How to Test Fiber Splice Loss with Optical Power Meter & Light Source – Complete Field Guide

Introduction: Why Splice Loss Testing Matters

Every fiber optic splice introduces a small amount of signal loss. In a long-haul backbone or a dense FTTH deployment, even a 0.1 dB excess loss per splice compounds across hundreds of joints — degrading link budget, increasing BER, and shortening the effective reach of your network. The Optical Power Meter (OPM) + Stable Light Source (SLS) method — also called the insertion loss method — is the most direct, cost-effective, and IEC-standardized way to verify splice quality in the field.

This guide walks you through the complete workflow: equipment selection, reference calibration, live measurement, and pass/fail evaluation.

1. Key Concepts

• Insertion Loss (IL): Total optical power lost through a splice. Expressed in dB. Lower is better.
• Splice Loss Formula: Loss (dB) = P₀ (dBm) − P₁ (dBm)
• IEC 61300-3-4: The international standard governing insertion loss measurement for fiber optic splices.
• Test Wavelengths: 1310 nm (O-band, SM) and 1550 nm (C-band, long-haul). Multimode: 850 nm and 1300 nm.

2. Equipment Required

2.1 Stable Light Source

The SLS injects a stable, calibrated optical signal into the fiber under test. Look for dual-wavelength output (1310/1550 nm), stability of ±0.05 dB or better, and connector type matching your fiber plant.

The EXFO ELS-50 Optical Light Source delivers dual-wavelength (1310/1550 nm) output with ±0.10 dB stability and 60-hour battery life — ideal for all-day FTTH and backbone field campaigns.

2.2 Optical Power Meter

The OPM measures received optical power at the far end. Key specs: InGaAs detector for SM (1260–1650 nm), measurement range typically −70 to +10 dBm, accuracy ≤ ±0.2 dB.

For entry-level FTTH work, the TC-100 Optical Power Meter is compact and reliable. For demanding backbone and data center environments, the EXFO FiberBasix EPM-50 offers 300-hour battery life and high-accuracy InGaAs detection for both SM and MM fibers.

2.3 Supporting Accessories

• Reference launch cables (mandrel-wrapped, 1–2 m SM or MM patch cords)
• Fiber optic cleaning kit: IPA wipes, lint-free swabs, one-click cleaners
• Connector adapters: SC/APC, SC/UPC, LC/UPC, FC/PC — match your plant

3. Pre-Test Preparation

Step 1: Warm Up Your Instruments

Power on both the SLS and OPM and allow them to stabilize for at least 5 minutes before taking any reference measurement. Laser output power drifts during warm-up; skipping this step introduces systematic error into every subsequent reading.

Step 2: Clean All Connectors

Use a one-click cleaner or IPA-dampened lint-free swab on every connector end-face. Inspect with a fiber scope (200× minimum) before mating. A single contaminated connector can add 0.5–3 dB of apparent loss, completely masking the true splice loss.

Step 3: Set Wavelength on the OPM

Set the OPM to the same wavelength as the SLS output (e.g., 1310 nm). Using the wrong wavelength calibration introduces a systematic offset error due to the detector’s wavelength-dependent responsivity curve.

Step 4: Select Measurement Units

Set the OPM to display in dBm for the reference step, then switch to dB (relative loss) for the splice measurement step. Most modern OPMs have a Set Reference function that automates this.

4. Establishing the Reference Power Level

Method A: One-Cord Reference (IEC 61300-3-4 Method B)

1. Connect a single reference launch cable directly from the SLS output port to the OPM input port.
2. Allow the reading to stabilize (5–10 seconds).
3. Press the Set Reference / Zero button on the OPM. The display should now read 0.00 dB.
4. Do not disconnect the SLS end of the reference cable after setting the reference.

⚠️ Important: Any movement or reconnection after setting the reference invalidates the baseline measurement.

Method B: Two-Cord Reference (IEC 61300-3-4 Method A)

1. Connect Launch Cable A from SLS → mating adapter → Receive Cable B → OPM.
2. Set reference with both cables mated.
3. Insert the splice under test between the two cables.
4. The OPM now reads the loss of the splice only.

5. Performing the Splice Loss Measurement

Step 5: Connect the Fiber Under Test

1. Without disturbing the SLS connection, disconnect the reference cable from the OPM.
2. Connect the fiber path: SLS → Launch Cable → [Splice Under Test] → Receive Cable → OPM.
3. Ensure all connectors are clean and fully seated.

