If you work with signals that have relatively long idle times between low-duty-cycle pulses or bursts of signal activity, then you'll love the segmented memory feature available in today's scopes. Segmented memory allows you to capture more selective signal details with less memory. With segmented memory, the scope’s acquisition memory is divided into multiple smaller memory segments. This enables your scope to capture a whole bunch of successive single-shot waveforms with a very fast re-arm time — without missing any important signal information. In simpler terms it is a way for you to optimize your scope memory usage by only capturing the signal segments you are interested in. The second advantage segmented memory provides is a user interface that allows you to easily view the signal segments to check overall signal quality or to quickly find that needle in a hay stack bug that you know is out there. After a segmented memory acquisition is performed, you can easily view all captured waveforms overlaid in an infinite-persistence display and quickly scroll through each individual waveform segment. Common applications for this type of oscilloscope acquisition include high-energy physics measurements, laser pulse measurements, radar burst measurements, and packetized serial bus measurements. Below is an example of using segmented memory in a high-energy physics application.
High-energy physics and laser pulse applications
Segmented memory acquisition in an oscilloscope is commonly used for capturing electrical pulses generated by high-energy physics (HEP) experiments, such as capturing and analyzing laser pulses. With segmented memory acquisition, the scope is able to capture every consecutive laser pulse, even if the pulses are widely separated. the figure below shows the capture of 300 successive laser pulses with a pulse separation time of approximately 12 µs and an approximate pulse width of 3.3 ns. All 300 captured pulses are displayed in the infinite-persistence gray color, while the current selected segment is shown in the channel’s assigned color (yellow for channel 1).
Note that the 300th captured pulse occurred exactly 3.62352380 ms after the first captured pulse, as indicated by the segment time-tag shown in the lower left-hand region of the scope’s display. With the scope sampling at 4 GSa/s, capturing this amount of time would require more than 14 Megapoints of conventional acquisition memory. If these laser pulses were separated by 12 ms, the amount of conventional acquisition memory to capture nearly 4 seconds of continuous acquisition time would be more than 14 Gigapoints. Unfortunately, there are no oscilloscopes on the market today that have this much acquisition memory. But since segmented memory only captures a small and selective segment of time around each pulse while shutting down the scope’s digitizers during signal idle time, scopes can easily capture this much information using just 8 Megapoints of memory
Agilent’s InfiniiVision Series oscilloscopes are the only scopes in the industry that not only provide segmented memory acquisitions simultaneously on all analog channels (up to four analog channels) and logic channels (up to 16 digital channels) of acquisition, but they also are the only scopes that provide hardware based serial decoding on packetized serial data for each captured waveform segment. The InfiniiVision Series includes the scope I have on my bench, the MSO7054A (although now they are on the B model so looks like I need to upgrade).
Click here for more on Agilent's segmented memory feature
Click here for more info on the scope I have on my bench (the B model though)