Loss control coupling method for light receiving optical path of high-speed silicon optical module and high-speed silicon optical module

Abstract

The invention provides a return loss control coupling method for a light receiving optical path of a high-speed silicon optical module and the high-speed silicon optical module. The return loss control coupling method comprises the following steps: bonding a TIA (Tungsten Inert Architecture) on the surface of a PCB (Printed Circuit Board); the integrated lens PD is mounted on the surface of the trans-impedance amplifier TIA through eutectic bonding, the integrated lens PD comprises a condensing lens and a PD, and the PD is a photoelectric detector; the whole FA is rotated by taking the front end of the FA as a rotating shaft and being perpendicular to the surface of the PCB, so that an outgoing beam of the FA is perpendicular to the surface of the photoelectric detector PD, and the FA is an optical fiber array; and moving the FA in a plane parallel to the PD, monitoring return loss of each channel and PD response current at the same time, and when the return loss and the PD response current meet certain conditions, dispensing and fixing the FA to complete coupling. The invention also provides a high-speed silicon optical module coupled by adopting the method, and when the integrated lens PD is used, the requirements of reflection resistance and high coupling efficiency can be met at the same time.

Description

Technical Field

[0001] The present invention relates to optical module coupling technology, and in particular to a return loss control coupling method for a high-speed silicon optical module optical receiving optical path with a single wavelength of 200G or above, and also to a high-speed silicon optical module coupled using the return loss control coupling method. Background Art

[0002] With the rapid expansion of AI computing power requirements and the upgrade of data center architecture, the data transmission rate requirements of optical modules have also increased from the traditional 400G to 800G and 1.6T. In order to meet the high-speed transmission requirements of optical modules, PD manufacturers have launched back-illuminated PDs with integrated lenses and small photosensitive surfaces. Conventional single-wavelength direct-illuminated PDs below 200G, such as the attached Figure 1 As shown in the optical path, the position of the external lens relative to the PD can be freely designed in the module optical path design. It is only necessary to consider whether the light beam reaches the PD surface. Because the prism or FA that emits the light beam will be made at a specific angle to reduce the proportion of reflected light from the PD surface returning to the optical fiber, and the PD used for single-wavelength 200G rates and above has an integrated lens, the position of the lens relative to the PD photosensitive surface is fixed. Under this premise, the design of the external optical path must enable the light beam to be efficiently converged into the PD photosensitive surface and avoid the reflected light from the PD surface returning to the optical fiber. The traditional optical path solution can only meet one of the requirements of anti-reflection and high coupling efficiency in the application of integrated lens PD, which affects the performance of the optical module and limits the application of integrated lenses. Summary of the Invention

[0003] To solve the problems in the prior art, the present invention provides a return loss control coupling method for the optical receiving optical path of a high-speed silicon photonic module, and also provides a high-speed silicon photonic module coupled using the return loss control coupling method. This overcomes the shortcomings of conventional optical paths in the application of integrated lens PDs, can improve the coupling efficiency of the integrated lens PD, and solve the return loss problem caused by the improved coupling efficiency.

[0004] The return loss control coupling method of the optical receiving optical path of the high-speed silicon photonic module of the present invention comprises the following steps:

[0005] Step 1: Glue the transimpedance amplifier (TIA) to the surface of the PCB.

[0006] Step 2: Mount the integrated lens PD on the surface of the transimpedance amplifier TIA through eutectic soldering. The integrated lens PD includes a focusing lens and a PD, which is a photodetector.

[0007] Step 3: Rotate the FA as a whole with the front end of the FA as the rotation axis, perpendicular to the surface of the PCB board, so that the FA output light beam is perpendicular to the surface of the photodetector PD. The FA is a fiber array.

[0008] Step 4: Move the FA in a plane parallel to the photodetector PD, while monitoring the return loss and PD response current of each channel. When the return loss and PD response current meet certain conditions, glue is applied to fix the FA to complete the coupling.

[0009] The present invention is further improved. In step three, the rotation angle of FA is greater than 4.5°.

