Iec Tr 62627-03-02-2011.Pdf

IEC/TR 62627 03 02 Edition 1 0 2011 12 TECHNICAL REPORT Fibre optic interconnecting devices and passive components – Part 03 02 Reliability – Report of high power transmission test of specified passiv[.]

Test conditions

The test details and performance measurements are presented in Tables 3 and 4 A step stress test was conducted, gradually increasing the incident power level, with each level maintained for five minutes to allow the temperature of the passive optical components to stabilize The testing temperature was set at 70 °C in accordance with IEC 61300-2-14:2005 During the test, changes in insertion loss (IL) and return loss (RL), as well as the outer surface temperature of the components, were monitored, based on the assumption that the high optical power absorbed by the components generates heat.

Input wavelength 1 480 nm (Raman laser)

Input power Maximum 4,4 W (forward direction test) and 5 W (backward direction test)

Test method Step stress test in which incident power level rises step by step

Duration Five minutes per each power level

Online monitoring IL (1 480 nm), RL (1 480 nm) and outer surface temperature of passive optical components Before and after the test IL, RL,

Polarisation dependent loss (PDL) for optical splitters and optical isolators and Isolation for optical isolators

The input light used for measurements had a wavelength of 1,480 nm, distinct from the signal wavelength, for several reasons Firstly, high-power light sources at this wavelength are widely available Secondly, the absorption coefficients of metal-doped fibers, utilized in optical attenuators, remain consistent between 1,480 nm and 1,550 nm Additionally, various wavelengths, including signal light and amplified spontaneous emission light, pass through the optical isolator, with the excitation wavelength of 1,480 nm being the most powerful due to the optical amplifier Notably, the absorption coefficient of the Faraday rotator at 1,480 nm is approximately 1% higher than at 1,550 nm, and the rotation angle of the Faraday rotator is temperature-dependent, varying from 0.07 °C to 0.1 °C.

1 480 nm in the forward direction is slightly larger than that of wavelength of 1 550 nm

When assessing high power light, utilizing a wavelength of 1480 nm is more suitable Additionally, the absorption coefficient of the adhesive at the junctions between the optical fiber and the waveguide in the optical splitter remains constant, showing no dependency on wavelength.

Moreover, in the light going through the optical splitter, the light energy is stronger in the remaining excitation light wavelength of 1 480 nm.

Apparatus and measurement conditions

The measurement setup adhered to the specifications outlined in IEC 61300-2-14:2005, as illustrated in Figures 1 and 2 Enhancements included the integration of a RL monitoring coupler and an optical power meter Additionally, two 2 × 2 20 dB optical couplers were employed for effective high power monitoring at both the input and output terminals of the device under test.

GPIB connection SMF: single mode fiber (ITU-TG.652.D)

Table 5 outlines the testing conditions, where IL, PDL, and isolation were assessed using the specific signal wavelength of each device The RL was measured with an instrument designed for RL measurement, utilizing a light source wavelength of 1,310 nm post-test Notably, the designs of the DUTs in this test ensure that there is no dependency of RL on the wavelength.

Table 5 – Measurement conditions in the test

Measurement Samples Measurement performances Measurement Wavelengths

Optical Attenuators IL (dB) 1 550 nm

Optical Isolators IL (dB) 1 550 nm

Optical Splitters IL (dB) 1 310 nm

Test results

Table 6 presents the test results, indicating that the outer surface temperature of the optical isolators and splitters stabilized after four minutes Additionally, there was no observed correlation between the attenuation of the fixed optical attenuator and the incident light power during the RL reduction.

However, in all the fixed optical attenuators, the change in return power was more than 10 dB

The return loss (RL) of an optical isolator decreased in the reverse direction at an input power of 5 W Detailed results of the backward direction tests for optical isolators are discussed in section 3.5 No damage to the optical splitters was noted until the incident power reached 4.4 W in the forward direction and 5 W in the backward direction.

