Fujikura FSM-45PM Fusion splicer repair

Fujikura FSM-45PM fusion splicer repair

1–3 business days
NIST-traceable calibration
Free evaluation
Ships worldwide

If you splice PM fiber in a production environment, you already know the FSM-45PM. If yours is down, you also know the replacement cost. Let's fix it instead.

The Fujikura FSM-45PM is a factory-floor PM fusion splicer that ships with a lot of capability packed into one unit. PAS core alignment, automatic theta alignment for PM fiber stress rod positioning, a V-groove driving system that handles cladding diameters from 80 to 400µm, sweep arc for dissimilar fiber combinations, crosstalk estimation after every PM splice, and an automatic splice mode covering 50 PM fiber types. GPIB for power monitor feedback alignment. USB and serial port for PC communication. SpliceMaster and SpliceLab software compatibility, plus a full LabVIEW driver set for custom automation builds.
That last part is why this instrument ends up in production environments that other splicers do not. Splicing based on real-time attenuation or polarization meter feedback is a different level of process control than running fixed arc parameters and checking results after the fact. Labs doing high-value PM component assembly, gyroscope fiber winding, or coherent telecom component manufacturing use that feedback capability regularly.
The FSM-45PM was issued in 2006 and 2007. Units have been working in production for close to two decades. When they fail, the fault modes reflect that age and use profile.
Where These Units Actually Fail
Theta alignment accuracy is the performance metric that matters most on a PM splicer, and it is also the first thing to drift when the Z-axis mechanism develops wear. The FSM-45PM uses automatic theta alignment to rotate the fiber and locate the stress rod position from the PAS image before firing the arc. When the rotational drive wears or the encoder feedback loses accuracy, theta positioning error climbs. You see it in crosstalk readings that are running 2 to 5dB worse than spec on fiber prep that used to be clean. Some operators compensate by adjusting theta offset parameters in software. That covers the symptom, not the problem.
V-groove wear and contamination on a unit that has handled the full 80 to 400µm cladding diameter range is a different problem than on a standard 125µm splicer. The V-grooves on the FSM-45PM are designed for wide-diameter fiber and the surface geometry matters for accurate fiber seating. Contamination from stripped coatings and fiber debris builds up faster when you are running through large-diameter specialty fibers. We clean and inspect the V-grooves on every unit that comes in.
Arc electrode degradation is straightforward. The FSM-45PM runs a harder arc for large-diameter and dissimilar fiber work than a standard telecom splicer. Electrode wear is faster than on a unit used only for 125µm SMF. High crosstalk, elevated splice loss, inconsistent prefusion behavior, and arc shape irregularity visible on the LCD are all signals that electrode condition needs attention before the next diagnosis goes to the alignment system.
Camera and imaging system degradation affects the PAS alignment quality and the crosstalk estimation accuracy directly. The FSM-45PM uses its CCD imaging to locate fiber core position and stress rod geometry. A degraded image gives the PAS system bad data to work from. Units where the LCD image looks soft or low-contrast on known-good fiber prep need the imaging path checked before anything else.
GPIB interface failures are common on instruments this old. The GPIB port is the primary path for power meter feedback alignment in automated production setups. A dead GPIB interface does not stop the splicer from working manually, but it takes power feedback alignment off the table completely.
The heating oven on high-cycle units eventually shows inconsistent shrink quality, usually from element drift rather than outright failure. Short heat times or uneven sleeve shrink are the symptoms.
What We Do
We run splice cycles on SM and PM fiber before touching anything. Theta alignment accuracy gets verified against spec. Arc performance gets checked. V-groove condition gets inspected on both sides. GPIB and USB interfaces get tested. Camera image quality on both axes gets assessed. Heating oven timing gets confirmed.
Repair is component level throughout. Arc calibration runs after any electrode work. Theta alignment accuracy is verified after any drive or encoder work. NIST-traceable calibration with a certificate comes with every completed repair. Test report with pre and post repair measurements ships with the unit.

Service Specifications

ParameterValue
Alignment MethodPAS core alignment + automatic theta alignment
Cladding Diameter Range80µm to 400µm
PM Fiber Auto Modes50 types
Fiber TypesSM, MM, DS, NZ-DSF, EDF, DCF, PM (PANDA, Bowtie, and others)
Sweep ArcYes, for dissimilar fiber combinations
Crosstalk EstimationYes, by profile observation after each PM splice
PC InterfaceUSB, RS-232 serial
Remote ControlGPIB (power monitor feedback alignment)
Software CompatibilityAFL SpliceMaster, SpliceLab, LabVIEW drivers
Cleave Length9mm standard, 10mm, short cleave capability

Service specifications

ParameterValue
Fiber typePM fiber, single-mode, specialty
AlignmentCore-to-core with theta (rotational) axis
Arc systemAutomatic arc calibration

Frequently asked questions

How does power meter feedback alignment work on the FSM-45PM?
The GPIB interface lets the splicer receive real-time power readings from an external optical power meter during the alignment phase. Instead of relying solely on PAS image-based alignment, the splicer can iteratively adjust fiber position to maximize coupled power as measured by the meter. This is used in high-value PM component assembly where you need the splice optimized for actual transmission performance rather than estimated loss from fiber images.
My FSM-45PM crosstalk numbers have been climbing for months. We adjusted theta offset to compensate. Is the splicer fixable?
Yes, and the theta offset workaround confirms the diagnosis. When operators start adjusting theta offset to get acceptable crosstalk results, the Z-axis rotational mechanism is the almost certain cause. The offset parameter can mask maybe 2 to 3dB of angular error before the splice quality starts showing other problems. We check the drive and encoder accuracy during diagnostics and correct the mechanical source of the drift rather than leaving the software compensation in place.
Is the FSM-45PM still worth repairing or should we look at a current Fujikura PM splicer?
If the optical and mechanical systems are sound, repair is almost always the right answer economically. The FSM-45PM's 80 to 400µm cladding range and GPIB power feedback capability are not standard features on current field splicers. If your process depends on those, replacing the FSM-45PM means replacing the process too. If you just need standard PANDA splicing and those specialized features are not in use, that is a different conversation.

Documents