[0010] The present invention is further improved, and the rotation angle of the FA is 7.2°.

[0011] The present invention is further improved in that a pad is provided between the FA and the PCB board, and the upper surface of the pad is an inclined surface with a matching rotation angle. In step three, the coupling table fixture automatically clamps the FA, rotates the FA so that the FA cover and the upper surface of the pad are parallel, and then the coupling table automatically moves the FA so that the FA cover and the pad contact. After the pressure sensor on the coupling table senses the contact between the FA cover and the pad, the FA is lifted by the coupling table to a set height.

[0012] The present invention is further improved. In step 4, the coupling platform moves the FA in a plane parallel to the PD surface, that is, on the pad, monitors the changes in the PD response current, couples the PD response current of each channel to the best, and the coupling platform records the position coordinates of the FA on the coupling platform at this time.

[0013] The present invention is further improved. After the PD response current is determined, a return loss adjustment step is performed to control the return loss within a set range. The processing method of the return loss adjustment step is:

[0014] (1) Taking the recorded position coordinates as the starting point, make FA move along a circular trajectory with a radius of R around the starting point, where R is a positive number;

[0015] (2) Record the return loss value every set angle to determine whether the return loss value of each channel meets the set conditions. If yes, stop the movement and the position of FA is the optimal coupling position. If not, continue to move along the motion trajectory.

[0016] (3) If there is no position that meets the requirements after one circle of movement, then the radius R of the circle is increased by d, and FA is allowed to move along the new radius R+d. Then, step (2) is executed until a suitable coupling position is found.

[0017] The present invention also provides a high-speed silicon photonic module, which uses the return loss control coupling method of the optical receiving optical path of the high-speed silicon photonic module to achieve coupling. The high-speed silicon photonic module includes a PCB board, an FA, an integrated lens PD, a transimpedance amplifier TIA and a pad. The lower surface of the pad and the transimpedance amplifier TIA are bonded to a set position on the surface of the PCB board, the integrated lens PD is mounted on the surface of the transimpedance amplifier TIA by eutectic welding, the FA is arranged above the pad, and the FA output light beam is perpendicular to the surface of the integrated lens PD.

[0018] The present invention is further improved in that the upper surface of the pad is an inclined surface with a matching rotation angle, and the thickness of the pad is between 150 and 500 μm.

[0019] The present invention is further improved in that the integrated lens PD is a back-illuminated high-speed PD, and the light-incoming surface of the photodetector PD is integrated with a focusing lens, the curvature radius of the focusing lens is between 80 and 140 μm, and the center of the focusing lens is aligned with the center of the photosensitive surface of the photodetector PD.

[0020] The present invention is further improved in that the optical fiber light-emitting end face of the FA is polished at an angle of 42.5°, so that the light beam is totally reflected on the polished surface and the forward direction of the light beam is deflected. The distance between the optical fiber light-emitting end face and the surface of the photodetector PD is 50 to 300 μm, so that the light beam emitted by the optical fiber can be completely converged into the photosensitive surface of the photodetector PD.

[0021] Compared with the prior art, the beneficial effects of the present invention are: by rotating the FA as a whole by a certain angle so that the FA output light beam is perpendicular to the surface of the photodetector PD, the present invention can significantly improve the coupling efficiency of the integrated lens PD and solve the return loss problem caused by the improved coupling efficiency, so that the present invention can simultaneously meet the anti-reflection and high coupling efficiency requirements when using the integrated lens PD in a high-speed silicon photonic module with a single wave of more than 200G, thereby effectively improving the performance of the high-speed silicon photonic module. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 This is a schematic diagram of the optical path using an external lens in the prior art;

[0023] Figure 2 This is a flow chart of the return loss control coupling method of the present invention;

[0024] Figure 3 This is a schematic diagram of the light receiving optical path of the high-speed silicon photonic module of the present invention;

[0025] Figure 4 Schematic diagram of the effect of FA rotation angle on PD receiving efficiency;

[0026] Figure 5Schematic diagram of the effect of FA rotation angle on the efficiency of light reflected from the PD surface back to the optical fiber;

[0027] Figure 6 Schematic diagram of the effect of the FA-PD distance on the PD receiving efficiency when the FA is rotated at different angles;

[0028] Figure 7 This is a simulation curve showing the effect of the lateral offset of the FA relative to the PD on the PD receiving efficiency and return loss when the FA is rotated 7.2°.

