Skip navigation
All Places > Keysight Blogs > Oscilloscopes Blog > Blog > 2017 > January
2017

Let’s be honest, oscilloscopes aren’t exactly something you can just pick up and use without some sort of help or prior experience. Typically, it takes a bit of time and practice to understand how to use one correctly and make accurate measurements. We don’t want you wasting time trying to figure out how to use your oscilloscope – you should be able to spend your valuable time testing so you can get your designs to market faster. This means you need the right material that can quickly teach you how to use your scope. The Keysight Oscilloscopes team is constantly working to provide you with the materials you need, and expert help so you can spend more time characterizing your designs. We have a lot of resources for you, some of which you may not even know exist.

 

The Educators Training Kit

 

When I was learning how to use an oscilloscope as a young engineer, the resource I found to be most helpful is the Educator’s Training Kit. Don’t let the name fool you - this is not something that only applies to professors and students. It’s helpful for anyone who wants to learn how to use an oscilloscope. The Training Kit includes fifteen hands-on labs that walk you through how to use various features and applications. The labs range from basic concepts, such as triggering and probe compensation, to more advanced measurements such as capturing infrequent events, gated measurements, and using acquisition modes like peak detect. If you complete all of these labs and you still want more, you can also check out the Advanced Training Guide for your oscilloscope model. This guide starts out by reviewing basics, but also covers more advanced labs, including triggering on logic patterns and decoding serial buses. Most of these labs can be completed using the 11 built-in training signals that come with the Educator’s Training Kit, but some guides also include one or two labs that require you to build a very basic RLC circuit as your device under test (DUT). The kit also includes a slide-set on scope fundamentals so you can make sure you really understand the basics before diving into the hands-on labs. The Educators Training Kit is now a FREE option for all Keysight InfiniiVision oscilloscopes so you can learn more about how to properly use an oscilloscope without having to find extra money in the budget.

 

Built-in Help

 

Throw away that user’s manual! Well, maybe not - it could still come in handy. But seriously, everyone hates having to dig through a user’s manual just to figure out something like what one of the trigger options should be used for. There are so many different options and measurement types on an oscilloscope, including some you probably didn’t even know were there. For example, did you know you can get help in seconds right on the screen of your oscilloscope? If you don’t know what something is, just hold down the front panel key or menu button associated with it for a few seconds, and a built-in HELP screen will appear. These HELP screens provide quick information and set-up tips. Bye-bye, user’s manual.

 

Scope Tips from Experts

 

We’re here to help you. That means not only do we provide you with the most accurate test equipment on the market, but we are also available to help you use it. We are regularly posting new videos, blog posts, webcasts, and application notes to help you characterize your designs more easily.  

 

Our Oscilloscope YouTube channel has quick how-to videos, like the Keysight 2-Minute Guru series, along with detailed demonstrations, like Johnnie Hancock’s demo on NFC Testing. The Digital Design and Test Webcast Series is another free resource that offers live, one-hour presentations by some of our experts on various measurement techniques. If you don’t have time to watch the live stream during the workday, you can always watch (or re-watch) it on-demand at a time that works for you. Some of these recordings can also be found on our YouTube channel.

 

When the experts here aren’t working on new videos for you, they are spending their time writing up new blog posts and application notes. These blog posts are another quick way to learn about a topic, while application notes dive deep into the specifics of a certain measurement applications. The Oscilloscope Learning Center offers help on various topics such as scope basics, probing, and applications, and provides links to related videos and application notes for your convenience.


 

I know what you’re probably thinking right now, “Wow, those sound like great resources, but that’s a lot of websites to constantly be checking for new content.” To make your life easier, you can just like our Facebook page. Every time we post new content on all of those free resources I discussed, we usually post a link on our Facebook page as well. The Keysight Oscilloscope Facebook page is the perfect central place to stay up-to-date on all of the new content available.

 

BenchVue

 

BenchVue is a free software tool that can quickly and easily connect your Keysight instruments to your PC in seconds. You can use it to control your instruments remotely and analyze data from various tests. Not only is it a great extension of your oscilloscope’s functionality, but it also has a helpful education feature, the Library tab. The Library tab allows you to type in any Keysight product number and it will return any documentation it can find on that product such as a data sheet, application notes, video demos, user guides, etc.  This eliminates all of that time you would spend searching the web to find these documents and puts them in one convenient location for you.

