Measuring the insertion loss of a Single Mode jumper would seem to be a simple matter but there are a few complications to consider.

Insertion Loss is a relative measure, it’s the reduction in power when an additional passive element is added to an optical circuit. So, measuring insertion loss is a two part process - Measure the power coming through a reference path, then power after the DUT has been inserted. The difference is IL.

The loss of an unmated connector can be mathematically estimated but it cannot be measured, only a connection, with a mated pair of connectors can be measured.

In practical measurement setups, test cables are used to deliver (and sometimes collect) the light to (from) the Device Under Test. So, the quality of these test cables has a direct bearing of the IL reported for the DUT, as the test cable is always part of the connection measured. Unfortunately, it’s not possible in any meaningful way to divide up the measured test connector/DUT connector loss and assign some part of the loss to the test cable and the rest to the DUT. If the test connector were perfect, in every detail, then all loss could then be attributed to the DUT, but no fiber optic cable is perfect, all connectors introduce some loss however small, including test cables.

A good part of the loss across the connection is due to a mismatch of one or more geometrical parameters of the connectors. Consider “Lateral Core Offset” where the fiber core is not perfectly concentric with the ferrule.  If mating connectors have a different offset, or offset in a different direction, some light will be directed into the cladding and lost. This offset can be due to a fiber in which the core is not centered, poor ferrules with fiber holes that are not concentric or poor assembly where debris is pushing the fiber to one side of the ferrule hole. For a single mode connection, an offset of 0.3 micron will produce a loss of about 0.4 dB

An important aspect of this offset loss across a pair of connectors is that it’s not a simple scalar quantity, it depends on the direction of the offset.  If offsets are of a similar degree and direction, the cores will match up and the loss could be quite small, but oriented in the opposite direction produces a high loss.

This directional effect can produce some disturbing results. In a dual test cable measurement, if the launch and receive fibers have offsets in opposite directions, the reference path (when they would be connected together) will have significant loss.

Drawing 1. REFERENCE


Subsequently a DUT could be connected between these connectors, who’s offsets (on either end) match those of the test cables to which they are connected. Measured power could now be higher than the reference power. The measurement of this DUT will show negative loss or gain, which is clearly impossible.


Drawing 2. NEGATIVE IL


Swapping the DUT around, so both ends are now badly mismatched with the test cables will show a significant loss. Only if the test connectors are perfect, can this situation be avoided. Good quality test cables should be ordered or made and they should be tested for geometry and loss before use and periodically inspected for wear and scratches.


Drawing 3. HIGH LOSS




Additional resources from the FOC team include:


Follow Chris at @TestExpert_FOC

English English Español Español