[0029] Figure 8 This is a simulation curve showing the effect of the lateral offset of the FA relative to the PD on the PD receiving efficiency and return loss when the FA is rotated 5°.

[0030] Figure 9 Schematic diagram of the coupling curve simulation of the coupling platform when the PD patch position deviates when the FA is rotated 5°;

[0031] Figure 10 Schematic diagram of the top view of the high-speed silicon photonic module of the present invention;

[0032] Figure 11 Schematic diagram of the side view of the high-speed silicon photonic module of the present invention;

[0033] Figure 12 This is a schematic diagram of the structure of the high-speed silicon photonic module of the present invention after the TIA and PD are mounted;

[0034] Figure 13 This is a schematic diagram of the structure of the high-speed silicon photonic module after the pad is mounted;

[0035] Figure 14 This is a schematic diagram of the optical fiber line layout of the coupling station with return loss monitoring function. DETAILED DESCRIPTION

[0036] The present invention is further described in detail below with reference to the accompanying drawings and embodiments.

[0037] like Figure 1 As shown, the return loss control coupling method of the optical receiving path of the high-speed silicon photonic module of the present invention includes the following steps:

[0038] Step 1: Glue the transimpedance amplifier (TIA) to the surface of the PCB.

[0039] Step 2: Mounting the integrated lens PD on the surface of the transimpedance amplifier TIA (hereinafter referred to as TIA) through eutectic soldering. The integrated lens PD includes a focusing lens and a PD, which is a photodetector.

[0040] Step 3: Rotate the FA as a whole with the front end of the FA as the rotation axis, perpendicular to the surface of the PCB board, so that the FA output light beam is perpendicular to the surface of the photodetector PD. The FA is a fiber array.

[0041] Step 4: Move the FA in a plane parallel to the photodetector PD, while monitoring the return loss and PD response current of each channel. When the return loss and PD response current meet certain conditions, glue is applied to fix the FA to complete the coupling.

[0042] The effect of FA's rotation angle on PD receiving efficiency is as follows: Figure 4 As shown in the figure, when the rotation angle is above 4.5°, the receiving efficiency no longer increases significantly with the rotation angle, so in this example, it is preferred to rotate above 4.5°. The FA rotation angle and the corresponding peak return loss are shown in the figure. Figure 5 shown.

[0043] Preferably, in step 4, the return loss and PD response current in this example need to meet the following conditions: the return loss is reduced to below -30 dB and the response current is above 0.7 mA.

[0044] like Figure 6 As shown in the figure, the present invention simulates PD reception efficiency with 5° and 7.2° rotations and compares this with the prior art, which uses no FA rotation. It can be seen that when the FA is rotated 5°, the coupling efficiency reaches the level of when the FA is rotated 7.2°. The FA-to-PD distance and PD reception efficiency curves for 5° and 7.2° rotations overlap. Compared with the prior art, this method can effectively improve PD reception efficiency.

[0045] Depend on Figure 7 and Figure 8 As shown, the present invention compares the PD coupling curve simulation corresponding to FA rotation of 7.2° and 5°. Figure 7 It can be seen that when the FA rotates 7.2°, the position where the PD has the best receiving efficiency and the position where the return loss is the worst coincide with each other. Figure 8 It can be seen that when the FA is rotated 5°, the position with the best PD receiving efficiency and the position with the worst return loss are separated. It seems that the solution of rotating the FA 5° should be the best.

