Quick Reference
Multimode: 850nm + 1300nm wavelengths, refractive index ~1.4818, pulse 5-30 ns for short links (under 2 km).
Why SM and MM OTDR Settings Differ
Single-mode and multimode fibers have fundamentally different physics. Single-mode fiber has a 9-micron core that supports only one optical mode at the operating wavelength, with very low attenuation (0.2-0.35 dB/km) and the ability to carry signals 100+ km. Multimode fiber has a 50- or 62.5-micron core that supports many modes, with higher attenuation (1-3 dB/km) and limited reach (under 2 km).
The OTDR has to use different laser wavelengths matched to each fiber's design. It has to apply different refractive indices for accurate distance calculations. It has to use different pulse widths and event detection algorithms because backscatter levels and dead zones differ. Setting up an OTDR for the wrong fiber type produces traces that may look plausible but are wildly inaccurate.
Wavelength Selection
Single-Mode Wavelengths
- 1310nm: Original SM operating window. Higher attenuation (0.32-0.40 dB/km) but immune to most macrobend losses. Test here for connector and splice loss baseline.
- 1550nm: Lowest-attenuation window for SM (0.18-0.25 dB/km). Standard for long-haul and the wavelength most sensitive to macrobends. Always test at 1550 to catch bend issues invisible at 1310.
- 1625nm: Used for L-band, DWDM, and live-fiber testing on networks running 1310/1550 in service. The 1625nm filter on the OLT lets you test active fiber without disrupting traffic.
- 1490nm: Downstream wavelength for GPON systems. Test here when characterizing GPON splitter networks.
Multimode Wavelengths
- 850nm: VCSEL operating wavelength for 10G/25G/40G/100G short-reach Ethernet over MM. Higher attenuation (~3 dB/km) but matches the laser source. Test at 850 for any modern data center MM link.
- 1300nm: LED operating wavelength for legacy MM systems and longer MM links. Lower attenuation (~1 dB/km). Test at 1300 for systems that operate at this wavelength.
For any link, the rule is: test at every wavelength the fiber will carry in service, plus the wavelength most likely to reveal problems (1550 for SM, 850 for MM data center).
Refractive Index (IOR)
The OTDR calculates distance from the time-of-flight of returning light. Light travels through fiber at c/n, where c is the speed of light in vacuum and n is the fiber's refractive index. If the OTDR uses the wrong n, the calculated distances are off proportionally.
Common IOR Values
| Fiber Type | Wavelength | Typical IOR |
|---|---|---|
| Single-mode (G.652) | 1310nm | 1.4675 |
| Single-mode (G.652) | 1550nm | 1.4681 |
| Single-mode (G.652) | 1625nm | 1.4683 |
| Multimode 50/125 OM3/OM4 | 850nm | 1.4818 |
| Multimode 50/125 OM3/OM4 | 1300nm | 1.4790 |
| Multimode 62.5/125 OM1/OM2 | 850nm | 1.4960 |
| Multimode 62.5/125 OM1/OM2 | 1300nm | 1.4910 |
The cable manufacturer's spec sheet has the exact IOR for that fiber. If you do not have it, the typical values above are accurate to within 0.05% for distance estimation. For pinpoint fault location on long links, use the manufacturer's exact value.
Pulse Width Selection
Pulse width controls the trade-off between dynamic range (how far the OTDR can see) and resolution (how close two events can be and still be distinguished).
Single-Mode Pulse Widths
- 5-30 ns: FTTH drops, patch panels, short links under 1 km. High event resolution.
- 30-100 ns: Distribution cables, links 1-10 km, typical building-to-building.
- 100-300 ns: Backbone links 10-30 km, OSP runs.
- 300-1000+ ns: Long-haul links over 30 km.
- 10-20 microseconds: Ultra long-haul, typically only available on premium long-haul OTDRs.
Multimode Pulse Widths
- 5-10 ns: Short MM links under 200m, data center. Best resolution for closely spaced events.
- 10-30 ns: Standard MM links 200m to 2 km.
- 30-100 ns: Long MM runs over 2 km (rare, MM is usually short).
MM fiber typically uses shorter pulses than SM because MM links are shorter and need higher resolution to distinguish closely spaced events at patch panels and connector cassettes.