Table 6 – Results of high power damage threshold test

Components Directions Numbers of samples Surface

Temperatures Duration times for surface temperature stability

Attenuators Plug to socket 6 89 °C > 5 min RL reduction at

Optical Isolators Forward 5 86 °C 4 min No failure by 4,4

Backward 4 174 °C 4 min RL reduction at 5

Optical Splitters Forward 3 87 °C 4 min No failure by 4,4

Backward 3 158 °C 4 min No failure by 5 W

The test results for the 10 dB attenuator are illustrated in Figures 2(a), (b), and (c) Notably, in Figure 3(a), a significant change in return power was observed when the incident power reached 2.3 W, corresponding to a surface temperature of 89 °C as shown in Figure 2(b) Additionally, the insertion loss (IL) varied within ± 0.5 dB.

C hange of re tur n los s ( dB )

Figure 2(a) – PM3 monitor output for the measurement of RL

C hange of t em per at ur e ( °C )

Figure 2(b) – Temperature sensor data for the surface temperature

Figure 2(c) – PM2 monitor output for the measurement of IL

Figure 2 – Results for high power transmission test of 10 dB attenuator

Table 7 presents the test results for IL and RL, highlighting that the RL varied by over 10 dB during the test with fixed optical attenuators, yet the difference between the pre-test and post-test data remains minimal.

Table 7 – Characteristics changes before and after the test

10 dB Attenuators Before 10,11 dB 53,0 dB

Optical Isolators Before 0,54 dB 63,1 dB > 55 dB

Optical Splitters Before 9,55 dB 0,04 dB > 50 dB

NOTE The values are the averages of measured samples The measuring conditions are as shown in Table 5.

Ferrule endfaces of the attenuators

To investigate the cause of RL change during testing, fixed attenuators were examined post-test A three-dimensional measurement instrument was used to assess the ferrule endface, with Table 8 detailing the fiber position Figure 3 illustrates the ferrule endface images for the 30 dB attenuator Prior to testing, the fibers were slightly protruding in all samples, but post-test, they had all retracted, with the maximum withdrawal measured at 0.15 μm on the input side.

Table 8 – Fibre protrusion and withdrawal in the fixed optical attenuator before and after the high power test

Sample Input side Output side

No.5 10 dB Attenuator + 0,02 àm – 0,10 àm No-data – 0,01 àm

NOTE The positive number means fibre protrusion from the ferrule, and the negative number means fibre withdrawal

Figure 3 – Pictures of ferrule endfaces in the input side of 30 dB attenuator

Change of characteristics in the backward direction incidence for optical

In Erbium Doped Fibre Amplifiers (EDFA), optical isolators are installed before and after the

Erbium Doped Fibre (EDF) is utilized to minimize signal light reflectance and ensure stable output power The optical isolator, positioned before the EDF, captures both the leakage of pumping light and the amplified spontaneous emission from the EDF Testing was performed using high power light incident in the backward direction.

Table 9 shows the test results In the optical isolator, the RL only changed in sample No 1

Figure 4 shows the monitoring result of optical isolator in this case It was observed that the

IL in the backward direction (same as isolation) changed between 30 and 40 dB as shown in

Figure 4(b) It is considered that this is due to the fact that the light with a wavelength of

The Faraday rotator absorbed light at 480 nm, resulting in a change in its angle of rotation due to elevated temperatures An analysis of the light absorption ratio of garnet and the temperature dependence of the Faraday rotator's rotation angle indicated that the temperature rose by 100 °C or more.

To investigate the cause of return loss degradation, the reflection point of the optical isolator was examined using a high-precision reflectometer from Agilent The analysis revealed that the reflection point was located 7 mm from the ferrule endface, as illustrated in Figure 5 This finding indicated that the optical fiber was broken at the ferrule.

