Dirty endface that won't clean up
very common
“I cleaned it twice and it still reads high. Every time I plug it back in I get a different number.”
Likely causes
-
The dust cap was the contaminant
most commonBrand-new factory jumper, dirty on the very first scope. Caps hold mould-release compound and shed plastic. A ring of fine particles around the ferrule edge is the signature.
-
You are only cleaning one side of the connection
most commonJumper scopes clean, you mate it, unmate it, and it is dirty again. The contamination lives on the far connector or inside the adapter sleeve and transfers on every mate.
-
Dry-cleaning a wet contaminant
commonScope shows the debris smeared wider than before, or drag marks radiating out from the core. Oil and cable gel need wet-then-dry; a dry wipe just spreads them.
-
Finger oil or skin contact
commonA smear or arc with no hard particles, usually off centre. Comes off completely on the first wet-dry cycle.
-
Permanent endface damage: pits, chips or scratches through the core zone
occasionalIdentical image after three clean cycles; the defects never move between cleans. Usually paired with a reflectance spike on the OTDR. Caused by mating a gritty ferrule, or by an APC/UPC mismatch.
-
Epoxy or polish residue on a field-terminated connector
rareHaze or a crescent of residue, and only on connectors somebody polished on site. Factory pigtails don't do this.
What to bring
- fiber inspection scope with LC, SC, MPO and bulkhead tips
- cleaning cassette
- 1.25 mm and 2.5 mm cleaning sticks
- 99% IPA or fiber-grade solvent, lint-free wipes
- spare adapters and jumpers
Safety
Never look into a connector, an adapter or a fiber end, and never put a direct-view eyepiece microscope on a fiber you have not proven dark with a power meter: a live PON or DWDM port delivers invisible energy to the retina with no warning and no pain. A video probe images onto a camera sensor, so it protects your eye but not itself: live power destroys the sensor. Either way, prove it dark first, then inspect, and cap every port you open. ANSI Z136.2.
Steps
-
Step 1: Scope it before you clean it
You need to know whether you have a particle, a smear or damage: they have three different fixes. You also want the before image if this turns into a dispute. Core-zone contamination matters far more than a cladding speck; inspect against the zone criteria in IEC 61300-3-35.
If that doesn’t do it
Scope won't focus or shows nothing: check you have the right tip for that connector and adapter, and that you are looking at the ferrule and not the adapter face.
-
Step 2: Clean wet-then-dry, then scope again
99% IPA or a purpose-made fiber solvent on lint-free, then dry. Dry-only on a smear makes it worse. Never canned air, never your shirt, never a wipe you already used.
If that doesn’t do it
Improved but not clean: repeat. Three full cycles, then stop and treat it as damage.
-
Step 3: Clean both sides of every connection, including inside the adapter
1.25 mm stick for LC, 2.5 mm for SC, ST and FC. Bulkhead-mounted connectors and the back side of every adapter are where contamination hides. On an MPO path clean the trunk end too: one dirty position takes the whole link down.
If that doesn’t do it
Loss returns after every mate: the mating connector or the adapter sleeve is the source. Replace the adapter first, it's cheaper. If a known-good cord still reads high in that one port with a clean adapter bore, the sleeve may be cracked or sprung rather than dirty: the adapter-sleeve sequence.
-
Step 4: Scope the tester's own reference cords
Reference cords get mated hundreds of times a day. They are the most-used and least-inspected fiber on site, and a dirty one poisons every result you took today.
If that doesn’t do it
Reference cords dirty or damaged: replace them, re-reference, and retest anything you certified since the last known-good reference.
-
Step 5: Clean, mate, unmate, then scope again
This one move separates a one-off speck from a transfer source. If it comes back dirty, stop cleaning the jumper and start on what it mates into.
If that doesn’t do it
Still transferring: replace the adapter, then the mating jumper.
-
Step 6: Replace or re-terminate anything that stays dirty
Three failed cleans is damage, not dirt. Scrubbing a pitted face grinds the defect in and contaminates your sticks for the next twenty connectors.
If that doesn’t do it
Re-terminated and still high: the endface isn't the fault. Work the loss-budget sequence.
References
- IEC 61300-3-35 (endface inspection criteria and zones)
- ANSI/TIA-568.3 (connection performance)