WaveExpert Series Sampling Oscilloscopes

WaveExpert 100H

Normalized Q-Scale analysis is performed on each edge of the data pattern. The slope of the linear portion is a measure of the random jitter while the separation of the lines at Q=0 gives the amount of Bounded Uncorrelated Jitter (BUj).

• 230 fs rms intrinsic timebase jitter
• Accurate total jitter analysis at any data rate
• Jitter breakdown using Q-Scale analysis
• Random jitter
• Data Dependent Jitter (DDj)
• Bounded Uncorrelated Jitter (BUj)
• Analysis of ALL edges in a waveform
• One-button Jitter Measurements

Conventional sampling oscilloscopes employ a sequential acquisition method which relies on an accurate time delay component to position the samples of the waveform in time. In addition to being slow, this type of sampling has high intrinsic jitter and requires a low jitter trigger signal. The patented HCIS timebase in the WaveExpert samples at rates 100 times faster and with 230 fs_rms intrinsic jitter.

The technology behind HCIS employs a phase-locked loop in the timebase which recovers the instrument's sampling clock from the bit clock of the signal under test. The advantages of this approach are fast sampling, high linearity, and low jitter over a wide frequency range. The fast sampling rate and long waveform memory of the HCIS timebase are essential elements for jitter analysis using the normalized Q-Scale technique.

The innovative normalized Q-Scale jitter analysis software used in the WaveExpert oscilloscope provides the most accurate measurements, regardless of the jitter scenario. Conventional oscilloscope-based jitter analysis relies on the accurate measurement of the jitter spectrum. This method can become inaccurate, and can overestimate jitter in cases where there is crosstalk or power supply noise. The Normalized Q-Scale method does not rely on the jitter spectrum but, instead, uses the measured jitter distribution to determine the random and bounded jitter components. When a repeating data pattern is used, the data dependent jitter can be removed from the jitter measurement, resulting in the first instrument that can measure Bounded Uncorrelated Jitter (BUj).

Complete jitter measurements utilize the coherent interleaved sampling timebase. Analysis includes total jitter, random jitter, deterministic jitter, and the components of deterministic jitter; DDj, ISI, and DCD.

Jitter analysis uses all edges in the data pattern. The slope and mean displacement from nominal is used to measure the data dependent jitter. All individual edges can be separately viewed, as shown in the center of the eye above.

The high stability coherent interleaved time base (HCIS) provides a significantly lower jitter noise floor compared to a conventional sequential sampling time base over a wide frequency range. The chart above shows the jitter performance of the standard and high stability coherent interleaved time bases over a range of bit rates.

The HCIS timebase combined with Normalized Q-Scale jitter analysis provides the highest accuracy jitter measurements regardless of the type of jitter present. This chart shows a comparison of jitter measurements on a calibrated jitter source using Q-scale and the spectral method. The WaveExpert 100H gives the most accurate measurements even in cases where large SJ (sinusoidal jitter) and BUj (bounded, uncorrelated jitter) are present. The HCIS timebase has the lowest jitter noise floor thus providing more accurate measurements than even a BERT.

Jitter analysis uses a pattern-locked signal waveform and measures every edge in the pattern. The combined jitter histogram from all edges provides the random and uncorrelated jitter.