The method is limited in three ways: - Reference fiber - Empirical cutoff wavelength equations - OTDR nonlinearity The measurement is made by comparison to a reference pigtail fiber with known values of MFD and cutoff wavelength. This reference fiber should be of a similar single-mode design as the fiber that is being characterized. An example of a fiber design difference is that of matched clad fiber vs. depressed clad fiber. Note: An empirical mapping can sometimes be used for characterization of a fiber of one design with a reference fiber of another design. This mapping is specific to the design pair. The equations for cutoff wavelength are derived from considering ideal fiber designs and the theoretical cutoff wavelength definition. Measured cutoff wavelength is not a theoretical value@ and varies with the definition@ i.e.@ fiber cutoff (FOTP-80) or cable cutoff (FOTP-170). Empirical relationships that follow the theoretical forms are evaluated for each fiber design and cutoff wavelength definition. The measurement is limited to MFD and cutoff wavelength at the reference-sample joint because OTDRs are nonlinear. This attribute is often specified by instrument manufacturers. Although typical specification values are sufficient for attenuation coefficient measurements@ they are not sufficiently stringent to allow accurate characterization of MFD and cutoff wavelength over the entire fiber length.