Trigger-Happy Scopes
by Tom Lecklider, Senior Technical Editor
Ideally, a scope would automatically examine its input signals, configure its trigger circuits, and display the signal artifact you wanted to see. Instead, you generally need to choose the triggering conditions yourself based on experience.
In a simple analog cathode-ray oscilloscope, the displayed trace is vertically deflected by an amplified version of the input signal. Horizontally, a sweep circuit applies a sawtooth signal to move the trace from the left edge to the right edge at a constant speed. Triggering synchronizes the two axes by timing the start of the sawtooth waveform to coincide with some aspect of the input signal, typically a level crossing or edge.
Because there is a huge number of different signal conditions, many distinct trigger capabilities have been developed, quickly leading to confusion unless terminology is clearly defined. Most manufacturers refer to the trigger point as the exact instant that begins a displayed trace or stored record.
Even this definition requires qualification because pre- and post-trigger delays reposition the trace or record relative to the actual trigger point. Nevertheless, whatever processing may occur, the final outcome is called the trigger, and it has a fixed relationship to the stored or displayed waveform. Many scopes indicate the trigger position by a small T on the display.
Another term used further forward in the triggering process is events; for example, trigger after 39 events on channel 1 where an event could be a level crossing or a certain serial data word. Event counting is one kind of computation within the process that eventually produces the trigger. In modern scopes with serial bus triggering, an event may be the successful match between a user-defined protocol state and a decoded state.
Trigger processing has grown in complexity and often is shared between hardware- and software-based techniques. Typically, software-based trigger processing examines acquired data. This means that some event must first occur, such as a positive-going edge of a data signal, to begin the acquisition process. In a simple system, the memory fills, acquisition stops, the acquired data is examined, and if conditions matching a predefined trigger state are found, a suitable trigger is established with the data displayed relative to it.
Many scopes have software-based mask triggering to qualify telecom signals. For a certain protocol and bit rate, the traces constituting an eye diagram must not fall within no-go areas defined by the mask. After an eye diagram waveform is acquired, software compares the data with the mask limits. If all the data words compare OK, another store of data is acquired, and the comparison process is repeated. Should a comparison fail, you may have options such as stopping acquisition, plotting the failed waveform, or sending an email with the failed data.
Some recently introduced scopes use special hardware to make the mask comparison, greatly reducing the time required when only software is used. Still other scopes claim on-the-fly hardware monitoring that can recognize complex trigger conditions in real time.
Agilent Technology’s Brig Asay, a senior product manager, commented on the use of FPGAs in trigger processing, “The Infiniium 9000 incorporates an FPGA dedicated to providing a rich set of hardware-based protocol triggers. In real time, the FPGA reconstructs serial packets and compares the packets against user-specified protocol triggers.”
New Agilent Oscilloscopes with Breakthrough Technology Deliver More Scope for the Same Budget
InfiniiVision 2000 and 3000 X-Series scopes offer 26 models from 70 to 500 MHz starting at $1,230 USD. Entry models offer industry-exclusive options like 8-channel Mixed Signal Oscilloscope and integrated function generator. Advanced scopes change to 16-channel MSOs and add serial bus debug options.