Effect of Fiber Weave Structure in Printed Circuit Boards on Signal Transmission Characteristics
Round 1
Reviewer 1 Report
My judgement of this paper is very ambiguous. Finally, after having understood what the paper was about, I found its content interesting. But it took me a while to understand that in the paper the signal transmission was not in the glass fibers and that the glass fibers are a building element of printed circuits.
In my opinion the authors must find another title with less ambiguity and with a clear hint to the application which is in the field of electric circuits.
This holds also for the abstract and the introduction.
The introduction is also the best place to give some more details on the structure of the boards and on the peculiarities of using 2 different materials.
Once these points are clearly explained, the readers can more easily follow the subsequent parts of the paper.
In the methods (2.2) the authors should state more clearly if the set-up presented in Fig. 1 is the only variant considered (of course it is not, but then indicate the other variats also graphically). Here it would be valuable for a general understanding to have an indication of the typical distances between the metallic lines and the modulated regions.
The results presented in section 3 are interesting but their value can be even higher if they can be associated to clear experimental conditions (absolute distances of the fibers, periodicity).
Since the discussion is interwoven with the experimental results, the most important aspects should be recalled in the conclusions, which, in my opinion, should be given some more space.
Author Response
Response to Reviewer 1 Comments
Point 1: In my opinion the authors must find another title with less ambiguity and with a clear hint to the application which is in the field of electric circuits. This holds also for the abstract and the introduction. The introduction is also the best place to give some more details on the structure of the boards and on the peculiarities of using 2 different materials. Once these points are clearly explained, the readers can more easily follow the subsequent parts of the paper.
Response 1: We accept comments from reviewers, The title we changed to " Effect of fiber Weave structure in printed circuit board on signal transmission characteristics", and the abstract and introduction have been revised too, please refer to the following words(The underlined part is where the changes were made) .Thank you!
Abstract：In this paper we characterize and compare signal transmission performance of traces with different specifications of fiber weave. Measurements demonstrate that dielectric constant, impedance fluctuation and differential skew are all affected by fiber weave style. For flatted fiber weave, the dielectric constant fluctuation will reach 0.18 , impedance fluctuation amplitude is 1.0ohm and differential skew is 2ps/inch, while conventional fiber weave the three parameters are 0.44, 2.5ohm, 4 ps/inch. The flatted fiber weave is more favorable for high-speed signal control. We also discussed the other methods to improve fiber weave effect, it is revealed that NE-glass fiber weave also has better performance in reducing impedance fluctuation and differential skew. Furthermore, making signal traces and fiber weave bundles with an angle or design the long signal line paralleled to the weft direction is a simple and effective method to solve this problem.
1. Introduction
It is known that in today’s printed circuit board laminates is a plate-like material made of electronic glass fiber cloth, impregnated with epoxy resin, covered with copper foil of a certain thickness on one or both sides, and then hot pressed. The fiber weave and resin have relative permittivity’s (Dielectric Constant, εr or Dk) of ～6 and ～3, respectively, presenting a non-homogeneous medium for signal transmission. The performance of differential transmission traces is dependent upon the underlying material properties of laminates used [1]-[7]. At high edge rate, especially transmission rate exceeded 25 Gbps, impact of fiber weave effect on high speed signal transmission becomes non-negligible. For 25Gbps transmission rate or coming 56Gbps, fiber weave effect continues to challenge the PCB industry to provide novel solution. The research for impact of fiber weave effect on ≥25 Gbps high speed materials becomes more meaningful.
In this work, dielectric constant Fluctuation of different types of fiber weave are studied by vector network analyzer. Then the impact of dielectric constant fluctuation on the fluctuation of single-ended line and the signal propagation delay of differential line are discussed. Furthermore, we also discussed the effect of trace direction on impedance fluctuation. Finally, we investigate different methods to improve the fiber weave effect. This paper allows high-speed digital designers to have a more-depth assessment of fiber weave effect.
Point 2: In the methods (2.2) the authors should state more clearly if the set-up presented in Fig. 1 is the only variant considered (of course it is not, but then indicate the other variats also graphically). Here it would be valuable for a general understanding to have an indication of the typical distances between the metallic lines and the modulated regions.
Response 2: We accept comments from reviewers, please refer to the following words. Thank you!
The graphic design for εr test is as follows, X is the width of fiber weave bundle, and x0 is the transmission line width. most of x0 on the circuit board is below 10mil at present, significantly less than most of the size of the fiber weave (see table 1, x2 and y2), and this will cause the transmission line directly above the warp/weft fiber weave or in the middle of two fiber weave. That will cause εr vary with the location of transmission line. From table 1, we can see the space of fiber weave are (x3-x2) and (y3-y2), most space are below 10mil （0.4mm）， we can design the space of single-ended microstrip lines to X+0.4mil, so as to obtain the single-ended impedance across different positions of the fiber weave.
