How to select fibre optic test equipment
are used to measure optical power or optical power loss over a fibre optic path. The measurement of optical power is the most fundamental measurement in fibre optic systems.
- Optical Power Meters (OPM)
The OPM is the workhorse of fibre optic measurements, much like the digital multimeter of electronics. An OPM can assess the performance of the optical terminal equipment by measuring the absolute power being injected into or emerging from the network.
When an OPM is used together with a stabilised light source, the combination can measure link loss to verify continuity and help assess the quality of the transmission path through the optical fibre.
launch light of known power and wavelength into the optical system. An SLS is used along with a power meter to measure the optical loss of a fibre optic system. If the system is already installed, you can often use the system transmitter (XMTR) as the SLS.
- Stabilised Light Sources (SLS)
Only when the system XMTR is inaccessible, is a separate SLS required. The wavelength of the SLS should match as closely as possible the wavelength of the system XMTR.
Frequent measuring of end-to-end loss of a system after installation is necessary to determine if the link loss, including connector, splice and fibre losses, meet the design specification.
are used to measure optical power loss over a fibre optic path. There are two types of OLTSs: (1) Component equipment of separately packaged OPM and SLS instruments and (2) integrated test sets that combine the OPM and SLS into one unit.
- Optical Loss Test Sets (OLTS)
With short LAN spans within walking and talking distance of the ends, the technicians at either end can use the economy of a single component OLTS with the SLS on one end and the OPM at the other end. For long distance networks, technicians should each be equipped with complete component or integrated OLTS configurations.
are used to simulate system losses to measure system margin, receiver operating range and receiver linearity. System margin is the difference between the actual received power versus the minimum received power level needed to operate reliably.
- Variable Optical Attenuators (VOA)
Performance reliability is usually characterised as bit-error rate (BER) in high end systems when it is necessary to characterise the system performance under various conditions. BER is expressed in the number of failures per bit.
A typical error rate of 10-9 means that one bit out of 100,000,000 may be wrong.
and fault locators are used to characterise fibre loss as a function of distance. With an OTDR, the technician can profile discrete components of the entire system, identifying and measuring fibre spans, splices and connectors. OTDRs are the most sophisticated as well as the most expensive instruments in fibre troubleshooting equipment.
- Optical Time Domain Reflectometers (OTDR)
Loss measurements can be made with access to only one end of the fibre unlike two ends required by OPMs and OLTSs. An OTDR trace shows the fibre attenuation for the system and the location and magnitude of any losses from connectors, splices, fibre anomalies, or fibre discontinuities.
OTDRs can be used for three applications: (1) to characterise (length and attenuation) of a cable before placement, (2) to obtain a signature trace of the fibre span and (3) to locate catastrophic failure points when trouble arises and the link goes down. A fault locator is a specialised version of an OTDR which attempts to automatically find fibre faults in a fibre link without the user having to learn the complexities of OTDR operation.
The selection process
Four steps always lead to the most-informed buying decision:
1) Identify system parameters;
2) Specify operating environment;
3) Compare performance factors;
4) Instrument maintenance.
Operating wavelength(s) (expressed in nm). The three major transmission windows are nominally 850, 1300 and 1550 nm.
Source type (LED or laser). Most low speed LANs (less than 100 MBs) use LED sources for their economy and usefulness for short haul applications. Most high speed systems greater than 100 MBs use laser sources to extend the signal over long distances.
Fibre type (SM/MM) and corelcladding diameter (in lint units). Standard singlemode (SM) fibre is 9/125 mm, although other special SM fibres exist and should be properly identified. Typical multimode (MM) fibre sizes include 50/125, 625/125, 1001140 and 200/230.
Popular connectors include Biconic, D4, FC-APC, FC-PC, FDDI, SC, SMA, ST, DIN, HP and Diamond.
- Connector(s) used (# of connectors and types)
Temperature is probably the most stringent criterion when selecting an instrument whereas packaging style, looks, etc may be more subjective to the buyer/user. Often, field measurements have to be performed under severe environmental conditions. Field portable equipment should operate over a range of - 18oC (uncontrolled humidity) to 50oC (95% humidity), as well as storage conditions down to -40oC (uncontrolled humidity) and up to +60oC (95% humidity). Lab instrumentation need be specified over a narrower, controlled range of 5oC to 50oC.
Unlike lab instrumentation requiring AC line operation, the power requirements of competitive field portable instruments can be quite diverse and have a direct impact on overall work efficiency.