Table 9 – Test result in the backward direction

Components Sample number Fluctuation of backward

Degradation of forward IL RL after the test Power at which damage occurred

Isolators No.1 < 10 dB > 30 dB < 0,1 dB > 55 dB 5 W

No.2 < 10 dB < 1 dB < 0,1 dB > 55 dB – No.3 < 10 dB < 1 dB < 0,1 dB > 55 dB – No.4 < 10 dB < 1 dB < 0,1 dB > 55 dB –

Laser Power (W) Las er pow er ( W )

Figure 4(a) – Temperature sensor data for the surface temperature

Change of insertion loss (dB)

Laser power (W) Las er pow er ( W )

C hange of i ns er tio n l os s ( dB )

Figure 4(b) – PM2 monitor output for the measurement of IL

Figure 4 – High power test result for backward direction for optical isolator (example)

Figure 5 – Measurement result for the ferrule point of reflection in the optical isolator

4 Thermal simulation of passive optical components

Thermal simulation in the high power light

General

High power light can lead to material degradation due to increased temperatures from light absorption To address this issue, thermal simulations were performed on each passive optical component.

Fixed optical attenuator

The internal structure of a fixed optical attenuator consists of metal doped fibre (MDF) with a specific absorption coefficient, securely bonded in a ferrule with adhesive The input and output fibre endfaces are polished using a physical contact (PC) method Desired attenuation levels are achieved by modifying both the absorption coefficient and the length of the MDF.

Thermal simulation assumes that light is absorbed depending on the absorption coefficient of

MDF and converted into the heat ANSYS Multi-physics Ver 9.1 simulation software was used

The contact point diameter between the optical fiber and the MDF in the connector measures 0.25 mm For ease of calculation, all components were modeled as cylindrical shapes, utilizing standard physical properties such as thermal conductivity, specific heat, and density.

Figure 6 – Thermal distribution of fixed optical attenuator by thermal simulation

Optical output power; 1 W Contact surface diameter; 0,25 mm

Figure 7 – Maximum internal temperature of fixed optical attenuator by thermal simulation (input power: 1 W)

Figure 6 illustrates the thermal distribution of a simplified cross section of a fixed optical attenuator, with a 1 W light power entering from the indicated direction The upper side is omitted due to its centrosymmetry The attached face refers to the area where the ferrule on the input side contacts the one on the MDF side Figure 7 displays the temperature changes in the MDF core area, highlighting that the maximum temperature for the 10 dB attenuator reached 200 °C, while the 20 dB attenuator recorded a temperature of 215 °C.

Figure 8 illustrates the temperature variations in the core area, sleeve area, and outer package of the optical attenuator when subjected to a light power of 2 W The outer package, in contact with ambient air, exhibited a slower temperature change compared to the core and sleeve areas Notably, the temperature in the sleeve area rapidly exceeded 200 °C within a brief timeframe.

The simulation estimated the outer package temperature to reach around 150 °C, while actual measurements indicated it was approximately 90 °C (refer to Table 6) The discrepancy between the simulation results and the actual data is attributed to thermal conduction from the ambient air surrounding the outer package.

Figure 8 – Thermal simulation of fixed optical attenuator (input power: 2 W)

Optical isolator

The thermal simulation results of an optical isolator, depicted in Figure 9, indicate that with an input power of 5 W, the maximum temperature reached around 120 °C, while the outer surface temperature was approximately 110 °C The absence of materials with high light absorption rates in the optical path contributed to a lower temperature compared to that observed in the fixed optical attenuator.

In a simulation of reverse direction with an incident light power of 3 W, the ferrule reached a maximum temperature of around 400 °C This elevated temperature, exceeding 400 °C, is believed to have led to the fiber breakage discussed in section 3.5.

Figure 9 – Thermal simulation of optical isolator (forward direction, input power: 5 W)

Optical splitter

The thermal simulation results of an optical splitter, depicted in Figure 10, indicate that under an insertion loss (IL) of 1 dB and a light input power of 5 W, the maximum temperature reached 150 °C This temperature exceeds the actual data recorded in Table 6, which shows a maximum of 87 °C The discrepancy is attributed to heat conduction effects from the outer case to the surrounding air.