[0046] However, the return loss calculated by simulation Figure 5 From the results shown, FA rotation of 5° is slightly better than FA rotation of 7.2°. Figure 8 and Figure 9 It can be seen that rotating the FA by 5° may lead to the risk of return loss exceeding the protocol. This risk comes from the position deviation of multiple PDs when the PDs are eutectic soldered to the TIA surface. The theoretical design is to make the four PDs strictly concentric. However, strict concentricity is difficult to achieve in actual process production. Generally, the patch can only guarantee a maximum offset of 5μm. Figure 9 The coupling curves are shown when the deviation between the two PD patches is 5μm. The coupling stage will eventually stop in the middle of the overlapping area of ​​the two PD coupling curves, which corresponds exactly to Figure 8The worst return loss position is the reason why the return loss may exceed the protocol when the FA is rotated 5°. Figure 8 It can be seen that the coupling efficiency curve is inclined with the position of FA on the horizontal plane, which means that in the subsequent return loss coupling, the PD receiving efficiency needs to be sacrificed to balance the return loss. Figure 7 It can be seen that when the FA is rotated 7.2°, the PD receiving efficiency curve has a flat area as the FA position changes on the horizontal plane. Moreover, the position where the return loss is reduced to meet the -30dB requirement is still within the coupling efficiency flat area. There is no need to sacrifice PD receiving efficiency during return loss coupling. Therefore, rotating the FA 7.2° is the optimal solution in this example.

[0047] like Figure 10 and Figure 11 As shown, the high-speed silicon photonic module of the present invention adopts the return loss control coupling method of the optical receiving optical path of the high-speed silicon photonic module to achieve coupling. The high-speed silicon photonic module includes a PCB board 5, FA 3, an integrated lens PD 4, a transimpedance amplifier TIA 6 and a pad 7, wherein the lower surface of the pad 7 and the transimpedance amplifier TIA 6 are bonded to a set position on the surface of the PCB board 5 by silver paste, the integrated lens PD 4 is mounted on the surface of the transimpedance amplifier TIA 6 by eutectic welding, the FA3 is arranged above the pad 7, and the FA 3 output light beam is perpendicular to the surface of the integrated lens PD 4.

[0048] The FA includes an optical fiber 31, an FA cover plate 34, a V-groove plate 33, and a FA tail glue 32 arranged between the FA cover plate 34 and the optical fiber 31. The FA tail glue 32 is used to fix and protect the optical fiber 31 and is preferably made of silicone.

[0049] In this example, the optical fiber's light-emitting end face is polished at a 42.5° angle. This means that the front end a of the V-groove plate 33 is angled at 42.5°. This causes the light beam to undergo total internal reflection at the polished surface, deflecting its forward direction. In this example, FA 3 is rotated 7.2 degrees, making the emitted light beam perpendicular to the surface of the integrated lens PD 4. If the front end a of the V-groove plate 33 is polished at a different angle, the corresponding rotation angle will be necessary to make the light beam perpendicular to the PD surface.

[0050] Preferably, the distance between the optical fiber light-emitting end face and the PD surface is 50-300 μm, so that the light beam emitted by the optical fiber can be completely converged into the photosensitive surface of the PD.

[0051] Preferably, the top surface of the pad 7 in this example is a rotation-angle-matching bevel, so that when the FA is placed on the top surface of the pad 7, it can automatically achieve the set rotation angle. In this example, the top surface is designed with a 7.2° bevel to ensure uniform adhesive thickness between the pad and the FA cover, avoiding light loss problems caused by high and low temperatures and ensuring device reliability. The thickness of the pad can be adjusted to match the curvature radius of the lens integrated by the PD, with a typical thickness between 150 and 500 μm.

[0052] The PD in this example is a back-illuminated high-speed PD. The PD light-incoming surface is integrated with a focusing lens. The curvature radius of the focusing lens is 80 to 140 μm, and the center of the focusing lens is aligned with the center of the PD photosensitive surface.

[0053] The position between FA 3 and integrated lens PD 4 is determined by active coupling, which monitors the PD response current and the reflected light on the PD surface. As an embodiment of the present invention, this example performs active coupling through a coupling stage.

