2025 Volume 66 Issue 4 Pages 248-254
Electrical connectors connecting contact wires and messenger wires are sometimes subject to fatigue-failure due to vibration caused by train passage. It is therefore desirable to establish a method for evaluating the fatigue resistance of the connectors. Therefore, the authors proposed a test method consisting of two types of vibration test that take into account the two fatigue factors of the connectors: the relative displacement of the contact wire and their resonance. The test conditions were determined by analyzing overhead contact line vibration using an OCL-pantograph simulation. Furthermore, the authors carried out vibration tests on real connectors and confirmed that the test results were consistent with the actual failure status of the connectors.
An overhead contact line (OCL) connector (Fig. 1; henceforth, “connector”) is a metal fitting that electrically connects OCL wires and eliminates the potential difference in order to prevent unnecessary current from flowing through the wires and metal fittings. The conducting wires that constitute the connector are made of easily bent wires such as soft copper stranded wires. These wires are subject to fatigue-failure due to OCL vibrations from passing trains. This has been a long-standing issue when using connectors.
Fatigue durability evaluations of OCL fittings, including connectors, have been conducted in accordance with the vibration test specified in the Japanese Industrial Standards (JIS) [1, 2]. However, the JIS vibration test is mainly defined as a confirmation test for the loosening of bolts, and fatigue-failure has occurred in the field even in connectors that have passed the test. The vibration conditions have also not been changed since their establishment in the former Japanese National Railways Standards in 1968. Consequently, there has been demand for a new test method that appropriately evaluates the fatigue durability of connectors.
To this end, the authors proposed two types of test method that focus on the fatigue modes of connectors, namely, a vertical vibration test and horizontal vibration test. The vibration conditions for these tests were set based on the results of OCL vibration analysis. In this paper, the authors use M-T connectors that connect the contact wire to the messenger wire in a simple catenary as an example in order to explain the specific test method. In the following text, “dropper” is referred to as “hanger,” which is abbreviated as “H.”
The fatigue modes of connectors in the field need to be considered in order to determine which vibration test method is suitable. Figure 2 shows an example of the vertical vibration waveform of an OCL and two types of connector fatigue modes reported in past research [3]. The maximum strain in the M-T connector shown in Fig. 2 occurs at the ear of the conducting wire in each mode.
The first fatigue mode is fatigue caused by deformation of the conducting wire due to the difference in vertical vibration displacement between the wires connected by the connector (“OCL relative displacement”). This mainly occurs at the moment when the pantograph passes. Figure 3 shows the results of the measurement of the vertical vibration displacement of an OCL on a certain line and the structural analysis results of the conducting wire strain when a connector is attached to that OCL [3]). The agreement between the OCL relative displacement and the strain waveform indicates that the relative displacement has a large effect on the fatigue of the conducting wire.
The second fatigue mode is fatigue caused by the deformation of the conducting wire due to the resonance of the connector. Connector resonance occurs when the OCL vibrates horizontally due to residual vibration of the OCL after a train passes, or when a viaduct or catenary pole vibrates, and the frequency of this vibration matches the natural frequency of the connector. The conducting wires are subjected to a large number of bending strains when resonance occurs. This may lead to fatigue-failure in a short period of time.
In the present study, the authors proposed two types of vibration test corresponding to each fatigue mode as follows:
(1) Vertical vibration test (corresponding to fatigue caused by OCL relative displacement)
Fatigue durability is evaluated for the case where the messenger wire side of the connector is fixed, and the contact wire side is repeatedly pushed up in the vertical direction.
(2) Horizontal vibration test (corresponding to fatigue caused by connector resonance)
Fatigue durability is evaluated with consideration given to the primary vibration mode of the M-T connector when the connector is subjected to horizontal (sleeper direction) vibration.
Note that “fatigue durability” here refers to the whether the conducting wire fails after each test. The evaluation concept and specific test method are described in the following sections.