M ax im um tem per at ur e ( °C )

Figure 10 – Maximum internal temperature of optical splitter by thermal simulation (forward direction, input power: 5 W)

Temperature rise simulation in the medium power light

The simulation was performed within the optical power range anticipated for the operational use of optical attenuators, with the specific conditions detailed in Table 10.

Figure 11 shows the maximum internal temperature at each attenuation value The horizontal axis shows the attenuation converted into the absorption rate (30 dB = 0,999, 20 dB = 0,99,

Table 10 – Conditions of optical attenuator for simulation

Input Powers 1 dB attenuator 5 dB attenuator 10 dB attenuator

Condition: A With Housing, Ambient Temperature 70 °C

Condition: B Without Housing, Ambient Temperature 70 °C

Condition: C With Housing, Ambient Temperature 25 °C

Figure 12 shows the relationship between the input light power and the maximum internal temperature All the data can be approximated by a line with a slope of 70 °C on the y-axis

(y = ax + 70) It is necessary to use the components at a temperature under 100 °C so that the glass transition temperature of adhesive (approximately 120 °C) is not exceeded

The estimated light power that remains below the glass transition temperature is 200 mW with a 10 dB attenuator and 300 mW with a 5 dB attenuator.

Figure 11 – Ambient temperature dependency of maximum temperature in the thermal simulation of fixed optical attenuator

Figure 12 – Relationship between input light power and maximum temperature in the thermal simulation of fixed optical attenuator

5 Long-term test of high power light

To assess the long-term reliability under high power conditions, a long-term test was performed This test was designed to be non-destructive, utilizing a specific input light power level.

60 % of that used in the damage threshold test The measuring performance is the same as those in Table 4 The test conditions are shown in Table 11

Table 11 – Conditions of long-term test

Items Optical attenuators Optical isolators Optical splitters

Out surface temperature monitored monitored monitored

Figure 13 illustrates the variations in insertion loss (IL), return loss (RL), and the outer surface temperature of a fixed optical attenuator over time In sample B, a temperature change was observed after approximately 100 hours, attributed to the detachment of the temperature sensor, which was subsequently replaced to continue the testing Likewise, Figures 14 and 15 depict the IL, RL, and outer surface temperature for an optical isolator and an optical splitter, respectively.

No characteristic changes were observed in either case

IL and RL variation (dB)

Temperature (deg.C) Tem per at ur e (°C )

IL and R L v ar ia tion ( dB )

Temperature (deg.C) Tem per at ur e (°C )

Figure 13 – Change of IL and RL of fixed optical attenuator

Temperature (deg.C)Tem per at ur e (°C )

Figure 14 – Change of IL and RL of optical isolator

Temperature (deg.C)Tem per at ur e (°C )

Figure 15 – Change of IL and RL of optical splitter

Tables 12, 13, and 14 present the optical characteristics of a fixed optical attenuator, an optical isolator, and an optical splitter before and after a long-term test All samples exhibited no changes in optical characteristics Additionally, the protrusion of the optical fiber in the fixed optical attenuator was measured before and after the test, revealing some protrusion; however, it was not significant enough to affect the optical characteristics.

Table 12 – Measurement result of optical characteristics and protrusion before and after the test of fixed optical attenuator

Samples IL (dB) RL (dB) Fibre protrusion

Table 13 – Measurement result of optical characteristics before and after the test of optical isolator

Samples IL (dB) PDL (dB) Isolation

Table 14 – Measurement result of optical characteristics before and after the test of optical splitter

No IL (dB) PDL (dB) RL (dB)

1 310 nm 1 550 nm 1 310 nm 1 550 nm 1 310 nm 1 550 nm

Optical fibre withdrawal was observed in the high power test of the optical attenuator In the fixed optical attenuator, a metal doped fibre is fixed in a zirconia ferrule with adhesive

The glass transition temperature of the adhesive is approximately 120 °C, which is significantly higher than the product's storage temperature Testing indicated that the temperature of the adhesive securing the metal-doped optical fiber in the ferrule became sufficiently high to soften, leading to the withdrawal of the optical fiber from the ferrule's head.

