Urea Dosing Anomaly Root-Cause Tree: Pump Calibration Drift, Injector Coking, Line Crystallization, and CAN Bus Timing Errors

Urea dosing is governed by a closed-loop control architecture where the ECM calculates required NH₃ based on NOₓ sensor feedback, exhaust temperature, and flow rate, then commands the dosing pump and injector via CAN. The pump delivers pressurized AdBlue® (typically 5–10 bar), while the injector—electrically pulsed—atomizes it into hot exhaust. Anomaly detection begins when commanded vs. actual flow diverges beyond tolerance bands.

Deeper inspection reveals three physical domains of failure: fluidic (pump calibration, line blockage), thermal (crystallization kinetics, injector heating), and digital (CAN timing integrity). For example, crystallization isn’t just about temperature—it’s governed by Fickian diffusion of water vapor out of stagnant urea pockets, accelerated by surface roughness and metal ion catalysis (Fe³⁺, Cu²⁺). Similarly, injector coking follows Arrhenius kinetics: biuret formation accelerates exponentially above 140°C exhaust gas temperature near the nozzle.

At the systems level, CAN timing errors expose architectural weaknesses: many Tier 4 Final engines use single-wire CAN for cost reasons, sacrificing deterministic latency. A 30 µs jitter may seem negligible—but at 100 Hz dosing frequency, it shifts pulse edges by 3% of the period, misaligning NH₃ injection with the optimal 0.8–1.2 s window for NOₓ–NH₃ reaction in the catalyst brick. This demands not just component replacement, but bus topology redesign—including stub-length optimization and common-mode choke placement per SAE J1708 Annex B.