Comparative Analysis of Digital Oscilloscope and Digitizer

What is an oscilloscope, really? Certainly, features such as display resolution, brightness, and ease of use are must-haves, and sampling speed, triggering, and bandwidth are also important considerations. Some digitizers offer most of these features, but are best described as data acquisition systems or transient recorders. Some oscilloscopes do not offer all of the above features, but they are still clearly oscilloscopes, albeit slower or more difficult to use.

The Newton Press Telecommunications Dictionary defines a digitizer as a device that converts an analog signal into a digital representation, the conversion function usually being accomplished by sampling the analog signal at a prescribed rate and encoding each sample into a digital representation of the sample's amplitude.

The dictionary defines an oscilloscope as a device that can display waveforms and other information on a cathode ray tube like a television. An oscilloscope is fundamentally different from a digitizer because it is an analog instrument used for testing. The waveform display is the most important factor.

Digitizers are embedded modules, such as those controlled by a system computer of a VXI bus or similar test system. Small digitizers are physically and electrically embedded in a laptop computer and provide similar performance to an oscilloscope, but the virtual user interface is not as easy to understand and use as the actual knobs and buttons. However, data analysis and reporting are done on the same instrument, and the digitized waveform is displayed on a digital screen.

Oscilloscope

Triggered sweep analog oscilloscopes are still sold in the tens of thousands each year. They are very affordable and fast, and make it easy to observe the relationship between two or more input signals. In order to use an analog oscilloscope, the signal must be nearly repetitive.

What if the signal is not repetitive at all? Or what if the signal changes so slowly that the signal waveform displayed with 1 second/grid is just a slowly moving blip? A digital storage oscilloscope (DSO) can handle both of these special cases.

The sampling rate of a DSO can be much higher than the frequency contained in the measured signal, and a single transient can be captured without recurring, similar to refreshing the display phosphor of a cathode ray tube. In a DSO, digitized data is continuously replayed from a semiconductor memory, and its speed is determined by the time required by the display device to produce a waveform image.

Early DSOs displayed waveforms on an electrostatic cathode-ray tube by driving digital converters on each axis. Many of these instruments also functioned as ordinary analog oscilloscopes. Engineers who were not fully confident in the capabilities of the DSO would need to switch to analog mode to verify that the waveforms they saw were correct.

There is a better reason. Analog oscilloscopes provide excellent characteristics for a variety of signals. Indeed, using trace enhancement or color to enhance the display information density of DSO overcomes some of the shortcomings. However, very high trigger rate, high vertical display resolution, high horizontal and brightness resolution, and very high transient response control are all superb characteristics of analog oscilloscopes. This is even more obvious when observing modulated and complex signals, which poses difficult problems to the data acquisition, compression and display capabilities of DSO.

The strength of a DSO is in displaying low or high speed signals, where analog oscilloscopes are limited by the need to use special cathode ray tubes. Analog oscilloscopes combine the advantages of ease of use, high time and amplitude display resolution, and the ability to display modulated waveforms where envelope characteristics play an important role. If archiving or further analysis of digitized signals is required, a DSO is the only valid solution, even though an analog oscilloscope has a better waveform representation.

The biggest improvement in DSOs is the increased sampling rate, which ensures that most signals are sufficiently oversampled. As a result, the waveform displayed on a modern DSO is usually the input signal, not an interesting but completely false aliased signal. Of course, signals have also become faster, and aliasing will still appear. Early DSOs had low sampling rates and often exhibited aliasing, causing engineers to spend hours looking for problems that did not exist.

Today’s DSOs use sophisticated digital signal processing to balance the long memory and fast display refresh rates required by users without generating spurious aliased signals in space or time. The problem is exacerbated when high acquisition rates are required simultaneously.

Signal integrity was the buzzword of the 1990s, coinciding with the explosive growth of digital data communications and wireless services. Fast trigger circuits allowed DSOs to capture only abnormal signal characteristics, so the oscilloscope had to process less information. Since the bit error rate of a perfect communication link was very low, a large amount of information had to be examined before a trigger was generated. As a result, acquisition rates remained impressive.

Many DSO designs require compromises because oscilloscopes are still visual test instruments. Oscilloscopes are display-focused by definition, but acquired waveform data can also be transferred to external devices such as printers or computers. As more computing power is added to DSOs to increase their capabilities, on-board measurements and analysis results are also available on the data I/O ports.

Oscilloscope Error

Signal integrity is a function of the oscilloscope and its inputs. Most DSOs have a gain inaccuracy of 1% to 5% at DC. Absolute gain at high frequencies is rarely specified, but the overall Gaussian roll-off of the oscilloscope ensures that transient response is good. The relative gain accuracy of the DSO display is affected by the preamplifier, attenuator, and analog-to-analog converter (ADC), unless an analog oscilloscope with electrostatic deflection or cathode ray tube is used, in which case the accuracy is not affected by the display system. Analog oscilloscopes have a total gain error of 2% to 3% due to errors introduced by the deflection amplifier and cathode ray tube.

