Printed circuit boards, printed wiring boards and electronic devices
By designing signal lines with specific structures on the printed circuit board and adjusting the line width and line length, the problem of mismatch between the characteristic impedance of the signal lines and the pad is solved, and the transmission quality and stability of the digital signal are improved.
Patent Information
- Application Number
- CN202010767058.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Priority Date
- 2020-07-01
- Filing Date
- 2020-08-03
- Publication Date
- 2024-11-19
- Estimated Expiration
- 2040-08-03
AI Technical Summary
The mismatch between the characteristic impedances of the signal lines and pads in printed circuit boards leads to a decrease in the quality of the signal waveform. Especially after the signal transmission speed increases, it is difficult for the prior art to effectively suppress such mismatch, affecting the transmission quality of the digital signal.
A printed circuit board is designed, and the signal line is composed of a continuous first line part, a second line part, a third line part and a fourth line part. By adjusting the line width and length, the signal terminal and the fourth line part are formed. In the overall structure, the characteristic impedance of the second line part is lower than that of the first line part, and the characteristic impedance of the third line part is higher than that of the first line part, and the characteristic impedance is matched by controlling the line width and line length to reduce signal reflection and disturbance.
Effectively reduce the characteristic impedance mismatch between the signal line and the pad, improve the transmission quality of the digital signal, reduce the voltage disturbance of the signal, improve the overall characteristic impedance matching of the signal line, and improve the stability of signal transmission and quality.
Smart Images

Figure CN112351577B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to wiring technology for printed circuit boards. Background Art
[0002] A printed circuit board includes a signal line for transmitting a signal. The signal line is designed under some restrictions, and one restriction is related to the shape of a pad on which an electrical component is mounted. Typically, the line width of the pad is greater than the line width of the main line of the corresponding signal line. Therefore, since the characteristic impedance of the pad becomes lower than the characteristic impedance of the main line, a mismatch in the characteristic impedance occurs at the boundary between the main line and the pad. The mismatch in the characteristic impedance affects the quality of the signal waveform. In order to suppress the mismatch in the characteristic impedance of the pad, Japanese Patent Application Publication No. 2014-116541 proposes a technology for forming an opening in a ground plane facing the pad.
[0003] However, the speed of signals transmitted through signal lines tends to increase. As the transmission speed of signals is increased, the signals are increasingly required to have higher quality, and therefore, printed circuit boards need to be improved. Summary of the invention
[0004] According to a first aspect of the present invention, a printed circuit board includes an electrical component having a signal terminal and a printed wiring board on which the electrical component is mounted. The printed wiring board includes a signal line connected to the signal terminal. The signal line includes a first line portion, a second line portion, a third line portion, and a fourth line portion that are sequentially and continuously disposed. The signal terminal is joined to the fourth line portion so that the signal terminal and the fourth line portion form an integral structure. The second characteristic impedance of the second line portion is lower than the first characteristic impedance of the first line portion. The third characteristic impedance of the third line portion is higher than the first characteristic impedance. The fourth characteristic impedance of the integral structure formed by the fourth line portion and the signal terminal is lower than the first characteristic impedance.
[0005] According to a second aspect of the present invention, a printed circuit board includes an electrical component having a signal terminal and a printed wiring board on which the electrical component is mounted. The printed wiring board includes a signal line connected to the signal terminal. The signal line includes a first line portion, a second line portion, a third line portion, and a fourth line portion that are sequentially and continuously disposed. The signal terminal is joined to the fourth line portion. The line width of the second line portion is greater than the line width of the first line portion. The line width of the third line portion is less than the line width of the first line portion. The line width of the fourth line portion is greater than the line width of the first line portion.
[0006] According to a third aspect of the present invention, a printed wiring board on which an electrical component is mounted includes a signal line connected to a signal terminal of the electrical component. The signal line includes a first line portion, a second line portion, a third line portion, and a fourth line portion formed continuously. The signal terminal is configured to engage with the fourth line portion. The characteristic impedance of the second line portion is lower than the characteristic impedance of the first line portion. The characteristic impedance of the third line portion is higher than the characteristic impedance of the first line portion. The characteristic impedance of the fourth line portion is lower than the characteristic impedance of the first line portion.
[0007] According to a fourth aspect of the present invention, a printed wiring board on which an electrical component is mounted includes a signal line connected to a signal terminal of the electrical component. The signal line includes a first line portion, a second line portion, a third line portion, and a fourth line portion formed continuously. The signal terminal is configured to engage with the fourth line portion. The line width of the second line portion is greater than the line width of the first line portion. The line width of the third line portion is less than the line width of the first line portion. The line width of the fourth line portion is greater than the line width of the first line portion.
[0008] Further features of the present invention will become apparent from the following description of exemplary embodiments with reference to the attached drawings. BRIEF DESCRIPTION OF THE DRAWINGS
[0009] Figure 1 is a diagram showing an electronic device according to the first embodiment.
[0010] Figure 2 is a perspective view of the processing module of the first embodiment.
[0011] Figure 3A It is a plan view of a part of the processing module of the first embodiment.
[0012] Figure 3B is a cross-sectional view of a portion of the process module of the first embodiment.
[0013] Figure 4A It is a plan view of a part of the processing module of the second embodiment.
[0014] Figure 4B is a cross-sectional view of a portion of a process module according to a second embodiment.
[0015] Figure 5 is a perspective view of a processing module of the third embodiment.
[0016] Fig. 6A It is a plan view of a part of the processing module of the third embodiment.
[0017] Figure 6B is a cross-sectional view of a portion of a process module according to a third embodiment.
[0018] Fig. 7A It is a plan view of a part of the processing module of the fourth embodiment.
[0019] Figure 7B is a cross-sectional view of a portion of a process module according to a fourth embodiment.
[0020] Fig. 8A is a graph showing simulation results obtained in the example.
[0021] Figure 8B is a diagram showing signals used in the example.
[0022] Fig. 9A is a graph showing simulation results obtained in the example.
[0023] Fig. 9B is a diagram showing signals used in the example.
[0024] Fig. 10A This is a plan view of a portion of the processing module of Comparative Example 1.
[0025] Fig. 10B This is a plan view of a portion of a processing module of Comparative Example 2. DETAILED DESCRIPTION
[0026] Hereinafter, some embodiments of the present invention will be described in detail with reference to the accompanying drawings.
[0027] First embodiment
[0028] Figure 1 6 is a diagram showing a digital camera 600, which is an image pickup device as an example of an electronic device of the first embodiment. The digital camera 600 as an image pickup device is a digital camera with a replaceable lens; and includes a camera body 601. A lens unit (lens barrel) 602 including a lens is detachably attached to the camera body 601. The camera body 601 includes a housing 611, a processing module 300, and a sensor module 900. The processing module 300 and the sensor module 900 are disposed in the housing 611. The processing module 300 is an example of a printed circuit board, and is also an example of a semiconductor module. The processing module 300 and the sensor module 900 are electrically connected to each other via a cable (not shown).
[0029] The sensor module 900 includes an image sensor 700 as an image pickup element and a printed wiring board 800. The image sensor 700 is mounted on the printed wiring board 800. The image sensor 700 may be a complementary metal oxide semiconductor (CMOS) image sensor or a charge coupled device (CCD) image sensor. The image sensor 700 has a function of converting light passing through the lens unit 602 into an electrical signal.
[0030] The processing module 300 includes a semiconductor device 100, a connector 400, and a printed wiring board 200. The connector 400 is an example of an electrical component. The semiconductor device 100 and the connector 400 are mounted on the printed wiring board 200. The printed wiring board 200 is a rigid board. The semiconductor device 100 may be a digital signal processor; and has a function of receiving an electrical signal from the image sensor 700, correcting the electrical signal, and creating image data.
[0031] The connector 400 serves as an interface between the processing module 300 and an external device, and may be a universal serial bus (USB) connector or a high-definition multimedia interface (HDMI) (registered trademark) connector. The connector 400 includes an insertion opening 401 into which a USB cable connector (not shown) or an HDMI (registered trademark) cable connector (not shown) is inserted, or from which a USB cable connector or an HDMI cable connector is removed. The connector 400 is disposed in the housing 611 so that the insertion opening 401 is exposed to the outside from the housing 611 of the camera body 601.
[0032] Figure 2 1 is a perspective view of the processing module 300 of the first embodiment. The semiconductor device 100 is a semiconductor package. In the first embodiment, the semiconductor device 100 is a ball grid array (BGA) semiconductor package.
[0033] The printed wiring board 200 includes a signal line 250 that electrically connects the semiconductor device 100 and the connector 400. The signal line 250 is a line through which a digital signal is transmitted from the semiconductor device 100 to the connector 400. Figure 2 In the printed wiring board 200 , lines other than the signal line 250 , such as a power supply line, a ground line, and a signal line other than the signal line 250 , are not shown.
[0034] In addition, although Figure 2 A single signal line 250 (through which a signal is transmitted) is shown in FIG. 1 , but a plurality of signal lines 250 may be deployed. In this case, the plurality of signal lines 250 may constitute a bus.
[0035] The signal line 250 extends in the wiring direction, that is, in the X direction which is the longitudinal direction of the signal line 250. The thickness direction of the signal line 250 is defined as the Z direction, and the width direction of the signal line 250 is defined as the Y direction. The Z direction is equal to the thickness direction of the printed wiring board 200, and is a viewing direction when the printed wiring board 200 is viewed in a plane.
[0036] The signal line 250 includes an end 251 and an end 252. The end 251 is a first end formed in the X direction, and the end 252 is a second end formed in the X direction and opposite to the end 251. The semiconductor device 100 includes a terminal 101 that outputs a digital signal. The terminal 101 is connected to the end 251 of the signal line 250. The transmission speed of the digital signal transmitted by the signal line 250 is 1Gbps (gigabit per second) or higher. In the first embodiment, the digital signal is a single-ended signal. The material of the signal line 250 includes a conductive metal material such as copper or gold.
[0037] Figure 3Ais a plan view of a portion of the processing module 300 of the first embodiment. Figure 3B 2 is a cross-sectional view of a portion of the processing module 300 of the first embodiment. The printed wiring board 200 includes four conductor layers 221, 222, 223 and 224 and insulator layers 231, 232 and 233. The conductor layers 221, 222, 223 and 224 are arranged so that each of the insulator layers 231, 232 and 233 is sandwiched between adjacent conductor layers. The conductor layers 221 and 224 are surface layers or outer layers, and the conductor layers 222 and 223 are inner layers. The conductor layer 221 is a surface layer on which the semiconductor device 100 and the connector 400 are mounted. Note that a solder resist (not shown) may be formed on each of the conductor layers 221 and 224.
[0038] The signal line 250 is disposed in the conductor layer 221. The thickness of the signal line 250 in the Z direction is constant in the X direction and in the Y direction. In the conductor layer 222 adjacent to the conductor layer 221 via the insulator layer 231, a planar ground pattern 262 is disposed. In the conductor layer 223, a planar ground pattern 263 is disposed. In the conductor layer 224, a signal line 264 other than the signal line 250 is disposed.
[0039] The connector 400 is a surface mounted electrical component and includes a terminal 414 as a signal terminal. The terminal 414 is pin-shaped. The terminal 414 is connected to the end 252 of the signal line 250. The terminal 414 receives a signal. The transmission direction of the signal through the signal line 250 is defined as the X1 direction.
[0040] The signal line 250 includes line portions 211, 212, 213, and 214 formed continuously in the X direction. Line portion 211 is a first line portion, line portion 212 is a second line portion, line portion 213 is a third line portion, and line portion 214 is a fourth line portion. In the present embodiment, line portion 211, line portion 212, line portion 213, and line portion 214 are formed in sequence and continuously toward the X1 direction. The connector 400 includes a line portion 415 continuous with the terminal 414. Line portion 415 is a fifth line portion and is an internal line of the connector 400.
[0041] The line portion 211 is a main line; and is the longest in the X direction among the line portions 211 to 214. The line portion 214 is a pad that can be joined to the terminal 414 of the connector 400 via solder. When the line portion 214 is joined to the terminal 414, an integral structure 314 is formed by the line portion 214 and the terminal 414. The line portion 214 as a pad is rectangular when viewed from the Z direction.
[0042] The characteristic impedance of the line portion 211 is represented by Z1, the characteristic impedance of the line portion 212 is represented by Z2, and the characteristic impedance of the line portion 213 is represented by Z3. In addition, the characteristic impedance of the integral structure 314 formed by the line portion 214 and the terminal 414 is represented by Z4. The characteristic impedance of the line portion 415 is represented by Z5. In addition, the characteristic impedance of only the line portion 214 is represented by Z14. The characteristic impedance Z1 is a first characteristic impedance, the characteristic impedance Z2 is a second characteristic impedance, and the characteristic impedance Z3 is a third characteristic impedance. In addition, the characteristic impedance Z4 is a fourth characteristic impedance, and the characteristic impedance Z5 is a fifth characteristic impedance.
[0043] The line width of the line portion 211 in the Y direction is represented by W1, the line width of the line portion 212 in the Y direction is represented by W2, the line width of the line portion 213 in the Y direction is represented by W3, and the line width of the line portion 214 in the Y direction is represented by W4. In addition, the length of the line portion 211 in the X direction is represented by L1, the length of the line portion 212 in the X direction is represented by L2, the length of the line portion 213 in the X direction is represented by L3, and the length of the line portion 214 in the X direction is represented by L4.
[0044] For comparison, the processing module of Comparative Example 1 will be described. Fig. 10A 1 is a plan view of a portion of a processing module 300X of Comparative Example 1. The processing module 300X includes a connector 400 and a printed wiring board 200X. The connector 400 is the same as that of the first embodiment, and the printed wiring board 200X of Comparative Example 1 is different from the printed wiring board 200 of the first embodiment.
