Variable memory depth and segmented memory of LOTO oscilloscope

LOTO2018

Variable memory depth and segmented memory of LOTO oscilloscope [Copy link]

Customers often ask and do not understand why the storage depth of the LOTO oscilloscope is variable, and also express their misunderstanding of the segmented storage function of the LOTO oscilloscope. This article makes a complete review of the storage mechanism of the LOTO oscilloscope to help our customers better use the oscilloscope.

Digital oscilloscopes all have dead time. As shown in the figure below, there must be a dead time between two acquisition time periods. The signal waveform within the dead time cannot be collected and displayed.

Therefore, oscilloscope manufacturers are trying their best to shorten the dead time period and lengthen the acquisition time period (the gray part in the above figure). However, the dead time period cannot be infinitely reduced, and the acquisition time period cannot be infinitely increased. Both methods need to be weighed in the actual use of the product.

Reducing the sampling rate or increasing the storage depth are two common methods to extend the acquisition time period. A large storage depth can ensure that the oscilloscope uses the highest possible sampling rate to collect waveforms for the longest possible period of time at a time, but it cannot eliminate the dead time period. Larger storage will reduce the processing and waveform capture rate. This will reduce the response rate of the oscilloscope and increase the dead time between each acquisition. The method of reducing the sampling rate reduces the number of stored data points, which is beneficial to improving the waveform capture rate and waveform refresh rate. The disadvantage is that when the sampling rate is reduced to a certain extent, the number of sampling points in each waveform cycle will be insufficient, causing distortion.

High-end oscilloscopes have a third method: segmented memory. Segmented memory divides the memory into smaller segments. The user can specify how many segments the memory should be divided into, with each segment being of equal length.

When the oscilloscope sees the first trigger event, it starts storing sample points in the first segment of acquisition memory until the first segment is full. The oscilloscope then re-arms for triggering and starts looking for the next trigger event to occur. When the trigger event occurs, the oscilloscope stores sample points in the next memory segment. The oscilloscope repeats this process until all memory segments are full. Segmented memory mode is especially useful when capturing bursty signals with long dead times. Many serial buses and communication signals fall into this category. With segmented memory, the oscilloscope can maintain a high sample rate while capturing for minutes, hours, or days.

LOTO oscilloscopes provide a variety of flexible storage methods for OSCA02 and higher models, making trade-offs and considerations on product performance. They can refresh quickly with a small storage depth and maintain a high waveform update rate, and can also be set to a long-time acquisition mode for continuous acquisition, and can also perform segmented storage.

Let's introduce each function one by one:

1Oscilloscope mode:

In most application scenarios of most customers, the LOTO oscilloscope mode should be selected, which is also the default mode of our software. This mode gives priority to using a fixed 128K storage depth to ensure fast waveform update, display and calculation.

In this mode, customers can adjust the frame rate of waveform refresh by themselves. They can also freely set the trigger mode and position and get fast waveform response.

In oscilloscope mode, customers do not need to consider sampling rate and storage depth, but only need to consider the time span of the current screen display, that is, the time slot. The oscilloscope will automatically adjust the sampling rate and storage depth according to the current time slot to achieve the most appropriate effect, taking into account both efficiency and accuracy. Therefore, the storage depth of the LOTO oscilloscope is variable.

In this mode, the LOTO oscilloscope also provides a 500-frame PC cache function and an afterglow function. These two functions record multiple frames of data and waveforms, and display them horizontally and vertically. This is equivalent to 500 current storage depths. For example, if the current storage depth is 128K, turning on the PC cache function is equivalent to a depth of 60M. And these data can be exported to a computer file, or imported into the LOTO oscilloscope software to reproduce the waveform for analysis, or exported frame by frame into a text file or spreadsheet file.

There is still a dead time between these 500 storage depths. When using a high waveform refresh rate, this function can greatly increase the probability of finding abnormal waveforms. This is a bit like the next mode, segmented storage mode, which we will introduce later.

2 Capture card mode (traditional long storage mode ) :

The LOTO oscilloscope can choose to switch the oscilloscope mode to the acquisition card mode, as shown in the following figure:

,

In this mode, the customer manually selects the sampling rate and the appropriate buffer (equivalent to storage depth) size for acquisition. After selecting, click on the acquisition, and the acquisition will start for a long time until the entire selected buffer is filled. The buffer can be set to 250 megabytes at most. If it is a dual-channel acquisition, the time required for the entire acquisition process is the buffer size divided by 2 and then divided by the sampling rate. For example, using a 2.4M sampling rate, a 250M buffer, and acquiring a 1KHZ sine wave, it will continue to collect for 60 seconds, about 1 minute of waveform data and draw it on the screen. The acquisition process will appear very stuck. If the waveform is not updated for a long time, there will be an indicator in progress:

You need to wait patiently for the collection to end, as shown in the following figure.

In most cases, one minute of data will appear very dense on the screen. For example, there are more than 60,000 cycles of waveforms in the figure. We can zoom in and observe and analyze them one by one. Generally, with the naked eye and the patience of engineers, people can only see 1,000 cycles at most before they collapse. That is to say, the maximum waveform cycle you can view is 1,000, which is a probability of one in a thousand. For an abnormality of 1ppm, you will definitely not be able to see it. Now you can search according to the conditions, but what are the conditions? Is it a pulse, an edge defect, or an overshoot? You cannot predict it, so you cannot search. Searching according to various conditions will take a lot of time and will also make mistakes. In some cases, use the oscilloscope mode of mode 1 to quickly refresh the waveform. Once you see an abnormal waveform flash, you can immediately stop the acquisition and find it in the past 500 frames of cache.

Long storage will increase the computing requirements. If the same processor has more data, the processing time will be longer and the waveform capture rate will become very slow. In order to facilitate the later analysis of such long-term and large-volume data collection, we can export the collected data into a data file, as shown in the following figure:

Then import it in 1 oscilloscope mode and turn it into a 500-frame segmented waveform. The total data volume is 250M bytes, which can be easily viewed and analyzed one by one, as shown in the following figure:

As shown in the figure above, the design of the LOTO oscilloscope takes into account the advantages of long-term acquisition and tries to make up for the inconvenience of subsequent analysis and viewing. The LOTO oscilloscope can not only use the acquisition card mode in the case of 2-channel oscilloscope analog signal input, but also in the case of dual-channel combined acquisition, and can also be used in the case of logic analyzer digital channel input, and can export files and import them into the PC cache for segmented analysis. Once imported into the PC cache, the waveform can be reproduced for analysis, and it can also be exported frame by frame into a text file or spreadsheet file.

3 Segment storage mode:

As mentioned earlier, in oscilloscope mode, turning on the PC cache function is basically equivalent to having a 500-segment segment storage capability. This segment storage mode has a particularly typical application scenario: low duty cycle pulses or burst signals. There is a long idle time between signals. In many cases, even with a larger storage or by reducing the sampling rate, it is difficult to achieve the desired acquisition time. Imagine that it may appear 100 times in a day. Each time the signal appears is very sharp and short, requiring a high sampling rate to capture. No matter how large the storage of the oscilloscope is, it is impossible to store a day's data. No matter how low the sampling rate is, it is impossible to capture all 100 signals. However, segment storage can accomplish this very well.

As shown in the figure above, segmented storage is triggered multiple times during high sampling rate acquisition, and the data obtained from each trigger sampling is stored in a small storage that divides the storage space into segments. The oscilloscope triggers once to fill a segment, and the idle signals or uninteresting parts of the signal between segments are not collected and stored.

Another common scenario where the segmented storage function is particularly suitable is serial bus analysis. The serial bus transmits in the form of data packets. The idle time between packets will occupy the valuable storage resources of the oscilloscope. With segmented storage, the oscilloscope can only collect data packets and not sample during idle time. While maintaining a high sampling rate, more data packets can be collected, which is convenient for decoding and analysis. Next, let's look at a case of a LOTO oscilloscope using segmented storage to collect and decode such a serial bus.

As shown in the figure above, the RS232 serial port sends a string of numbers every 200ms (you can set it longer, such as 1 minute interval, for a more obvious effect), 0000000 to 9999999999 and sends it in a loop. We start the trigger and the PC buffer function of 500 frames as segmented storage.

As shown in the figure above, the data of the 6th frame of segmented storage is 0000000, and it is decoded. In the figure below, the data of the 7th frame is 1111111, and it is decoded.

The 8th frame of the segmented storage in the figure below has data 2222222, and it is decoded...

In this way, until the end, each frame is captured as a data packet and decoded, and 500 frames can be captured without omission, no matter how long the entire sending process is.