Optical detection method for evaluating the quality of optical fiber end faces

The problem at the optical fiber connection is the main cause of network failure. Therefore, the detection of the optical fiber end is crucial. This article discusses three main face detection methods.

The optical fiber end surface processing quality has a great impact on the overall performance of the optical fiber communication system. It is estimated that more than half of the loss in the network is caused by unsatisfactory optical fiber connection.

The optical fiber end face detection technology can identify two main processing problems: geometric problems and cleaning problems. Geometric Problems are usually caused by polishing or processing, and the impact of Optical Fiber operation will not change. This problem can be detected by Optical Interference Microscopy and specialized software that executes the End Face Detection Program. The hardware and software for implementing the interference detection process are now relatively complete and follow a series of widely accepted standards in the industry.

The term "clean" is widely used to describe permanent damage to the optical fiber end face, such as scratches, cracks or concave points) and temporary contamination of dirt, oil stains, water or cleaning agent residues ). Keeping the connector clean is a problem that needs to be paid attention to throughout the process of optical fiber production. Any link in the assembly process may cause damage and pollution to the connector.

Due to the lack of appropriate standards, subjective cognitive differences, low testing accuracy, and the absence of repeated testing methods, it is very difficult to determine acceptable end face pollution levels. However, additional damage may lead to data loss and network connectivity. Face Detection is critical for communications and data applications. For high-power applications, these damages may cause catastrophic consequences, and serious damages may cause the connector to be completely invalid.

This article will introduce the current production, R & D, and end-user methods for cleaning detection, and discuss how much of the three 2D light detection techniques can evaluate the end face processing quality, the advantages and disadvantages of each method are compared. These three methods include manual testing by the operator through the microscope, semi-manual testing by the operator using the "Auxiliary" software to operate the microscope, and a fully automated testing system.

Detected content, time, and location: cDn Optical Communication

End detection EFI) needs to be applied throughout the entire supply chain system. The optical cable production process usually has the following detection points: After the polishing process ends, the intermediate testing process, and the final testing. The QA department needs to apply EFI during stain detection, new processes or product R & D, authentication, or routine maintenance. The end user applies EFI during QA, routine maintenance, and reliability testing.

Face defects of optical fiber connectors include scratches, concave points, cracks, loose or fixed pollution. Typical defects can be divided into Scratch and particle contamination. Scratches are defined as damages much larger than the diameter of the End Face, usually> 30: 1), and all other damages are defined as particle contamination or damage.

Scratches are usually formed during the polishing process, but they may also occur during online business operations such as fiber insertion and removal. In a certain image with a given focal length, it is difficult to distinguish between scratches and cracks with the naked eye. Concave points are permanent and irregular material damage, usually caused by irregular operations, or in the production and connector plugging process.

The potential permanent damage of the joint face is the primary concern for the cleaning and detection process before the optical joint connection in the industry standard engineering process. Temporary pollution, such as dirt, dust, oil stains, or other material pollution, can be removed through a series of cleaning procedures, and permanent pollution is defined as pollution that cannot be removed, unless it is re-polished) including epoxy residue, dirt or embedded particle impurities.

Detection criteria cDn Optical Communication

Detection standards were initially developed by optical systems vendors and recently accepted as IPC-8497-1 international standards. These criteria have their own goals. The difference is that the indicators for passing tests are different. The Detection criteria define a series of areas centered on the core of the optical fiber. The importance of different areas varies. The number of areas and the specific diameter value depend on the type of the fiber single-mode or multimode) and the casing.

The most critical area is the fiber core. The standard specifies that the single-mode optical fiber cannot have visible scratches in this area, and the multi-mode optical fiber only allows a very small number of scratches. Such damage has been proven to cause greater insertion loss, reduced reflection, and reduced the quality of optical link transmission. The IPC standard relaxed the limitations on the other three areas, namely, the contact layer connected to Single-Mode Optical Fiber by the coating, epoxy layer, and casing for ceramics, the normative principles for these regional damages are more related to the loss caused to the core when the connector is connected, and the impact of processing quality becomes secondary. There is a widely accepted conclusion that large particles may lead to improper connection head matching, thus reducing the reflection loss index.

In specific circumstances, especially in high-power applications, the standards are more stringent, so as to prevent the failure of the joints caused by heat accumulation. In this case, it is critical to check the reliable detection mechanism of particle impurities in all regions. These applications need to use the appropriate industrial and user-defined standards.