LCD screens and magnetically deflected cathode ray oscilloscopes are driven at television rates, and the composite signal includes all logos, menu text, graphics and waveforms. Therefore, the relative accuracy of the waveform to the grid is not affected by the linearity of the display device.

The absolute accuracy of the grid lines may be inaccurate, but the signal will be accurately positioned on each grid line. In contrast, the grid of an electrostatically deflected cathode ray tube in an analog oscilloscope is etched on glass, so any nonlinearity in the deflection amplifier or cathode ray tube will add to the overall gain error. [page]

Most DSOs have only 8-bit resolution, and a few DSOs offer 10 or 12 bits. For example, for biomedical signals, which contain unknown offset voltages, the larger dynamic range of a 10 or 12-bit system is beneficial. If complex signals require vertical magnification to inspect fine details, higher resolution is also required.

Higher resolution can be obtained by averaging, which is based on the fact that a large amount of inorganic noise is present in the processing process, resulting in a typical RMS relationship, for example, collecting the signal 16 times can improve the signal resolution by 4 times.

A more reliable method is to add a certain amount of specially weighted noise to disturb the ADC input. In each sampling process, the ADC input is offset a little from the actual input signal value. The offset, direction and distribution of the disturbance are preset to ensure that the average high-resolution data of the disturbance process will not produce spectral distortion.

The problem is that the disturbance reduces the peak-to-peak range of the ADC by an amount equal to the disturbance signal value. Because the improvement in resolution is proportional to the disturbance signal and the number of samples averaged, there is a limit to the improvement that can be achieved.

Digitizer

The output data of the digitizer can be analyzed to find trends, anomalies, parameter distribution, and maximum and minimum values. The design of the digitizer focuses on the reliability of the signal, data transmission rate or number of channels, which is particularly suitable for certain applications. The digitizer has a display part.

Oscilloscopes and digitizers have different purposes. Oscilloscopes are the most useful troubleshooting tools and are often used in R&D environments. Oscilloscopes can be used to observe real-time signals and determine the performance of the device under test. Waveform digitizers are more useful when the problem is already known and more specific information is needed.

The basis of digitizers and ADCs is the ADC. For slower very high-resolution applications, Σ-Δ converters can provide resolutions above 20 bits, or 1ppm. Of course, noise and amplifier nonlinearity quickly degrade performance below 16 bits, but even then, 16 bits of resolution is 256 times higher than an 8-bit oscilloscope.

A typical 20MS/s digitizer has a basic accuracy of 0.5%, a common mode rejection ratio (CMRR) of 68dB, and a resolution of 12 bits. In contrast, a representative 100MHz, 8-bit DSO has a channel isolation specification of greater than 40dB at 0-20MfZ, with an accuracy of 1.5%.

Digitizers use finite bandwidth filters and are best used in the time or frequency domain at the input. Using abrupt passband or stopband filters such as Chebyshev filters is very effective in reducing aliasing, but can introduce ringing in the time domain.

In the frequency domain, the errors associated with the filter can be corrected to obtain an accurate spectrum. The slower Bessel filter has much better signal characteristics in the time domain and does not need correction, but it cannot completely suppress aliasing.

Oscilloscope as Digitizer

The fastest oscilloscopes and digitizers typically use parallel flash converters and 8-bit resolution. 8 bits or 256 levels of digitization are sufficient to produce a relatively smooth and easy to understand waveform display. Therefore, why not use a DSO as a digitizer, especially for high-speed signals, it is difficult for both instruments to achieve resolutions above 8 bits.

In fact, the results of doing so are satisfactory, but there are exceptions. Oscilloscopes are non-continuous acquisition instruments, while digitizers are not. After the oscilloscope captures a signal, it needs a place to store the data before capturing more signals, unless continuous waveform acquisition similar to the TV frame rate is used to store the data in a pixel image. This acquisition and equivalent display rate is very high, but the data format makes the amount of data for further external analysis very large.

Except for the special processing mentioned above, the oscilloscope can only continuously acquire and display signals at a very low speed. The digitizer can achieve a continuous throughput of 100MS/s or higher, limited only by the speed of the memory bus. For example, a digitizer card for the PCI bus has a data transfer rate of 100MB/s, and the PCI bus can operate at 66MS/s (132MB/s).

The throughput rate of the oscilloscope is limited by the data processing speed of the slower, low I/O capabilities. Slower digitizers and data recorders can write data directly to the hard disk and archive several GB of data, while oscilloscopes generally only have a maximum of 16MB. If you look at the data transfer rate from another perspective, many applications only need to capture sporadic data, but these bursts may be close together. At this time, it is very important to transfer data records quickly. Such signals include scanning radars with high repetition pulse frequencies (PRF), time-resolved ultrasonic sonars, time-of-flight mass spectrometers, and nucleon counting applications.