Saturday 28 January 2012

Ten Points To Consider When Purchasing Your Next Oscilloscope

Ten Points To Consider When Purchasing Your Next Oscilloscope
by Robert Lashlee, Learning Products Engineer, Agilent Technologies


You need to choose an oscilloscope that enables you to complete your work in an efficient and
accurate manner. However, selecting such a scope can be a daunting task.

Comparing the specifications and features of oscilloscopes made by different manufacturers is
often time-consuming and confusing. Here are ten concepts that can speed the selection process and
help you avoid some common pitfalls.


Point One: Bandwidth

Bandwidth is the single most important property of an oscilloscope. It determines the range of
signals that can be displayed. It also dictates, to a large extent, the price of the oscilloscope.

To ensure an oscilloscope has enough bandwidth, take into account the bandwidths of the signals
you expect to display with it. The system clock is usually the highest-frequency signal the scope is
likely to display. As such, your oscilloscope should have a bandwidth at least three times greater
than your clock frequency, in order to obtain a reasonable display of the waveform.

Another characteristic that dictates bandwidth needs is signal rise times. Since it's likely you'll be
displaying more than pure sine waves, your signals will contain harmonics at frequencies beyond
the fundamental. If you don't ensure proper oscilloscope bandwidth, rounded edges will be
displayed, instead of the clean fast edges you're expecting. This, in turn, will affect the accuracy of
your measurements.

Fortunately, there are three very simple equations that will help in determining the proper
oscilloscope bandwidth, given your signal characteristics (see Fig. 1).



Fig. 1. Three Key Equations To Figure Out Your Required Bandwidth Needs
It's important to project into the future when determining your bandwidth needs, as that will
probably change over the lifetime of the oscilloscope. Agilent Technologies Infiniium Series
oscilloscopes, for example, mitigate this problem by letting you upgrade to various bandwidths as
your needs change.


Point Two: Number Of Channels

It's important to analyze your work to accurately predict the number of channels that will be
required. Digital content is everywhere in modern designs and conventional 2-channel and 4-channel oscilloscopes don't always provide the required number of channels.

For today's digital world, a new breed of oscilloscope enhances use in digital and embedded debug
applications. These mixed-signal oscilloscopes tightly interleave an additional sixteen logic timing
channels with the two or four channels of a traditional oscilloscope. The result is a fully functional
oscilloscope with up to twenty channels of time-correlated triggering, acquisition, and viewing.


Point Three: Sample Rate

It's important to note that most oscilloscopes can increase sample rate by incorporating a form of
interleaving. This is accomplished when two or more channels couple their ADCs to provide a
maximized sample rate on only one or two channels of a 4-channel oscilloscope. Keep in mind that
the banner specification of the oscilloscope will usually emphasize this maximized sample rate: it
will not state that the sample rate applies to only one channel.

From Equation 3 (see Fig. 1, again) the sample rate of an oscilloscope should be, at a minimum,
four times greater than its bandwidth. For a 12-GHz oscilloscope, the minimum per-channel sample
rate to support the full bandwidth on each channel equals 4 x 12 GHz, or 48 Gsample/s/channel.

A 12-GHz oscilloscope maker may advertise a maximum 64-Gsample/s sampling rate, but fail to
point out that that sampling rate is applicable on one channel only. The per-channel sample rate of
this oscilloscope, when using either three or four channels, would be insufficient to support the
12-GHz bandwidth on more than a couple of channels.

So, make sure the oscilloscope you consider has enough sample rate per channel for every channel
that may be used simultaneously.


Point Four: Memory Depth

In an oscilloscope, an ADC digitizes the input waveform. The resulting data is stored into high-speed memory. An important selection factor is to understand how the oscilloscope uses this stored
information.

Many engineers assume an oscilloscope's maximum sampling rate spec applies to all timebase
settings. In actuality, the memory depth is limited and, therefore, all oscilloscopes must reduce their
sampling speed as the timebase is set to wider ranges.

The deeper the oscilloscope's memory, the more time can be captured at full sampling speed. You
need to check the oscilloscope you're considering to see how its sampling speed is affected by the
timebase setting.

The required memory depth you need is dependent on the amount of time you want displayed, as
well as the sample rate you want to maintain. If you're interested in looking at longer periods of
time with high resolution between points, you need deep memory. A simple equation can tell you
how much memory you'll need, given time span and sample rate (see Fig. 2).



Fig. 2: An Equation Relating Memory Depth To Sample Rate

Once you've determined your memory depth, it's equally important to see how the scope operates
when using the deepest memory setting. Oscilloscopes often respond sluggishly, which can
negatively impact productivity. Before you purchase an oscilloscope, make sure to evaluate its
responsiveness in its deepest memory setting.


Point Five: Update Rate

One major factor in the quality of an oscilloscope display is its update rate. Update rate refers to the
rate at which the scope is able to acquire and update the display of a waveform.

Faster update rates improve the probability that infrequent events, such as glitches, are captured.
Agilent's InfiniiVision 7000 Series oscilloscopes, for example, have an update rate of up to 
100 k waveforms/s.