[0045] The printed wiring board 200X includes a signal line 250X having a structure different from that of the signal line 250 of the first embodiment. The other structures of the printed wiring board 200X are the same as those of the printed wiring board 200 of the first embodiment. The signal line 250X includes a line portion 211X and a line portion 214X. The line portion 214X is continuous with the line portion 211X. The line portion 214X is a pad that can be joined to the terminal 414 of the connector 400 via solder. When the line portion 214X is joined to the terminal 414, an integral structure 314X is formed by the line portion 214X and the terminal 414.
[0046] The line width W4X of the line portion 214X is greater than the line width W1X of the line portion 211X. Therefore, the characteristic impedance Z14X of the line portion 214X is lower than the characteristic impedance Z1X of the line portion 211X. As a result, mismatching of characteristic impedance occurs between the line portion 211X and the line portion 214X.
[0047] When the terminal 414 of the connector 400 is engaged with the wire portion 214X, the characteristic impedance Z4X of the integral structure 314X formed by the wire portion 214X and the terminal 414 is lower than the characteristic impedance Z14X of the wire portion 214X without the electrical component installed. This is because the thickness of the integral structure 314X (formed by the wire portion 214X and the terminal 414) in the Z direction is greater than the thickness of the terminal 414 in the Z direction. The thicker integral structure 314X increases the impedance of the integral structure 314X and the ground pattern 262 ( Figure 3B ) between the characteristic impedance Z1X and the characteristic impedance Z4X. Therefore, the difference Z1X-Z4X between the characteristic impedance Z1X and the characteristic impedance Z4X is greater than the difference Z1X-Z14X between the characteristic impedance Z1X and the characteristic impedance Z14X.
[0048] Furthermore, in the first embodiment, Figure 3A The line width W4 of the line portion 214 shown in FIG is greater than the line width W1 of the line portion 211. Therefore, the characteristic impedance Z14 of the line portion 214 is lower than the characteristic impedance Z1 of the line portion 211. When the terminal 414 of the connector 400 is engaged with the line portion 214, the characteristic impedance Z4 of the integral structure 314 formed by the line portion 214 and the terminal 414 is lower than the characteristic impedance Z14 of the line portion 214 without the electrical component installed. Therefore, the difference Z1-Z4 between the characteristic impedance Z1 and the characteristic impedance Z4 is greater than the difference Z1-Z14 between the characteristic impedance Z1 and the characteristic impedance Z14.
[0049] In the first embodiment, the line portion 212 and the line portion 213 are disposed between the line portion 211 and the line portion 214. The line portion 212 and the line portion 213 are disposed to control the disturbance of the voltage of the digital signal caused by the difference Z1-Z4 between the characteristic impedance Z1 and the characteristic impedance Z4. The two line portions 212 and 213 constitute the control line portion 210.
[0050] In the first embodiment, the line portions 211 to 214 have the same thickness in the Z direction. The line width W2 of the line portion 212 is greater than the line width W1 of the line portion 211. Therefore, the characteristic impedance Z2 of the line portion 212 is lower than the characteristic impedance Z1 of the line portion 211. The line width W3 of the line portion 213 is less than the line width W1 of the line portion 211. Therefore, the characteristic impedance Z3 of the line portion 213 is higher than the characteristic impedance Z1 of the line portion 211. Since the control line portion 210 is disposed between the line portion 211 and the line portion 214, the reflection of the signal caused by the mismatch of the characteristic impedance between the integral structure 314 and the line portion 213 can be reduced compared to the configuration of the comparative example in which the integral structure 314X and the line portion 211X are disposed. In addition, since the characteristic impedance of the line portion 212 is lower than the characteristic impedance of the line portion 213 of the control line portion 210, the reflection of the signal caused by the mismatch of the characteristic impedance between the integral structure 314 and the line portion 213 can be further reduced. In this way, the disturbance of the voltage of the digital signal, that is, the difference between the maximum peak value and the minimum peak value is reduced. Therefore, the quality of the digital signal transmitted through the line is improved.
[0051] The characteristic impedance Z2 may be equal to or lower than the characteristic impedance Z4. The characteristic impedance Z2 is preferably higher than the characteristic impedance Z4 for stabilizing the voltage of the digital signal transmitted through the line. The characteristic impedance Z3 may be equal to or higher than the characteristic impedance Z5. The characteristic impedance Z3 is preferably lower than the characteristic impedance Z5 for stabilizing the voltage of the digital signal transmitted through the line.
[0052] Second embodiment
[0053] Next, the configuration of a digital camera as an example of an electronic device of the second embodiment will be described. In the second embodiment, the configuration of the processing module of the digital camera is different from that of the first embodiment. For this reason, the processing module will be described. Figure 4A It is a plan view of a part of the processing module 300A of the second embodiment. Figure 4B is a cross-sectional view of a part of a process module 300A of the second embodiment. In the second embodiment, the same components as those of the first embodiment are given the same symbols, and description thereof will be omitted.
[0054] The processing module 300A as one example of a printed circuit board includes a printed wiring board 200A and a connector 400 mounted on the printed wiring board 200A. As in the first embodiment, the processing module 300A includes Figure 2 The semiconductor device 100 (the semiconductor device 100 is Figure 4A and Figure 4B The semiconductor device 100 is mounted on the printed wiring board 200A.
[0055] As in the first embodiment, the printed wiring board 200A includes four conductor layers 221, 222, 223, and 224 and insulator layers 231, 232, and 233. The conductor layers 221, 222, 223, and 224 are arranged so that each of the insulator layers 231, 232, and 233 is sandwiched between adjacent conductor layers.
[0056] In the second embodiment, as in the first embodiment, the signal line 250 is disposed in the conductor layer 221. In the conductor layer 222, a planar ground pattern 262A is disposed. The ground pattern 262A is different from the ground pattern 262 ( Figure 3B )different.
[0057] The ground pattern 262A includes an opening H1. In a plan view of the printed wiring board 200A, that is, when viewed from the Z direction, the opening H1 overlaps at least a portion of the line portion 214 of the signal line 250. In the second embodiment, the opening H1 overlaps the entire line portion 214 when viewed from the Z direction.
[0058] Since the opening H1 is formed in the ground pattern 262A, the characteristic impedance Z4 of the overall structure 314 of the second embodiment is higher than the characteristic impedance Z4 of the overall structure 314 of the first embodiment. That is, in the second embodiment, the difference between the characteristic impedance Z1 and the characteristic impedance Z4 can be made lower than the difference between the characteristic impedance Z1 and the characteristic impedance Z4 in the first embodiment. Therefore, the quality of the digital signal is further improved.
[0059] Third embodiment
[0060] Next, the configuration of a digital camera as an example of an electronic device of the third embodiment will be described. In the third embodiment, the configuration of the processing module of the digital camera is different from that of the first embodiment. For this reason, the processing module will be described. Figure 5 3 is a perspective view of a process module 300B of the third embodiment. In the third embodiment, the same components as those of the first embodiment are given the same symbols, and description thereof will be omitted.
[0061] Processing module 300B includes semiconductor device 100B, connector 400B and printed wiring board 200B. Connector 400B is an example of an electrical component. Semiconductor device 100B and connector 400B are mounted on printed wiring board 200B. Printed wiring board 200B is a rigid board. Semiconductor device 100B can be a digital signal processor. Connector 400B includes insertion opening 401B, USB cable connector (not shown) or HDMI (registered trademark) cable connector (not shown) is inserted into the insertion opening 401, and USB cable connector or HDMI cable connector is removed from the insertion opening 401.
[0062] The printed wiring board 200B includes a pair of signal lines, each of which is the same as the signal line 250 of the first embodiment. That is, the printed wiring board 200B includes a pair of signal lines 2501 and 2502 that electrically connect the semiconductor device 100B and the connector 400B. Each of the signal lines 2501 and 2502 has the same structure as the signal line 250 of the first embodiment. However, the digital signal transmitted by the signal lines 2501 and 2502 is different from the digital signal transmitted by the signal line 250. That is, the digital signal transmitted by the signal line 250 in the first embodiment is a single-ended signal, while the digital signal transmitted by the pair of signal lines 2501 and 2502 in the second embodiment is a differential signal. The signal lines 2501 and 2502 constituting a pair are arranged adjacent to each other.
[0063] Note that Figure 5 In the printed wiring board 200B, lines other than the signal lines 2501 and 2502 such as a power line, a ground line, and signal lines other than the signal lines 2501 and 2502 are not shown. In addition, a plurality of pairs of signal lines 2501 and 2502 may be arranged. In this case, the plurality of pairs of signal lines 2501 and 2502 may constitute a bus.
[0064] The signal line 2501 includes an end 2511 and an end 2521. The end 2511 is a first end formed in the X direction, and the end 2521 is a second end formed in the X direction and opposite to the end 2511. Similarly, the signal line 2502 includes an end 2512 and an end 2522. The end 2512 is a first end formed in the X direction, and the end 2522 is a second end formed in the X direction and opposite to the end 2512. The semiconductor device 100B includes terminals 1011 and 1012 that output differential signals. The terminal 1011 is connected to the end 2511 of the signal line 2501, and the terminal 1012 is connected to the end 2512 of the signal line 2502. The transmission speed of the digital signal transmitted through the signal lines 2501 and 2502 is 1 Gbps or higher. The material of the signal lines 2501 and 2502 contains a conductive metal material such as copper or gold.
[0065] Fig. 6A It is a plan view of a part of the processing module 300B of the third embodiment. Figure 6B 2 is a cross-sectional view of a portion of a processing module 300B of the third embodiment. As in the first embodiment, a printed wiring board 200B includes four conductor layers 221, 222, 223, and 224 and insulator layers 231, 232, and 233. The conductor layers 221, 222, 223, and 224 are arranged so that each of the insulator layers 231, 232, and 233 is sandwiched between adjacent conductor layers.
[0066] In the third embodiment, a pair of signal lines 2501 and 2502 are disposed in the conductor layer 221. In the conductor layer 222 adjacent to the conductor layer 221 via the insulator layer 231, a planar ground pattern 262 is disposed. In the conductor layer 223, a planar ground pattern 263 is disposed. In the conductor layer 224, a signal line 264 other than the signal lines 2501 and 2502 is disposed.
[0067] The connector 400B is a surface mounted electrical component and includes a pair of terminals 4141 and 4142 as signal terminals. The terminals 4141 and 4142 are pin-shaped. The terminal 4141 is connected to the end 2521 of the signal line 2501, and the terminal 4142 is connected to the end 2522 of the signal line 2502. The pair of terminals 4141 and 4142 receive differential signals. The transmission direction in which the differential signal is transmitted through the pair of signal lines 2501 and 2502 is defined as the X1 direction.
[0068] The signal line 2501 includes line portions 2111, 2121, 2131, and 2141 formed continuously in the X direction. The signal line 2502 includes line portions 2112, 2122, 2132, and 2142 formed continuously in the X direction. The line portion 2111 and the line portion 2112 are adjacent to each other in the Y direction and are spaced apart from each other. The line portion 2121 and the line portion 2122 are adjacent to each other in the Y direction and are spaced apart from each other. The line portion 2131 and the line portion 2132 are adjacent to each other in the Y direction and are spaced apart from each other. The line portion 2141 and the line portion 2142 are adjacent to each other in the Y direction and are spaced apart from each other. The connector 400B includes a line portion 4151 continuous with the terminal 4141 and a line portion 4152 continuous with the terminal 4142.
[0069] Next, the signal line 2501 will be described. The line portion 2111 is a main line; and is the longest in the X direction among the line portions 2111 to 2141. The line portion 2141 is a pad that can be joined to the terminal 4141 of the connector 4001 via solder. When the line portion 2141 is joined to the terminal 4141, an integral structure 3141 is formed by the line portion 2141 and the terminal 4141. The line portion 2141 as a pad is rectangular when viewed from the Z direction. Since the signal line 2502 has the same structure as that of the signal line 2501, its description will be omitted.
[0070] The differential impedance of a pair of line portions 2111 and 2112 is represented by Z1. The differential impedance of a pair of line portions 2121 and 2122 is represented by Z2. The differential impedance of a pair of line portions 2131 and 2132 is represented by Z3. The differential impedance of a pair of line portions 4141 and 4142 is represented by Z4. The differential impedance of a pair of line portions 4151 and 4152 is represented by Z5. In addition, the differential impedance of a pair of line portions 2141 and 2142 to which the terminal is not engaged is represented by Z14. The differential impedance Z1 corresponds to the first characteristic impedance, the differential impedance Z2 corresponds to the second characteristic impedance, and the differential impedance Z3 corresponds to the third characteristic impedance. In addition, the differential impedance Z4 corresponds to the fourth characteristic impedance, and the differential impedance Z5 corresponds to the fifth characteristic impedance.
[0071] The line width of each of the line portions 2111 and 2112 in the Y direction is represented by W1, the line width of each of the line portions 2121 and 2122 in the Y direction is represented by W2, the line width of each of the line portions 2131 and 2132 in the Y direction is represented by W3, and the line width of each of the line portions 2141 and 2142 in the Y direction is represented by W4. In addition, the length of each of the line portions 2111 and 2112 in the X direction is represented by L1, the length of each of the line portions 2121 and 2122 in the X direction is represented by L2, the length of each of the line portions 2131 and 2132 in the X direction is represented by L3, and the length of each of the line portions 2141 and 2142 in the X direction is represented by L4.
[0072] The line width W4 is greater than the line width W1. Therefore, the differential impedance Z14 is lower than the differential impedance Z1. In addition, when the terminals 4141 and 4142 of the connector 400B are engaged with the line portions 2141 and 2142, the differential impedance Z4 of the overall structures 3141 and 3142 is lower than the differential impedance Z14 of only the line portions 2141 and 2142. Therefore, the difference Z1-Z4 between the differential impedance Z1 and the differential impedance Z4 is greater than the difference Z1-Z14 between the differential impedance Z1 and the differential impedance Z14.
