Intermittent CAN Bus Fault During Gear Retraction
An aircraft has a 250 kbit/s CAN bus linking an engine data concentrator, two display units and a data-link node. The bus is 18 m long with 120 Ω terminators at each end. Intermittently, both displays show 'ENG DATA BUS' and the concentrator logs bus-off errors. Fault appears during gear retraction and sometimes on rough runway. With power off at the diagnostic connector, resistance is 60 Ω. During a vibration test, a capture shows CAN_H and CAN_L both near 2.5 V with 0.0 V differential for about 80 ms; at the same moment the resistance across the bus falls to 38 Ω. One technician wants to replace the left display because it is the first node to flag the fault. The harness passes within 20 mm of a hydraulic line.
1. Is 60 Ω normal for this bus, and what does the 38 Ω reading indicate?
Read the worked answer
Yes, 60 Ω is normal for two 120 Ω terminators in parallel: 120 ÷ 2 = 60 Ω. That reading alone shows the two end terminators are present and the bus is not open-circuit.
A fall to 38 Ω is abnormal. The extra parallel resistance is:
1/Rx = 1/38 − 1/60 = 0.02632 − 0.01667 = 0.00965
Rx = 103.6 Ω.
That is close to another 120 Ω termination, or more likely a partial short or chafed path between CAN_H and CAN_L. A chafed harness near the hydraulic line can produce a resistance that changes with flexing and vibration. The 38 Ω reading therefore points to a physical media fault, not to a display software fault.
2. Explain why the 0.0 V differential capture matters and why replacing the left display is not the first action.
Read the worked answer
A healthy CAN bus has a recessive state near 2.5 V on both lines, giving about 0 V differential, and a dominant bit with roughly 2 V differential (for example CAN_H near 3.5 V and CAN_L near 1.5 V). A 0.0 V differential for 80 ms during attempted traffic means the bus is not producing valid dominant bits. Nodes then log errors and can go bus-off.
The fault appears when the gear retracts and the harness moves, and the bus resistance changes at the same moment. That is strong evidence of an intermittent short or chafe in the harness. Replacing the left display would not correct a bus-wide media fault. The display is only reporting that it lost engine data; it is not necessarily the defective unit.
3. Give a practical isolation and repair plan.
Read the worked answer
Power the bus down and disconnect nodes one at a time. At each step measure CAN_H to CAN_L (should return toward 60 Ω with both terminators, or 120 Ω with one), CAN_H to ground and CAN_L to ground (should be high resistance). A node that makes the H-L resistance drop, or a segment that changes when flexed, is suspect.
Inspect the harness near the hydraulic line for chafe, pinched insulation or fluid contamination. Flex the harness while monitoring resistance and, if available, run a CAN analyzer. Check that both terminators are correct and that stub lengths are not excessive. Repair or replace the damaged section, restore at least 20 mm clearance from the hydraulic line and secure the harness. Repeat the vibration test and confirm no bus-off events before release.