Fusion Splicer Arc Calibration at Altitude with Worn Electrodes
At a 3,200 m elevation site, a crew is splicing 288 single-mode fibers in an outdoor cabinet. Morning temperature is 2°C and relative humidity is 35%. The fusion splicer has 4,850 arc counts on a set of electrodes rated for 5,000 arcs; the arc power is left at the factory value of 100. After 40 splices, the splicer's estimated losses are 0.01-0.03 dB, but OTDR testing shows ten splices with bidirectional averages of 0.25-0.45 dB. Splice images show slight waist thinning and occasional bubbles. The crew proposes increasing arc power to 120 and continuing. The project specifies 0.15 dB maximum bidirectional splice loss. You have the splicer maintenance log, the splice images, and the OTDR results.
1. What are the likely root causes of the elevated splice losses, and why is simply raising arc power from 100 to 120 a poor decision?
Read the worked answer
The likely causes are a combination of electrode wear, high-altitude arc behavior, and cold ambient conditions. At 4,850 arcs the electrodes are near the end of their rated life, so arc stability and shape are degraded. At 3,200 m the lower air density changes arc characteristics, so a factory arc power calibrated at low elevation may not produce the correct heat profile. The images support this: waist thinning suggests insufficient or poorly distributed heat, while bubbles suggest contamination or unstable arc energy.
Raising arc power to 120 without calibration is a poor decision because it may over-fuse the fiber, enlarge bubbles, increase taper or bulge, and reduce splice strength. It also treats the symptom while leaving worn electrodes and an uncalibrated arc in service. The correct response is maintenance and calibration, not a blind power increase.
2. Give a corrective sequence before the crew resumes production splicing.
Read the worked answer
Stop production splicing and quarantine the affected splices. Replace the electrodes, clean the V-grooves and fiber clamps, and confirm that the cleaver is producing good angles. Set the splicer's altitude or ambient correction if it has that function, then run the manufacturer's arc calibration at the actual site conditions. Use test splices from the same fiber type to verify arc shape and estimated loss before returning to production.
After calibration, splice a small test batch and measure it bidirectionally with the OTDR. Do not rely on the splicer's estimated loss alone; the mismatch between 0.01-0.03 dB estimated and 0.25-0.45 dB measured shows why. Resume only when the test batch meets the 0.15 dB bidirectional limit. Record the electrode change, arc calibration result, and test-batch data.
3. How should the already-installed splices be dispositioned?
Read the worked answer
All 40 installed splices should be measured bidirectionally at the project wavelength and compared with the 0.15 dB limit. The ten splices already averaging 0.25-0.45 dB should be cut out and re-spliced after the splicer is corrected. Do not accept them because the splicer displayed a low estimate, and do not try to compensate by increasing arc power on the existing splices.
Splices that were not measured should not be assumed good. If time is limited, test every splice in the affected batch, prioritize the worst events, and document which splices were reworked. After rework, retest both directions and update the splice record with the new average, date, and splicer maintenance status.