Step 6: Read and Record the Loss

1. Allow the reading to stabilize (5–15 seconds).
2. Record the displayed loss value in dB — this is your insertion loss at the splice.
3. If no reference-set function: Loss = P₀ (dBm) − P₁ (dBm).
4. Repeat 3 times and average the results to reduce random noise.

Step 7: Test at Both Wavelengths

Switch the SLS to 1550 nm, update the OPM wavelength setting, re-establish the reference, and repeat. A splice that passes at 1310 nm may show higher loss at 1550 nm due to macro-bend sensitivity.

6. Bidirectional Measurement (Best Practice)

Splice loss is not always symmetric. The true splice loss is the average of both directions:

True Splice Loss = (Loss A→B + Loss B→A) ÷ 2

For acceptance testing on backbone and long-haul links, always perform bidirectional measurements and report the average. For routine FTTH drop testing, unidirectional measurement is typically sufficient.

7. Pass / Fail Criteria

If a splice exceeds the maximum acceptable loss, it must be re-spliced. Do not attempt to accept a high-loss splice by adjusting the link budget — it will degrade further over time.

8. Troubleshooting High Splice Loss Readings

• Dirty connectors: Re-clean all end-faces and re-measure. This is the #1 cause of false high-loss readings.
• Wrong wavelength setting: Confirm OPM wavelength matches SLS output.
• Instrument not warmed up: Allow 5+ minutes and re-establish reference.
• Reference cable damaged: Inspect and replace if necessary.
• Fiber type mismatch: Confirm both fibers are the same type (e.g., G.652D to G.652D).
• Poor cleave quality: Inspect cleave angle (<0.5° for SM) before re-splicing.
• Macro-bend in fiber: Check for tight bends (<30 mm radius for SM) near the splice closure.
• Splice protector not fully shrunk: Re-heat the heat-shrink sleeve if necessary.

9. Documenting Your Test Results

Professional splice loss documentation should include: date and technician name, fiber type and cable ID, instrument make/model and calibration date, test wavelength(s), reference power level (dBm), measured loss (dB), pass/fail result vs. project specification, and any corrective actions taken.

10. OPM + SLS vs. OTDR: When to Use Which

Best practice: Use OPM + SLS for acceptance testing of every splice, and OTDR for periodic network health checks and fault diagnosis.

Recommended Test Equipment

• EXFO ELS-50 Optical Light Source — 1310/1550 nm, ±0.10 dB stability, 60-hour battery.
• EXFO FiberBasix EPM-50 Optical Power Meter — 300-hour battery, high accuracy, SM & MM.
• TC-100 Optical Power Meter — Compact, affordable, field-ready.
• HSV-300P Optical Power Meter — Wide dynamic range, multi-wavelength calibration.

Frequently Asked Questions

What is an acceptable splice loss for a fusion splice?

For single-mode fiber, the industry target is <0.05 dB per fusion splice, with a maximum acceptable value of 0.10 dB per IEC 61300-3-4. Modern fusion splicers routinely achieve 0.01–0.03 dB on well-prepared fibers.

Can I use an OPM without a stable light source?

No. An OPM is a passive receiver and requires an active light source to inject a signal. Without a calibrated SLS, you cannot establish a reference power level and cannot measure insertion loss.

How often should I clean connectors during testing?

Clean every connector every time you mate it — no exceptions. One-click cleaners make this fast and consistent in the field.

Why does my splice loss reading change when I wiggle the fiber?

Movement-sensitive readings indicate a poorly seated connector, a damaged end-face, or a macro-bend near the splice. Re-clean, re-seat, and inspect all connectors before re-measuring.

Do I need to test at both 1310 nm and 1550 nm?

For single-mode links, yes — especially for backbone and long-haul applications. A splice may pass at 1310 nm but show elevated loss at 1550 nm due to macro-bend sensitivity.

Conclusion

The OPM + Stable Light Source insertion loss method is the gold standard for verifying fiber splice quality in the field. By following the reference calibration procedure carefully, testing at both operating wavelengths, and applying the correct pass/fail criteria, you can confidently accept or reject every splice in your network — protecting link budget, ensuring service quality, and building a defensible test record for network handover.

For questions about instrument selection or test methodology, contact our technical team — we are here to help you choose the right tools for your specific fiber plant and application.