[0073] As in the first embodiment, the control line portion 2101 formed of the line portion 2121 and the line portion 2131 is disposed between the line portion 2111 and the line portion 2141. Similarly, the control line portion 2102 formed of the line portion 2122 and the line portion 2132 is disposed between the line portion 2112 and the line portion 2142. The pair of control line portions 2101 and 2102 are disposed for controlling the disturbance of the voltage of the differential signal caused by the difference Z1-Z4 between the differential impedance Z1 and the differential impedance Z4.
[0074] In the third embodiment, the line portions 2111 to 2141 and the line portions 2112 to 2142 have the same thickness in the Z direction. The line width W2 is greater than the line width W1. Therefore, the differential impedance Z2 is lower than the differential impedance Z1. The line width W3 is less than the line width W1. Therefore, the differential impedance Z3 is higher than the differential impedance Z1. In this structure, the reduction of the differential impedance in the overall structures 3141 and 3142 is suppressed by the line portions 2131 and 2132, the differential impedance of which is higher than the differential impedance of the overall structures 3141 and 3142. In addition, the increase of the differential impedance of the line portions 2131 and 2132 is suppressed by the line portions 2121 and 2122, the differential impedance of which is lower than the differential impedance of the line portions 2131 and 2132. In this way, the reflection of the signal caused by the integral structures 3141 and 3142 is reduced, so that the disturbance of the voltage of the differential signal, that is, the difference between the maximum peak and the minimum peak, is reduced. Therefore, the quality of the differential signal transmitted through the line is improved.
[0075] The differential impedance Z2 may be equal to or lower than the differential impedance Z4. The differential impedance Z2 is preferably higher than the differential impedance Z4 to stabilize the voltage of the differential signal transmitted through the line. The differential impedance Z3 may be equal to or higher than the differential impedance Z5. The differential impedance Z3 is preferably lower than the differential impedance Z5 to stabilize the voltage of the differential signal transmitted through the line.
[0076] Fourth embodiment
[0077] Next, the configuration of a digital camera as an example of an electronic device of the fourth embodiment will be described. In the fourth embodiment, the configuration of the processing module of the digital camera is different from that of the third embodiment. For this reason, the processing module will be described. Fig. 7A It is a plan view of a part of the processing module 300C of the fourth embodiment. Figure 7B It is a cross-sectional view of a part of a process module 300C of the fourth embodiment. In the fourth embodiment, the same components as those of the third embodiment are given the same symbols, and the description thereof will be omitted.
[0078] The processing module 300C as one example of a printed circuit board includes a printed wiring board 200C and a connector 400B mounted on the printed wiring board 200C. As in the third embodiment, the processing module 300C includes Figure 5 The semiconductor device 100B (the semiconductor device 100B is Fig. 7A and Figure 7B The semiconductor device 100B is mounted on the printed wiring board 200C.
[0079] As in the third embodiment, the printed wiring board 200C includes four conductor layers 221, 222, 223, and 224 and insulator layers 231, 232, and 233. The conductor layers 221, 222, 223, and 224 are arranged such that each of the insulator layers 231, 232, and 233 is sandwiched between adjacent conductor layers.
[0080] In the fourth embodiment, as in the third embodiment, the signal lines 2501 and 2502 are arranged in the conductor layer 221. In the conductor layer 222, a plane ground pattern 262C is arranged. The ground pattern 262C is different from the ground pattern 262 ( Figure 6B ) of the third embodiment.
[0081] The ground pattern 262C includes an opening H2. In the plan view of the printed wiring board 200C, that is, when viewed from the Z direction, the opening H2 overlaps at least a part of the line portions 2141 and 2142. In the fourth embodiment, when viewed from the Z direction, the opening H2 overlaps the entire pair of line portions 2141 and 2142.
[0082] Since the opening H2 is formed in the ground pattern 262C, the differential impedance Z4 of the pair of overall structures 3141 and 3142 in the fourth embodiment is higher than the differential impedance Z4 of the pair of overall structures 3141 and 3142 in the third embodiment. That is, in the fourth embodiment, the difference between the differential impedance Z1 and the differential impedance Z4 can be made smaller than the difference between the differential impedance Z1 and the differential impedance Z4 in the third embodiment. Therefore, the quality of the differential signal is further improved.
[0083] Example
[0084] Hereinafter, the lines for single-ended signals will be described in Example 1, Example 2, and Comparative Example 1, and the lines for differential signals will be described in Example 3, Example 4, and Comparative Example 2.
[0085] Example 1
[0086] Reference will be made to Figure 3A and Figure 3B to describe Example 1. Example 1 indicates the specific numerical values of the first embodiment. In order to achieve the characteristic impedance relationship of Z4 < Z2 < Z1 < Z3 < Z5, a printed circuit board is designed by using a simulation device and the following parameters, and the characteristic impedance is calculated. To calculate the characteristic impedance, HyperLynx manufactured by Mentor is used.
[0087] The thickness of the conductor layer 221 is set to 37 μm. The thickness of the conductor layer 222 is set to 35 μm. The thickness of the conductor layer 223 is set to 35 μm. The thickness of the conductor layer 224 is set to 37 μm. The thickness of the insulator layer 231 is set to 100 μm. The thickness of the insulator layer 232 is set to 1200 μm. The thickness of the insulator layer 233 is set to 100 μm. The relative dielectric constant of the insulator layers 231, 232 and 233 is set to 4.3, and the dielectric tangent of the insulator layers 231, 232 and 233 is set to 0.02. In addition, the surfaces of the conductor layers 221 and 224 are covered with a solder resist (not shown). The thickness of the solder resist (not shown) in the Z direction is set to 20 μm. The relative dielectric constant of the solder resist (not shown) was set to 3.0, and the dielectric tangent of the solder resist was set to 0.02.
[0088] The line width W1 of the line portion 211 in the Y direction is set to 150 μm, and the length L1 of the line portion 211 in the X direction is set to 28.6 mm. The line width W2 of the line portion 212 in the Y direction is set to 280 μm, and the length L2 of the line portion 212 in the X direction is set to 0.4 mm. The line width W3 of the line portion 213 in the Y direction is set to 85 μm, and the length L3 of the line portion 213 in the X direction is set to 1.0 mm. The line width W4 of the line portion 214 in the Y direction is set to 250 μm, and the length L4 of the line portion 214 in the X direction is set to 2.0 mm.
[0089] The width of the terminal 414 joined to the line portion 214 in the Y direction is set to 250 μm, and the length of the terminal 414 in the X direction is set to 2.0 mm. The thickness of the terminal 414 in the Z direction is set to 200 μm. The length of the line portion 415 in the X direction is set to 2.0 mm. In addition, the signal output portion of the line portion 415 is terminated by a 50Ω resistor (not shown).
[0090] After performing calculations under the above conditions, the following characteristic impedances Z1 to Z4 are obtained. Characteristic impedance Z1 is 50.8Ω. Characteristic impedance Z2 is 36.4Ω. Characteristic impedance Z3 is 64.2Ω. Characteristic impedance Z4 is 36.1Ω. Characteristic impedance Z5 is 65Ω. Although the width of the integral structure 314 formed by the line portion 214 and the terminal 414 is smaller than the width of the line portion 212, the characteristic impedance Z4 is lower than the characteristic impedance Z2. The characteristic impedance Z14 is 40.3Ω.
[0091] In addition, a time domain reflectometry (TDR) analysis simulation was performed on the structure of Example 1. Fig. 8A is a graph showing simulation results obtained in the example. Fig. 8AIn the , the vertical axis represents characteristic impedance (Ω), and the horizontal axis represents time (seconds). TDR analysis can determine the characteristic impedance of a signal line at a location separated by a certain distance from the signal source. In addition, TDR analysis can evaluate the quality of the voltage waveform of a digital signal.
[0092] exist Fig. 8A , a waveform 1001 indicates a TDR analysis result obtained in Example 1. For the TDR analysis, HSPICE manufactured by Synopsys, Inc. was used. In addition, a pulse signal was used as a digital signal to be applied to a signal line.
[0093] Figure 8B is a diagram showing a pulse signal applied from a signal source to one end of a main line in an example. In Example 1, the main line is line portion 211. One end of the main line is Figure 2 The end 251 shown in FIG. Figure 2 The voltage amplitude of the pulse signal applied to the line portion 211 is determined by V in The rise time of the pulse signal is represented by tr. The rise time tr is related to the voltage amplitude V in The voltage amplitude V in The internal impedance of the signal source is 50Ω.
[0094] The analysis results in Example 1 will be described by comparing the TDR analysis results and the calculation results of the characteristic impedance. In Example 1, a pulse signal is applied to one end of the line portion 211. In the TDR analysis results, the characteristic impedance Z1 of the line portion 211 is 52Ω.
[0095] The characteristic impedance Z2 of the line portion 212 obtained by calculation is 36.4Ω, which is lower than the characteristic impedance Z1 of the line portion 211. The characteristic impedance Z3 of the line portion 213 is 64.2Ω, which is higher than the characteristic impedance Z1. The characteristic impedance Z4 of the overall structure 314 is 36.1Ω, which is lower than the characteristic impedance Z1.
[0096] The length L1 of the line portion 211 in the X direction is set to 28.6 mm. The relative dielectric constant of the solder resist formed on the line portion 211 is set to 3, and the relative dielectric constant of the insulator formed on the line portion 211 is set to 4.3. The propagation velocity of the electromagnetic wave is calculated using the relative dielectric constant of the solder resist and the relative dielectric constant of the insulator. To calculate the propagation velocity of the electromagnetic wave, HyperLynx manufactured by Mentor is used. The propagation velocity of the electromagnetic wave is determined to be approximately 1.654×10 8 m / s, the propagation speed is the speed of light (≈3.0×10 8Therefore, the time for the signal to propagate through the line portion 211 is 172.9 ps.
[0097] The length L2 of the line portion 212 in the X direction is set to 0.4 mm. The relative dielectric constant of the solder resist formed on the line portion 212 is set to 3, and the relative dielectric constant of the insulator formed on the line portion 212 is set to 4.3. The propagation velocity of the electromagnetic wave is calculated using the relative dielectric constant of the solder resist and the relative dielectric constant of the insulator. To calculate the propagation velocity of the electromagnetic wave, HyperLynx manufactured by Mentor is used. The propagation velocity of the electromagnetic wave is determined to be approximately 1.617×10 8 m / s, the propagation speed is the speed of light (≈3.0×10 8 Therefore, the time for the signal to propagate through the line portion 212 is 2.5 ps.
[0098] The characteristic impedance decreases to 36.4Ω within the rise time tr (i.e., 35ps) of the pulse signal. However, in the line portion 212, the signal wave goes back and forth within 5.0ps (=2.5×2). Therefore, the effective change from the internal impedance (50Ω) of the signal source to the characteristic impedance Z2 (36.4Ω) of the line portion 212 is -1.9Ω (≈(36.4-50)×(5.0 / 35)).
[0099] The length L3 of the line portion 213 in the X direction is set to 1.0 mm. The relative dielectric constant of the solder resist formed on the line portion 213 is set to 3, and the relative dielectric constant of the insulator formed on the line portion 213 is set to 4.3. The propagation velocity of the electromagnetic wave is calculated using the relative dielectric constant of the solder resist and the relative dielectric constant of the insulator. To calculate the propagation velocity of the electromagnetic wave, HyperLynx manufactured by Mentor is used. The propagation velocity of the electromagnetic wave is determined to be approximately 1.681×10 8 m / s, the propagation speed is the speed of light (≈3.0×10 8 Therefore, the time for the signal to propagate through the line portion 213 is 5.9 ps.
[0100] The characteristic impedance increases to 64.2Ω within the rise time tr (i.e., 35ps) of the pulse signal. However, in the line portion 213, the signal wave goes back and forth within 11.8ps (=5.9×2). Therefore, the effective change from the internal impedance (50Ω) of the signal source to the characteristic impedance Z3 (64.2Ω) of the line portion 213 is 4.8Ω (≈(64.2-50)×(11.8 / 35)).
[0101] The length of the overall structure 314 in the X direction is set to 2.0 mm. Therefore, the time for the signal to propagate through the overall structure 314 is 11.3 ps. The characteristic impedance is reduced to 36.1 Ω within the rise time tr (i.e., 35 ps) of the pulse signal. However, in the overall structure 314, the signal wave goes back and forth within 22.6 ps (= 11.3 × 2). Therefore, the effective change from the internal impedance (50 Ω) of the signal source to the characteristic impedance Z4 (36.1 Ω) of the overall structure 314 is -9.0 Ω (≈ (36.1-50) × (22.6 / 35)).
[0102] In Example 1, the line portion 211, the line portion 212, the line portion 213, and the overall structure 314 are formed successively in sequence, and the characteristic impedance changes at the line portion 211, the line portion 212, the line portion 213, and the overall structure 314. Therefore, in the TDR analysis, the characteristic impedance of each line portion is obtained by adding the effective change in the line portion to the characteristic impedance of the previous line portion. If the characteristic impedance Z1 of the line portion 211 obtained in the TDR analysis is added to the effective changes in the three line portions (obtained by calculation), the resulting value is calculated to be 45.9Ω (=52-1.9+4.8-9.0), which is lower than the characteristic impedance Z1. In the TDR analysis, this value is 46.9Ω.
[0103] exist Fig. 8A In the waveform 1001 shown, twice the time for the signal wave to propagate through the line portion 211 is 345.8 ps; and the time period from 0 to 345.8 ps corresponds to the line portion 211. In addition, twice the time for the signal wave to propagate through the line portion 212 is 5 ps; and the time period from 345.8 to 350.8 ps (=345.8+5) corresponds to the line portion 212. In addition, twice the time for the signal wave to propagate through the line portion 213 is 10.8 ps; and the time period from 350.8 to 361.6 ps (=350.8+10.8) corresponds to the line portion 213. Finally, twice the time for the signal wave to propagate through the overall structure 314 is 22.6 ps; and the time period from 361.6 to 384.2 ps (=361.6+22.6) corresponds to the overall structure 314. The characteristic impedance corresponding to 384.2 ps is 46.9Ω.
