[Tektronix Application Sharing] Three things to consider when synchronizing oscilloscopes for higher channel counts

When building a test system, you may need to measure multiple signals, and it may not be possible to fully capture them all with just the available channels of an oscilloscope. To increase the number of oscilloscope channels in a test system, a common approach is to group multiple oscilloscopes together. Multichannel measurements are useful in a variety of scenarios, such as capturing complex particle physics experiment data, measuring large numbers of power rails, and analyzing three-phase power converters.

These measurements cover tasks such as detecting power supply crosstalk on a serial bus, analyzing radio frequency interference, and verifying the integrity of incoming input/output signals. In multi-channel applications or measurement scenarios, maintaining precise synchronization between channels is critical to accurately analyzing timing relationships within the entire system under test.

Figure 1: TekScope PC analysis software

When many signals need to be captured simultaneously, there are several ways to synchronize your oscilloscopes. Let's talk about three ways to synchronize a multi-oscilloscope measurement system using 5 and 6 Series B MSO oscilloscopes and TekScope PC analysis software (Figure 1).

1. Low latency synchronization method using external sources

The most accurate synchronization method relies on a single trigger source to distribute the trigger signal to multiple oscilloscopes. This is achieved by utilizing a power divider (BNC or SMA) to feed the trigger signal evenly to all instruments. To ensure accurate synchronization, the cables connecting the power splitter to each instrument must be of the same length and preferably phase matched. This approach minimizes lags due to propagation delay variations. By keeping propagation delays consistent through routing and power dividers, instruments can achieve synchronized trigger conditions that accurately replicate the channel-to-channel timing of a single oscilloscope.

Therefore, using a high-quality power divider is critical to ensuring the integrity of the trigger signal. The power divider acts as a balancing voltage divider by connecting the 50 ohm trigger source to the 50 ohm cable and then connecting the cable to the oscilloscope's 50 ohm input port. The divider reduces the amplitude of the trigger signal applied to each scope, so this needs to be considered when setting the trigger levels.

Remember to pay attention to the crossover specifications and trigger signal requirements. For example, for an auxiliary trigger input driving a 5 or 6 Series B MSO, the optimal signal amplitude would be greater than 500 mV. Providing a trigger signal with a larger amplitude can improve the oscilloscope's trigger system response and stability, resulting in better time lag results. Therefore, using a suitable crossover and ensuring a trigger signal of appropriate amplitude will help maintain excellent trigger signal integrity.

2. Probe-based synchronization method

If no external trigger source is available or the 50 ohm divider cannot be driven, another way to achieve synchronization is to probe the same trigger source on each oscilloscope. While this approach provides excellent timing accuracy, it comes at the cost of one channel per oscilloscope. But the skew caused by the propagation delay difference is still within the oscilloscope's lag settings.

To minimize overall skew, it is recommended to use an active probe such as the Tektronix TAP4000 (see Figure 2). The probe offers a pulse rise time of less than 115 picoseconds, making it ideal for reducing trigger jitter. In addition, the TAP4000 probe has an input capacitance as low as 0.8 pF. Note that the capacitance of each probe is cumulative, so the circuit must handle the additional load.

Figure 2: Tektronix TAP4000 single-ended low voltage probe

To use a probe-based synchronization method between two oscilloscopes, connect an oscilloscope probe (preferably a TAP4000) to the channels of each oscilloscope. Probe the same signal using the same length and type of probe tip on both probes. The trigger signal being detected must have a relatively fast rise time (approximately 50-100 picoseconds). Enable simple edge triggering on each scope, setting the trigger level to midpoint. Note that any difference in trigger levels may introduce additional lag.

Depending on the trigger source used, it should be possible to control the overall time lag in the tens of picoseconds range when using this approach. This approach is a viable synchronization solution when no external trigger source is available or when specific triggering requirements cannot be met.

3. Simplified synchronization method suitable for applications with less stringent timing requirements

This approach can provide greater setup flexibility when a multi-oscilloscope system does not require minimal time lag.

Figure 3: The auxiliary trigger output of the main oscilloscope acts as a signal source, feeding the signal to other oscilloscopes through a splitter

In the configuration above (Figure 3), the auxiliary trigger output of the main oscilloscope acts as a signal source, feeding the signal through the oscilloscope to the other oscilloscopes. In a 5 or 6 Series B MSO, there is a nominal 900 nanosecond lag between the trigger event and the auxiliary output signal. Any additional skew to the rest of the scope can be minimized by using a splitter and matching cables.

If the record length is long enough, the trigger delay setting in the horizontal marker settings can be applied to correct the lag between the trigger and auxiliary outputs. The advantage of this setup is that any channel on the main scope can act as a trigger source, freeing up all channels on the remaining scopes for signal acquisition.

These options help achieve oscilloscope synchronization when ensuring simultaneous capture of a large number of signals is critical.