Be careful when comparing update rates. Vendors quote the best possible update rate their
oscilloscopes can achieve. However, special acquisition modes are often required to obtain these
banner specifications.

These special modes can severely limit the performance of an oscilloscope in areas such as memory
depth, sample rate, and waveform reconstruction.


Point Six: Triggering Capabilities

Edge triggering is widely use by general-purpose oscilloscope users. However, it may be useful to
have additional triggering power in some applications.

For serial designers, some oscilloscopes are equipped with serial triggering protocols, meeting
standards such as SPI (Serial Peripheral Interface), CAN (Controller Area Network), USB
(Universal Serial Bus), I
2
C (Inter-Integrated Circuit, pronounced I-squared-C), FlexRay, or LIN
(Local Interconnect Network). Advanced triggering options can save a significant amount of time
in day-to-day debugging.

What if you need to capture an infrequent event? Glitch triggering permits you to trigger on a
positive-going or negative-going glitch, or on a pulse greater than, or less than, a specified width.
Additionally, many scopes provide triggering capability for TV, HDTV, and video applications.

Agilent's 90000A Series oscilloscopes will eventually have the industry's only 3-level sequence
trigger, in the company's InfiniiScan Plus system. This is a hardware/software trigger system.

Overall, it's important to anticipate the kind of events you will need to trigger on and then make
sure your oscilloscope has these triggering capabilities.


Point Seven: Probing Capabilities

The probe you choose is important because system bandwidth – the bandwidth of the
oscilloscope/probe combination – is limited by the lesser of the two bandwidths. Consider, for
example, a 1-GHz oscilloscope coupled with a 500-MHz passive probe. With this combination, you
will not be able to obtain the full bandwidth of the 1-GHz oscilloscope, but will instead be limited
to the 500-MHz allowed by the probe.

Additionally, every time you connect a probe, the probe becomes part of the circuit under test. The
probe tip is basically a short transmission line that can load your DUT (device under test).

Consider active probes. They not only provide greater bandwidths than passive probes, but can also
mitigate some of the transmission-line effects. Some Agilent Technologies probes minimize signal
loading, and resulting signal distortion, by incorporating resistive damped tips. These damped
accessories prevent resonant LC tank-circuit impedance from going too low – thus preventing
ringing and signal distortion caused by loading a signal.

You also want to ensure your probe is capable of full bandwidth even when you're using probe
head accessories. Agilent InfiniiMax probes use a single amplifier. They let you connect a variety
of differential or single-ended probe heads, while still obtaining full system bandwidth.

Point Eight: Connectivity Capabilities

Many digitizing oscilloscopes now have numerous connectivity capabilities. These can include
IEEE-488/GPIB, RS-232, LAN, and USB 2.0 interfaces. If you transfer oscilloscope data often to
your PC, it will be important for your chosen scope to have at least one of these connectivity
options. A built-in DVD-RW or CD-RW drive can also help in transferring data.

Some oscilloscopes, such as the InfiniiVision 7000,  let you export waveform data as an .alb file.
You can then import this file into a logic analyzer application loaded on your PC. This can be very
useful if you're working in a group that is geographically dispersed, as you can send waveforms
captured on your oscilloscope to fellow team members and they can analyze the signals on their
PCs.

Determining, ahead of time, your connectivity options, can drastically reduce the amount of time
you spend transferring and storing data.


Point Nine: Application Software

Automatic measurements, built-in analysis capability, and additional application software can save
time and make your job easier. Math functions, measurement statistics, and FFT (fast Fourier
transform) capabilities are available on most oscilloscopes.

For the power-user interested in waveform analysis, some manufacturers offer software packages
that enable you to customize complex measurements or math functions, and do post-processing
directly from the oscilloscope's user interface. 

Application software can let you make measurements that would otherwise be very difficult.
Agilent's InfiniiScan software, for example, is an identification package that can identify signal
integrity problems by scanning through thousands of waveforms and then isolating anomalies in a
signal.

It's important to investigate what additional software is available, so you don't find yourself needing
a function or measurement that your oscilloscope's software can't handle.


Point Ten: Ease of Use

Think through the previous nine considerations. They can help you narrow the field to a limited
number of oscilloscopes that meet your selection criteria.

Then try them out and perform a side-by-side comparison. If you can, borrow the oscilloscopes for
a few days so you have time to thoroughly evaluate them. This will give you the opportunity to
analyze each oscilloscope's ease of use.

When evaluating ease of use, there are several questions you should ask. Are there dedicated knobs
for often-used adjustments such as vertical sensitivity, timebase speed, trace position, or trigger
level? How many buttons do you need to push to go from one operation to the next? Can you
operate the oscilloscope intuitively while concentrating on your circuit under test? Finding an
oscilloscope that is easy to use can save you a great deal of frustration later.


About the Author

Robert Lashlee joined Agilent Technologies in 2007 as a Learning Products Engineer working on
high-performance oscilloscopes and probes.  Robert obtained his BS in physics from the University
of Central Missouri and his MS in physics from the University of Georgia’s  Center for
Simulational Physics under the direction of Dr David P Landau. His pastimes include fishing and
reading.

1 comment: