Top 10 factors to consider when choosing an oscilloscope

1. How much bandwidth do you need?
We are already in the era of digital oscilloscopes. Compared with the bandwidth of analog amplifiers, the bandwidth of the oscilloscope should be considered more. In order to ensure that the oscilloscope provides enough bandwidth for the application, you must consider the bandwidth of the signal that the oscilloscope will examine.
Bandwidth is the most important feature of an oscilloscope because it determines the range of signals that can be displayed. It also largely determines the price that users need to pay. When making bandwidth decisions, you must balance the current limited budget with the expected needs during the use of the oscilloscope in the laboratory.
In current digital technology, the system clock is usually the highest frequency signal that an oscilloscope may display. The bandwidth of the oscilloscope should be at least three times higher than this frequency to reasonably display the shape of this signal.
Another signal characteristic in the system that determines the bandwidth requirement of the oscilloscope is the rise time of the signal. Since you may not only see a pure sine wave, the signal will contain harmonics at frequencies beyond the fundamental frequency of the signal. For example, if you are examining a square wave, the signal contains frequencies at least 10 times higher than the fundamental frequency of the signal. If you don't ensure the appropriate oscilloscope bandwidth when examining signals such as square waves, you will see rounded edges on the oscilloscope display instead of the sharp and fast edges you expect. This will in turn affect the accuracy of your measurements.
Fortunately, we have some very simple formulas that can help you determine the appropriate oscilloscope bandwidth based on the characteristics of your signal.
1. Signal bandwidth = 0.5/signal rise time
2. Oscilloscope bandwidth = 2 x signal bandwidth
3. Oscilloscope real-time sampling rate = 4 x oscilloscope bandwidth
After you have determined the appropriate oscilloscope bandwidth, you need to consider the sampling rate of each channel that the oscilloscope intends to use simultaneously. As listed in Formula 3 above, for each channel you intend to use, you must ensure that the sampling rate is four times the oscilloscope bandwidth so that these channels can fully support the rated bandwidth of the oscilloscope. We will discuss this in more detail later.

2. How many channels do you need?
At first glance, the number of channels seems to be a simple question. After all, don't all oscilloscopes come with two or four channels? Nothing more! Digital content is everywhere in today's designs, and no matter how high or low the proportion of digital content in the design, traditional 2-channel or 4-channel oscilloscopes cannot always provide the number of channels required to trigger and view all signals of interest. If you have encountered this situation, you will understand the problems involved in building external hardware or writing dedicated software to isolate the activities of interest.
For today's increasingly digital field, a new type of oscilloscope has enhanced the application of oscilloscopes in digital applications and embedded debugging applications. Mixed signal oscilloscopes (commonly called MSOs) closely insert another 16 logic timing channels in addition to the 2 or 4 oscilloscope channels of a typical oscilloscope. The result is a full-featured oscilloscope that provides up to 20 time-correlated triggering, acquisition, and viewing channels.
We will use the common SDRAM application as an example to introduce how to use mixed signal oscilloscopes for daily debugging. To isolate the SDRAM write cycle, you must trigger the system on a combination of five different signals—RAS, CAS, WE, CS, and clock. A 4-channel oscilloscope alone is not sufficient for this basic measurement requirement.
As shown in Figure 2, the 16 logic timing channels are used to set up the system to trigger on RAS high, CAS low, WE high, and CS. Oscilloscope channel 1 is used to view and extract the rising edge of the logic clock. Unlike a combined logic analyzer and oscilloscope solution, where the logic analyzer can only cross-trigger the oscilloscope or vice versa, a mixed signal oscilloscope can trigger full width on both the oscilloscope and logic timing channels.

3. What is the sampling rate you require?
As mentioned earlier, sampling rate is a very important consideration when evaluating an oscilloscope. Why? Most oscilloscopes use interpolation, which provides the maximum sampling rate on only one or two channels in a four-channel oscilloscope when two or more channels are coupled to the analog-to-digital converter, thereby increasing the sampling rate. Many manufacturers only emphasize this maximum sampling speed in the main technical specifications of the oscilloscope, but do not tell the user that the sampling rate only applies to one channel! If you want to buy a 4-channel oscilloscope, then in fact you want to use and get the full bandwidth on more than just one channel.
Recall the formula given in the second consideration that the sampling rate of the oscilloscope should be at least 4 times the bandwidth of the oscilloscope. When the oscilloscope uses some form of digital reconstruction, it is best to use a 4x multiplier, such as sin(X)/X interpolation. When the oscilloscope does not use digital reconstruction, the multiplier should actually be 10x. Since most oscilloscopes use some form of digital reconstruction, a 4x multiplier should be sufficient.
Let's look at an example using a 500MHz oscilloscope that uses sin(X)/X interpolation. For this oscilloscope, to support a full 500MHz bandwidth on each channel, the minimum sample rate required per channel is 4 x (500MHz), or 2GSa/s per channel. Some 500MHz oscilloscopes on the market today claim a maximum sample rate of 5GSa/s, but fail to specify that the 5GSa/s sample rate applies to only one channel. When using three or four channels, these oscilloscopes actually only sample 1.25GSa/s per channel, which is insufficient to support 500MHz bandwidth on several channels.
Another way to think about the sample rate is to determine the desired resolution between application points. The sample rate is the inverse of the resolution. For example, suppose you want to achieve a resolution of 1ns between sample points. The sample rate that would improve this resolution is 1/(1ns) = 1GSa/s.
In summary, make sure the oscilloscope you are considering provides enough sampling rate per channel for all the channels you want to use simultaneously, so that each channel can support the rated bandwidth of the oscilloscope.
4. How much memory depth do you need?
As mentioned earlier, bandwidth and sampling rate are closely related. Memory depth is also closely related to sampling rate. The analog-to-digital converter digitizes the input waveform and the resulting data is stored in the oscilloscope's high-speed memory. An important factor in choosing an oscilloscope is to understand how the oscilloscope uses this stored information. Memory technology allows users to capture acquired data, zoom in to see more details, or perform math operations, measurements, and post-processing functions on the acquired data.
Many people believe that the maximum sampling rate specification of an oscilloscope applies to all timebase settings. This is certainly a good thing, but it may require a very large memory, and almost no one can afford an oscilloscope with such a large memory. Because the memory depth is limited, all oscilloscopes must reduce the sampling rate as people set the timebase to a wider and wider range. The deeper the memory of an oscilloscope, the more time can be captured at the full sampling rate. There is a popular oscilloscope on the market today that has a sampling rate of several gigasamples per second and a memory of 10,000 samples. This oscilloscope is forced to reduce the sampling rate to a few thousand samples per second when the timebase is set to 2ms/div and slower. You must examine the oscilloscope in question to understand the effect of the timebase setting on its sampling rate. The oscilloscope in question will only provide a few kilohertz of bandwidth when operating at the required sweep rate to display the entire system operating cycle.
The depth of memory you need depends on the number of displays you want to view and the sample rate you want to maintain. If you want to view longer periods of time with higher resolution between different sample points, you need deep memory. A simple formula can tell you how much memory you need, taking into account the time interval and the sample rate:
Memory Depth = Sample Rate x Display Time
If you need to zoom in and view waveforms more closely, maintaining a high sample rate at all time settings on the oscilloscope can prevent aliasing and provide more detailed information about the waveform.
Once the memory depth has been determined, it is also important to examine how the oscilloscope operates when using the deepest memory setting. Oscilloscopes with traditional deep memory architectures are slow to respond, which can have a negative impact on productivity. Because of the slow response, oscilloscope manufacturers often relegate deep memory to a dedicated mode, and engineers typically use it only when deep memory is absolutely necessary. Although oscilloscope manufacturers have made great progress in deep memory architectures over the years, some deep memory architectures are still slow and time-consuming to operate. Before purchasing an oscilloscope, be sure to evaluate the oscilloscope's responsiveness at the deepest memory setting.

5. What display features do you need?
All oscilloscope vendors know that they sell waveform images. Back in the days of analog oscilloscopes, the design features of the oscilloscope's CRT display determined the quality of the image. In today's digital world, the actual performance of an oscilloscope depends largely on digital processing algorithms rather than the physical features of the display device. Some oscilloscope manufacturers have added dedicated display modes to their products to overcome some of the differences between traditional analog oscilloscope displays and digital displays. There is no good way to determine which oscilloscope is best suited for a user's laboratory environment by studying the oscilloscope's technical specifications. Only when the user demonstrates it in real time on their workbench and uses their own waveforms can they determine which oscilloscope is best suited to meet their needs.
Current digital oscilloscopes are divided into two categories: waveform viewing instruments and waveform analyzers. Oscilloscopes designed for viewing waveforms are usually used in testing and problem diagnosis applications, where waveform images will provide all the information the user needs.
In waveform analysis applications, Microsoft Features such as the Windows operating system and advanced analysis capabilities can apply additional levels of abstraction to determine the performance of the system under test. In this regard, it is also difficult to determine whether an oscilloscope can meet user needs based solely on product data sheets. Real-time demonstrations in the lab are required to determine whether the oscilloscope under investigation can display what the user needs to see.