The user should consider the following and the potential tradeoffs:
(1) The internal battery pack should be easily accessible and removable for replacement.
(2) The minimum service life should be 10 hours (equal to one extended work shift) either on a full charge or new batteries. However, the objective should be 40-50 hours to allow for optimum operating efficiency of both technician and instrument over an entire week.
(3) Use of popular battery cell styles, such as the common 9 V transistor and 1.5 V AA, C and D cells which are easy to obtain from local sources on a moment's notice, even from a local hardware store.
(4) Standard disposable batteries are preferred over rechargeable types (ie lead-acid, and NiCad types). The latter have become less desirable due to their non-standard packages, lack of availability, "memory" problems and emerging environmental public policy requiring users to implement special handling and disposal procedures.
A few years ago, finding portable test equipment which could meet all four objectives was virtually impossible. Today, state-of-the-art OPMs with the most modern CMOS circuitry can operate over 100 hours on just two common AA alkaline cells. Some lab models offer dual power capability (both AC line and internal battery operation).
This is the third step in the selection process. It involves detailed analysis of each of the optical test set types.
Optical power meters
These are used to measure optical power of lasers and LEDs. They are also used for the verification of the power budget of optical fibre links. Most important of all, the power meter is the key instrument in the characterisation and performance test of components (fibre, connectors, splices, attenuators, etc).
In selecting the appropriate OPM for a given application, the user should select optimum detector type and interface; evaluate calibration accuracy and manufacturer's calibration procedure for suitability to your range of fibre and connector requirements; identify those models with suitable measurement range and display resolution; need for dB function for direct insertion loss measurements.
The optical detector is the most critical element. It is a solid state photodiode which receives light coupled in from the network and converts it to an electrical signal. Input to the detector can be either a dedicated interface which can accept only one connector type, or a universal interface which, via screw-on adaptors, can accept a wide range of industry standard connectors.
The OPM circuitry translates the detector output signal, applies a calibration factor based on the selected calibration wavelength, and displays the optical power on a digital readout in dBm units (absolute dB referenced to 1 mW, ie 0 dBm = 1 mW). Figure 1 is a block diagram of the OPM.
The most important OPM selection criterion is matching the appropriate optical detector type with the expected range of operating wavelengths. It is important to note that InGaAs is best overall choice for making measurement in all three transmission windows. The overall preference for InGaAs over Ge include InGaAs's superior spectral flatness (uniformity) across all three windows, superior measurement accuracy within the 1550 nm transmission window as well as its temperature stability and low noise characteristics.
The next factor concerns calibration accuracy.
Is the meter calibrated in a manner consistent with your application?
What is the measurement uncertainty of using different connector adaptors?
It is important to consider other potential sources of error, such as the spectral uncertainty of the source, optical detector type and errors due to similar connectors from different manufacturers.
The third step is to identify models which meet your measurement range requirements. Expressed in dBm units, measurement range is a comprehensive parameter which identifies the min/max range of input signal for which the OPM can guarantee sufficient overall accuracy and linearity.
A basic feature found in most (but not all) OPMs is the relative dB function for direct optical loss measurements. Lower cost OPMs do not usually offer this function. Without the dB function, the technician must write down individual reference and test values and then calculate the difference.
Stabilised light sources
During the loss measurement process, a stabilised optical source (SLS) launches light of known power and wavelength into the optical system. A power meter/optical detector calibrated for use as the wavelength of SLS, receives light from the network and converts it into an electrical signal.
To ensure accuracy in the loss measurement, the SLS should simulate, as far as possible, the operating characteristics of the transmission equipment that will be used: operate at the same nominal wavelength and preferably of same type (LED, laser); stable with respect to time and temperature in output power and spectral characteristics over the duration of the measurement; offer the same connector interface and fibre launch characteristics; magnitude of its output power sufficient to measure worst case system loss.
When a separate SLS is needed independent of the transmission system, the optimum selection of source should simulate the system's optical transmitter characteristics and measurements requirements at an affordable cost.
Choices in light sources include:
1) Laser diode (LD). Light emitted from an LD has a narrow band of wavelengths. It is nearly monochromatic - that is, of a single wavelength (see Figure 2).
