1.Optical Power Meter

Fiber optic power meters measure the average optical power emanating from an optical fiber. They typically consist of a solid state detector (silicon for short wavelength systems, germanium or InGaAs for long wavelength systems), signal conditioning circuitry and a digital display of power. To interface to the large variety of fiber optic connectors in use, some form of removable connector adapter is usually provided. Power meters are calibrated to read in linear units (milliwatts, microwatts and nanowatts) and/or dB referenced to one milliwatt or one microwatt optical power. Some meters offer a relative dB scale also, useful for laboratory loss measurements. (Field measurements more often use adjustable sources set to a standard value to reduce confusion. See sources below.) Power meters cover a very broad dynamic range, over 1 million to 1, so some form of automatic range switching is provided in the signal conditioning circuitry to allow reasonable display resolution. Although most fiber optic power and loss measurements are made in the range of 0 dBm to -50 dBm, some power meters offer much wider dynamic ranges. For testing analog CATV systems or fiber amplifiers, on needs special meters with extended high power ranges up to +20 dBm (100 mW). Although no fiber optic systems operate at very low power, below about -50 dBm, some lab meters offer ranges to -70 dBm or more, which can be useful in measuring optical return loss or spectral loss characteristics with a monochromator source. Power meters measure the time average of the optical power, not the peak power, so the meters are sensitive to the duty cycle of an input digital pulse stream. One can calculate peak power if one knows the duty cycle of the input, by dividing the average power by the duty cycle. For most loss measurements, one uses a test source with CW (continuous wave) or 2 kHz modulated output. As long as the source modulation doesn't change, no compensation needs to be made. When testing link transmitter power or receiver sensitivity, it is necessary to establish a standard test pattern, generally a 50% duty cycle, called a square wave, to allow accurate measurement of transmitter output or receiver sensitivity. Fiber optical power meters have a typical measurement uncertainty of +/-5%, when calibrated to transfer standards provided by national standards laboratories like the US National Institute of Standards and Technology (NIST). Sources of errors are the variability of coupling efficiency of the detector and connector adapter, reflections off the shiny polished surfaces of connectors, unknown source wavelengths (since the detectors are wavelength sensitive), nonlinearities in the electronic signal conditioning circuitry of the fiber optical power meter and detector noise at very low signal levels. Since most of these factors affect all power meters, regardless of their sophistication, expensive laboratory meters are hardly more accurate that the most inexpensive handheld portable units.

2.Optical Light Source

In order to make measurements of optical loss or attenuation in fibers, cables and connectors, one must have a standard signal source as well as a fiber optics power meter. The source must be chosen for compatibility with the type of fiber in use (singlemode or multimode with the proper core diameter) and the wavelength desired for performing the test. Most sources are either LED's or lasers of the types commonly used as transmitters in actual fiber optic systems, making them representative of actual applications and enhancing the usefulness of the testing. Some tests, such as measuring spectral attenuation of fiber requires a variable wavelength source, which is usually a tungsten lamp with a monochromator to vary the output wavelength. Typical wavelengths of sources are 665 nm (plastic fiber), 820, 850 and 870 nm (short wavelength glass fiber ) and 1300 and 1550 nm (long wavelength ). LED's are typically used for testing multimode fiber and lasers are used for singlemode fiber, although there is some crossover, especially in older telecom systems which used multimode fiber with lasers and the testing of short singlemode jumper cables with LED's. The source wavelength can be a critical issue in making accurate loss measurements, since attenuation of the fiber is wavelength sensitive especially at short wavelengths. Thus all test sources should be calibrated for wavelength. Adaptability to a variety of fiber optic connectors is important also, since over 70 styles of connectors exist, although the types most commonly used are SMA, ST, FDDI and ESCON for multimode fiber and Biconic, FC, SC and D4 for singlemode fiber. Some LED sources use modular adapters like power meters to allow adaptation to various connector types. Lasers almost always have fixed connectors. If the connector on the source is fixed, hybrid test jumpers with connectors compatible with the source on one end and the connector being tested on the other must be used. Other source-related factors affecting measurement accuracy are the stability of the output power and the modal distribution launched into fiber. For extremely accurate measurements, the source may need optical feedback stabilization to maintain output power at a precise level for long times required for some measurements. And mode scramblers, filters and strippers may be required to adjust the modal distribution in the fiber to approximate actual operating condition

