Understanding The Oscilloscope Jitter Specification

Introduction

Jitter is defined as the unwanted phase modulation of a digital signal, and is considered one of the most important specifications for measuring a device’s quality. To better understand why jitter is so important, consider today’s wireless communications devices, which rely on digital transmission to avoid cumulative noise-induced degradation. These devices should provide users with error-free communications, but this is often not the case. Instead, issues like mis-timing inside transmission equipment can result in jitter that degrades performance. As a result, today jitter performance is mandated by the various communications standard technologies, referenced by equipment specifications, and taken as a minimum requirement. Compliance requirements specify random jitter, deterministic jitter, time-interval error, and total jitter. In addition, some standard technologies specify measurements like bounded uncorrelated jitter and J2/J9 jitter. A jitter measurement that is too high can have a significant impact on a product–in some cases forcing it to be redesigned and, in the worst case scenario, keeping it from being shipped altogether.

Jitter can be precisely measured using a common, yet powerful tool like the oscilloscope. While most

Windows-based oscilloscopes offer analysis software to help you measure jitter, such instruments are limited by their jitter measurement floor, which can erode crucial design margins and even cause devices to fail tests unnecessarily. If the jitter of the device-under-test (DUT) is lower than the oscilloscope’s jitter measurement floor, it cannot be measured. It is incumbent on engineers, therefore, to fully understand an oscilloscope’s jitter specifications before making any purchase. Doing so will not only ensure you select the right oscilloscope, but also enable more accurate measurements and faster time to market.

In an oscilloscope, the most important jitter specification is the jitter measurement floor. This specification is a combination of the oscilloscope’s sample clock jitter and noise floor, and results in random jitter within the oscilloscope that can affect its accuracy. Because oscilloscope vendors specify their instrument’s jitter measurement floor in different ways, making an educated purchasing decision requires the engineer to understand what each possible specification means. These specifications include: intrinsic jitter, jitter measurement floor, and long-term jitter.

• Intrinsic Jitter

The oscilloscope’s intrinsic, or sample clock jitter, is defined as the amount of jitter it transmits using internal timing. To better understand this definition, consider that real-time oscilloscopes sample data very fast, up to 120 GSa/s. Because of this, ensuring data points are in alignment is a critical task. This can be accomplished in one of two ways, either by using a chip or a time-base system that provides the necessary tight-time correlation required between the sampled input signals delivered to the analog-to-digital (A-to-D) converter and the internal clock. Like the oscilloscope, the internal clock has its own jitter specification. The internal clock also is characterized by how well it can align the oscilloscope’s sample points through its time base. Consequently, the jitter specification of the entire oscilloscope time base is what is referred to as the sample clock jitter.

Intrinsic jitter is often specified differently by different oscilloscope vendors. Keysight Technologies, Inc., for example, specifies the intrinsic jitter of its Infiniium 90000 X-Series oscilloscope as 150 fs. Here, intrinsic jitter is meant to show the oscilloscope’s theoretical best-case jitter measurement in the presence of no other variables. Some vendors refer to this as the oscilloscope’s jitter measurement floor. Regardless, the intrinsic jitter specification alone is not sufficient to tell the engineer exactly how much jitter an oscilloscope will contribute to a measurement.