[0054] The optical fiber line layout of the coupling station with return loss monitoring function is as follows Figure 14 As shown, the optical module in this example has four optical channels. Therefore, the first branches 8 of four 50 / 50 beam splitters 10 are connected to a four-channel light source, which provides light for the device's optical power (PD). The second branches 9 are connected to a four-channel power meter, which measures the optical power of light reflected from the PD surface back into the optical fiber, also known as return loss. The combined end 101 of the 50 / 50 beam splitter 10 connects to the FC end of an MPO-FC patch cord via an FC-FC flange 11. The MPO connector 12 of the MPO-FC patch cord connects to the MPO optical port 14 of the optical device via an MPO adapter 13.

[0055] The return loss control coupling method in this example is as follows:

[0056] (1) TIA 6 is bonded to PCB 5 with silver paste. After the glue is cured, the integrated lens PD 4 is mounted on the TIA surface by eutectic soldering. In this example, it is a 4-channel optical device. Four separate integrated lens PDs are mounted on the TIA surface in sequence. The equipment for mounting the integrated lens PDs needs to ensure that the spacing error between the four integrated lens PDs is within 5μm. Figure 12 As shown in the figure, when the patch position deviation between the four PDs is controlled within 5μm, and the deviation between the PD's photosensitive surface and the integrated lens center is within 3μm, it can be ensured that the PD receiving efficiency will not be affected by the patch position deviation.

[0057] (2) The design of the mounting position of the pad 7 needs to take into account the overflow of glue when TIA 6 is mounted in the previous step. The front plane of the pad should be kept at a distance of more than 200μm from the TIA to avoid the distance being too small, which will cause the front end of the pad to be attached to the glue overflowing from the TIA, resulting in the pad 7 not being able to be flatly attached to the PCB board. Figure 13 shown.

[0058] (3) The coupling stage fixture automatically grips the FA and rotates it 7.2° to align the FA cover and the upper surface of the pad. The coupling stage then automatically moves the FA so that the FA cover and the pad contact. After the pressure sensor on the coupling stage senses contact between the FA cover and the pad, the FA is lifted 50 μm (the distance from the FA to the PD in this example is designed to be 50 μm). The coupling stage then moves the FA in a plane parallel to the PD surface, monitoring changes in the PD response current. The PD response currents of the four channels are coupled to the optimal state, and the coupling stage records the position coordinates of the FA on the coupling stage at this time.

[0059] (4) The coupling program jumps to the return loss adjustment step. With the FA position recorded in step (3) as the starting point, the FA motion trajectory is a circle with a radius of 5μm, centered at the starting point. A return loss value is recorded every 45°. If the scan has found a point where the return loss of all four channels is below -30dB, the FA position stops at the corresponding position. If the first scan does not find a position where the return loss meets the requirements, the radius of the circle is enlarged by 2μm, and the scan is repeated until the return loss of all four channels meets the requirements below -30dB. According to the feedback from the actual operation results, the first circle with a radius of 5μm can generally find a point where the return loss meets the requirements. In some cases, the radius of the circle needs to be enlarged by 2μm. The return loss scan mentioned in this step records a value every 45°, which is only a suggestion. It can also be recorded at intervals of 20°, 30°, etc. The smaller the scanning angle interval, the more accurate the FA position is found. However, the time consumed by the return loss scan will also increase accordingly.

[0060] As can be seen from the above, the present invention has the following innovations:

[0061] By rotating the FA as a whole by a certain angle so that the FA's output light beam is perpendicular to the surface of the photodetector PD, the present invention can significantly improve the coupling efficiency of the integrated lens PD and solve the return loss problem caused by the improved coupling efficiency. When the integrated lens PD is used in a high-speed silicon photonic module with a single wave of more than 200G, the present invention can simultaneously meet the anti-reflection and high coupling efficiency requirements, thereby effectively improving the performance of the high-speed silicon photonic module.

[0062] The specific implementation manner described above is a preferred implementation manner of the present invention, and is not intended to limit the specific implementation scope of the present invention. The scope of the present invention includes but is not limited to this specific implementation manner. All equivalent changes made in accordance with the present invention are within the protection scope of the present invention.