Distance Range
Set the distance range to roughly 1.5-2x your expected fiber length. Too short and the trace cuts off before the far end. Too long wastes acquisition time and compresses events horizontally.
- SM short (FTTH drop, patch): 500m - 2.5 km range
- SM medium (distribution): 2.5 - 25 km range
- SM long-haul: 25 - 250 km range
- MM short (data center): 250m - 2.5 km range
- MM long: 2.5 - 10 km range
Averaging Time
Longer averaging produces cleaner traces with better event detection. The trade-off is acquisition time.
- 15-30 sec: Quick troubleshooting, link verification.
- 30-60 sec: Routine acceptance testing.
- 60-180 sec: Documentation-grade traces, long links, low-loss event detection.
- 3+ minutes: Very long links, dynamic range limited, looking for faint events.
MM fiber has higher backscatter levels so MM traces are typically cleaner than SM at the same averaging time. SM long-haul benefits most from extended averaging.
Event Detection Thresholds
The OTDR identifies events by looking for changes in the backscatter slope or sudden reflections. Detection thresholds determine the smallest event the OTDR will flag.
- Loss threshold: Typical 0.05 dB. Lower (0.02 dB) on premium SM testing where you need to catch every fusion splice. Higher (0.10 dB) on MM where backscatter noise is higher.
- Reflectance threshold: Typical -65 dB for SM, -55 dB for MM. Below these levels, reflective events get filtered out.
- End-of-fiber threshold: Typical 5 dB. Above this drop the OTDR considers the fiber ended.
- Splitter loss threshold: For PON testing, the OTDR needs a higher loss threshold (around 5-15 dB depending on splitter ratio) to recognize the splitter as a single event rather than misinterpreting it.
Side-by-Side Setup
| Setting | Single-Mode | Multimode |
|---|---|---|
| Wavelengths | 1310, 1550 (1625, 1490 optional) | 850, 1300 |
| IOR (typical) | 1.4675-1.4683 | 1.4790-1.4960 |
| Pulse width range | 5 ns - 20 us | 5 ns - 100 ns typical |
| Distance range | 500m - 250 km | 250m - 10 km |
| Typical attenuation | 0.18-0.4 dB/km | 1-3 dB/km |
| Splice loss target | under 0.10 dB | under 0.30 dB |
| Connector loss target | under 0.5 dB | under 0.75 dB |
| Dead zone (event) | 0.5-3m | 1-5m |
| Common applications | FTTH, OSP backbone, long-haul | LAN, data center, building backbone |
Tools for Both Fiber Types
Fiber Ranger OTDR
Dual-mode OTDR supporting both SM (1310/1550) and MM (850/1300) wavelengths in one instrument.
Optical Power Meter LC
End-to-end power measurements complementing OTDR characterization at all common SM and MM wavelengths.
OM4 MM Patch Cord
For MM launch fibers and patch testing in OM3/OM4 environments.
SM LC/UPC Patch Cord
For SM launch fibers and SM patch testing.
WiFi Fiber Microscope
Inspect connectors before testing — endface contamination shows on traces regardless of fiber type.
LC Connector Cleaner
Clean LC connectors common to both SM and MM patch panels.
Frequently Asked Questions
What wavelengths should I use for SM OTDR?
1310nm and 1550nm minimum. Add 1625nm for long-haul, DWDM, and live-fiber testing. Add 1490nm for GPON characterization.
What wavelengths for MM?
850nm and 1300nm. Test 850 for VCSEL-based modern data center systems, test 1300 for legacy LED-based systems and longer MM runs.
Why do MM and SM traces look different?
MM has higher attenuation (steeper slope), higher backscatter (brighter trace), wider dead zones (less event resolution), and shorter usable distances. The interpretation rules are the same; the numbers differ.
Do I need separate OTDRs for SM and MM?
Most modern OTDRs are dual-mode. Check the spec sheet — entry-level units may be SM-only or MM-only. Carrier and FTTH techs typically need only SM; data center techs typically need both.
What refractive index do I use?
Use the cable manufacturer's spec value if you have it. Otherwise the standard values: SM ~1.4675-1.4683, MM 50/125 ~1.4790-1.4818, MM 62.5/125 ~1.4910-1.4960.
Related Reading
OTDR and Test Equipment
Dual-mode OTDRs, power meters, patch cords, and connector cleaners for both single-mode and multimode fiber characterization.