3.Visual Fault Locator

Many of the problems in connection of fiber optic networks are related to making proper connections. Since the light used in systems is invisible, one cannot see the system transmitter light. By injecting the light from a visible source, such as a LED or incandescent bulb, one can visually trace the fiber from transmitter to receiver to insure correct orientation and check continuity besides. The simple instruments that inject visible light are called visual fault locators. If a powerful enough visible light ,such as a HeNe or visible diode laser is injected into the fiber, high loss points can be made visible. Most applications center around short cables such as used in telco central offices to connect to the fiber optic trunk cables. However, since it covers the range where OTDRs are not useful, it is complementary to the OTDR in cable troubleshooting. This method will work on buffered fiber and even jacketed single fiber cable if the jacket is not opaque to the visible light. The yellow jacket of singlemode fiber and orange of multimode fiber will usually pass the visible light. Most other colors, especially black and gray, will not work with this technique, nor will most multifiber cables. However, many cable breaks, macrobending losses caused by kinks in the fiber , bad splices etc. can be detected visually. Since the loss in the fiber is quite high at visible wavelengths, on the order of 9-15 dB/km, this instrument has a short range, typically 3-5 km.

4.Fiber Identifier

If one carefully bends the fiber enough to cause loss, the light that couples out can also be detected by a large area detector. A fiber identifier uses this technique to detect a signal in the fiber at normal transmission wavelengths. These instruments usually function as receivers, able to discriminate between no signal, a high speed signal and a 2 kHz tone. By specifically looking for a 2 kHz "tone" from a test source coupled into the fiber, the instrument can identify a specific fiber in a large multifiber cable, especially useful to speed up the splicing or restoration process. Fiber identifiers can be used with both buffered fiber and jacketed single fiber cable. With buffered fiber, one must be very careful to not damage the fiber, as any excess stress here could result in stress cracks in the fiber which could cause a failure in the fiber anytime in the future.

5.Optical Talk Sets

While technically not an measuring instrument, fiber optics talksets are useful for fiber optics installation and testing. They transmit voice over fiber optic cables already installed, allowing technicians splicing or testing the fiber to communicate effectively. Talksets are especially useful when walkie-talkies and telephones are not available, such as in remote locations where splicing is being done, or in buildings where radio waves will not penetrate. The way to use talksets most effectively is to set up the talksets on one fiber (or pairs appropriate) and leave them there while all testing or splicing work is done. Thus, there will always be a communications link between the working crew, which facilitates deciding which fibers to work with next. The continuous communications capability will greatly speed the process. Recent developments in talksets include talksets for networking multi-party communications, especially helpful in restoration, and system talksets for use as intercoms in installed systems. They do not allow continuous communications during testing as discussed above, since they must be moved to the next pair of fibers of interest each time. There are no standards for the way talksets communicate. Some use simple AM (amplitude modulations) transmission, some FM (frequency modulations) and some proprietary digital schemes. Thus no two manufacturers' talksets can communicate with each other. Bellcore has addressed this matter in a technical advisory that proposes a FM method at 80 and 120 kHz, but it will take years before a standard has been set and manufacturers offer compatible instruments.

1.Ge Dectector.

Ge (Germanium) detector is to measure loss and power with moderate accuracy over 850 -1550 nm. Ge is good to measure high power (< +10 dBm), since it provides about 5 dB higher range than an InGaAs detector (< +5 dBm). Ge is unsuitable for accurate work on WDM systems above 1550 nm and is also very temperature sensitive. Ge is unsuitable for 1550 nm systems if very cold conditions are expected. Ge is unsuitable for precision or laboratory grade accuracy. It is inherently non-linear by about 0.04 dB and has some temperature sensitivity, For precision measurement, use Si or InGaAs.

2.InGaAs Detector

InGaAs (Indium Gallium Arsenide) detector is for precision measurements over 1000 - 1650 nm. InGaAs gives the most stable reading over the CWDM or DWDM bands. The InGaAs is suitable for work on CWDM and DWDM systems above 1550 nm because it it very stable.InGaAs is unsuitable for multimode, since it's very wavelength sensitive at 850 nm.

3.Si Dector

Si (Silicon) detector for precision measurement at 600 - 1000 nm up to 0 dBm Graphs for your reference: 1. Wavelength Dependence. 2. General Detector responsivity graphs The above graph shows the room temperature response of power meters with Ge & InGaAs detectors as the wavelength is changed beyond 1500 nm. The Ge meter is unsuitable for work on CWDM and DWDM systems above 1550 nm, the InGaAs meter is obviously a much better choice, since it is very stable. This graph uses real measurement data. The above graph shows how the 1580 nm thermal response of a power meter with a Ge detector changes with temperature. This instability makes Ge power meters basically unsuitable for field work on CWDM and DWDM systems above 1550 nm. The thermal stability below 1550 nm is much better, around 0.2 dB, however it's never as good as InGaAs. This graph uses real measurement data.