Fabrication Method for the High-Accuracy Optical Fiber Delay Line with Specified Length

Abstract

We propose a novel scheme for fabricating high-accuracy optical fiber delay lines (OFDLs). The fabrication system integrates a self-designed optical fiber cutting device and a high-accuracy fiber length measurement module based on optical frequency domain reflectometry. The optical fiber cutting device can cleave optical fibers to a specific length with the help of the motorized stage. The accuracy of fiber-cutting was determined by the positional accuracy of the motorized stage, which can reach several microns or even lower. This integrated design significantly reduces the errors and uncertainties introduced by fiber-cutting. To test, a set of OFDLs of a certain length was fabricated by this system. The deviation from the desired fiber length was kept below 50 μm, thus proving high fabrication accuracy and repeatability.

1. Introduction

Optical fiber delay line (OFDL) is an important component in many fields, such as fiber sensing, optical measurement, optical communications, and microwave photonics [1,2,3,4]. The OFDL is usually a fiber pigtail or fiber-optic patch cable that is added to optical links to provide a pre-deterministic optical delay. Strict requirements for precision, repeatability, and efficiency are placed on OFDL fabrication. Usually, patch cable OFDLs are fabricated by splicing two fiber pigtails of a specific length. Therefore, the fabrication of these fiber pigtails of a specific length is of utmost importance. The accuracy of the fiber length measurement method and the optical fiber cutting device (OFCD) both affect the final accuracy of the OFDL. There are two methods for measuring the optical fiber length: an electronic and an optical method. The electronic method is based on electronic–optical–electronic conversion [5,6]. This method is complex and cannot be used in the fabrication of the pigtail-type OFDL. The optical method has a simplified measurement system structure and can achieve measurement accuracy of a few microns [7,8]. With traditional OFCDs, the length of the fiber cut is manually controlled; achieving cutting accuracy can only reach up to several millimeters. In addition, it is difficult to meet the fiber-splicing requirements with the cutting angle of the fiber pigtail when fabricated manually. To improve the fiber cutting accuracy, optical low-coherence reflectometry technology is used to measure the optical fiber length in the fabrication scheme proposed by S. Li et al., which achieves a measurement accuracy of tens of microns [9,10]. However, poor accuracy caused by manual positioning and fiber-cutting results in errors in fiber lengths greater than 0.2 mm, thus limiting the fabrication accuracy of the OFDL.

In this paper, we propose a novel scheme to fabricate high-accuracy OFDLs and describe their process in detail. In our scheme, the fiber length cutting accuracy is completely determined by the positioning accuracy of the motorized stage, achieving accuracies of 1 μm or even lower. Additionally, fiber length is measured by an optical vector analyzer (OVA) based on optical frequency domain reflectometry (OFDR) technology. Therefore, the fabrication accuracy of OFDL is mainly limited by the fiber length measurement accuracy and the error from the OFCD can be minimized. OFDLs were fabricated based on this method and the effectiveness of the scheme was verified.

2. OFDL Fabrication Method

The schematic diagram of our high-accuracy OFDL fabrication system is shown in Figure 1. The integrated system mainly consists of an OVA, an automated OFCD, and an optical fiber fusion splicer (OFFS). The OVA is the instrument used for the ultra-high-accuracy length measurement of the optical fiber based on the OFDR technology [11,12]. It is also employed as a test light source, connected to a fiber-optic polarizer (FP) that is used to adjust the optical polarization of the light emitted by the OVA. At this time, the length value measured by the OVA is introduced by the measuring link and should be subtracted when calculating the fiber length. When the output port of the FP is connected to the optical fiber, the fiber length is measured by the OVA immediately. The FP is not required if the optical fiber measured by the OVA is a single-mode fiber. The length of the fiber to be cut is then calculated according to the measured length by the computer, which instructs the OFCD to cut the fiber pigtail automatically. Pigtail-type OFDLs of a specified length can thus be fabricated by cutting a fiber pigtail twice though the integrated device. However, if the required OFDL is a fiber-optic patch cable, two fiber pigtails of a specific length are to be spliced via the OFFS. The effect of fiber splicing on the fiber length is negligible.

The structural diagram of the OFCD is shown in Figure 2. It is mainly composed of a motorized stage, an optical fiber cleaver, a pair of fiber clamps, and a self-designed adapter. The motorized stage and the optical fiber cleaver are both installed on the metal platform. The displacement direction of the stage should be perpendicular to the cutting blade surface of the optical fiber cleaver. Additionally, an adapter is required to fix the fiber clamp on the motorized stage, and the fiber pigtail to be cut is clamped by the fiber clamp. Controlled by the computer, when the motorized stage moves, the adapter also moves along the chute of the optical fiber cleaver with it until it arrives at the correct position. Then, the end of the fiber pigtail is squeezed by the right rubber pressure pad of the optical fiber cleaver, and the fiber is cut by the optical fiber cleaver.

