You need to test, we're here to help.

You need to test, we're here to help.

08 March 2021

TDME Primer: Serial Trigger and Sequence Mode Sampling

Figure 1: Sequence mode sampling packs multiple acquisitions
into memory with very little “dead time” between them.
Sequence mode sampling, also referred to as segmented acquisition, is a sampling mode that divides the oscilloscope’s acquisition memory into a user-defined number of equal length segments. Each segment stores a single acquisition of the triggering event, with as much buffer zone as will fit into that segment, given the total number of segments requested. Only after all segments have been acquired is the data processed and displayed. 

The real power of sequence mode becomes evident when you combine it with intelligent triggers, such as the serial data triggers delivered with TDME options

Sequence mode minimizes the occurrence of “dead time” between trigger events, resulting in a high acquisition rate of only waveforms of interest.  This makes sequence mode sampling a natural complement to the packetized data found in the serial data standards supported by the TDME options, since packets have clearly defined boundaries and other identifying characteristics that can be easily accessed through serial triggers. By using sequence mode sampling with serial trigger, you gather only the data you actually want to analyze, saving precious memory that can be used to extend the duration of time acquired.

By extension, sequence mode automatically groups segments by their triggering characteristics, eliminating the need to search a long record for what you wish to analyze. For example, sequence mode is a good way to capture protocol errors. You could use this technique to run a test overnight, seeking an error that occurs very infrequently, then compare segments to find out when the errors occur, what the time is between them, etc. 

Figure 2: Using sequence mode sampling to isolate and
capture only CAN messages containing errors.
Figure 2 shows a CAN trigger combined with sequence mode acquisition, capturing only CAN packets containing ACK error conditions. Another acquisition shows packets containing CRC errors. The red waveform overlay indicates a protocol error.

The Trigger Times table at the bottom of the display shows the time stamps for the first ten trigger events. The first segment was acquired on June 5th at 4:47. The next segment was acquired 2.8 seconds after that.  The fourth column of the table shows the "between Segment" time—which, in this case, tells us the time between errors.

Figure 3: Using serial trigger and sequence mode sampling to
acquire only CAN engine messages, ignoring other sources.
Figure 3 shows an example of using sequence mode to capture data from just a specific message source, such as when a designer working on engine messages wants to capture just engine messages while ignoring other things that are happening on the bus.

Acquiring in sequence mode while triggering on engine-related ID fields uses 5000 samples for each engine message, while all the other non-engine related messages are ignored.  This is in contrast to capturing millions of samples only a few of which are of interest, then having to sort through all that data to find what is needed. 

This technique could be refined to capture only engine-related messages with specific data content by triggering on the message ID plus a particular Data Value. Each TDME option will have different criteria that can be used to refine the trigger, but they will always be relevant to the types of messages and bus events that are of interest in that protocol.

Note: While many serial triggers support sequence mode sampling, some do not. See the software option instruction manual for any exclusions to sequence mode acquisition. In general, even if you cannot use a serial trigger, you can use sequence mode sampling with another type of trigger, and decode the segmented acquisition.

Unlike a real-time acquisition, where the total acquisition is displayed across the width of the grid, sequence mode segments can be displayed in different ways to facilitate their comparison, another advantage of sequence mode acquisition. They can be displayed side-by-side to form a traditional “waveform,” in a waterfall, tiled, and so on. 

For more on this topic, see our on-demand webinar, Measuring and Debugging CAN and CAN FD. Much of the information in it is relevant to any protocol.

Also see:

No comments:

Post a Comment