What is the FFT on the oscilloscope? First, with a digital oscilloscope, we are not simply processing the waveform, and we no longer just stay in the shape of the waveform, no longer meet only a few parameters measured. We always think about doing more processing on the data we have collected. The oscilloscope understands it more accurately. It is more like a waveform analyzer. It is the dissatisfaction of engineers that we have the constant drive to push the limits because we often underestimate us. Potential, where is the limit? Who is the first to use the FFT (Fast Fourier Transform) in a digital oscilloscope? There are many claims. It seems that all of a sudden, everyone has FFT function on the oscilloscope, and they are all standard configurations. Although they all have this function, the results are very different, the speed and the indicators are all different, everything is the same at the beginning. All are first pursued and then talk about differentiation. Moreover, the oscilloscope itself is a qualitative tool, who cares about the oscilloscope's index accuracy in the frequency domain, in addition to our lovely R & D engineers. The situation is changing. In many cases, users want to solve all problems through an instrument. To tell the truth, many engineers do not have the conditions to place potentiometers, spectrum analyzers, oscilloscopes, and vector networks on the desk. In most cases, the oscilloscope samples the time-domain data collected by the oscilloscope, performs software fft operations, and becomes a sample in the frequency domain. Then, the oscilloscope displays the samples in the frequency domain through data reorganization. The fft's ability depends on a few indicators: memory size, software operating speed, dynamic significance ENOB, and noise floor. Because these indicators directly determine the refresh rate after fft, dynamic range, sensitivity, resolution bandwidth RBW. Second, the oscilloscope fft can solve what problem? Constrained by the tools at hand (all engineers dream of having state-of-the-art oscilloscopes and spectrum analyzers on their desks), and many times engineers need to make qualitative observations when debugging circuits, and fft becomes a good tool for watching the spectrum. To be honest, many manufacturers have done a poor job with fft functions. There are only two types of reasons. One is that they do not have the ability to do a good job. Doing a good job of spectrum analysis still requires a lot of DSP experts and radio frequency technology; , But subjectively, I don't really want to make fft too strong. It's so good. So how do I sell the spectrum analyzer? There is a question of the opportunity cost. But fft can still solve some problems, such as looking at spectral range, looking at the harmonic components, looking at the proportion of harmonics, taking a cursory look at the spectrum interference, etc., but often it also brings some problems, such as sampling chips. When multiple pieces are stacked together, the overlapping lines will be exposed, and the processing speed will be so slow that the noise floor will be a bit too outrageous, and the jitter component will be a bit messy. To avoid these problems, of course, we will come up with some good ones. Methods, such as restricting the fft analysis sample, do not crash when storing fft for a long time, such as lowering the noise floor of the waveform on average. Third, the oscilloscope fft is tasteless? Needless to say, sometimes it's really bad, and the processing speed is too slow. A slightly larger sample is almost the same as a crash. The RBW is too outrunning, the harmonic rejection ratio is poor, and the noise often submerges the harmonics, and the dynamic range is not enough. But in fact, on many occasions in our country, if the fft function is good enough, it is not chicken, it is a chicken leg. For example, test the filter and the system's impulse response (characteristics), identify and locate noise sources, determine spurious emissions, jitter analysis, harmonic power analysis, and EMI analysis. This is a great place to look at fft. Fourth, we do the oscilloscope on the spectrum analysis function to achieve the ultimate, how to do it? First of all, we must increase the speed of spectrum analysis and refresh it in real time. So you can't stand to suffer similar crashes when the oscilloscope fft transforms. Secondly, we do RBW up to 1Hz. This level can only be achieved by spectrum analyzer. Our interface The design and operation of the spectrum analyzer are exactly the same, the center frequency, the spectrum range, the start spectrum, the cutoff frequency, the RBW setting, the window function setting, and almost all the settings of the spectrum analyzer. Here are four ways to demonstrate how we can achieve the ultimate in fft functionality: 1, the special digital down converter DDC Traditionally, oscilloscopes have collected signal samples and then used software algorithms to perform software operations. The speed is very slow. Our approach uses a dedicated hardware-accelerated integrated circuit (ASIC) to hand off the fft function to this hardware circuit. The speed is so fast that it hardly affects the refresh rate of the original waveform. Of course, this ASI needs to spend a lot of money on R&D. The core comparison uses a dedicated DDC circuit. Let's take a look at how traditional oscilloscopes are Our oscilloscope fft principle A comparison of the above figure shows that a DDC processing is performed before the window function. The user sets the center frequency and sets the initial and cutoff frequencies. The result of the processing is to process only the frequency band concerned, or the set frequency band. The traditional method must perform fft calculation on all frequency ranges, and then select a certain frequency to display. The amount of data for operation is very large. In principle, our principle is to process only the frequency band you are interested in or the initial frequency and cutoff frequency range you choose. Of course, the extreme case is to select the full frequency band for processing, so that you have the opportunity to reduce the amount of processing and processing capacity. Focus on the range after DDC. The two figures below clearly show the difference between the traditional way and our way. This approach brings two benefits: a) Faster speed, frequency conversion to baseband processing will result in higher update rate and faster processing speed, saving processing time. 2, the use of hardware accelerators In the traditional solution, it has been implemented using software processing such as statistical histogram function, template test function, and fft function. In the RS oscilloscope, all hardware-specific circuits are used to achieve the freedom of the processor, so the histogram function, the template test function, or the abnormal resource-consuming fft function still maintains a high refresh rate, usually exceeding 60,000 times/s, this speed exceeds the refresh rate when all oscilloscopes on the market do not perform any operation. This can guarantee that when doing complicated waveform analysis, the refresh rate is still very high, and the high refresh rate ensures the fast display rate of the real-time spectrum. 3, overlapping FFT algorithm application The traditional oscilloscope's fft operation method collects one segment, processes one segment, and then collects and then processes. Therefore, continuous intermittent acquisition, continuous processing, but the frequency of the accidental signal is also very easy to lose, can not be found. RS's oscilloscope performs segment processing on the collected samples and divides the collected signal into many small segments for processing so that the spectral content changes in one acquisition can be seen. However, the fragment processing cannot avoid the loss. Because before the operation of fft, there was a window function processing, and there was inevitable loss of spectral information at the positions of two adjacent frames. Therefore, we adopted another, more innovative method. Using the overlap algorithm of FWT, the influence of the window function is greatly improved, and the loss of the abnormal spectrum. With the display of analog persistence, the real-time spectrum display is more reliable and reliable. Benefit summary: a) It facilitates the monitoring of abnormal signals 4, similar to the traditional spectrum analyzer control interface and control methods The previous oscilloscope control method was nothing more than to influence the resolution bandwidth by adjusting the length of the acquisition time, and then select the frequency band of interest to observe. The current practice is to select the center frequency first, or select the start and cutoff frequencies, and adjust the spectrum observation mode by directly adjusting RBW so that users of custom spectrum analyzers are also accustomed to oscilloscopes. There is also a table to help understand what window functions are used in what situations. 5, using the template approach to achieve the frequency domain trigger settings Many people who are accustomed to oscilloscopes like the oscilloscope's trigger function. They use various triggering methods to isolate various events, stabilize the display, and observe abnormalities. It is difficult to achieve triggering on traditional spectrum analyzers, but when we find that the oscilloscope's template triggering method is very suitable, we can change the real-time spectrum of the time domain waveform to the frequency domain to observe. With the help of some MASK testing tools, we actually Easy to set up and easy to trigger. Because the shape of the template is freely edited and the triggered actions are freely combined, such waveform analysis has completely spanned the usage habits of the time domain and the frequency domain, and completely integrates the time domain and frequency domain signal thinking methods. Red template area trigger instance Fifth, the development trend of spectrum analysis on the oscilloscope The oscilloscope's analysis speed is faster and faster, the algorithm is more and more scientific, the storage depth is getting bigger, the fft function is no longer as dispensable as before, the ability of spectrum analysis depends on the fft capability, depends on the dynamic range, depends on Noise size. Spectral analysis done by the principle of the oscilloscope needs to increase the dynamic range. It is nothing more than doing some time-domain averaging before fft to reduce the noise, or increase the storage depth, increase the RBW, reduce the asynchronous noise, and achieve the purpose of improving the dynamic range. Pharmaceuticals,2-Methyl- Propanoic Acid Monohydrate Price,2-Methyl- Propanoic Acid Monohydrate Free Sample,Pure 2-Methyl- Propanoic Acid Monohydrate Zhejiang Wild Wind Pharmaceutical Co., Ltd. , https://www.wild-windchem.com
Second, the oscilloscope FFT can solve the problem?
Third, the oscilloscope's FFT often becomes the taste of the user's hands, where is the problem?
Fourth, we do the oscilloscope on the spectrum analysis function to achieve the ultimate, how to do it?
Fifth, the development trend of spectrum analysis function on the oscilloscope 
b) Better resolution bandwidth because better magnification factors will be used. 
b) show rare occurrences in the short term
c) Increased refresh rate of the spectrum (because the fft of a new frame has already started before the fft of one frame is completed)
d) It is possible to distinguish between multiple spectrum events in an fft frame 