[0104] Characteristic impedance Z3 is higher than characteristic impedance Z1. Preferably, characteristic impedance Z3 is lower than characteristic impedance Z5, which is the highest among characteristic impedance Z2, characteristic impedance Z4 and characteristic impedance Z5. The reason is to make the variation range of characteristic impedance of line portion 213 obtained by TDR analysis lower than the highest characteristic impedance Z5 in the circuit.
[0105] In addition, it is preferable that the length L3 of the line portion 213 in the X direction satisfies the following expression (1),
[0106]
[0107] Here, tr is the rise time of the digital signal, vo is the propagation rate of the digital signal, Z1 is the characteristic impedance of the line portion 211 , and Z5 is the characteristic impedance of the line portion 415 .
[0108] The characteristic impedance of the line portion 211 as the main line is controlled to be about 50Ω. In expression (1), the ratio is calculated by dividing the amount of change in the characteristic impedance allowed in the line portion 213 (Z1×0.10) by the difference Z5-Z1 between the characteristic impedance of the line portion 211 and the characteristic impedance of the line portion 415. Then, the ratio is multiplied by half the rise time of the pulse signal and the propagation rate. Through this calculation, the length of the line portion 213 required to maintain the quality of the voltage waveform of the digital signal is determined. Note that the line portion 213 is a structure for suppressing the disturbance of the voltage waveform of the digital signal caused in the overall structure 314, and the characteristic impedance of the line portion 213 is determined so that the line portion 213 does not generate any new disturbance.
[0109] If the line has a length in which the round trip time of the electromagnetic wave is long enough to be longer than the rise time of the pulse signal, the characteristic impedance of the line can be measured in the TDR analysis after the rise time of the pulse signal has passed. In contrast, if the line has a length in which the round trip time of the electromagnetic wave is shorter than the rise time of the pulse signal, that is, if the line has a length shorter than the time resolution of the pulse signal, the characteristic impedance of the line does not change to the value obtained in the line having a sufficient length. If Z3=Z5, then the time in which the characteristic impedance changes by the amount of Z5-Z1 is equal to the rise time of the pulse signal. If the characteristic impedance Z3 is limited to Z1×0.10, then expression (1) is obtained.
[0110] In addition, in order to maintain the quality of the voltage waveform of the digital signal, the board is generally manufactured so that the amount of change in the characteristic impedance allowed in the line portion 213 is about ±5 to ±15% of the characteristic impedance of the main line. That is, considering the amount of change in the characteristic impedance, it is preferred that the length L3 of the line portion 213 in the X direction satisfies the following expression (2).
[0111]
[0112] The characteristic impedance Z2 is lower than the characteristic impedance Z1. Preferably, the characteristic impedance Z2 is higher than the characteristic impedance Z4, which is the lowest among the characteristic impedance Z3, the characteristic impedance Z4 and the characteristic impedance Z5. The reason for this is to make the variation range of the characteristic impedance of the line portion 212 obtained by TDR analysis relative to the characteristic impedance of the line portion 211 lower than the characteristic impedance of the overall structure 314.
[0113] In addition, it is preferable that the length L2 of the line portion 212 in the X direction satisfies the following expression (3).
[0114]
[0115] The characteristic impedance of the line portion 211 as the main line is controlled to be about 50Ω. In expression (3), the ratio is calculated by dividing the amount of change in the characteristic impedance allowed in the line portion 212 (Z1×0.10) by the difference Z1-Z4 between the characteristic impedance of the line portion 211 and the characteristic impedance of the overall structure 314. Then, the ratio is multiplied by half the rise time of the pulse signal and the propagation rate. Finally, the length of the line portion 212 is determined by multiplying the resulting ratio by 0.5. Note that the line portion 212 is a structure for suppressing disturbances of the voltage waveform of the digital signal caused in the overall structure 314, and the characteristic impedance of the line portion 212 is determined so that the line portion 212 does not generate any new disturbances.
[0116] Since the line portion 212 is disposed in front of the line portion 213, the increase in characteristic impedance caused by the line portion 213 can be suppressed. However, if the characteristic impedance is excessively reduced by the line portion 212 disposed in front of the line portion 213, the difference between the characteristic impedance Z1 of the line portion 211 reduced by the line portion 213 and the characteristic impedance Z4 of the overall structure 314 increases again. Therefore, in Example 1, the length of the line portion 212 is corrected. In the above expression (3), the correction coefficient is 0.5.
[0117] In addition, in order to maintain the quality of the voltage waveform of the digital signal, the board is manufactured so that the amount of change in the characteristic impedance allowed in the line portion 212 is about ±5, ±10, or ±15% of the characteristic impedance of the main line. That is, considering the amount of change in the characteristic impedance, it is preferred that the length L2 of the line portion 212 in the X direction satisfies the following expression (4).
[0118]
[0119] It is preferable that the length L4 of the line portion 214 in the X direction satisfies the following expression (5).
[0120]
[0121] If the line has a length in which the round trip time of the electromagnetic wave is sufficiently longer than the rise time of the pulse signal, then the characteristic impedance of the line that is substantially the same as the characteristic impedance of a line having an infinite length can be measured in the TDR analysis after the rise time of the pulse signal has passed. In contrast, if the line has a length in which the round trip time of the electromagnetic wave is shorter than the rise time of the pulse signal, that is, if the line has a length shorter than the time resolution of the pulse signal, then the characteristic impedance of the line does not change to the value obtained in a line having a sufficient length. If the line portion 214 has a length in which the round trip time of the electromagnetic wave is longer than the rise time of the pulse signal, then the effect of suppressing the disturbance of the voltage waveform of the digital signal caused in the overall structure 314 by the arrangement of the line portions 212 and 213 is not produced. Therefore, the length L4 is determined as described above.
[0122] Example 2
[0123] Will refer to Figure 4A and Figure 4B Example 2 is described. Example 2 indicates specific numerical values of the second embodiment. First, the characteristic impedance is calculated. To calculate the characteristic impedance, HyperLynx manufactured by Mentor is used as in Example 1. Hereinafter, the numerical value of each component used to calculate the characteristic impedance will be described. The structure of each layer of the printed wiring board 200A of Example 2 is the same as the structure of each layer of the printed wiring board of Example 1. In the following description, the features of Example 2 that are different from those of Example 1 will be described.
[0124] The line width W1 of the line portion 211 in the Y direction is set to 150 μm, and the length L1 of the line portion 211 in the X direction is set to 28.2 mm. The line width W2 of the line portion 212 in the Y direction is set to 210 μm, and the length L2 of the line portion 212 in the X direction is set to 0.8 mm. The line width W3 of the line portion 213 in the Y direction is set to 85 μm, and the length L3 of the line portion 213 in the X direction is set to 1.0 mm. The line width W4 of the line portion 214 in the Y direction is set to 250 μm, and the length L4 of the line portion 214 in the X direction is set to 2.0 mm. The width of the opening H1 in the Y direction is set to 250 μm, which is equal to the line width W4, and the length of the opening H1 in the X direction is set to 2.0 mm, which is equal to the length L4. That is, the area of the line portion 214 is equal to the area of the opening H1.
[0125] The width of the terminal 414 joined to the line portion 214 in the Y direction is set to 250 μm, and the length of the terminal 414 in the X direction is set to 2.0 mm. The thickness of the terminal 414 in the Z direction is set to 200 μm. The length of the line portion 415 in the X direction is set to 2.0 mm. In addition, the signal output portion of the line portion 415 is terminated by a 50Ω resistor (not shown).
[0126] After performing the calculation under the above conditions, the following characteristic impedances Z1 to Z4 are obtained. The characteristic impedance Z1 is 50.8 Ω. The characteristic impedance Z2 is 42.9 Ω. The characteristic impedance Z3 is 64.2 Ω. The characteristic impedance Z4 is 42.6 Ω. The characteristic impedance Z5 is 65 Ω. Therefore, the relationship of the characteristic impedances is Z4 < Z2 < Z1 < Z3 < Z5. The width of the overall structure 314 formed by the line part 214 and the terminal 414 is greater than the width of the line part 212, and the characteristic impedance Z4 is lower than the characteristic impedance Z2. The characteristic impedance Z14 is 48.8 Ω.
[0127] Note that in Example 2, although the case where the line width W3 is set to 85 μm so that the characteristic impedance Z3 of the line part 213 becomes 64.2 Ω has been described, the present disclosure is not limited thereto. For example, a slit may be formed in the ground pattern 262A facing the line part 213. In this case, if the line width W3 is 100 μm and the slit is formed in the ground pattern 262A at the position facing the line part 213, then the characteristic impedance Z3 becomes 62.8 Ω.
[0128] In addition, a TDR analysis simulation is performed on the structure of Example 2. In Fig. 8A Waveform 1002 indicates the TDR analysis result obtained in Example 2. For the TDR analysis, as in Example 1, HSPICE manufactured by Synopsys is used. In addition, a pulse signal is used as the digital signal to be applied to the signal line. The pulse signal is the same as the pulse signal used in Example 1.
[0129] The analysis result in Example 2 will be described by comparing the TDR analysis result and the calculation result of the characteristic impedance. In Example 2, a pulse signal is applied to one end of the line part 211. In the TDR analysis result in Example 2, the characteristic impedance Z1 of the line part 211 is 52 Ω.
[0130] The characteristic impedance Z2 of the line part 212 obtained by calculation is 42.9 Ω, and this characteristic impedance Z2 is lower than the characteristic impedance Z1 of the line part 211. The characteristic impedance Z3 of the line part 213 is 64.2 Ω, and this characteristic impedance Z3 is higher than the characteristic impedance Z1 of the line part 211. The characteristic impedance Z4 of the overall structure 314 is 42.6 Ω, and this characteristic impedance Z4 is lower than the characteristic impedance Z1. When viewed from the Z direction, the line part 214 overlaps with the opening H1 of the ground pattern 262A. Therefore, the characteristic impedance Z4 is higher than the characteristic impedance Z4 of Example 1.
[0131] The length L1 of the line portion 211 in the X direction is set to 28.2 mm. The relative dielectric constant of the solder resist formed on the line portion 211 is set to 3, and the relative dielectric constant of the insulator formed on the line portion 211 is set to 4.3. The propagation velocity of the electromagnetic wave is calculated using the relative dielectric constant of the solder resist and the relative dielectric constant of the insulator. To calculate the propagation velocity of the electromagnetic wave, HyperLynx manufactured by Mentor is used. The propagation velocity of the electromagnetic wave is determined to be approximately 1.654×10 8 m / s, the propagation speed is the speed of light (≈3.0×10 8 Therefore, the time for the signal to propagate through the line portion 211 is 170.4 ps.
[0132] The length L2 of the line portion 212 in the X direction is set to 0.8 mm. The relative dielectric constant of the solder resist formed on the line portion 212 is set to 3, and the relative dielectric constant of the insulator formed on the line portion 212 is set to 4.3. The propagation rate of the electromagnetic wave is calculated using the relative dielectric constant of the solder resist and the relative dielectric constant of the insulator. To calculate the propagation rate of the electromagnetic wave, HyperLynx manufactured by Mentor is used. The propagation rate of the electromagnetic wave is determined to be approximately 1.635×10 8 m / s, the propagation speed is the speed of light (≈3.0×10 8 m / s). Therefore, the time for the signal to propagate through the line portion 212 is 4.9 ps. The characteristic impedance is reduced to 42.9Ω within the rise time tr of the pulse signal (i.e., 35 ps). However, in the line portion 212, the signal wave goes back and forth within 9.8 ps (=4.9×2). Therefore, the effective change from the internal impedance of the signal source (50Ω) to the characteristic impedance Z2 (42.9Ω) of the line portion 212 is -2.0Ω (≈(42.9-50)×(9.8 / 35)).
[0133] The length L3 of the line portion 213 in the X direction is set to 1.0 mm. The relative dielectric constant of the solder resist formed on the line portion 213 is set to 3, and the relative dielectric constant of the insulator formed on the line portion 213 is set to 4.3. The propagation velocity of the electromagnetic wave is calculated using the relative dielectric constant of the solder resist and the relative dielectric constant of the insulator. To calculate the propagation velocity of the electromagnetic wave, HyperLynx manufactured by Mentor is used. The propagation velocity of the electromagnetic wave is determined to be approximately 1.681×10 8 m / s, the propagation speed is the speed of light (≈3.0×10 8m / s). Therefore, the time for the signal to propagate through the line portion 213 is 6.0 ps. The characteristic impedance increases to 64.2Ω within the rise time tr of the pulse signal (i.e., 35 ps). However, in the line portion 213, the signal wave goes back and forth within 12 ps (=6.0×2). Therefore, the effective change from the internal impedance of the signal source (50Ω) to the characteristic impedance Z3 (64.2Ω) of the line portion 213 is 4.9Ω (≈(64.2-50)×(12 / 35)).
[0134] The length of the overall structure 314 in the X direction is set to 2.0 mm. The relative dielectric constant of the air on the overall structure 314 is set to 1, and the relative dielectric constant of the insulator formed on the overall structure 314 is set to 4.3. The propagation velocity of the electromagnetic wave is calculated using the relative dielectric constant of the air and the relative dielectric constant of the insulator. To calculate the propagation velocity of the electromagnetic wave, HyperLynx manufactured by Mentor is used. The propagation velocity of the electromagnetic wave is determined to be approximately 1.835×10 8 m / s, the propagation speed is the speed of light (≈3.0×10 8 m / s). Therefore, the time for the signal to propagate through the overall structure 314 is 10.9 ps. The characteristic impedance is reduced to 42.6 Ω within the rise time tr of the pulse signal (i.e., 35 ps). However, in the overall structure 314, the signal wave goes back and forth within 21.8 ps (=10.9×2). Therefore, the effective change from the internal impedance of the signal source (50 Ω) to the characteristic impedance Z4 (36.1 Ω) of the overall structure 314 is -4.6 Ω (≈(42.6-50)×(21.8 / 35)).
