What is the total jitter budget for a high speed serial link and how do I allocate it?
Serial Link Jitter Budget
The jitter budget is the fundamental performance metric for high-speed serial link design. A link that exceeds its jitter budget at any point in the channel will have an elevated BER.
- Performance verification: confirm specifications against the application requirements before finalizing the design
- Environmental factors: temperature range, humidity, and vibration affect long-term reliability and parameter drift
- Cost vs. performance: evaluate whether the application demands premium components or standard commercial grades
- Interface compatibility: verify impedance, connector type, and mechanical form factor match the system architecture
Frequently Asked Questions
What is the biggest jitter contributor?
In most high-speed serial links: ISI (Inter-Symbol Interference) is the dominant jitter source. ISI is caused by: channel loss (the frequency-dependent attenuation of the PCB traces, connectors, and vias), reflections (impedance discontinuities at connectors, vias, and package transitions), and crosstalk (signal coupling from adjacent lanes). ISI is deterministic and pattern-dependent: certain bit patterns (e.g., isolated transitions after long runs of identical bits) experience more ISI than others. The equalizer's job is to remove ISI: CTLE (Continuous-Time Linear Equalizer) at the receiver compensates for channel loss, and DFE (Decision Feedback Equalizer) cancels post-cursor ISI. A properly equalized link reduces ISI from 0.3-0.5 UI to less than 0.05-0.1 UI.
How do I measure jitter?
Jitter measurement methods: real-time oscilloscope: captures the waveform and directly measures the timing of each edge. Jitter separation: decomposes TJ into DJ and RJ components using algorithms (e.g., RJDJ separation per MJSQ). Instruments: Keysight UXR (110+ GHz BW), Tektronix DPO70000SX (70 GHz). Sampling oscilloscope: triggers off a recovered clock and builds a histogram of the edge timing over many acquisitions. Lower noise floor than real-time scopes. Instruments: Keysight 86100D, Tektronix DSA8300. BERT (Bit Error Rate Tester): measures the actual BER as a function of sampling position (bathtub curve). The most definitive measurement (directly shows the jitter margin at the target BER). Instruments: Keysight M8040A, Anritsu MP2110A.
What is a bathtub curve?
The bathtub curve plots the BER as a function of the sampling time position within the UI (unit interval). Shape: high BER at the UI boundaries (where transitions occur), low BER in the center of the UI (where the data is stable), and the BER transitions follow a Gaussian roll-off (from the RJ component) overlaid on sharp DJ boundaries. The horizontal opening of the bathtub curve at the target BER (e.g., 10^-12) gives the jitter margin: the timing window within which the receiver can sample the data and achieve the required BER. A wider bathtub opening = more margin = more robust link.