In this section, the authors show the results of OCL vibration frequency and amplitude analysis. This vibration is the basis for the vibration conditions for the vertical and horizontal vibration tests.
3.1 OCL relative displacementThe frequency of OCL relative displacement is expressed by the equation consisting of train speed v and hanger interval l, that is, v/l, as shown in Eq. (1) in Fig. 4 [4]. The graph in Fig. 4 compares the above equation with the analysis results of the vibration frequency of a simple catenary for Shinkansen. The equation and results were roughly in agreement at the center of the span ((ⅰ) in the graph in Fig. 4). However, a hanger tends to float relative to the messenger wire near the support points of a simple OCL. This results in the tendency of the hanger interval effectively increasing and the frequency of the OCL relative displacement becoming lower ((ⅱ) in the graph in Fig. 4).
Past research [3] has indicated that the amplitude of the OCL relative displacement also varied depending on the position within the span. Figure 5 shows the relationship between the position within the span and the OCL relative displacement amplitude and frequency for a simple Shinkansen catenary. In contrast to the frequency, the amplitude of the OCL relative displacement had an increasing tendency near the support point. This was because the messenger wire was fixed at the support point, which resulted in a smaller messenger wire vibration displacement, and a larger difference in vibration displacement between the contact wire and messenger wire. Therefore, the OCL relative displacement frequency and amplitude vary depending on the position within the span. This in turn suggests that the vibration conditions for the vertical vibration test must be set for each position where the connector is attached.
Figure. 6 shows the results of calculating the OCL relative displacement frequency and amplitude for a simple catenary for Shinkansen line currently used in Japan by OCL/pantograph simulation. Section 5 describes the procedure for setting the test conditions for the vertical vibration test using the same figure. For safety reasons, the vibration amplitude in the vertical vibration test should be set considering the maximum OCL relative displacement amplitude that can occur in the OCL. Previous research [3] has reported that “increase in train speed” and “decrease in OCL tension” were conditions that increase the OCL relative displacement amplitude. The simulation conditions in Fig. 6 consider the maximum train speed in operation, and the tension reduction (−10%) that can occur due to temperature changes and other factors. As previously mentioned, the figure shows a tendency of a low frequency and large amplitude near the support point, and a high frequency and small amplitude in the center of the span, as well as the distribution of the graph as a whole to be downward sloping to the right. Table 1 shows the frequency range of the OCL relative displacement for each connector installation position extracted from Fig. 6.
| Connector installation position | Frequency range (Hz) |
| Support point - 1H | 2 - 7 |
| 1 - 2H | 6 - 16 |
| 2 - 9H | 7 - 20 |
| 9 - 10H | 4 - 11 |
| 10H - Support point | 1 - 7 |
In this section, the authors analyze the OCL vibration in a section of a simple catenary and viaduct for Shinkansen, where fatigue-failure of connector conducting wires due to resonance has been reported [4].
Previous research [5] has shown that the vibration frequency input to the viaduct is dominated by “running speed (m/s) / car length (m)”, and that the vibration of this viaduct is input to utility poles and metal fittings. The vibration frequency range of the OCL to be considered in the horizontal vibration test (“OCL frequency range”) can be calculated to be 1.1 - 4.0 Hz when the running speed is 100 - 360 km/h and the car length is 25 m.
The OCL horizontal vibration amplitude was calculated using a structural analysis program. Figure 7 shows the constructed OCL model. This analysis involves the calculation of the OCL horizontal vibration displacement by inputting the pole vibration displacement at the support point of the model. The validity of the analysis results was confirmed by conducting a pole vibration test on the OCL equipment at RTRI (Fig. 8). The analytical results tended to produce slightly larger displacements, but the analytical values and the measured values were generally consistent with each other, so the analysis was judged as valid. Figure 9 shows an example of an analysis of the OCL horizontal vibration for a simple catenary for Shinkansen on an elevated section. The pole vibration displacement to be input to the OCL model was calculated using a structural analysis method [5] using a coupled model of the viaduct and the pole. The dashed lines in each graph in Fig. 9 indicate the maximum amplitude at each connector installation position. Figure 9 shows that the OCL horizontal vibration has a peak frequency near the pole's natural frequency (2.5 Hz), and many other peaks due to the OCL's natural vibration mode according to the span length were also confirmed.
