Safety first
Never look into a fibre, connector or transceiver port. Optical light can be invisible and still harmful. Use appropriate inspection equipment and follow workplace safety procedures. Do not disconnect a production link without change approval and a documented recovery plan.
Step 1: document the symptom
Note the exact time, which ports and services are affected and whether the link is completely down, flapping or has bit errors. Save logs, alarms and DDM/DOM values before rebooting or replacing components. Then you can compare before and after.
Check if the fault coincides with cable work, software upgrade, temperature change, power outage or new patching. A clear timeline reduces the search area.
Step 2: verify port, speed and compatibility
Confirm that both ports are enabled and configured for the same speed. Check auto-negotiation where relevant, FEC, breakout and platform alarms for incompatible transceiver.
Read the transceiver ID from both ends and compare form factor, standard, wavelength and range. For BiDi, Tx/Rx wavelengths should be complementary. For duplex, the transmitting fibre at one end should reach the receiver at the other.
Step 3: inspect and clean
Contamination on the contact surface is a common cause of faults. Follow the principle of inspect, clean if necessary and inspect again. Even new factory-assembled patch cables should be inspected before connection. A dirty connector can contaminate the clean counterpart.
Use cleaning methods and tools that are intended for the type of connector. Never replace a protective cap that is not clean and store cleaned ends protected. MPO/MTP requires equipment that can inspect and clean the entire multi-fibre surface.
Step 3: inspect and clean
Contamination on the contact surface is a common cause of faults. Follow the principle of inspect, clean if necessary and inspect again. Even new factory-assembled patch cables should be inspected before connection. A dirty connector can contaminate the clean counterpart.
Use cleaning methods and tools that are intended for the type of connector. Never replace a protective cap that is not clean and store cleaned ends protected. MPO/MTP requires equipment that can inspect and clean the entire multi-fibre surface.
Step 4: read DDM/DOM at both ends
Compare Tx power, Rx power, temperature, voltage and laser bias against the module's specified limits. A low Rx value with normal Tx at the opposite end indicates attenuation, incorrect patching or receiver issues. Deviant Tx may indicate transmitter, port feeding or temperature.
DDM values are diagnostic and have measurement uncertainty. Confirm critical levels with a calibrated power meter when necessary. Save values in both directions – a fibre duplex link may have faults only in one direction.
Step 5: isolate with controlled swaps
Change one thing at a time and document the result. Start with the least intrusive component: known working patch cable, cleaned adapter or verified spare module. Do not move multiple components at the same time, as you will lose evidence of the root cause.
If the fault follows the transceiver, optics or coding is suspect. If the fault remains at the port, the port or configuration is suspect. If the fault remains on the fibre run, proceed with measurement.
Step 6: measure the cable installation
Measure insertion loss with a light source and power meter according to defined reference method. Compare with the project's loss budget. Use OTDR with launch and receive fibre to locate events and assess splices and connectors when the method is appropriate.
An approved OTDR trace does not always replace end-to-end insertion loss, and an approved insertion loss does not indicate where a fault is located. The methods answer different questions and complement each other.
Quick fault picture
- No Tx power: check port management, module support, feeding and transmitter.
- Normal Tx but very low Rx: check cleaning, polarity, cable path, splices and incorrect fibre/wavelength.
- Too high Rx: check optical class, short link and need for a dimensioned attenuator.
- Link flaps with temperature: check the module's temperature class, cooling, DDM and margin.
- One direction works: check individual fibre, duplex polarity or BiDi pair.
After action
Verify stable port status, error counter and DDM over a relevant observation period. Document replaced parts, measurement values and root cause. Update labeling and spare parts list. If the root cause was contamination, the inspection and cleaning procedures should be improved – not just the individual connector.
Next step
Is the fault time-critical? Verify alarms, device model, images of the transceiver labels, DDM values and link length. We will help you to narrow down the fault and choose the next checkpoint.