6. What trigger functions do you need?
Many general-purpose oscilloscopes use edge triggering. However, other trigger functions may be required in some applications. Advanced triggering functions enable you to isolate the events you want to view. For example, in digital applications, triggering on a certain pattern in the channel can be very helpful. As mentioned earlier, mixed signal oscilloscopes can trigger on logic channels and oscilloscope channel patterns, while in oscilloscope/logic analyzer combination solutions, users can only cross-trigger the two instruments by connecting their respective input/output trigger signal cables together.
For serial designers, some oscilloscopes are even equipped with serial trigger protocols for standards such as SPI, CAN, USB, I2C, and LIN. Advanced triggering options can save a lot of time in daily debugging tasks. What if you need to capture rare events? Glitch triggering allows triggering on positive or negative glitches. Or trigger on pulses greater than or less than a specified width. These features are particularly useful when diagnosing a problem. You can trigger on the problem, look back in time (using the delay or horizontal position knob), look back in time (using the delay or horizontal position knob), and see what caused the problem.
Many oscilloscopes on the market today also offer triggering for TV and video applications. By using the oscilloscope's TV triggering feature, you can trigger the system on the specific line where you need to see it.
7. What is the best way to probe a signal?
Signals begin to change at rates exceeding 1GHz. Since passive probes are generally limited to 600MHz, getting the full bandwidth of the oscilloscope can be a problem. The system bandwidth (that is, the combined oscilloscope/probe bandwidth) is the lower of these two bandwidths. For example, consider a 1GHz oscilloscope with a 500MHz passive probe. The combined system bandwidth is 500MHz. It is not worth buying a 1GHz oscilloscope if you get 500MHz of bandwidth because of the probe!
In addition, every time you connect a probe to a circuit, the probe becomes part of the circuit being measured. The probe tip is essentially a short transmission line. A transmission line is an LC resonant circuit, and at a frequency that is 1/4 wave of the transmission line, the impedance of the LC resonant circuit will become low, close to zero, and will load the device under test. The loading of the LC resonant circuit can be easily seen in the slow rise time and ringing of the signal.
Active probes not only provide more bandwidth than passive probes, but they also eliminate some of the transmission line effects when the probe is connected to the device under test (DUT). By using resistive "attenuated" probe tips and accessories in active probes, Agilent Technologies minimizes signal loading and the resulting signal distortion. These attenuated accessories prevent the impedance of the LC resonant circuit from becoming too low, thereby preventing the loading signal from causing ringing and signal distortion.
In addition, the attenuated accessories allow the probe's frequency response to remain flat across the entire bandwidth of the probe. With a flat frequency response, signal distortion is prevented across the entire bandwidth of the probe.
Now that the signal distortion issue has been resolved, if you are probing high-speed signals, the next step is to ensure that the full bandwidth is still achieved even when using probe accessories. Agilent InfiniiMax probes optimize probe bandwidth by using a controlled transmission line between the probe amplifier and the probe tip. By using one amplifier, you can connect a variety of differential or single-ended probes, including browsing probes, probes with sockets, soldered probes, and SMA probes, and get the full system bandwidth. In addition, because the probe amplifier is actually separated from the probe tip by a controlled transmission line, it is easy to access tight probe spaces.
The key here is to understand the rated bandwidth of the probe when using various probes and accessories. Accessories can reduce the performance of the probe, and users certainly don't want to spend thousands of dollars unnecessarily to buy a high-bandwidth active probe that seriously reduces system performance when the user's preferred probing configuration.

8. What archiving and connectivity features do you need?
Many digital oscilloscopes now come with the same interfaces as personal computers, including GPIB,
RS-232, LAN, and USB. It is much easier to send images to a printer or transfer data to a PC or server than in the past. Do you often transfer oscilloscope data to a PC? It is very important that the oscilloscope has at least one of the interface options listed above. A built-in floppy drive or optical drive can also help you transfer data, but using a floppy drive or optical drive usually requires more work than sending files from the oscilloscope via a USB or LAN connection. For economical oscilloscopes that do not have more advanced interface options such as LAN and USB, oscilloscope manufacturers usually provide software that allows waveform images and data to be easily transferred to a PC via GPIB or RS-232. If the PC does not have a GPIB card installed, or the user wants an easier way to transfer waveforms to a laptop, you may want to consider a GPIB to USB converter. Many oscilloscopes also come with a hard drive of several GB, which the user can also use to store data. You should determine in advance what level of connectivity and archiving features you need from the oscilloscope. If you need to connect an oscilloscope as part of an automated test system, be sure to ensure that the oscilloscope is equipped with sufficient software and driver level to adapt to your programming environment.