In contrast to the LED, laser light is not continuous across the band of its spectral width (less than 5 nm). Several distinct wavelengths are emitted on either side of the central wavelength. Although a laser provides more power than an LED, it is more expensive. Laser diodes are most often used to characterise long-haul SM data links with system losses exceeding 10 dB. In general, try avoiding measuring multimode fibre with a laser source.
2) Light-emitting diode (LED). An LED has a much wider spectral width than an LD, usually in the range of 50-200 nm. In addition, LED output is non-coherent and more stable in power.
LED sources are much less expensive than LD devices, but may lack the output power to measure worst case losses. LEDs are typically used in short-haul network and LAN applications with multimode fibre. LED sources can be used to make accurate loss measurements of LD-based SM fibre systems as long as sufficient output power can be launched into the fibre network.
3) White light. A tungsten lamp can be used as an alternative, inexpensive source. In applications where expected losses are moderate, the tungsten source is equivalent to an 850 nm source when used with a Si-based OPM.
In combination with an InGaAs OPM, the superposition of the detector's spectral response curve over the lamp's spectral output curve produces a centre wavelength at 1300 nm. Consequently, these devices can be used as inexpensive substitutes for LED or LD sources at 850 and 1300 nm. The white light source can also double as a visual fibre tracer for quick checks of continuity and fibre identification, without the hazards of laser-based systems.
Optical loss test sets
The combination of an OPM and SLS is called an optical loss test set, which are used to measure optical power loss over a fibre optic path. These instruments can either be two individual component devices, or a single, integrated unit. Overall, both types of OLTS can perform the same measurements to the same accuracy. The differences usually extend to cost and features. The integrated OLTS is usually much more expensive, sophisticated and laden with additional features.
Figure 2 is a typical system configuration for optical power and loss measurements.
To technically evaluate various OLTS configurations, the basic OPM and SLS critiera still apply. It is important to match the transmitter and detector for operation at the same wavelength and to have sufficient power and sensitivity to provide the required dynamic range to measure the optical system in question. The unifying parameter is the measurement range capability of the resultant OLTS system. Measurement range is expressed in dB, and reflects the difference between the maximum output power of the SLS and the minimum measurement power of the OPM.
The resultant OLTS measurement range specification must equal or exceed the expected link loss.
Integrated OLTS systems are generally more expensive but usually offer higher levels of sophistication and functionality than their component counterparts. For instance, integrated SM dual wavelength units are available which can simultaneously measure link loss at both 1300 and 1550 nm. Component systems would not be able to perform this measurement as easily, but this additional sophistication comes at a higher price.
Variable optical attenuators (VOA) are essential for testing optical receivers. When a system is installed, the System Engineer uses an attenuator to determine if the system will work over the specified range. With a BER generator connected to the electrical input of the optical transmitter, the VOA is used to reduce power (measured and verified with an OPM) and verify acceptable BER performance down to the minimum signal requirements of the optical receiver.
Attenuator evaluation requires focussing on these performance parameters: operating wavelength, fibre type/size and connector interface; residual loss and attenuation range; accuracy and resolution; reflectance.
Various VOA configurations are available - some are calibrated and feature calibrated readouts or dials, others are uncalibrated (where an OPM is employed to set the level). With regard to the optical interface, the user can select between the inflexibility of a fixed connector or the versatility of the universal connector interface (UCI).
The UCI system offers three advantages: accessibility to the ferrule end for routine cleaning and maintenance; the ability to accommodate all industry standard connectors via simple screw-on/screw-off adaptors; eliminates the need for hybrid patch cords. The second factor focuses on the loss characteristics of the VOA. Residual loss (or insertion loss) is the loss of the attenuator at the 0dB setting. Residual loss is more than expected connector mating loss which is dependent on the connector type. High residual losses (coupled with poor mating losses of particular connector types) limit the user's ability to measure performance at higher power levels at the 0dB starting point.
The attenuation range is defined as the full range of losses over which the attenuator can be set.
Third, attenuation accuracy is defined as the correctness of the optical loss setting (ie the difference between the display setting and the true value of the optical attenuation) that should be taken together with the VOA's resolution (ability to incrementally set the attenuator to a specified setting).
Actually, attenuators can be of three types: continuous, step or combinational step/continuous configuration where the stepping function momentarily blocks the optical signal. When BER measurements are being performed, both pure step and combinational models are undesirable due to the momentary blocking of the optical path between attenuation settings.
Another important performance factor is optical reflectance which is a critical performance factor in analogue/video transmission systems as well as digital systems as bit rates reach beyond 1GBs.
Optical reflectance (OR, expressed in dB) is a measurement of the amount of forward power reflected back when the optical signal encounters reflective events (splices, connectors, etc) within the fibre link. Reflectance is always expressed in negative dB units. Some people prefer the alternative terminology of optical return loss (RL) which is the negative of reflectance and thus, the numbers are always expressed as positive values.
It is desirable to have either smaller optical reflectance values or higher return loss. Different transmission technologies (analogue vs digital) have different RL requirements.
OTDRs are the most sophisticated form of instrumentation and provide the most information regarding the link under test. An OTDR is essentially a one-dimensional, closed circuit optical radar, requiring the use of only one end of a fibre to make measurements.
A high intensity, short duration light pulse is launched into the fibre while the high speed signal detector records the returned signal (backscattered light). This light results from two principal sources: Fresnel reflections at the fibre ends, breaks and connectors; and Rayleigh scattering from fibre imperfections.
The instrument provides a visual interpretation of the optical link. Splices, connectors and faults may be identified in both magnitude (dB loss) in distance away from the operator, as in Figure 3. The OTDR evaluation process shares many similarities to the requirements of the OLTS. In fact, an OTDR can be considered a very specialised test set comprising a stable, high speed pulsed source and a high speed optical detector.
The selection process should focus on: identifying operating wavelength, fibre type and connector interface; the expected link loss and required backscatter range; spatial resolution. For optimum performance, the operating wavelength, fibre type/size and connector interface must match the same characteristics of the optical link.
With MM systems, an OTDR of mismatched fibre size could be used for most MM link applications (such as 50/125 used for both 50/125 and 62.6/125) but the user should be cautioned that his measurement range will be reduced. The user can select the Diamond universal connector interface (UCI) which can be used on both SM and MM systems without any degradation in optical mating performance.
The second most important factor is matching the backscatter range of the prospective OTDR (module) to the expected link loss. The range is expressed in dB and represents the useful range over which loss measurements can be made of the fibre backscatter signal. This range must exceed the expected link loss to characterise the entire fibre span, especially for long-haul SM systems where system spans can easily exceed 10km.
As a rule of thumb, the maximum distance (expressed in meters and/or feet) should be greater than twice the fibre length.
The third critical factor is the event dead zone (EDZ expressed in meters). Two subfactors evaluate the spatial resolution performance of the OTDR with regard to interpreting features in the link. EDZ to the first connector specifies the minimum distance required to distinguish between two reflected connectors (based on the Fresnel reflections). Furthermore, EDZ to the first fusion splice characterises the OTDRs ability to distinguish between one reflective and one non-reflective fusion splice.
The industry divides OTDRs into two groups based on their spatial resolution performance high resolution and long-haul. Based on an instrument's ability to distinguish two connector events, a high resolution OTDR can resolve two connectors separated by 10m whereas the event dead zones for long-haul OTDRs well exceeds 10m.
In general, if the expected distance of the fibre segments, (ie distance between connectors) is expected to be short (less than 100 m), you must select a short haul, high resolution type OTDR. No one OTDR can do all things; there is always a tradeoff between backscatter range, distance and resolution.
It is obvious that optical systems will not perform if their interfaces are not clean. As a rule of thumb, the technician should clean the optical interface (particularly, connectors) before any measurement. Even a 1-2 mm dust particle can cause havoc in SM system performance. The user should evaluate instruments based on the accessibility of both ends of a mating interface.
The typical fixed connector set up leaves little to be desired since the internal connector is inaccessible between a fixed bulkhead. Fortunately, new technology is now available in the form of the universal connector interface (UCI) which features a male-type connector mounted on the front panel for direct access to the ferrule end for routine cleaning and maintenance.
The user should also consider the calibration requirements of different models. How often does the instrument have to be recalibrated? What is the cost for a routine calibration and turnaround time? Can the maintenance/recalibration be performed by the user or must the equipment be returned to the manufacturer? Can non-manufacturer calibration labs do the same work? Consequently, the user should closely evaluate the manufacturer and the performance of their service operation.
All these factors should be taken as seriously as the direct performance characteristics of the particular instrument.
Please me and tell me if you liked my technical information, if I've made any mistakes or even if you might have ideas on topics for me to include here.
To go back to my 'techo' page, please press or
to go back to my Home Page, or even to goto my employer's home page.
This page has been accessed
Last revised: Friday, 27 November 1998