Taking all these peaks into consideration, the horizontal vibration amplitudes were set within the OCL frequency range according to each speed uniformly at the maximum amplitude (Table 2).
| Maximum train speed (km/h) | Frequency range (Hz) | Vibration amplitude (mm) | ||||
| Support point - 1H | 1H - 2H | 2H - 3H | 3H - 4H | 4H - 5H | ||
| 260 | 1.1 - 2.9 | 25 | 35 | 60 | 80 | 90 |
| 320 | 1.1 - 3.6 | 45 | ||||
| 360 | 1.1 - 4.0 | |||||
The conventional JIS vibration test does not specify the height of the connector to be installed. However, differences in frequency characteristics that depend on the height of the connector need to be considered. Figure 10 shows the results of structural analysis of the relationship between the conducting wire strain and frequency when vibration is applied by pushing up the contact wire to connectors of different heights. The figure shows that the strain peak frequency and magnitude change depending on the connector height. This is because the frequency at which the connector is likely to deform changes depending on the length and inclination of the conducting wire. Therefore, conducting a safe evaluation in the vibration test requires setting a vibration frequency that causes the largest strain for the expected connector height.
Figure 11 shows the proposed implementation flow for the vertical vibration test and horizontal vibration test. A sample description is provided here, namely the procedure for setting the test conditions for both tests, where the assumption is that the connector for the Shinkansen is installed between 1H and 2H. The connector height selection range is 825-1475 mm when the connector is installed between 1H and 2H (including the overlapping area).
The test conditions are set using the relationship between the frequency and amplitude of the OCL relative displacement shown in Fig. 6. The procedure is shown below. To obtain a vibration amplitude that is safe and in line with the actual usage situation, it is considered that the value should be set on the line that passes through the maximum amplitude of the same frequency (red solid line in Fig. 6). Henceforth, the equation for this line is referred to as the maximum relative displacement equation.
The frequency range of OCL relative displacement at the target connector attachment position (1H-2H) is 6-16 Hz according to Table 1. As mentioned in Section 4, the frequency characteristics of the conducting wire strain vary depending on the connector height. Therefore, conducting the test on the safe side requires selecting the connector height at which the strain is greatest in the above frequency range, and the frequency at that time. Figure. 12 shows the analysis results of the frequency characteristics of the maximum strain for the selected connector height range. In this case, the connector with a height of 1225 mm (frequency 7.1 Hz) has the maximum strain at 6-16 Hz. Therefore, the test connector height was selected to be 1225 mm, and the vibration frequency was selected to be 7.1 Hz. Then, using the maximum relative displacement equation, the vibration amplitude was determined to be 30.8 mm (= −2×7.1 +45).
The number of vibrations in this test was set to at least 2×106 times, which is the same as in the conventional JIS test. A value of at least 2×106 times was considered to exceed the number of pantographs that pass during the replacement cycle of a typical connector (approximately 8-15 years).
5.2 Horizontal vibration testThe horizontal vibration test requires a vibration of the messenger wire side and the contact wire side of the connector in the horizontal direction at the same time as in the field, so a new vibration test machine was created (Fig. 13).
The vibration amplitude at the target connector attachment position (1H-2H) is 35 mm (assuming 260 km/h) according to Table 2. As in the vertical vibration test, ensuring a safe test requires selecting the connector height and vibration frequency at which the strain is maximum. Figure. 14 shows the analysis results of the frequency characteristics of the maximum strain when vibration that simulates a horizontal vibration test was input to the connector model. The connector with a height of 1425 mm (frequency of 1.8 Hz) showed the maximum strain in the OCL frequency range of 1.1-2.9 Hz (see Table 2). Therefore, a test connector height of 1425 mm and vibration frequency of 1.8 Hz were selected.