[0135] In Example 2, the line portion 211, the line portion 212, the line portion 213, and the overall structure 314 are formed successively in sequence, and the characteristic impedance changes at the line portion 211, the line portion 212, the line portion 213, and the overall structure 314. Therefore, in the TDR analysis, the characteristic impedance of each line portion is obtained by adding the effective change in the line portion to the characteristic impedance of the previous line portion. If the characteristic impedance Z1 of the line portion 211 obtained in the TDR analysis is added to the effective changes in the three line portions (obtained by calculation), the resulting value is calculated to be 50.3Ω (=52-2.0+4.9-4.6), which is lower than the characteristic impedance Z1. In the TDR analysis, this value is 51.2Ω.
[0136] exist Fig. 8AIn the waveform 1002 shown, twice the time for the signal wave to propagate through the line portion 211 is 340.8 ps; and the time period from 0 to 340.8 ps corresponds to the line portion 211. In addition, twice the time for the signal wave to propagate through the line portion 212 is 9.8 ps; and the time period from 340.8 to 350.6 ps (=340.8+9.8) corresponds to the line portion 212. In addition, twice the time for the signal wave to propagate through the line portion 213 is 12 ps; and the time period from 350.6 to 362.6 ps (=350.6+12) corresponds to the line portion 213. Finally, twice the time for the signal wave to propagate through the overall structure 314 is 21.8 ps; and the time period from 362.6 to 384.4 ps (=362.6+21.8) corresponds to the overall structure 314. The characteristic impedance corresponding to 384.4 ps is 51.2Ω.
[0137] Comparative Example 1
[0138] The specific numerical values of Comparative Example 1 will be described. First, the characteristic impedance is calculated. In order to calculate the characteristic impedance, HyperLynx manufactured by Mentor is used. Hereinafter, the numerical value of each component used to calculate the characteristic impedance will be described. The structure of each layer of the printed wiring board 200X of Comparative Example 1 is the same as the structure of each layer of the printed wiring board of Example 1. In the following description, the features of Comparative Example 1 that are different from those of Example 1 will be described.
[0139] The line width W1X of the line portion 211X in the Y direction is set to 150 μm, and the length L1X of the line portion 211X in the X direction is set to 30 mm. The line width W4X of the line portion 214X (which is a pad) in the Y direction is set to 250 μm, and the length L4X of the line portion 214X in the X direction is set to 2.0 mm. The other values are the same as those of Example 1. The characteristic impedance Z1X of the line portion 211X is 50.8Ω. The characteristic impedance Z4X of the line portion 214X is 36.1Ω.
[0140] In addition, TDR analysis simulation is performed on the structure of Comparative Example 1. Fig. 8A , waveform 1003 indicates the TDR analysis result in Comparative Example 1. For TDR analysis, HSPICE manufactured by Synopsys was used as in Examples 1 and 2. In addition, a pulse signal was used as a digital signal to be applied to the signal line. The pulse signal was the same as the pulse signal used in Examples 1 and 2.
[0141] The analysis result in Comparative Example 1 will be described. In Comparative Example 1, a pulse signal is applied to one end of the line portion 211X. In the TDR analysis result in Comparative Example 1, the characteristic impedance Z1X of the line portion 211X is 52Ω. The characteristic impedance Z4X of the overall structure 314X is lower than the characteristic impedance Z1X of the line portion 211X. Therefore, in the TDR analysis result, Fig. 8A As shown in waveform 1003, when the pulse signal reaches the overall structure 314X, the characteristic impedance is reduced to about 45Ω.
[0142] The length L1X of the line portion 211X in the X direction is set to 30 mm. The relative dielectric constant of the solder resist formed on the line portion 211X is set to 3, and the relative dielectric constant of the insulator formed on the line portion 211X is set to 4.3. The propagation rate of the electromagnetic wave is calculated using the relative dielectric constant of the solder resist and the relative dielectric constant of the insulator. To calculate the propagation rate of the electromagnetic wave, HyperLynx manufactured by Mentor is used. The propagation rate of the electromagnetic wave is determined to be approximately 1.654×10 8 m / s, the propagation speed is the speed of light (≈3.0×10 8 Therefore, the time it takes for the signal to propagate through the line portion 211X is 181.3 ps.
[0143] In addition, the length of the overall structure 314X in the X direction is set to 2.0 mm. The relative dielectric constant of the air on the overall structure 314X is set to 1, and the relative dielectric constant of the insulator formed on the overall structure 314X is set to 4.3. The propagation rate of the electromagnetic wave is calculated using the relative dielectric constant of the air and the relative dielectric constant of the insulator. In order to calculate the propagation rate of the electromagnetic wave, HyperLynx manufactured by Mentor is used. The propagation rate of the electromagnetic wave is determined to be approximately 1.772×10 8 m / s, the propagation speed is the speed of light (≈3.0×10 8 m / s). Therefore, the time for the signal to propagate through the overall structure 314X is 11.3 ps.
[0144] Therefore, the propagation time (11.3 ps) of the signal corresponding to the length of the overall structure 314X is shorter than the resolution, which is the rise time tr (i.e., 35 ps) of the signal applied to the overall structure 314X. Therefore, the signal reaches the line portion 415 of the connector 400 before the characteristic impedance obtained by TDR analysis is reduced to the actual characteristic impedance Z4X (36.1Ω).
[0145] In the following, the specific calculation and TDR analysis results will be described. The characteristic impedance is reduced to 36.1Ω within the rise time tr (i.e., 35ps) of the pulse signal. However, in the overall structure 314X, the signal wave goes back and forth within 22.6ps (=11.3×2). Therefore, the effective change from the internal impedance of the signal source (50Ω) to the characteristic impedance (36.1Ω) of the overall structure 314X is -9.0Ω (≈(36.1-50)×(22.6 / 35)). Therefore, in the calculation, the characteristic impedance is reduced to 43.0Ω (=52-9.0). In the TDR analysis, the characteristic impedance is reduced to 45.4Ω.
[0146] exist Fig. 8A In the waveform 1003 shown, twice the time for the signal wave to propagate through the line portion 211X is 362.6 ps; and the time period from 0 to 362.6 ps corresponds to the line portion 211X. In addition, twice the time for the signal wave to propagate through the overall structure 314X is 22.6 ps; and the time period from 362.6 to 385.2 ps (= 362.6 + 22.6) corresponds to the overall structure 314X. The characteristic impedance corresponding to 385.2 ps is 45.4Ω.
[0147] Comparison between Example 1, Example 2 and Comparative Example 1
[0148] As a result of the above-described TDR analysis, in Example 1, the change in the characteristic impedance of the overall structure 314 relative to the characteristic impedance of the line portion 211 is lower than that in Comparative Example 1. The reason why the change in the characteristic impedance of the overall structure 314 is lower in the TDR analysis is that the line portions 212 and 213 are disposed between the line portion 211 and the overall structure 314 to correct the mismatch in impedance between the line portion 211 and the overall structure 314. Similar to the pulse signal used in the TDR analysis, the digital signal output from the semiconductor device 100 to the signal line 250 also rises from a voltage of 0 V to a voltage V within a rise time tr. in Therefore, similar to a pulse signal used in TDR analysis, the digital signal transmitted through the signal line 250 is less reflected from the overall structure 314. Therefore, the quality of the digital signal transmitted to the connector 400 through the signal line 250 is improved.
[0149] In addition, as a result of the above-mentioned TDR analysis, in Example 2, the change in the characteristic impedance of the overall structure 314 relative to the characteristic impedance of the line portion 211 is lower than that in Comparative Example 1. The reason why the change in the characteristic impedance of the overall structure 314 is lower in the TDR analysis is that the line portions 212 and 213 are disposed between the line portion 211 and the overall structure 314 to correct the mismatch of the impedance between the line portion 211 and the overall structure 314. Similar to the pulse signal used in the TDR analysis, the digital signal output from the semiconductor device 100 to the signal line 250 also rises from a voltage of 0 V to a voltage V within a rise time tr. in Therefore, similar to a pulse signal used in TDR analysis, the digital signal transmitted through the signal line 250 is less reflected from the overall structure 314. Therefore, the quality of the digital signal transmitted to the connector 400 through the signal line 250 is improved.
[0150] In the above description, the simulation for the characteristic impedance is described. In order to measure the characteristic impedance of the signal line 250 of the actual printed wiring board, a TDR oscilloscope is used. In this measurement, a signal line with V in A step pulse with a voltage amplitude of 100 ps and a rise time tr of 35 ps is sent to one end of the line portion 211 via a probe connected to a TDR oscilloscope. If there is a mismatch point in the characteristic impedance, the signal is reflected from the mismatch point and returns to the probe, which is sent to one end of the line portion 211 via the probe. Therefore, the reflected voltage is added to the voltage of the signal observed at the probe. The characteristic impedance of the signal line 250 can be calculated based on the observed voltage.
[0151] If the voltage of the observation point is represented by Vr and the output impedance of the TDR oscilloscope is 50Ω, the characteristic impedance Z0 of the signal line 250 is calculated by using the following expression (6).
[0152]
[0153] In addition, the propagation velocity vo of the signal transmitted through the signal line 250 is obtained by calculating a time period measured between change points of the characteristic impedance observed on the TDR oscilloscope and multiplying the time period by 0.5.
[0154] In order to measure the rise time of the pulse signal output from the transmission circuit (not shown), an oscilloscope is used. First, the waveform of the voltage at one end of the line portion 211 connected to the transmission circuit (not shown) is measured by using a probe. In this measurement, the voltage is measured from the voltage amplitude V in A 20% change in voltage amplitude V in The voltage amplitude V is obtained from the matching condition of the characteristic impedance. inFor example, if the voltage amplitude is 400 mV in the matched state, a time period corresponding to a voltage change of 240 mV is measured. If this time period is represented by tr', the relationship between the time period tr' and the rise time tr is expressed by the following expression.
[0155] tr=tr' / 0.6
[0156] Note that the characteristic impedance can be measured by another method other than TDR analysis. For example, the characteristic impedance can be obtained by cutting the board, measuring the dimensions of the cross section of the wire (i.e., thickness and width), measuring the dielectric constant of the material and the conductivity of the conductor, and calculating the characteristic impedance using an electromagnetic field simulator.
[0157] Example 3
[0158] Will refer to Fig. 6A and Figure 6B Example 3 is described. Example 3 indicates specific numerical values of the third embodiment. First, differential impedance is calculated. To calculate differential impedance, HyperLynx manufactured by Mentor is used. Hereinafter, numerical values of each component used to calculate differential impedance will be described.
[0159] The thickness of the conductor layer 221 is set to 37 μm. The thickness of the conductor layer 222 is set to 35 μm. The thickness of the conductor layer 223 is set to 35 μm. The thickness of the conductor layer 224 is set to 37 μm. The thickness of the insulator layer 231 is set to 100 μm. The thickness of the insulator layer 232 is set to 1200 μm. The thickness of the insulator layer 233 is set to 100 μm. The relative dielectric constant of the insulator layers 231, 232 and 233 is set to 4.3, and the dielectric tangent of the insulator layers 231, 232 and 233 is set to 0.02. In addition, the surfaces of the conductor layers 221 and 224 are covered with a solder resist (not shown). The thickness of the solder resist (not shown) in the Z direction is set to 20 μm. The relative dielectric constant of the solder resist (not shown) was set to 3.0, and the dielectric tangent of the solder resist was set to 0.02.
[0160] The line width W1 in the Y direction of each of the line portions 2111 and 2112 is set to 100 μm, and the length L1 in the X direction of each of the line portions 2111 and 2112 is set to 28.34 mm. The gap between the line portions 2111 and 2112 is set to 135 μm. The line width W2 in the Y direction of each of the line portions 2121 and 2122 is set to 300 μm, and the length L2 in the X direction of each of the line portions 2121 and 2122 is set to 0.37 mm. The gap between the line portions 2121 and 2122 is set to 150 μm. The line width W3 in the Y direction of each of the line portions 2131 and 2132 is set to 90 μm, and the length L3 in the X direction of each of the line portions 2131 and 2132 is set to 1.29 mm. The gap between the line portions 2131 and 2132 is set to 410 μm. The line width W4 in the Y direction of each of the line portions 2141 and 2142 is set to 250 μm, and the length L4 in the X direction of each of the line portions 2141 and 2142 is set to 2.0 mm. The gap between the line portions 2141 and 2142 is set to 250 μm. The width in the Y direction of each of the terminals 4141 and 4142 is set to 250 μm, and the length in the X direction of each of the terminals 4141 and 4142 is set to 2.0 mm. The thickness in the Z direction of each of the terminals 4141 and 4142 is set to 200 μm. The length in the X direction of each of the line portions 4151 and 4152 is set to 2.0 mm. In addition, the signal output portions of the line portions 4151 and 4152 are terminated by a 100 Ω resistor (not shown).
[0161] After performing calculations under the above conditions, the following differential impedances Z1 to Z4 are obtained. The differential impedance Z1 is 100.1 Ω. The differential impedance Z2 is 61.7 Ω. The differential impedance Z3 is 121.3 Ω. The differential impedance Z4 is 60.6 Ω. The differential impedance Z5 is 123.4 Ω. Therefore, the relationship of the differential impedances is Z4 < Z2 < Z1 < Z3 < Z5. Although the width of the overall structures 3141 and 3142 is smaller than the width of the line portions 2121 and 2122, the differential impedance Z4 is lower than the differential impedance Z2. The differential impedance Z14 is 75.5 Ω.
[0162] In addition, a TDR analysis simulation is performed on the structure of Example 3. Fig. 9A is a graph showing the simulation results obtained in the example. In Fig. 9A the vertical axis represents the differential impedance (Ω), and the horizontal axis represents the time (seconds). TDR analysis can determine the differential impedance of a signal line at a position separated from the signal source by a certain distance. In addition, TDR analysis can evaluate the quality of the voltage waveform of a digital signal.