9. How do you analyze waveforms?
Automatic measurements and built-in analysis functions can save users time and make work easier. Digital oscilloscopes usually come with a range of measurement functions and analysis options that are not available on analog oscilloscopes.
Mathematical functions include addition, subtraction, multiplication, division, integration, and differentiation. Measurement statistics (minimum, maximum, and average) can verify measurement uncertainty, which is an important resource when verifying noise and timing margins. Many digital oscilloscopes also provide FFT functions.
For "high-demand users" who focus on waveform analysis, oscilloscope manufacturers are providing greater flexibility in mid-range and high-end oscilloscopes. Some manufacturers provide software that allows customizing complex measurements, performing mathematical functions and post-processing directly from the oscilloscope user interface. For example, measurement programs can be written in C++ or Visual Basic and then executed from the oscilloscope graphical user interface (GUI). With this function, users do not need to transfer data to an external PC, which can save a lot of time for users who focus on waveform analysis.

10. Last but not least: Demo, demo, and more demo!
If you have considered the previous nine factors, you have probably narrowed down the field to a small number of oscilloscopes that meet your criteria. Now is the time to try out these oscilloscopes and do a side-by-side comparison. Borrowing an oscilloscope for a few days will give you time to fully evaluate them. Some of the factors to consider as you use each oscilloscope include:
Ease of use: During your experimentation, evaluate each oscilloscope for ease of use. Does the oscilloscope have easy-to-use, dedicated knobs for common adjustments such as vertical sensitivity, timebase speed, trace position, and trigger level? How many buttons do you have to press to get from one operation to another? Can you run the oscilloscope intuitively while focusing on the circuit under test?
Display responsiveness: When evaluating an oscilloscope, pay attention to the oscilloscope's responsiveness, which is a key factor whether you are using the scope to diagnose a problem or to collect large amounts of data. Does the oscilloscope respond quickly to changes in V/div, time/div, memory depth, and position settings? Take another look at the oscilloscope's response speed when you turn on the measurement function. Is the response speed noticeably slow?
Conclusion
After thoroughly examining these issues and evaluating oscilloscopes, you should have a good idea of ​​which model will truly meet your needs. If you are not sure yet, you may want to discuss the product selection with other oscilloscope users or call the manufacturer's technical support.
Glossary Glitch
A signal (usually an electrical interface signal) that is sampled at a rate lower than the Nyquist rate (twice the signal's largest frequency component) and therefore erroneously rearranges the signal's frequency components.
CAN Controller Area Network, a robust serial communications bus standard popular in automotive and industrial applications.
Digital Oscilloscope An oscilloscope that uses a high-speed analog-to-digital converter (ADC) to measure a signal and then displays the signal on a screen (CRT or LCD) using standard computer graphics techniques.
GPIB General Purpose Instrumentation Bus, also known as the IEEE-488 bus, is a widely used interface for connecting test instruments to computers and providing programmable instrument control capabilities.
Harmonic A frequency component of a signal that is an integer multiple of the signal's fundamental harmonic.
I2C Successor to Inter-Circuit Bus, a short-distance serial communication bus standard consisting of two signals (clock and data) that is popular for communication between multiple integrated circuits on the same printed circuit board.
Interpolation A technique used in digital oscilloscopes where analog-to-digital converters of different analog channels are used together, generally, the fewer channels used, the higher the sampling rate and the deeper the memory depth.
LC resonant circuit A circuit consisting of an inductor and a capacitor that can store electrons continuously over a frequency band and distribute them roughly at a frequency where the circuit resonates or is tuned.
LIN Local Interconnect Network, a short-distance serial communication standard that is very common in systems that include the CAN bus. LIN is less fast and less complex than the CAN bus.
Mixed Signal Oscilloscopes (MSOs) Digital oscilloscopes with more channels than are commonly used to view analog and digital signals. MSOs typically have two or four analog channels with at least 8 bits of vertical resolution. They usually have 16 digital channels, but they generally have only 1 bit of vertical resolution.
SDRAM Synchronous Dynamic Random Access Memory, the most popular form of digital memory today, differs from previous generation DRAM in that all signal timing is relative to a clock.
SPI Serial Peripheral Interface, a very simple short-distance serial communications bus standard consisting of two signals (clock and data) or three signals (clock, data, and strobe), popular for applications such as reading data from microcontroller peripherals such as ADCs.
USB Universal Serial Bus, an interface used to connect peripherals (including test instruments) to computers.