In the simulation, the input side of the fibre core reached temperatures between 200 to 300 °C, while the output side ranged from 150 to 250 °C This resulted in internal temperatures surpassing the glass transition temperature of the adhesive The withdrawal of the optical fibre is attributed to the softening of the adhesive and the differing line expansion coefficients between zirconia and metal-doped fibre, which ultimately affected the reflectance level (RL).

When high power light is input in a backward direction into an optical isolator, it can cause the optical fiber to break The optical isolator operates by rotating the polarization angle through a Faraday rotator and adjusting the angle of incidence using the refractive index of bi-refringent crystals Consequently, the light's focus point in the backward direction does not align with the fiber core, preventing concentration within the optical fiber As a result, an optical power of 5 W is diffused within the fiber and absorbed by the ferrule, leading to a significant increase in temperature, ultimately causing the surface temperature of the optical isolator to rise sharply.

174 °C (actual data), and the temperature of ferrule was higher than that of the optical isolator

The stress that leads to the breakage of optical fibers is primarily attributed to the thermal distribution and the differences in thermal expansion coefficients among various materials, including glass, ferrule, adhesive, and air.

General

High optical power was applied to specific passive optical components, including fixed optical attenuators, optical isolators, and optical splitters, through various tests such as damage threshold, long-term, and thermal simulation.

Table 2, and the following results were obtained.

Fixed optical attenuator

When the incident optical power reached 2.3 W or higher, the return loss (RL) exhibited significant changes over a short duration In contrast, an incident optical power of 1 W maintained a stable RL over 500 hours Post-breaking test analysis revealed fiber withdrawal from the ferrule, prompting a thermal simulation to assess internal temperature This simulation indicated a correlation between internal temperature, incident optical power, and attenuation value Notably, the thermal simulation for medium power identified a threshold for incident optical power, with the limit for a 10 dB attenuator set at 200 mW.

Optical isolator

No deterioration in optical characteristics was observed when incident light power of up to 5 W was applied in the forward direction However, when high power light was input in the backward direction, one of the four optical isolators exhibited a decrease in return loss (RL), with the external case temperature reaching approximately 170 °C.

The internal reflection position was assessed using a reflect meter, revealing a break in the optical fiber located 7 mm from the ferrule's endface Additionally, thermal simulations indicated that the internal temperature increased to approximately

400 °C Based on these results, it was considered that the optical fibre broke due to the stress caused by the difference of thermal expansion coefficient of constituting materials.

Optical splitter

No deterioration of optical characteristic was observed with an incident light power up to 5 W for the optical splitter.

Conclusion

Based on the results of the tests and simulation, the following was found:

High power light can lead to the performance deterioration of optical passive components, primarily due to the temperature rise in materials from light absorption and the resulting thermal stress or distortion caused by uneven thermal distribution.

To ensure accurate testing of high power light, it is essential to monitor the reflection levels on-site Future studies should prioritize tracking the reflection loss (RL) before, during, and after the testing process.

To effectively estimate deterioration, conducting thermal simulations using a simplified model is essential Additionally, careful consideration must be given to the management of waste heat in the design of optical passive components intended for use under high-power light conditions.

4) It was found that it was necessary to maintain the internal temperature less than glass transition temperature of adhesive (approximately 120 °C) to avoid the deterioration of

RL caused by the withdrawal of fibre

5) In the high power light test, it is also useful to disassemble samples to estimate the deterioration mechanism

Optoelectronic Industry and Technology Development Association (OITDA) – Technical Paper

(TP), OITDA-TP04/SP_PD-2008, “Technical Paper of investigation of high-power reliability for passive optical components for optical communication application”

Tiêu đề	Report of high power transmission test of specified passive optical components
Trường học	International Electrotechnical Commission
Chuyên ngành	Electrotechnics and Optical Components
Thể loại	technical report
Năm xuất bản	2011
Thành phố	Geneva

Định dạng
Số trang	30
Dung lượng	1,76 MB