Fault fighting

The testing regimes for fibre are somewhat different to those for copper. In many ways fibre is much simpler – there are no complex issues like impedance.

The defining factor for any fibre network is the power (or loss) budget. The power budget is system dependent, with the needs of both transmitters and receivers varying greatly. This is expressed in dB and is the difference between the launch power of the transmission equipment and the power required by the receiver.

Good practice dictates that the loss requirements in a multimode system for a pair of connectors should be <0·3 dB and any fusion splice should be 0·1 dB. Power budgets have a safety margin to account for any future expansion, poor maintenance etc. For example, if the source transmits at 0 dB and the receiver sensitivity is -20 dB, allowing a safety margin of 7 dB the power budget is 13 dB to cover losses in the fibre, splices and the connectors that form the link.

Return loss is not an issue with multi-mode systems, however laser transmitters are used with single-mode rather than led, and the performance of the laser can be significantly affected by the amount of power (light) reflected back in from the fibre network.

Test equipment
The most useful tool for testing fibre networks is the light source and power meter. This passes a prescribed amount of light into the fibre under test and then measures the amount received at the far end, displaying the figure – the insertion loss – in dB. If the insertion loss is greater than the power budget then, simply, the system will not work.

A much more complex piece of test equipment is the optical time domain reflectometer (OTDR). This only requires access to one end of the fibre. The OTDR launches a pulse of light down the fibre and then measures the reflections from any change in refractive index that occurs along the way, for example from a splice, connector or the cable end. It calculates the distance of each such feature from the "echo" time of the returned light, and can display the insertion and return losses.

You may think that this is the only kit needed to test all fibre installations. Unfortunately, not so:

• OTDRs are expensive – prices range from £5000 to £30 000.

• The equipment has a dead zone. This means that a certain length at the start of the cable under test, often as much as 100 m, is not visible to the OTDR. Until the pulse of light launched has stabilised, the reflections caused by the first connectors are so severe that the machine cannot recognise them meaningfully.

• A reasonable level of skill is required to interpret the results – the OTDR displays a complicated trace, which to the layman can be meaningless. However, some have auto-test facilities to enable quick testing of fibres.

• The resolution of the kit is limited – its ability to differentiate between features that are close together is questionable. Although there are now OTDRs on the market with extremely fine resolution and zero dead zones, they can be extremely expensive.

One further piece of test kit available is the live fibre identifier. This is used to detect traffic along the length of a fibre without needing to access either end – simply apply a small bend to the fibre and any light escaping in either direction is detected. But caution must be taken as use of the live fibre identifier may adversely affect the operating fibre.

Faults to look out for
What exactly is all this equipment used to test for? The article has already mentioned insertion loss. Excessive insertion loss can be caused by many things – most commonly bad termination of connectors, broken fibres, dirty connector faces and extreme bends in the cable. These need to be traced and rectified.

Excessive return loss indicates there is a reflection problem due to changes in refractive index along the fibre link – caused perhaps by connector mating problems.

Fibre mismatches can occur when the wrong fibre size has been used, often during maintenance. This is easily identified by an OTDR or by testing using a hand-held powermeter and light source in each direction in turn.

Damaged cable and connectors are the most common form of fault within any system where a fibre may have failed within a connector or inside the cable itself. If the cable is too short to use an OTDR then it can be determined whether there is a break or crack by using a visible light source or a hand-held power meter/light source.

Always ensure that the connectors being tested are cleaned using isopropyl alcohol and lint-free tissues. Dirt on the connector faces will significantly reduce the performance of any fibre system.