[0163] In Fig. 9A, a waveform 2001 indicates a TDR analysis result obtained in Example 3. For the TDR analysis, HSPICE manufactured by Synopsys, Inc. was used. In addition, a pulse signal was used as a digital signal to be applied to a signal line.
[0164] Fig. 9B is a diagram showing a pulse signal applied from a signal source to one end of a main line in Example 3. In Example 3, the main line is line portions 2111 and 2112. One end of the main line is Figure 5 The ends 2511 and 2512 are shown. The signal source and Figure 5 The voltage amplitude of the differential pulse signal applied to one end of the main line is V in The rise time of the differential pulse signal is represented by tr. The rise time tr is related to the voltage amplitude V in The voltage amplitude V in The internal impedance of the signal source is 100Ω.
[0165] The analysis results in Example 3 will be described by comparing the TDR analysis results and the calculation results of the differential impedance. In Example 3, a differential pulse signal is applied to one end of a pair of line portions 2111 and 2112. In the TDR analysis results, the differential impedance Z1 of the line portions 2111 and 2112 is 104Ω.
[0166] The differential impedance Z2 of the pair of line portions 2121 and 2122 obtained by calculation is 61.7Ω, which is lower than the differential impedance Z1 of the pair of line portions 2111 and 2112. The differential impedance Z3 of the pair of line portions 2131 and 2132 is 121.3Ω, which is higher than the differential impedance Z1. The differential impedance Z4 of the pair of integral structures 3141 and 3142 is 60.6Ω, which is lower than the differential impedance Z1.
[0167] The length L1 of each of the line portions 2111 and 2112 in the X direction is set to 28.34 mm. The gap between the line portions 2111 and 2112 is set to 135 μm. The relative dielectric constant of the solder resist formed on the line portions 2111 and 2112 is set to 3, and the relative dielectric constant of the insulator formed on the line portions 2111 and 2112 is set to 4.3. The propagation rate of the electromagnetic wave is calculated using the relative dielectric constant of the solder resist and the relative dielectric constant of the insulator. In order to calculate the propagation rate of the electromagnetic wave, HyperLynx manufactured by Mentor is used. The propagation rate of the electromagnetic wave is determined to be approximately 1.759×10 8 m / s, the propagation speed is the speed of light (≈3.0×10 8Therefore, the time for the signal to propagate through the line portions 2111 and 2112 is 161.1 ps.
[0168] The length L2 of each of the line portions 2121 and 2122 in the X direction is set to 0.37 mm. The gap between the line portions 2121 and 2122 is set to 150 μm. The relative dielectric constant of the solder resist formed on the line portions 2121 and 2122 is set to 3, and the relative dielectric constant of the insulator formed on the line portions 2121 and 2122 is set to 4.3. The propagation rate of the electromagnetic wave is calculated using the relative dielectric constant of the solder resist and the relative dielectric constant of the insulator. In order to calculate the propagation rate of the electromagnetic wave, HyperLynx manufactured by Mentor is used. The propagation rate of the electromagnetic wave is determined to be approximately 1.688×10 8 m / s, the propagation speed is the speed of light (≈3.0×10 8 Therefore, the time for the signal to propagate through the pair of line portions 2121 and 2122 is 2.2 ps.
[0169] The differential impedance is reduced to 61.7Ω within the rise time tr (i.e., 35ps) of the pulse signal. However, in a pair of line portions 2121 and 2122, the signal wave travels back and forth within 4.4ps (=2.2×2). Therefore, the effective change from the internal impedance (100Ω) of the signal source to the differential impedance Z2 (61.7Ω) of a pair of line portions 2121 and 2122 is -4.8Ω (≈(61.7-100)×(4.4 / 35)).
[0170] The length L3 of each of the line portions 2131 and 2132 in the X direction is set to 1.29 mm. The gap between the line portions 2131 and 2132 is set to 410 μm. The relative dielectric constant of the solder resist formed on the line portions 2131 and 2132 is set to 3, and the relative dielectric constant of the insulator formed on the line portions 2131 and 2132 is set to 4.3. The propagation rate of the electromagnetic wave is calculated using the relative dielectric constant of the solder resist and the relative dielectric constant of the insulator. In order to calculate the propagation rate of the electromagnetic wave, HyperLynx manufactured by Mentor is used. The propagation rate of the electromagnetic wave is determined to be approximately 1.713×10 8 m / s, the propagation speed is the speed of light (≈3.0×10 8 Therefore, the time for the signal to propagate through the pair of line portions 2131 and 2132 is 7.5 ps.
[0171] The differential impedance is reduced to 121.3Ω within the rise time tr (i.e., 35ps) of the pulse signal. However, in a pair of line portions 2131 and 2132, the signal wave travels back and forth within 15.0ps (=7.5×2). Therefore, the effective change from the internal impedance (100Ω) of the signal source to the differential impedance Z3 (121.3Ω) of a pair of line portions 2131 and 2132 is 9.1Ω (≈(121.3-100)×(15.0 / 35)).
[0172] The length of the integral structures 3141 and 3142 in the X direction is set to 2.0 mm. The relative dielectric constant of the air on the integral structures 3141 and 3142 is set to 1, and the relative dielectric constant of the insulator formed on the integral structures 3141 and 3142 is set to 4.3. The propagation velocity of the electromagnetic wave is calculated using the relative dielectric constant of the air and the relative dielectric constant of the insulator. To calculate the propagation velocity of the electromagnetic wave, HyperLynx manufactured by Mentor is used. The propagation velocity of the electromagnetic wave is determined to be approximately 1.931×10 8 m / s, the propagation speed is the speed of light (≈3.0×10 8 Therefore, the time for the signal to propagate through the pair of integral structures 3141 and 3142 is 10.4 ps.
[0173] The differential impedance is reduced to 60.6Ω within the rise time tr (i.e., 35ps) of the pulse signal. However, in a pair of integral structures 3141 and 3142, the signal wave goes back and forth within 20.8ps (=10.4×2). Therefore, the effective change from the internal impedance (100Ω) of the signal source to the differential impedance Z4 (60.6Ω) of a pair of integral structures 3141 and 3142 is -23.4Ω (≈(60.6-50)×(20.8 / 35)).
[0174] In Example 3, a pair of line portions 2111 and 2112, a pair of line portions 2121 and 2122, a pair of line portions 2131 and 2132, and a pair of integral structures 3141 and 3142 are formed successively in sequence, and the differential impedance changes at the pair of line portions 2111 and 2112, the pair of line portions 2121 and 2122, the pair of line portions 2131 and 2132, and the pair of integral structures 3141 and 3142. Therefore, in the TDR analysis, the differential impedance of each pair of line portions is obtained by adding the effective change in a pair of line portions to the differential impedance of the previous pair of line portions. If the differential impedance Z1 of the pair of line portions 2111 and 2112 obtained in the TDR analysis is added to the effective change in the three pairs of line portions (obtained by calculation), the result value is calculated to be 84.9Ω (=104-4.8+9.1-23.4), which is lower than the differential impedance Z1. In the TDR analysis, the value was 87.9Ω.
[0175] exist Fig. 9A In the waveform 2001 shown, twice the time for the signal wave to propagate through the line portions 2111 and 2112 is 322.2 ps; and the time period from 0 to 322.2 ps corresponds to the line portions 2111 and 2112. In addition, twice the time for the signal wave to propagate through the line portions 2121 and 2122 is 4.4 ps; and the time period from 322.2 to 326.6 ps (= 322.2 + 4.4) corresponds to the line portions 2121 and 2122. In addition, twice the time for the signal wave to propagate through the line portions 2131 and 2132 is 15.0 ps; and the time period from 326.6 to 341.6 ps (= 326.6 + 15.0) corresponds to the line portions 2131 and 2132. Finally, twice the time for the signal wave to propagate through the overall structures 3141 and 3142 is 20.8 ps, and the time period from 341.6 to 362.4 ps (=341.6+20.8) corresponds to the overall structures 3141 and 3142. The differential impedance corresponding to 362.4 ps is 87.9Ω.
[0176] The differential impedance Z3 is higher than the differential impedance Z1. Preferably, the differential impedance Z3 is lower than the differential impedance Z5, which is the highest among the differential impedance Z2, the differential impedance Z4, and the differential impedance Z5. The reason for this is to make the variation range of the differential impedance of the pair of line portions 2131 and 2132 obtained by TDR analysis lower than the highest differential impedance Z5 in the circuit.
[0177] In addition, it is preferable that the length L3 of each of the line portions 2131 and 2132 in the X direction satisfies the following expression (7),
[0178]
[0179] Wherein, tr is the rise time of the digital signal, vo is the propagation rate of the digital signal, Z1 is the differential impedance of a pair of line portions 2111 and 2112, and Z5 is the differential impedance of a pair of line portions 4151 and 4152.
[0180] The differential impedance of the line portions 2111 and 2112 as the main line is controlled to be about 100Ω. In expression (7), the ratio is calculated by dividing the amount of change in the differential impedance allowed in the line portions 2131 and 2132 (Z1×0.10) by the difference Z5-Z1 between the differential impedance of the line portions 2111 and 2112 and the differential impedance of the line portions 4151 and 4152. Then, the ratio is multiplied by half the rise time of the pulse signal and the propagation rate. Through this calculation, the length of the line portions 2131 and 2132 required to maintain the quality of the voltage waveform of the digital signal is determined. Note that the line portions 2131 and 2132 are structures for suppressing disturbances of the voltage waveform of the digital signal caused in the overall structures 3141 and 3142, and the differential impedance of the line portions 2131 and 2132 is determined so that the line portions 2131 and 2132 do not generate any new disturbances.
[0181] If the line has a length such that the round trip time of the electromagnetic wave is longer than the rise time of the pulse signal, the characteristic impedance of the line can be measured in the TDR analysis after the rise time of the pulse signal has passed. In contrast, if the line has a length such that the round trip time of the electromagnetic wave is shorter than the rise time of the pulse signal, that is, if the line has a length shorter than the time resolution of the pulse signal, the characteristic impedance of the line does not change to the value obtained in the line with sufficient length. If Z3=Z5, the time for the characteristic impedance to change by the amount of Z5-Z1 is equal to the rise time of the pulse signal. If the differential impedance Z3 is limited to Z1×0.10, then expression (7) is obtained.
[0182] In addition, in order to maintain the quality of the voltage waveform of the digital signal, the board is generally manufactured so that the amount of change in the differential impedance allowed in the line portions 2131 and 2132 is about ±5 to ±15% of the differential impedance of the main line. That is, considering the amount of change in the differential impedance allowed in the line portions 2131 and 2132, it is preferred that the length L3 of the line portions 2131 and 2132 in the X direction satisfies the following expression (8).
[0183]
[0184] The differential impedance Z2 is lower than the differential impedance Z1. Preferably, the differential impedance Z2 is higher than the differential impedance Z4, which is the lowest among the differential impedance Z3, the differential impedance Z4, and the differential impedance Z5. The reason for this is that the differential impedance of the pair of line portions 2121 and 2122 obtained by TDR analysis has a variation range relative to the differential impedance of the pair of line portions 2111 and 2112 lower than the differential impedance of the pair of integral structures 3141 and 3142.
[0185] In addition, it is preferable that the length L2 of each of the line portions 2121 and 2122 in the X direction satisfies the following Expression (9).
[0186]
[0187] The differential impedance of the line portions 2111 and 2112 as the main line is controlled to be about 100Ω. In expression (9), the ratio is calculated by dividing the amount of change in the differential impedance allowed in the line portions 2121 and 2122 (Z1×0.10) by the difference Z1-Z4 between the differential impedance of the line portions 2111 and 2112 and the differential impedance of the overall structures 3141 and 3142. Then, the ratio is multiplied by half the rise time of the pulse signal and the propagation rate. Finally, the length of the line portions 2121 and 2122 is determined by multiplying the resulting ratio by 0.5. Note that the line portions 2121 and 2122 are structures for suppressing disturbances of the voltage waveform of the digital signal caused in the overall structures 3141 and 3142, and the differential impedance of the line portions 2121 and 2122 is determined so that the line portions 2121 and 2122 do not generate any new disturbances.
[0188] Since the line portions 2121 and 2122 are disposed in front of the line portions 2131 and 2132, the increase in the differential impedance of the line portions 2131 and 2132 can be suppressed. However, if the differential impedance of the line portions 2121 and 2122 disposed in front of the line portions 2131 and 2132 is excessively reduced, the difference Z1-Z4 between the differential impedance Z1 of the line portions 2111 and 2112 reduced by the line portions 2131 and 2132 and the differential impedance Z4 of the overall structures 3141 and 3142 increases again. Therefore, in Example 3, the lengths of the line portions 2121 and 2122 are corrected. In the above expression (9), the correction coefficient is 0.5.
[0189] In addition, in order to maintain the quality of the voltage waveform of the digital signal, the board is manufactured so that the amount of change in the differential impedance allowed in the line portions 2121 and 2122 is about ±5, ±10, or ±15% of the differential impedance of the main line. That is, considering the amount of change in the differential impedance allowed in the line portions 2121 and 2122, it is preferred that the length L2 of the line portions 2121 and 2122 in the X direction satisfies the following expression (10).
[0190]
[0191] In addition, it is preferable that the length L4 of each of the line portions 2141 and 2142 in the X direction satisfies the following expression (11).
[0192]
[0193] If the line has a length in which the round trip time of the electromagnetic wave is sufficiently longer than the rise time of the pulse signal, then the differential impedance of the line that is substantially the same as the differential impedance of a line having an infinite length can be measured in the TDR analysis after the rise time of the pulse signal has passed. In contrast, if the line has a length in which the round trip time of the electromagnetic wave is shorter than the rise time of the pulse signal, that is, if the line has a length shorter than the time resolution of the pulse signal, then the differential impedance of the line does not change to the value obtained in the line having a sufficient length. If the line portions 2141 and 2142 have a length L4 in which the round trip time of the electromagnetic wave is longer than the rise time of the pulse signal, then the effect of suppressing the disturbance of the voltage waveform of the digital signal caused in the overall structure 3141 and 3142 generated by the arrangement of the line portions 2121 and 2122 and the line portions 2131 and 2132 is not generated. Therefore, the length L4 is determined as described above.