Claims

1. A return loss control coupling method for a high-speed silicon photonic module optical receiving optical path, characterized in that: The steps include: Step 1: Glue the transimpedance amplifier (TIA) to the surface of the PCB. Step 2: Mount the integrated lens PD on the surface of the transimpedance amplifier TIA through eutectic soldering. The integrated lens PD includes a focusing lens and a PD, which is a photodetector. Step 3: Rotate the FA as a whole with the front end of the FA as the rotation axis, perpendicular to the surface of the PCB board, so that the FA output light beam is perpendicular to the surface of the photodetector PD. The FA is a fiber array. Step 4: Move the FA in a plane parallel to the photodetector PD, while monitoring the return loss and PD response current of each channel. When the return loss and PD response current meet certain conditions, glue is applied to fix the FA to complete the coupling.

2. The return loss control coupling method according to claim 1, wherein: In step 3, the rotation angle of FA is greater than 4.5°.

3. The return loss control coupling method according to claim 2, wherein: The rotation angle of the FA is 7.2°.

4. The return loss control coupling method according to claim 2, wherein: A pad is provided between the FA and the PCB board, and the upper surface of the pad is an inclined surface with a matching rotation angle. In step three, the coupling table fixture automatically clamps the FA, rotates the FA so that the FA cover and the upper surface of the pad are parallel, and then the coupling table automatically moves the FA so that the FA cover and the pad contact. After the pressure sensor on the coupling table senses the contact between the FA cover and the pad, the FA is lifted by the coupling table to a set height.

5. The return loss control coupling method according to claim 4, wherein: In step 4, the coupling platform moves the FA in a plane parallel to the PD surface, that is, on the pad, monitors the changes in the PD response current, and couples the PD response current of each channel to the best. The coupling platform records the position coordinates of the FA on the coupling platform at this time.

6. The return loss control coupling method according to claim 5, wherein: After the PD response current is determined, a return loss adjustment step is performed to control the return loss within a set range. The processing method of the return loss adjustment step is as follows: (1) Taking the recorded position coordinates as the starting point, make FA move along a circular trajectory with a radius of R around the starting point, where R is a positive number; (2) Record the return loss value every set angle to determine whether the return loss value of each channel meets the set conditions. If yes, stop the movement and the position of FA is the optimal coupling position. If not, continue to move along the motion trajectory. (3) If there is no position that meets the requirements after one circle of movement, then the radius R of the circle is increased by d, and FA is allowed to move along the new radius R+d. Then, step (2) is executed until a suitable coupling position is found.

7. A high-speed silicon photonic module, wherein coupling is achieved by using the return loss control coupling method for a high-speed silicon photonic module optical receiving path according to any one of claims 1 to 7, characterized in that: The device comprises a PCB, an FA, an integrated lens PD, a transimpedance amplifier TIA, and a pad. The lower surface of the pad and the transimpedance amplifier TIA are bonded to a set position on the surface of the PCB. The integrated lens PD is mounted on the surface of the transimpedance amplifier TIA by eutectic soldering. The FA is arranged above the pad, and the FA output light beam is perpendicular to the surface of the integrated lens PD.

8. The high-speed silicon photonic module according to claim 7, wherein: The upper surface of the pad is an inclined surface with a matching rotation angle, and the thickness of the pad is between 150 and 500 μm.

9. The high-speed silicon photonic module according to claim 7, wherein: The integrated lens PD is a back-illuminated high-speed PD. The light-incoming surface of the photodetector PD is integrated with a condenser lens. The curvature radius of the condenser lens is between 80 and 140 μm. The center of the condenser lens is aligned with the center of the photosensitive surface of the photodetector PD.

10. The high-speed silicon photonic module according to claim 7, wherein: The optical fiber light-emitting end face of the FA is polished at an angle of 42.5°, so that the light beam is totally reflected on the polished surface and the forward direction of the light beam is deflected. The distance between the optical fiber light-emitting end face and the surface of the photodetector PD is 50 to 300 μm, so that the light beam emitted by the optical fiber can be completely converged into the photosensitive surface of the photodetector PD.

Images