear together in a larger grid.

LeCroy WavePro 7100


This instrument is packaged in the largest enclosure of all three, and its front panel also has a
rather busy layout. However, this does mean that many functions are directly accessible. The
knobs appear simple, but they are quite pleasant to use. The user interface is well organised and
appears to be carefully thought through, down to the last detail. The screen is large, and the dis-
play is well matched to the screen resolution. All functions can be readily controlled using the
touch screen. The ultimate result is very nice. Once you’ve gotten accustomed to it, you won’t
want to use anything else. The scope can also be operated using an optional mouse and key-
board. Just like its little brother, the Wavesurfer, this instrument occasionally has a somewhat long
response time due to constant recalibration. This can of course be disabled, but by default it is enabled. This approach is differ-
ent from the other instruments, for which automatic calibration is disabled by default. LeCroy clearly assumes that you want the
measurement results to be accurate!
Just like the other two scopes, this instrument runs on a PC platform with Windows. This works quite well, but we do consider a
working memory of 256 MB for this LeCroy to be on the meagre side.
The input channels are suitable for 1 MΩ as well as 50 Ω. This means you have a bit more choice in your connection options. The
triggering options are quite extensive, with a particularly notable feature being that the Aux input can be used as an extra channel
for all triggering. The automatic measurement options are also very extensive, and they are divided into amplitude, time, custom,
disk, eye, jitter, power, pulse, statistical and miscellaneous measurements. The mathematical functions provide extensive options for
manipulating signals, functions, measurements, or data in the memory. This signals are displayed in up to eight separate grids.

Conclusion
It’s clear that all three instruments have both positive and negative aspects. The Agilent is quite fast with regard to bandwidth rela-
tive to the number of channels and sampling rate, but it falls short of the other two with regard to features, ease of use and finish.
The Tektronix is fast in operation and has an especially large number of features. Our only real complaint is that touchscreen opera-
tion could be improved, and it’s a pity that the instrument doesn’t do justice to the resolution of the screen. Finally, the LeCroy is
sometimes somewhat slow, but has more flexible features and is more nicely finished than the other two.
We’d like to again emphasise that these three instruments are more analysis instruments than oscilloscopes. They are intended to be
used to quickly execute highly complex measurements and calculations. There is a comparable instrument in the lab of one of our
test engineers, and even a year after acquiring it they are still discovering new functions! The descriptions of these instruments in
this article are thus only intended to serve as an initial introduction. We recommend that you allow yourself to be thoroughly
informed by the importer of manufacturer if you contemplate purchasing such an instrument.

10/2004 - elektor electronics 25


rate is significantly lower.
You also have to pay attention to the memory depth. The
larger the memory, the higher the sampling rate can be
for a relatively long time interval (time base range). But a
large memory is not always necessary, and sometimes it
can even be undesirable, since processing an enormous
amount of data generally slows down the operation of
the oscilloscope.
You should also pay attention to the trigger options, and
with digital oscilloscopes in particular, the functions that
allow infrequent events (glitches) to be detected. Here the
instrument’s performance also depends on the sampling
rate and resolution in relatively large time intervals.
Finally, you should also consider the (mathematical)
analysis functions and report-generation features. Several
basic functions are present with most of the manufactur-
ers, and it is often possible to use supplementary soft-
ware modules to expand the available features.
Both the analysis software and the firmware of most digi-
tal oscilloscopes are often subsequently upgraded, so it’s
a good idea to keep track of these updates.
Incidentally, in the more expensive instruments all of the
options are often already present, and they only have to
be activated using the appropriate software keys. With
digital oscilloscopes that run on a Windows platform, it is
often possible to also run ‘normal’ software, such as Mat-
lab, on the same platform. This can also form a way to
Although even more expensive oscilloscopes are avail- add extra functions.
able, for this test we considered them to be the top end
of the normal market. For this reason, we gave more
attention to them than to the rest of the selection. The Conclusion
results are presented in a separate inset. We don’t want to claim that choosing an oscilloscope is
an easy task, even with this survey to help you. As
already mentioned, it’s a good idea (or even essential) to
Recommendations base your choice on your specific measurement tasks,
To start with, it is important to determine how much band- taking future wishes into account as well. This survey and
width and how many channels you need for the measure- the reported results should then help you to more quickly
ments you will be making – both now and in the future. sort through the available options.
After all, an oscilloscope’s bandwidth and number of The test results show that in particular, all of the relatively
channels cannot be upgraded later on. The next thing to simple digital oscilloscopes are still surpassed by practi-
consider is that an oscilloscope cannot be ‘interfaced’ to cally all analogue instruments with regard to display
the object to be measured without the proper probes and capabilities. To start with, in our view buying a digital
adapters. You are thus well advised to reserve a portion oscilloscope is only justified if you have a particular need
of your budget for such items. for specific triggering and/or memory functions. This will
It is also important to pay attention to the ‘lifetime’ of the in particular be true if you have to make measurements
instrument, with regard to the availability of spare parts. on signals in which relatively infrequent special events
The question of whether you should select an analogue occur, or if you want to use the oscilloscope for auto-
scope or a digital scope depends on what it will be used mated measurements and reports, possibly even operat-
for. Digital instruments suffer from noise, which can cause ing under remote control.
waveform details to be lost. However, a good digital For ‘garden variety’ applications, if you’ll pardon our use
instrument can have quite extensive memory and trigger- of the term, buying a combined oscilloscope is worth seri-
ing functions, while analogue instruments are clearly lim- ous consideration.
ited in this regard. Regardless of which instrument you’re interested in, if at
If you decide on a digital storage oscilloscope, when all possible you should first try it out in your own mea-
determining the required sampling rate you should bear surement setup. With the more specialised instruments,
in mind that many of these instruments can measure in this is practically unavoidable, since the suppliers will be
two different modes. The real-time mode is specifically only to pleased to demonstrate all the features to you.
intended to be used for making single-shot measurements We certainly consider this to be a good idea, since it
up to the physical sampling rate limit of the A/D convert- should now be clear that although most oscilloscopes are
ers. In the repetitive mode, a periodic signal can be sam- handy and versatile instruments, each one has its own
pled at a higher rate than what the A/D converters can particular faults and shortcomings.
actually handle. This involves making repeated real-time (040274-1)
measurements at intervals that are successively offset by
a small fraction of the period of the waveform. As the sig- Elektor Electronics would like to thank the University of
nal is repetitive, the results of these measurements can be Twente, the Netherlands, and in particular the Chair of
combined. As a result, it appears that the signal was Signals and Systems and the Scintilla study association,
sampled at a (much) higher sampling rate. for their cooperation and contributions to this project.
But note that this is strictly limited to repetitive signals. Naturally, we would also like to thank all of the partici-
Some manufacturers are occasionally inclined to adver- pating manufacturers, importers and suppliers for making
tise this higher sampling rate, although the true physical the oscilloscopes available.

26 elektor electronics - 10/2004

You might also like