 

I hope all of these resources prove to be helpful when learning how to use your oscilloscope. We will continue working hard to give you new content to help you in your testing. If you have any specific questions or content requests, just let us know in the comment section below.

KennyJ

"Hey, I Need a Current Probe"

Posted by KennyJ Employee Jan 18, 2017

When I hear someone say this and I ask “why,” the answer is nearly always “I need to measure how much power this thing is using”.  Based on this experience I have come to consider current probe use as synonymous with power measurements. Usually someone is trying to figure out how much power they are using or determine if the supply they have designed or are using is working properly and efficiently. It seems that sooner or later nearly everyone needs to make a current measurement so I thought I would share some of my experience with current probes and making current measurements (I’ve designed some current probes and have some current probe patents). My goal is to help you understand your options, what they are useful for, and what their limitations or drawbacks are.

 

Of course, one of the most common methods for measuring current is to use a digital multimeter. For this article we’re going to skip the DMM because you probably want to see a waveform of how the current changes over time since you are using an oscilloscope.

 

Let’s start with a review of the types of current probes. There are two classes of measurement capabilities—AC only and AC/DC—and primarily two different measurement methods—magnetic field sensing and Ohm’s law. The most general purpose measurement is the AC/DC measurement so we’ll focus on that. When it comes to magnetic field sensing there is a wide range of options—Hall Effect sensors, transformers, Rogowski coils, giant magnetoresistance (GMR), tunnel magnetoresistance (TMR), and many more cool and exotic-sounding approaches, but the most common is to use a Hall Effect sensor teamed with an AC transformer.  For the Ohm’s law approach you throw down a resistor and measure the voltage drop across it. Some people balk at the idea of inserting a resistor in series because of the voltage drop it induces. This is called the burden voltage. These same people that balk at using a current sense resistor tend to not think twice about using a DMM. What these individuals are overlooking is that a DMM inserts a sense resistor in series, usually about 1-5 Ω’s, and measures the voltage drop across the resistor. There is also the contact resistance of the connection points and the resistance of the test leads. In my experience, the DMM series resistance is about 100X greater than the resistance typically used by the Ohm’s law approach.

 

I like to think about the two types of probes this way. If you are going to measure large current (5—100’s A) or fast current (like a switch mode power supply) then you want to use a magnetic field probe. If you need to measure small current or current in sub circuits or current in a physically small product then you want to use the Ohm’s law approach.

 

The most ubiquitous current probe and the one you have probably used or seen is the split-core AC/DC current probe (one example is shown here). These probes use a split ferrite core—a core that has two separable pieces. Usually one part of the core slides back-and-forth or opens like scissors to allow you to clamp it around the wire in your circuit carrying the current you want to measure. The core has a series of windings around it that are connected to the internals of the current probe for measuring the current. Most probes have a negative feedback loop inside that is used to generate an opposing magnetic flux in the core to keep the core from saturating. Sandwiched inside the non-moving portion of the core is the Hall Effect sensor. The Hall Effect sensor is necessary to measure the DC portion of the signal since a transformer only works for AC. In its simplest form a Hall Effect sensor is a conducting (or semiconducting) plate with a bias current flowing along it with voltage measurement points placed perpendicular to the current path. As a magnetic field hits the plate, perpendicular to its center, the electrons flowing across the plate from the bias current tend to shift towards one side or the other, depending on the direction of the magnetic field, and create a voltage potential across the plate. The probe measures this voltage which is proportional to the strength of the magnetic field which of course is proportional to the current flowing throw the conductor being measured.

 

There are a couple of things to be aware of when using this type of probe. First, remember back to when you were studying magnetism and current flowing through a conductor and they would always say “assume an infinitely long, straight conductor”. Well, that’s important to keep in mind. It turns out that changing the shape and position of the wire passing through the probe can change the measurement a little and this creates a repeatability problem with this type of probe. It’s usually not a big deal if you are measuring amperes of current but if you are measuring milli-amperes then it can affect the measurement.  Another issue is the air gap between the two halves of the core. If you change that gap just the slightest amount, by adding side pressure or letting the probe hang on the wire you are measuring you will have repeatability issues when measuring smaller current. There is also the issue of mechanical stress on the Hall Effect element.  Thermally or physically induced stress on the Hall Effect element can change its resistance or induce a piezo electric effect which will cause measurement inaccuracies. It’s always best to let the current probe warm up for 20 minutes or so before use to minimize thermal stresses. Finally there is the issue of residual magnetism. Most probes have a zero/degauss function to address this. The proper procedure then for getting the best results from you clamp-on current probe is to let it warm up, clamp it onto the unenergized wire that you want to measure, zero and degauss the probe and then energize your circuit.

 

The Ohm’s law approach for measuring current is very popular for measuring current in sub-circuits or in targets that can’t tolerate the big long wires for the clamp-on probes (either due to size or because the long wire adds inductance and acts like an antenna picking up external noise). Traditionally folks use a pair of single-ended probes or a differential probe to measure the voltage across the current sense resistor. I am aware of one dedicated Ohm’s Law style current probe, the Keysight N2820A. This probe is a differential probe with two outputs. Each output has different gain for measuring different current ranges. The user can connect one output to the oscilloscope and choose normal or high sensitivity and the probe automatically switches which output goes to the scope. If you have a signal with a large dynamic range, like something that has a sleep state, then you hook both outputs to the oscilloscope and can view both the zoomed-in (high sensitivity) and zoomed-out (normal sensitivity) simultaneously. In this mode the probe has a dynamic range of 20,000:1. This allows simultaneous measurement of very small sleep currents and large current spikes associated with the active state. To use the probe you simply select the current sense resistor value appropriate for the current you want to measure.  The probe will work with anything from 001Ω--1MΩ. You tell the oscilloscope the value of the resistor being used and the scope automatically scales and displays the results in amperes. As an example, using a .050Ω resistor the probe can measure from 100uA to 24A—assuming the resistor is rated to handle the power of the large current.

Screen shot of the N2820A showing zoomed-in data (green) and zoomed-out data (yellow) simultaneously. This is from an IoT weather station.

 

The drawback of the Ohm’s Law approach is that you need to either design in sense resistors or cut the trace and solder them in place. You’ll also need access to some current sense resistors. I usually just get mine from one of the many online component distributors. They stock a wide range of values and power ratings for various current ranges. One great thing to note about using these application specific current sense resistors is that they have very small thermal parts per million resistance variation—the resistance changes very little as the resistor heats up.

 

If you want to dig a little deeper into what is out there just hit our website. You’ll see the full range of current probes we offer. They are of the two main types I described above, magnetic field sensing split-core AC/DC current probes and Ohms law current probes. The web pages show the important specs like maximum and minimum current, bandwidth, price and so on. You’ll also find some application notes, data sheets and videos. If you are making power measurement such as testing a power supply or measuring low power, like a battery powered device, then you should visit the Keysight oscilloscope power page.

In urban slang, signal-to-noise ratio (SNR) is a simple enough concept: the ratio of useful to useless information. We all know people whose SNR is not as high as we might hope. Unfortunately there’s no technology yet available to boost their SNR.

So engineers can be happy that’s not true for RF signals. We can now extend SNR in wideband oscilloscope-based RF measurements through what’s known as “processing gain.” Digital down-conversion lets you see small pulsed RF signals next to large signals by reducing the noise level in a particular measurement—whether it’s RF pulse envelope characteristics or frequency or phase shift across a pulse.

Increase in pulsed RF capture dynamic range

So how does it work? The trick is adding vector signal analysis (VSA) software. VSA in conjunction with an oscilloscope can extend the SNR. First VSA shifts a captured signal down to baseband I/Q. Then it bandpass filters the acquired oscilloscope data and finally resamples the data at a lower sample rate. The result is lower noise, higher dynamic range, and a wider SNR.

Let’s look at an example: An 8 GHz-wideband oscilloscope captures a pulse train in which a large pulse is immediately followed by a small pulse that is 50 dB down from the first pulse. This corresponds to being 100,000 times lower in power and ~316 times smaller in voltage (sqrt[100,000]) than the first pulse. The two-pulse sequence then repeats.

The large pulse has a +6 dBm power level (~1.4 mW), which results in a peak voltage of around 633 mV into 50 ohms. This can be represented as a -4 dBVpk level (20log 0.633). It also corresponds to a 1266 mV peak-to-peak signal into 50 ohms.

In contrast, the small pulse, being 316 times smaller in voltage, is only 4 mV peak to peak (-44 dBm, -54 dBVpk).

The VSA software, which also controls the oscilloscope front-end sensitivity, is set to +6 dBm (633 mV peak). This corresponds to an oscilloscope vertical range of 1266 mV.  There are eight vertical divisions, so this also corresponds to a ~160 mV/div setting.

At the full 8-Hz bandwidth for this ~160 mV/div setting, the broadband RMS noise for the 8 GHz bandwidth oscilloscope is around 5 mV, interpolating from a noise chart in the data sheet, as shown in Table 1.  The 5 mV of noise translates roughly into a peak-to-peak noise that is three times the RMS noise (assuming Gaussian noise). In other words, we’re looking at 15 mV of peak-to-peak noise.

8GHz bandwidth oscilloscope RMS noise levels

Table 1. 8-GHz bandwidth oscilloscope RMS noise levels at various V/div settings

The small pulse (4 mV p-p) is masked by the noise in the measurement (15 mV p-p). (Think how easily a big-mouth can drown out softer-spoken colleagues.) The small pulse can’t be well-discerned in the full 8-GHz measurement of the oscilloscope, with a linear scale and no averaging, as shown in Figure 1.

8-GHz bandwidth oscilloscope capture of +6 dBm pulse next to a 50 dB down pulse

Figure 1. 8-GHz bandwidth oscilloscope capture of +6 dBm pulse next to a 50 dB down pulse (2nd pulse cannot be seen)

Import of real-time captured pulsed RF signals into analysis software and digital down-conversion

Basic pulsed RF measurements can be made natively on a high-bandwidth oscilloscope. And there are certainly times that measurements on directly sampled signals are desired. But this isn’t one of those times. Instead we’re looking for advantages available through external signal processing and analysis on captured signals.  For example, through a process called digital down-conversion, it’s possible to make a range of RF pulse measurements with higher accuracy. That’s due to the lower noise present by using processing gain. Let’s take a closer look.

Figure 2 shows the basic process of digital down-conversion.  Through digital signal processing, the oscilloscope samples are multiplied by the sine and cosine of an imaginary oscillator of frequency fc, where fc is generally chosen to be the center frequency of the signal of interest. In effect, we’re “tuning” to the frequency of the input signal. This process converts the time samples into real and imaginary number pairs that completely describe the behavior of the input signal. To reduce noise, these samples can be low-pass filtered and then re-sampled at a lower rate to reduce the size of the data set and allow FFT processing of the data at a later stage. The resulting digitally down-converted samples can then be placed into memory for further processing.

Oscilloscope-captured samples input to VSA software for digital down-conversion

Figure 2. Oscilloscope-captured samples input to VSA software for digital down-conversion

Some important demodulation information comes from this digital down-conversion process. First, consider what happens when the digital local oscillator frequency Fc is equal to the carrier frequency of a modulated signal. The output of the digital filters, which includes the real part I(t) and imaginary part Q(t), consists of time-domain waveforms that represent the modulation on the carrier signal.

Do you want that in math? Here’s a representation of the captured input signal:

=  A(t) * Cos[2pfct +q(t)]

where the following equation describes the amplitude modulation:

amplitude modulation equation

And the equation here describes the phase modulation:

phase modulation equation

Displaying the I-Q results in terms of magnitude coordinates gives us a view of the amplitude modulation.  Displaying the I-Q results in terms of phase coordinates offers a view of the phase modulation. Taking the derivative of phase modulation yields the frequency modulation.

frequency modulation equation

By adjusting the width of the low-pass filters, you can set a defined span around the center frequency where the filter width is just wide enough to pass the signal of interest, but narrow enough to filter out a lot of the noise.

 

Results of digital down-conversion and processing gain on the 50-dB down RF pulse

So in short, processing gain “tunes” to the center frequency of the signal and “zooms” into the signal to analyze the modulation.

In this example, the original 8-GHz-wide measurement with the associated noise is reduced to a 500-MHz wide measurement, centered on the 3.7-GHz carrier with an instantaneous measurement bandwidth slightly wider than the width of the signal modulation.  This corresponds to an improvement in SNR as follows:

10log*(ScopeBW/Span) = 10log*(8E+09/500E+6) = 12 dB.

Taking advantage of this processing gain, combined with VSA software’s ability to have a log magnitude scale, and using averaging, the 50-dB down pulse is now visible, as shown in Figure 3.

down pulse seen with Keysight VSA software

Figure 3. 50-dB down pulse seen with VSA software “Center Frequency” and “Span” set to 3.7 GHz and 500 MHz

The improvement in SNR realized through narrowing down the span is depicted graphically in Figure 4.

achievable SNR

 

Figure 4. Plot of SNR achievable in time view verses span adjustment in VSA software

You can draw a similar plot to see improvement in dynamic range possible when measuring narrow band signals, as shown in Figure 5.

Plot of dynamic range in FFT vs. resolution BW setting in Keysight VSA software

Figure 5. Plot of dynamic range in FFT vs. resolution BW setting in VSA software

Here the dynamic range improvement when measuring narrowband signals in an FFT view is described as:

10log*(ScopeBW/RBW)

This does not describe the spur-free dynamic range (SFDR) or harmonic distortion characteristics of the oscilloscope response, but it does give an idea of where the noise floor will be in an FFT measurement.  As the resolution bandwidth is decreased, and the noise is divided among smaller time buckets, the noise floor drops.

This graph does not account for limitations due to various spurs, so the spur-free dynamic range (SFDR) remains limited to around 50 dB.

Conclusion

Through the process of digital down-conversion, the SNR of an oscilloscope measurement can be significantly improved as a function of how much that measurement can be “spanned down” from the initial DC to 3dB bandwidth of the oscilloscope.  As our example showed, a 50-dB down pulse, not even visible on a normal scope screen, can be clearly seen once processed by VSA software and then displayed in a log-magnitude scale. This approach can be very helpful to speed system validation measurements on Aerospace/Defense pulsed-RF signals. With this process, you can significantly improve measurement accuracy when evaluating the spectral, pulse envelope, frequency chirp, and phase shift characteristics of an RF pulse train.

Keysight N2820A current probe

Many oscilloscope users measure current flowing through a circuit by measuring the voltage drop across the current sense resistor. But if you’re lucky enough to have a Keysight N2820A high-sensitivity current probe, it is even more than that. While the N2820A (and N2821A) is designed to measure low level current signals with high sensitivity, accuracy and dynamic range, the underlying architecture of the probe is actually a voltage probe. It essentially is an ultra-sensitive voltage probe with extra-low probe loading measuring the voltage drop across the sense resistor in the current flowing path and dividing the voltage by the sense resistor value to derive the current flow. Or simply put, I = V/R.

 

 

You can use the N2820A current probe as a voltage probe in just a few steps:

 

1. Don’t use the current sense resistor. Use the user-defined head (Keysight N2825A) with the probe and connect the +, - input of the probe head directly to the DUT like you would to measure voltage with a voltage probe.

Keysight N2825A with N2820A oscilloscope probe

 

connecting your current probe to the oscilloscope

2. Keep the common-mode voltages of the signal input to less than ±12V, and differential input should be less than ±1.2V. If you have a signal with some DC component that exceeds the differential input range of the probe, you will rather AC couple the signal on the oscilloscope.  

 

3. Choose 1Ω Rsense in the oscilloscope menu. That way, you can get the correct values on your measurement. E.g., 1 mA = 1mV.

Choose 1-ohm Rsense in the oscilloscope menu

 

Measuring voltage on a Keysight InfiniiVision oscilloscope

4. If you have a Keysight Infiniium oscilloscope, you can go a step further and add external scaling to turn the units back into volts (with 1:1 attenuation ratio). Check the External Scaling under the channel’s probe menu and select ‘volt’ as the units. If you are using a Keysight InfiniiVision oscilloscope, you would interpret the voltage reading in ‘Amphere’ as ‘Voltage’.

 

Measuring voltage on a Keysight Infiniium oscilloscope

 

When using the N2820A as a voltage probe, the key characteristics are:

  • Bandwidth (-3dB) : 500 kHz (zoom-in), 3 MHz (zoom-out)
  • Measurable input range: 3uV - 1.2V (differential), 12V (common mode)
  • Input impedance : 3 Gohm (giga ohm) differential, 1.5 Gohm single-ended
  • DC amplitude accuracy : 3% or 10 uV whichever is greater
  • Probe gain factor: 300X (zoom-in), 2X (zoom-out)
  • Offset : scope vertical positioning only (no probe offset)