[0194] Example 4
[0195] Will refer to Fig. 7A and Figure 7B Example 4 is described. Example 4 indicates specific numerical values of the fourth embodiment. First, the differential impedance as the characteristic impedance is calculated. In order to calculate the differential impedance, HyperLynx manufactured by Mentor is used as in Example 3. Hereinafter, the numerical value of each component used to calculate the differential impedance will be described. The structure of each layer of the printed wiring board 200C of Example 4 is the same as the structure of each layer of the printed wiring board of Example 3. In the following description, the features of Example 4 that are different from the features of Example 3 will be described.
[0196] The line width W1 of each of the line portions 2111 and 2112 in the Y direction is set to 100 μm, and the length L1 of each of the line portions 2111 and 2112 in the X direction is set to 28.13 mm. The gap between the line portions 2111 and 2112 is set to 135 μm. The line width W2 of each of the line portions 2121 and 2122 in the Y direction is set to 230 μm, and the length L2 of each of the line portions 2121 and 2122 in the X direction is set to 0.58 mm. The gap between the line portions 2121 and 2122 is set to 230 μm. The line width W3 of each of the line portions 2131 and 2132 in the Y direction is set to 90 μm, and the length L3 of each of the line portions 2131 and 2132 in the X direction is set to 1.29 mm. The gap between the line portions 2131 and 2132 is set to 410 μm. The line width W4 of each of the line portions 2141 and 2142 in the Y direction is set to 250 μm, and the length L4 of each of the line portions 2141 and 2142 in the X direction is set to 2.0 mm. The gap between the line portions 2141 and 2142 is set to 250 μm. The width of each of the terminals 4141 and 4142 in the Y direction is set to 250 μm, and the length of each of the terminals 4141 and 4142 in the X direction is set to 2.0 mm. The width of the opening H2 in the Y direction is set to 750 μm, and 750 μm is obtained by doubling the width W4 and adding the resulting value to the gap 250 μm between the line portions 2141 and 2142. The length L4 of the opening H2 in the X direction is set to 2.0 mm. The thickness of each of the terminals 4141 and 4142 in the Z direction is set to 200 μm. The length of each of the line portions 4151 and 4152 in the X direction is set to 2.0 mm. In addition, the signal output portions of the pair of line portions 4151 and 4152 are terminated by 100 Ω resistors (not shown).
[0197] After performing calculations under the above conditions, the following differential impedances Z1 to Z4 are obtained. The differential impedance Z1 is 100.1 Ω. The differential impedance Z2 is 75.4 Ω. The differential impedance Z3 is 121.3 Ω. The differential impedance Z4 is 74.8 Ω. The differential impedance Z5 is 123.4 Ω. Therefore, the relationship of the differential impedances is Z4 < Z2 < Z1 < Z3 < Z5. The width of the overall structures 3141 and 3142 is greater than the width of the line portions 2121 and 2122, and the differential impedance Z4 is lower than the differential impedance Z2. The differential impedance Z14 is 98.6 Ω.
[0198] In addition, a TDR analysis simulation is performed on the structure of Example 4. In Fig. 9A, waveform 2002 indicates the TDR analysis result obtained in Example 4. For the TDR analysis, HSPICE manufactured by Synopsys was used as in Example 3. In addition, a pulse signal was used as a digital signal to be applied to the signal line. The pulse signal was the same as the pulse signal used in Example 3.
[0199] The analysis results in Example 4 will be described by comparing the TDR analysis results and the calculation results of the differential impedance. In Example 4, a differential pulse signal is applied to one end of a pair of line portions 2111 and 2112. In the TDR analysis results, the differential impedance Z1 of the line portions 2111 and 2112 is 104Ω.
[0200] The differential impedance Z2 of the pair of line portions 2121 and 2122 obtained by calculation is 75.4Ω, which is lower than the differential impedance Z1 of the pair of line portions 2111 and 2112. The differential impedance Z3 of the pair of line portions 2131 and 2132 is 121.3Ω, which is higher than the differential impedance Z1 of the pair of line portions 2111 and 2112. The differential impedance Z4 of the pair of integral structures 3141 and 3142 is 74.8Ω, which is lower than the differential impedance Z1. When viewed from the Z direction, the pair of line portions 2141 and 2142 overlap with the opening H2 of the ground pattern 262C. Therefore, the differential impedance Z4 is higher than the differential impedance Z4 of the pair of line portions 2141 and 2142 of Example 3.
[0201] The length L1 of each of the line portions 2111 and 2112 in the X direction is set to 28.13 mm. The gap between the line portions 2111 and 2112 is set to 135 μm. The relative dielectric constant of the solder resist formed on the line portions 2111 and 2112 is set to 3, and the relative dielectric constant of the insulator formed on the line portions 2111 and 2112 is set to 4.3. The propagation rate of the electromagnetic wave is calculated using the relative dielectric constant of the solder resist and the relative dielectric constant of the insulator. In order to calculate the propagation rate of the electromagnetic wave, HyperLynx manufactured by Mentor is used. The propagation rate of the electromagnetic wave is determined to be approximately 1.759×10 8 m / s, the propagation speed is the speed of light (≈3.0×10 8 Therefore, the time for the signal to propagate through the line portions 2111 and 2112 is 159.9 ps.
[0202] The length L2 of each of the line portions 2121 and 2122 in the X direction is set to 0.58 mm. The gap between the line portions 2121 and 2122 is set to 250 μm. The relative dielectric constant of the solder resist formed on the line portions 2121 and 2122 is set to 3, and the relative dielectric constant of the insulator formed on the line portions 2121 and 2122 is set to 4.3. The propagation rate of the electromagnetic wave is calculated using the relative dielectric constant of the solder resist and the relative dielectric constant of the insulator. In order to calculate the propagation rate of the electromagnetic wave, HyperLynx manufactured by Mentor is used. The propagation rate of the electromagnetic wave is determined to be approximately 1.689×10 8 m / s, the propagation speed is the speed of light (≈3.0×10 8 Therefore, the time for the signal to propagate through the pair of line portions 2121 and 2122 is 3.4 ps.
[0203] The differential impedance is reduced to 75.4Ω within the rise time tr (i.e., 35ps) of the pulse signal. However, in a pair of line portions 2121 and 2122, the signal wave goes back and forth within 6.8ps (=3.4×2). Therefore, the effective change from the internal impedance (100Ω) of the signal source to the differential impedance Z2 (75.4Ω) of a pair of line portions 2121 and 2122 is -4.8Ω (≈(75.4-100)×(6.8 / 35)).
[0204] The length L3 of each of the line portions 2131 and 2132 in the X direction is set to 1.29 mm. The gap between the line portions 2131 and 2132 is set to 410 μm. The relative dielectric constant of the solder resist formed on the line portions 2131 and 2132 is set to 3, and the relative dielectric constant of the insulator formed on the line portions 2131 and 2132 is set to 4.3. The propagation rate of the electromagnetic wave is calculated using the relative dielectric constant of the solder resist and the relative dielectric constant of the insulator. In order to calculate the propagation rate of the electromagnetic wave, HyperLynx manufactured by Mentor is used. The propagation rate of the electromagnetic wave is determined to be approximately 1.713×10 8 m / s, the propagation speed is the speed of light (≈3.0×10 8 Therefore, the time for the signal to propagate through the line portions 2131 and 2132 is 7.5 ps.
[0205] The differential impedance increases to 121.3Ω within the rise time tr (i.e., 35ps) of the pulse signal. However, in a pair of line portions 2131 and 2132, the signal wave goes back and forth within 15.0ps (=7.5×2). Therefore, the effective change from the internal impedance (100Ω) of the signal source to the differential impedance Z3 (121.3Ω) of a pair of line portions 2131 and 2132 is 9.1Ω (≈(121.3-100)×(15.0 / 35)).
[0206] The length of each of the integral structures 3141 and 3142 in the X direction is set to 2.0 mm. The relative dielectric constant of the air on the integral structures 3141 and 3142 is set to 1, and the relative dielectric constant of the insulator formed on the integral structures 3141 and 3142 is set to 4.3. The propagation rate of the electromagnetic wave is calculated using the relative dielectric constant of the air and the relative dielectric constant of the insulator. In order to calculate the propagation rate of the electromagnetic wave, HyperLynx manufactured by Mentor is used. The propagation rate of the electromagnetic wave is determined to be approximately 2.043×10 8 m / s, the propagation speed is the speed of light (≈3.0×10 8 m / s). Therefore, the time for the signal to propagate through the overall structures 3141 and 3142 is 9.8 ps.
[0207] The differential impedance is reduced to 74.8Ω within the rise time tr (i.e., 35ps) of the pulse signal. However, in a pair of integral structures 3141 and 3142, the signal wave goes back and forth within 19.6ps (=9.8×2). Therefore, the effective change from the internal impedance (100Ω) of the signal source to the differential impedance Z4 (74.8Ω) of a pair of integral structures 3141 and 3142 is -14.1Ω (≈(74.8-100)×(19.6 / 35)).
[0208] In Example 4, a pair of line portions 2111 and 2112, a pair of line portions 2121 and 2122, a pair of line portions 2131 and 2132, and a pair of integral structures 3141 and 3142 are formed successively in sequence, and the differential impedance changes at the pair of line portions 2111 and 2112, the pair of line portions 2121 and 2122, the pair of line portions 2131 and 2132, and the pair of integral structures 3141 and 3142. Therefore, in the TDR analysis, the differential impedance of each pair of line portions is obtained by adding the effective change in a pair of line portions to the differential impedance of the previous pair of line portions. If the differential impedance Z1 of the line portions 2111 and 2112 obtained in the TDR analysis is added to the effective change in the three pairs of line portions (obtained by calculation), the result value is calculated to be 94.2Ω (=104-4.8+9.1-14.1), which is lower than the differential impedance Z1. In the TDR analysis, this value was 97.8Ω.
[0209] exist Fig. 9A In the waveform 2002 shown, twice the time for the signal wave to propagate through the line portions 2111 and 2112 is 319.8 ps; and the time period from 0 to 319.8 ps corresponds to the line portions 2111 and 2112. In addition, twice the time for the signal wave to propagate through the line portions 2121 and 2122 is 6.8 ps; and the time period from 319.8 to 326.6 ps (= 319.8 + 6.8) corresponds to the line portions 2121 and 2122. In addition, twice the time for the signal wave to propagate through the line portions 2131 and 2132 is 15.0 ps; and the time period from 326.6 to 341.6 ps (= 326.6 + 15.0) corresponds to the line portions 2131 and 2132. Finally, twice the time for the signal wave to propagate through the overall structures 3141 and 3142 is 19.6 ps, and the time period from 341.6 to 361.2 ps (=341.6+19.6) corresponds to the overall structures 3141 and 3142. The differential impedance corresponding to 361.2 ps is 97.8Ω.
[0210] Comparative Example 2
[0211] Next, the processing module of Comparative Example 2 will be described. Fig. 10B It is a plan view of a part of a process module 300Y of Comparative Example 2. The process module 300Y includes a connector 400B and a printed wiring board 200Y. The connector 400B is the same as the connector of the third embodiment, and the printed wiring board 200Y is different from the printed wiring board 200B of the third embodiment.
[0212] The printed wiring board 200Y includes signal lines 250Y1 and 250Y2 whose structures are different from those of the signal lines 2501 and 2502 of the third embodiment. The other structures of the printed wiring board 200Y are the same as those of the printed wiring board 200B of the third embodiment. The signal line 250Y1 includes a line portion 211Y1 and a line portion 214Y1 continuous with the line portion 211Y1. Similarly, the signal line 250Y2 includes a line portion 211Y2 and a line portion 214Y2 continuous with the line portion 211Y2. The line portions 214Y1 and 214Y2 are pads that can be joined to the terminals 4141 and 4142 of the connector 400B via solder. When the line portions 214Y1 and 214Y2 are joined to the terminals 4141 and 4142, integral structures 314Y1 and 314Y2 are formed.
[0213] Next, the specific numerical values of Comparative Example 2 will be described. First, the differential impedance as the characteristic impedance is calculated. In order to calculate the differential impedance, HyperLynx manufactured by Mentor is used. Hereinafter, the numerical value of each component used to calculate the differential impedance will be described. The structure of each layer of the printed wiring board 200Y of Comparative Example 2 is the same as the structure of each layer of the printed wiring board of Example 3. In the following description, the features of Comparative Example 2 that are different from those of Example 3 will be described.
[0214] The line width W1Y of each of the line portions 211Y1 and 211Y2 in the Y direction is set to 100μm, and the length L1Y of each of the line portions 211Y1 and 211Y2 in the X direction is set to 30mm. The gap between the line portions 211Y1 and 211Y2 is set to 150μm. The line width W4Y of each of the line portions 214Y1 and 214Y2 in the Y direction is set to 250μm, and the length L4Y of each of the line portions 214Y1 and 214Y2 in the X direction is set to 2.0mm. The gap between the line portions 214Y1 and 214Y2 is set to 250μm. The other numerical values are the same as those of Example 3. The differential impedance Z1Y of a pair of line portions 211Y1 and 211Y2 is 100.1Ω. The differential impedance Z4Y of the pair of line portions 214Y1 and 214Y2 is 60.6Ω.
[0215] In addition, TDR analysis simulation is performed on the structure of Comparative Example 2. Fig. 9A , waveform 2003 indicates the TDR analysis result obtained in Comparative Example 2. For TDR analysis, HSPICE manufactured by Synopsys was used as in Examples 3 and 4. In addition, a pulse signal was used as a digital signal to be applied to the signal line. The pulse signal was the same as the pulse signal used in Examples 3 and 4.