OCL horizontal vibration continues for a certain amount of time with the passage of one train, and the number of vibrations varies depending on the OCL conditions and train speed. Therefore, using this as a basis to determine the number of vibrations is difficult. As a result, the number of vibrations in the horizontal vibration test was set to at least 107, which is considered to be the upper limit sufficient for practical use in general vibration tests.
In this section, the authors conduct the above-mentioned test on a real connector and confirm whether the test results matched the connector's on-site disconnection status. As in the previous section, the authors assumed that the connector for Shinkansen would be installed between 1H and 2H, and performed the vibration test according to the test implementation flow in Fig. 11. Table 3 shows the test conditions defined in the previous section. Note that in both tests, one connector was used.
| Fatigue factor | OCL relative displacement | Resonance |
| Damage to actual equipment | Failure cases absent | Failure cases present |
| Corresponding test | Vertical vibration test | Horizontal vibration test |
| Vibration frequency | 6.3 Hz | 1.8 Hz |
| Vibration amplitude | 32.4 mm | 35 mm |
| Connector height at maximum strain |  |  |
First, the results of the vertical vibration test are described. Figure 15(a) shows the connector installed on the tester, and Fig. 15(b) shows the test results. Results after 2 × 106 vibrations showed that no wire failure occurred. These results suggest that OCL relative displacement-induced fatigue-failure will not occur for cases where the target connector is attached between 1H and 2H. No previous research has reported cases of the target connector breaking at the OCL relative displacement, and the test results are consistent with this.
Next, the results of the horizontal vibration test are described. Figure 16(a) shows the connector attached to the tester, and Fig. 16(b) shows the test results. The horizontal vibration test results showed that the conducting wires of the target connector failed completely prior to reaching 107 vibrations (687,000 times). These results suggest that attaching the target connector between 1H and 2H may result in fatigue-failure when the train speed and connector height conditions are met. This is consistent with the reported case in which the target connector failed due to resonance.
The above results confirmed that the results of this test were consistent with the damage to connectors in the field. Therefore, the results suggest that conducting this test allows for the evaluation of the practical fatigue durability of connectors.
The authors sought to appropriately evaluate the fatigue durability of connectors by focusing on two connector fatigue factors of connectors, namely OCL relative displacement and resonance. The authors then proposed a vibration test method consisting of a vertical vibration test and a horizontal vibration test corresponding to each of those fatigue factors. The main results were as follows.
● An OCL/pantograph simulation was used for the vertical vibration test in order to understand the relationship between the amplitude and frequency of OCL relative displacement according to the connector installation position. A relational equation was also proposed for setting the vibration conditions.
● For the horizontal vibration test, the horizontal OCL vibration waveform was determined by OCL structural analysis. The vibration amplitude and OCL frequency range were then proposed according to the position within the span.
● The connector's strain frequency characteristics and OCL frequency range were used to propose a method for selecting the test connector height at which the strain for OCL vibration is maximized.
● An implementation flow was created for the above test method. The same test was also conducted on an actual connector, which confirmed that the test results were consistent with the disconnection conditions of the connector reported at academic conferences and other venues.
|
Takuya OHARA Senior Researcher, Current Collection Maintenance Laboratory, Power Supply Technology Division Research Areas: Fatigue of Overhead Contact Line, Vibration Engineering |
|
Chikara YAMASHITA, Ph.D.Eng. Senior Chief Researcher, Head of Current Collection Maintenance Laboratory, Power Supply Technology Division (Former) Research Areas: Tribology under Current Flowing Condition, Fatigue of Overhead Contact Line |