The fabrication process for pigtail-type OFDL of a required length, $L$, is shown in Figure 3. First, an optical fiber of roughly 15 mm longer than $L$ is measured with a ruler and manually cut to obtain a fiber pigtail of a rough length. As shown in Figure 2, the distance between the cutting blade surface and the right rubber pressure pad of the optical fiber cleave is $d_1$. The rough value is determined by $d_1$ and should be more than twice $d_1$. The fiber pigtail is then clamped into the fabrication system. Next, coating from a small section at the end of the fiber pigtail is removed. The position of the fiber pigtail in the fiber clamp is then adjusted until it can be pressed by the rubber pressure pad of the optical fiber cleaver. Here, the fiber pigtail is cut by the optical fiber cleaver for the first time. Next, the fiber pigtail is connected to the FP and its length, $L^{\prime }$, is measured by the OVA. The second length of fiber to be cut, $\Delta L$, is the difference between $L^{\prime }$ and $L$. After that, the motorized stage is driven to move automatically towards the optical fiber cleaver by $\Delta L$. The fiber pigtail is now cut for the second time. The fiber pigtail after the second cut, $L^{″}$, is measured by the OVA. The error between $L^{″}$ and $L$ of the OFDL is then calculated. The system continues until fiber length with acceptable error and accuracy for the OFDL is reached.

When the required OFDL is a fiber-optic patch cable, two fiber pigtails of a specific length are fabricated first, then spliced by the OFFS. The sum of the lengths of the two fiber pigtails should be equal to the required length of the patch cable-type OFDL. After the OFDL is taken out from the optical fiber clamps, a heat-shrinkable tube is generally used to protect the fusion splicing joint. It should be noted that the optical fiber clamps used for fabricating the two fiber pigtails must be the standard pair of clamps matched with the OFFS, not the fiber clamp matched with the optical fiber cleaver. This is to keep the position of the fiber pigtail in the fiber clamp and ensure the cutting angle of the fiber pigtail meets the requirements of fiber splicing. As shown in Figure 2, the distance between the cutting edge and the top of the chute of the optical fiber cleaver is $d_2$. Before the fiber clamp is removed from the OFCD, the distance between the fiber clamp and the top of the chute of the optical fiber cleaver is $\Delta d$ after the second cleaving. Thus, the distance between the fiber clamp and the cutting end face of each fiber pigtail is $d_2+\Delta d_1$ and $d_2+\Delta d_2$, respectively. Both fiber pigtails are removed together with the fiber clamps from the OFCD and are placed on the clamp platform of the OFFS. Before splicing the fiber pigtails, relevant parameters of the OFFS should be set according to the fiber type and $\Delta d$. The fusion type of the OFFS is selected according to the fiber type. The distance between the discharge electrode and the clamp platform of the OFFS on both sides must be equal to $d_2+\Delta d_1$ and $d_2+\Delta d_2$, respectively.

A set of patch cable-type OFDLs with two fiber-optic connectors was experimentally fabricated by the proposed method above—slightly more complicated than the fabrication of the pigtail-type OFDL described above—and its setup is shown in Figure 4. Considering the desired length of the OFDL to be $L$, the fabrication procedures comprise three steps. In the first step, a fiber pigtail is cut roughly to the half of $L$, and its length, $L_1$, is measured by the OVA. The difference $\Delta L$ between $L_1$ and $L$ is then calculated. Next, in the second step, the other fiber pigtail of accurate length is fabricated by the integrated device shown in Figure 1 and Figure 2. Its length, $L_2$, should be exactly equal to $\Delta L$. Finally, in the third step, both fiber pigtails are spliced by the OFFS (Fujikura corp., Tokyo, Japan, FSM-100P+) to obtain the final OFDL. The relevant parameters of the OFFS are set according to the fiber type and $\Delta d$. One of the ports in the OFDL is connected with the FP, and its length, $L^{\prime }$, is measured by the OVA. Ideally, the length value $L^{\prime }$ should be equal to the sum of the lengths $L_1$ and $L_2$. The negligible difference in length was caused by fiber splicing. During OFDL fabrication, all devices were placed on a vibration-isolated optical platform, and the fluctuation of the ambient temperature did not exceed 0.5 °C.

As for the fabrication system, the fiber length measurement accuracy of the OVA (LUNA Corp., Roanoke, VA, USA, OVA5000) was 20 μm, and the fiber length measurement range of the OVA was 150 m approximately in transmission mode. The cutting accuracy of the OFCD was 1.5 μm, which was determined by the bidirectional repeatability of the motorized stage (Thorlabs Corp., Newton, NJ, USA, MT1-Z8). Therefore, the uncertainties due to cutting were significantly reduced, and the fabrication accuracy of OFDL was improved, which mainly depends on the accuracy of the fiber length measurement.

The desired length $L$ of the patch cable-type OFDLs was 12 m. The actual lengths of the OFDLs were measured by OVA, and each OFDL was measured ten times. The length measurement results of 20 OFDLs are shown in Figure 5. The error between the measured and desired lengths of the OFDLs was below 50 μm, which shows high fabrication accuracy and reproducibility.

3. Conclusions

We propose and experimentally demonstrate a novel method to fabricate high-accuracy OFDLs by using the system-integrated optical fiber length measurement and cutting device. The proposed method can fabricate both types of OFDL—pigtail and patch cable—to the desired length. The fiber length cutting accuracy was determined by the positional accuracy of the motorized stage, which can reach 1 μm or even lower. The real-time fiber length measurement was achieved by using an OVA based on the OFDR technology. A set of OFDLs was fabricated and the length deviation from the desired fiber length was kept below 50 μm. Thus, the proposed fabrication system features a high degree of automization, high fabrication accuracy, high fabrication efficiency, and reliable fabrication quality.