[0216] The analysis results in Comparative Example 2 will be described. In Comparative Example 2, a differential pulse signal is applied to one end of a pair of line portions 211Y1 and 211Y2. In the TDR analysis results, the differential impedance Z1 of the line portions 211Y1 and 211Y2 is 104Ω. The differential impedance Z4Y of the pair of integral structures 314Y1 and 314Y2 is lower than the differential impedance Z1Y. Therefore, in the TDR analysis results, as Fig. 9A As shown in waveform 2003, when the pulse signal reaches a pair of integral structures 314Y1 and 314Y2, the differential impedance is reduced to about 86Ω.
[0217] The length L1Y of each of the line portions 211Y1 and 211Y2 in the X direction is set to 30 mm. The relative dielectric constant of the solder resist formed on the line portions 211Y1 and 211Y2 is set to 3, and the relative dielectric constant of the insulator formed on the line portions 211Y1 and 211Y2 is set to 4.3. The propagation rate of the electromagnetic wave is calculated using the relative dielectric constant of the solder resist and the relative dielectric constant of the insulator. To calculate the propagation rate of the electromagnetic wave, HyperLynx manufactured by Mentor is used. The propagation rate of the electromagnetic wave is determined to be approximately 1.759×10 8 m / s, the propagation speed is the speed of light (≈3.0×10 8 Therefore, the time for the signal to propagate through the line portions 211Y1 and 211Y2 is 170.6 ps.
[0218] In addition, the length of the integral structures 314Y1 and 314Y2 in the X direction is set to 2.0 mm. The relative dielectric constant of the air on the integral structures 314Y1 and 314Y2 is set to 1, and the relative dielectric constant of the insulator formed on the integral structures 314Y1 and 314Y2 is set to 4.3. The propagation rate of the electromagnetic wave is calculated using the relative dielectric constant of the air and the relative dielectric constant of the insulator. In order to calculate the propagation rate of the electromagnetic wave, HyperLynx manufactured by Mentor is used. Based on the volume ratio of air to insulator, the propagation rate of the electromagnetic wave is determined to be approximately 1.931×10 8 m / s, the propagation speed is the speed of light (≈3.0×10 8 m / s). Therefore, the time for the signal to propagate through the overall structures 314Y1 and 314Y2 is 10.4 ps.
[0219] Therefore, the propagation time (10.4 ps) of the signal corresponding to the length L4Y of the integral structures 314Y1 and 314Y2 is shorter than the resolution, which is the rise time tr (i.e., 35 ps) of the signal applied to the integral structures 314Y1 and 314Y2. Therefore, before the differential impedance obtained by TDR analysis is reduced to the actual differential impedance Z4Y (60.6Ω), the signal reaches the line portions 4151 and 4152 of the connector 400B.
[0220] The specific calculation will be described below. The differential impedance is reduced to 60.6Ω within the rise time tr (i.e., 35ps) of the differential pulse signal. However, in a pair of integral structures 314Y1 and 314Y2, the signal wave travels back and forth within 20.8ps (=10.4×2). Therefore, the effective change from the internal impedance (100Ω) of the signal source to the differential impedance Z4Y (60.6Ω) of a pair of integral structures 314Y1 and 314Y2 is -23.4Ω (≈(60.6-100)×(20.8 / 35)). Therefore, in the calculation, the differential impedance is reduced to 80.6Ω (=104-23.4). In the TDR analysis, the differential impedance is reduced to 85.7Ω.
[0221] exist Fig. 9A In the waveform 2003 shown, twice the time for the signal wave to propagate through the line portions 211Y1 and 211Y2 is 341.2 ps; and the time period from 0 to 341.2 ps corresponds to the line portions 211Y1 and 211Y2. In addition, twice the time for the signal wave to propagate through the overall structures 314Y1 and 314Y2 is 20.8 ps; and the time period from 341.2 to 362.0 ps (= 341.2 + 20.8) corresponds to the overall structures 314Y1 and 314Y2. The differential impedance corresponding to 362.0 ps is 85.7Ω.
[0222] Comparison between Example 3, Example 4 and Comparative Example 2
[0223] As a result of the above-described TDR analysis, in Example 3, the change in the differential impedance of the pair of integral structures 3141 and 3142 relative to the differential impedance of the pair of line portions 2111 and 2112 is lower than that in Comparative Example 2. In Example 3, the line portions 2121 and 2122 and the line portions 2131 and 2132 are disposed between the line portions 2111 and 2112 and the integral structures 3141 and 3142. The reason why the change in the differential impedance of the integral structures 3141 and 3142 is reduced in the TDR analysis is that the mismatch in impedance between the line portions 2111 and 2112 and the integral structures 3141 and 3142 is corrected by the line portions 2121 and 2122 and the line portions 2131 and 2132. Similar to the differential pulse signal used in the TDR analysis, the differential signal output from the semiconductor device 100B to the pair of signal lines 2501 and 2502 also rises from a voltage of 0 V to a voltage V within a rise time tr. in Therefore, similar to the differential pulse signal used in TDR analysis, the differential signal transmitted through the signal lines 2501 and 2502 is also less reflected from the pair of integral structures 3141 and 3142. Therefore, the quality of the digital signal transmitted to the connector 400B through the signal lines 2501 and 2502 is improved.
[0224] In addition, as a result of the above-mentioned TDR analysis, in Example 4, the change in the differential impedance of the pair of integral structures 3141 and 3142 relative to the differential impedance of the pair of line portions 2111 and 2112 is lower than that in Comparative Example 2. In Example 4, the line portions 2121 and 2122 and the line portions 2131 and 2132 are disposed between the line portions 2111 and 2112 and the integral structures 3141 and 3142. The reason why the change in the differential impedance of the integral structures 3141 and 3142 is reduced in the TDR analysis is that the mismatch in impedance between the line portions 2111 and 2112 and the integral structures 3141 and 3142 is corrected by the line portions 2121 and 2122 and the line portions 2131 and 2132. Similar to the differential pulse signal used in the TDR analysis, the differential signal output from the semiconductor device 100B to the pair of signal lines 2501 and 2502 also rises from a voltage of 0 V to a voltage V within a rise time tr. in Therefore, similar to the differential pulse signal used in TDR analysis, the differential signal transmitted through the signal lines 2501 and 2502 is also less reflected from the pair of integral structures 3141 and 3142. Therefore, the quality of the digital signal transmitted to the connector 400B through the signal lines 2501 and 2502 is improved.
[0225] In the above description, the simulation for differential impedance has been described. In order to measure the differential impedance of a pair of signal lines 2501 and 2502 of an actual printed wiring board, a TDR oscilloscope is used. In this measurement, a voltage with V in A step pulse with a voltage amplitude of 100 ps and a rise time tr of 35 ps is sent to one end of a pair of line portions 2111 and 2112 via a probe connected to a TDR oscilloscope. The step pulse consists of a positive signal and a negative signal, and the phases of the signals are opposite to each other. If there is a mismatch point in the differential impedance, the signal is reflected from the mismatch point and returns to the probe, which is sent to one end of the line portions 2111 and 2112 via the probe. Therefore, the reflected voltage is added to the voltage of the signal observed at the probe. The differential impedance of a pair of signal lines 2501 and 2502 can be calculated based on the observed voltage.
[0226] If the voltage at the observation point is represented by Vr and the output impedance of the TDR oscilloscope is 100Ω, the differential impedance Z0 of a pair of signal lines 2501 and 2502 is calculated by using the following expression (12).
[0227]
[0228] The propagation velocity vo of the signal transmitted through the pair of signal lines 2501 and 2502 is obtained by calculating the time period measured between the change points of the differential impedance observed on the TDR oscilloscope and multiplying the time period by 0.5.
[0229] In order to measure the rise time of the pulse signal output from the transmission circuit (not shown), an oscilloscope is used. First, the waveform of the voltage between one end of a pair of line portions 2111 and 2112 connected to the transmission circuit (not shown) is measured by using a probe. In this measurement, the voltage is measured from the voltage amplitude V in A 20% change in voltage amplitude V in The voltage amplitude V is obtained from the matching condition of the differential impedance. in For example, if the voltage amplitude is 400 mV in the matched state, the time period corresponding to the voltage change of 240 mV is measured. If the time period is represented by tr', the relationship between the time period tr' and the rise time tr is expressed by the following expression.
[0230] tr=tr' / 0.6
[0231] Note that the differential impedance may be measured by another method besides TDR analysis. For example, the differential impedance may be obtained by cutting the board, measuring the dimensions of the cross section of the wire (i.e., thickness and width), measuring the dielectric constant of the material and the conductivity of the conductor, and calculating the differential impedance using an electromagnetic field simulator.
[0232] The present invention is not limited to the above-described embodiments, and can be modified within the technical spirit of the present invention. In addition, the effects described in the embodiments are merely examples of the most suitable effects produced by the present invention. Therefore, the effects of the present invention are not limited to the effects described in the embodiments.
[0233] In addition, in the above-mentioned embodiment, the case where the electrical component is the connector 400 or 400B has been described. However, the present disclosure is not limited thereto. For example, the electrical component may be an integrated circuit (IC) having a ball grid array (BGA) or a land grid array (LGA), as long as the electrical component is a surface mounted electrical component.
[0234] In the above embodiment, the case where the characteristic impedances Z2 and Z3 are adjusted by changing the line widths W2 and W3 of the line portion has been described. However, the present disclosure is not limited thereto. For example, the characteristic impedances Z2 and Z3 may be adjusted by changing the thickness of the line portion, or the thickness of the line portion and the line width.
[0235] While the present invention has been described with reference to exemplary embodiments, it is to be understood that the invention is not limited to the disclosed exemplary embodiments.The scope of the following claims is to be accorded the broadest interpretation so as to encompass all such modifications and equivalent structures and functions.
Claims
1. A printed circuit board, characterized in that: include: an electrical assembly including a signal terminal; and a printed wiring board on which electrical components are mounted, wherein the printed wiring board includes signal lines connected to signal terminals, The signal line includes a first line portion, a second line portion, a third line portion and a fourth line portion which are arranged successively in sequence. The signal terminal is bonded to the fourth line portion so that the signal terminal and the fourth line portion form an integral structure. wherein the second characteristic impedance of the second line portion is lower than the first characteristic impedance of the first line portion, wherein a third characteristic impedance of the third line portion is higher than the first characteristic impedance, and The fourth characteristic impedance of the overall structure formed by the fourth line portion and the signal terminal is lower than the first characteristic impedance.
2. The printed circuit board according to claim 1, wherein: The electrical component includes a fifth line portion configured to be continuous with the signal terminal, and The fifth characteristic impedance of the fifth line portion is higher than the first characteristic impedance.
3. The printed circuit board according to claim 2, wherein: The second characteristic impedance is higher than the fourth characteristic impedance, and The fifth characteristic impedance is higher than the third characteristic impedance.
4. The printed circuit board according to claim 2 or 3, wherein: and is satisfied, wherein tr is the rise time of the pulse signal applied to the first line portion, vo is the propagation velocity of the pulse signal, Z1 is the first characteristic impedance, Z4 is the fourth characteristic impedance, Z5 is the fifth characteristic impedance, L2 is the length of the second line portion, and L3 is the length of the third line portion.
5. The printed circuit board according to any one of claims 1 to 3, wherein: is satisfied, wherein tr is the rise time of the pulse signal applied to the first line portion, vo is the propagation velocity of the pulse signal, and L4 is the length of the fourth line portion.
6. The printed circuit board according to any one of claims 1 to 3, wherein: The electrical components are mounted on the surface layer of the printed wiring board, and Among them, the first line portion, the second line portion, the third line portion, and the fourth line portion are disposed on a surface layer of the printed wiring board.
7. The printed circuit board according to claim 6, wherein: The printed wiring board includes a ground pattern disposed in an inner layer adjacent to a surface layer, and The ground pattern includes an opening configured to overlap with the fourth line portion in a plan view.
8. A printed circuit board, characterized in that: include: an electrical assembly including a signal terminal; and a printed wiring board on which electrical components are mounted, wherein the printed wiring board includes signal lines connected to signal terminals, The signal line includes a first line portion, a second line portion, a third line portion and a fourth line portion which are arranged successively in sequence. The signal terminal is connected to the fourth line portion. The line width of the second line portion is greater than the line width of the first line portion. wherein the line width of the third line portion is smaller than the line width of the first line portion, and The line width of the fourth line portion is greater than the line width of the first line portion.
9. The printed circuit board according to claim 1 or 8, wherein: The signal line is one signal line of a pair of signal lines configured to transmit a differential signal.
10. The printed circuit board according to claim 1 or 8, wherein: The electrical component is a connector.
11. The printed circuit board according to claim 1 or 8, further comprising a semiconductor device mounted on the printed wiring board, in, The semiconductor device includes a terminal connected to a signal line and configured to output a signal to the signal line.
12. A printed wiring board having an electrical component mounted thereon, characterized in that: The printed wiring board includes signal lines connected to signal terminals of the electrical components, The signal line includes a first line portion, a second line portion, a third line portion and a fourth line portion which are formed successively in sequence. wherein the signal terminal is configured to be joined to the fourth line portion, wherein the characteristic impedance of the second line portion is lower than the characteristic impedance of the first line portion, wherein the characteristic impedance of the third line portion is higher than the characteristic impedance of the first line portion, and The characteristic impedance of the fourth line portion is lower than the characteristic impedance of the first line portion.
13. A printed wiring board having an electrical component mounted thereon, characterized in that: The printed wiring board includes signal lines connected to signal terminals of the electrical components, The signal line includes a first line portion, a second line portion, a third line portion and a fourth line portion which are formed successively in sequence. wherein the signal terminal is configured to be joined to the fourth line portion, The line width of the second line portion is greater than the line width of the first line portion. wherein the line width of the third line portion is smaller than the line width of the first line portion, and The line width of the fourth line portion is greater than the line width of the first line portion.
14. An electronic device, characterized in that: include: case; as well as A printed circuit board according to any one of claims 1 to 11, wherein the printed circuit board is disposed in a housing.