Manual, semi-automatic, and automatic detection of cDn Optical Communication

Manual detection is currently the most commonly used detection method, because the traditional PC system does not have sufficient hardware and software capabilities and lacks effective algorithms to accurately detect and distinguish small defects, especially the shallow scratches. Many people think that manual detection is still the most cost-effective solution for most applications, and the detection process is simple and quick. Although it takes a certain amount of manpower, the detection results are subjective, detection personnel must be trained at a higher level to obtain reproducible detection results.

A video optical fiber microscope, a joint fixture, and a video monitor are required for manual testing. The microscope scales up the image of the optical fiber end face and displays it on the monitor. The typical test environment is shown in Figure 1.

Photo 1: manual optical fiber detection

Once the connector is manually installed, the tester performs the detection process in the following steps: 1. adjust the focal length; 2. identify damage; 3. determine the damage size and quantity in each region; 4. determine whether the detection is successful or not.

A Probe-type microscope will be used in the field or pipeline environment, rather than the above-mentioned desktop microscope.

Despite the help of polyester film, it is very difficult for the operator to repeatedly judge the size and position of each flaw. The detection effect depends on the focal length specified by the operator), the microscope Resolution, and the contrast of the video monitor used for display. These factors, coupled with the lack of detailed inspection records, usually lead to inefficient repetitive work between different Testers in all stages of the supply chain. Studies have shown that the proportion of repetitive labor accounts for about 60%.

In the past 10 years, some semi-automatic or "software-assisted" testing methods have gradually become mature and commercially available. The microscope used for semi-automated testing is the same as that used for manual testing. The difference is that it uses computer image processing software to analyze the optical fiber end face. Compared with analog camera-based devices, most digital cameras are configured with USB or FireWire interfaces, saving the cost of frame capture.

Similar to manual testing, the semi-automatic method also requires the operator to insert the joint into a fixed testing platform, locate the fiber to be tested using a multi-fiber connection head such as MPO) and adjust the focal length of the microscope. Once a satisfactory image is obtained, the operator starts the software to capture and analyze the image. The EFI software collects face images, performs inspection, classification, measurement, and judgment of the damage location, and compares them with the preset standard indicators of the software to quantitatively determine the region information, determine whether the connection is qualified.

The semi-automatic method has a correlation with the software capability, the microscope performance, and the operator's focus and positioning image skills. It has proved that its accuracy, repeatability and reproducibility are superior to manual testing. This solution can provide specific records of detection results, including face image and damage detection data. Photo 2 shows an optical fiber end face with scratches detected.

Photo 2: defective face images show detected scratches

The completely automatic detection system adopts the same process as semi-automatic detection. This type of system uses a computer to control the detection action. In some cases, it also uses a multi-resolution camera to quickly locate, focus, and collect images of multiple optical fiber ends on multiple connectors. The testing device is configured, 3 ).

Photo 3: Automatic Optical fiber detection system in the Cleanroom

The system eliminates the uncertainty of manual focus and positioning optical fiber, and the control of software, microscope, lighting device and motion control equipment can be controlled by the manufacturer, the performance of the entire system can be verified. The detection is highly reproducible, and the results of repeated detection are consistent. The total damage detection consistency exceeds 99%, and the edge Damage Detection consistency exceeds 95%. Some users of automatic detection systems also reflect that multiple testing systems have consistent results for the "6 Σ" sample size of millions of optical fibers.

Compared with manual and semi-automatic detection, automatic detection has an important advantage. Focus, control, and light calibration processes are fully automated by tracking the calibration process with primary optical fibers and NIST. The automatic method improves the accuracy of the test, reduces the test cost, enhances the test capability, reduces the training workload of the operator, increases the data size, and reduces the supply chain restrictions. At the same time, the fixed detection device in the automatic detection system can automatically check the type of optical fiber, optical cable or product in the detection area. The tester only needs to install and remove the product to be tested. You can install a device that requires testing for several hours in the system. It takes only a few minutes for the tester to install, disassemble the product to be tested, open the collected data, and locate the specific data according to the serial number. In the automatic detection, the part that the operator needs to participate in is greatly reduced, thus greatly reducing labor costs.

The automatic detection system performs 2D geometric measurement, MT casing positioning detection, optical consistency detection, and potential integrated cleaning capabilities. Computer-based testing makes it possible to introduce a simpler face quality evaluation method, for example, iNEMI of the international electronics manufacturers alliance) the optical device Cleaning Technology Working Group has demonstrated that the joint insertion loss is closely related to GWpOAGaussian Weighted percent Occluded Area. When using a computer-aided detection method, this conclusion may greatly simplify the parameters for determining the cleanliness of the optical joint.

The application of optical fiber is more and more extensive, and the rapid increase in usage makes automatic detection more attractive. As the quality requirements of Optical Fiber applications are getting higher and higher, automatic detection will become a more popular solution because of its improved repeatability and accuracy.