Point 3: The results presented in section 3 are interesting but their value can be even higher if they can be associated to clear experimental conditions (absolute distances of the fibers, periodicity).
Response 3: We accept comments from reviewers, according to our measured values, the difference is quite large indeed, please refer to the following table. Thank you!
Point 4: Since the discussion is interwoven with the experimental results, the most important aspects should be recalled in the conclusions, which, in my opinion, should be given some more space.
Response 4: We accept comments from reviewers, and the discussion please refer to the following words. Thank you!
Conclusions
The dielectric constant fluctuation of different types of fiber weave are inconsistent, 1080, 3313 and 2116 are much larger than that of 1035 and 1078 (1080 is 0.44 and 1078 is 0.18). This fluctuation characteristic will fed back to the effect on impedance and signal propagation delay. 1035 and 1078 have relatively lower impedance fluctuation which is below 1.0 ohm for 50 ohm designed trace(1080, 3313 and 2116 can reach 2.0-2.5 ohm) , and lower differential skew with various differential transmission lines(1078 is 2 ps/inch and 1080 can reach 4.0 ps/inch).
For the methods to improve fiber weave effect, we advised high-speed digital designers to use flatted fiber weave (like 1078 and 1035) and NE-glass fiber weave（like 2116NE) . We also suggest to design the long signal line paralleled to the weft direction or rotates the image with an proper angle( if don't know the fiber weave pitch, 5o is recommend).
Author Response File: 
Author Response.docx
Reviewer 2 Report
This is an interesting work concerning the impact of fiber weave effect on signal transmission. However, the theoretical part of manuscript should be enriched and the authors should explain better the novelty of their work. More specifically, in order to increase the impact of their manuscript, the authors should explain better the following issues:
1. The authors should explain better the novelty of their work especially in sections, “Abstract”, “Introduction” and “Conclusions”.
2. The outcomes which appear in Figures 7 and 8 should be explained much better in the text. Additionally, the Figures 7, 8 and 14 should be redrawn in order to improve their visibility.
3. The theoretical part of the manuscript is very poor. The authors should explain more detailed the methods which are used and their novelty, comparing with previous works.
4. The authors could add a few more recent and significant works in the section “References”.
Author Response
Response to Reviewer 2 Comments
Point 1: The authors should explain better the novelty of their work especially in sections, “Abstract”, “Introduction” and “Conclusions”.
Response 1: We accept comments from reviewers, the abstract ,introduction and conclusion all have been revised, please refer to the following words(The underlined part is where the changes were made) .Thank you!
Abstract：In this paper we characterize and compare signal transmission performance of traces with different specifications of fiber weave. Measurements demonstrate that dielectric constant, impedance fluctuation and differential skew are all affected by fiber weave style. For flatted fiber weave, the dielectric constant fluctuation will reach 0.18 , impedance fluctuation amplitude is 1.0ohm and differential skew is 2ps/inch, while conventional fiber weave the three parameters are 0.44, 2.5ohm, 4 ps/inch. The flatted fiber weave is more favorable for high-speed signal control. We also discussed the other methods to improve fiber weave effect, it is revealed that NE-glass fiber weave also has better performance in reducing impedance fluctuation and differential skew. Furthermore, making signal traces and fiber weave bundles with an angle or design the long signal line paralleled to the weft direction is a simple and effective method to solve this problem.
1. Introduction
It is known that in today’s printed circuit board laminates is a plate-like material made of electronic glass fiber cloth, impregnated with epoxy resin, covered with copper foil of a certain thickness on one or both sides, and then hot pressed. The fiber weave and resin have relative permittivity’s (Dielectric Constant, εr or Dk) of ～6 and ～3, respectively, presenting a non-homogeneous medium for signal transmission. The performance of differential transmission traces is dependent upon the underlying material properties of laminates used [1]-[7]. At high edge rate, especially transmission rate exceeded 25 Gbps, impact of fiber weave effect on high speed signal transmission becomes non-negligible. For 25Gbps transmission rate or coming 56Gbps, fiber weave effect continues to challenge the PCB industry to provide novel solution. The research for impact of fiber weave effect on ≥25 Gbps high speed materials becomes more meaningful.
In this work, dielectric constant Fluctuation of different types of fiber weave are studied by vector network analyzer. Then the impact of dielectric constant fluctuation on the fluctuation of single-ended line and the signal propagation delay of differential line are discussed. Furthermore, we also discussed the effect of trace direction on impedance fluctuation. Finally, we investigate different methods to improve the fiber weave effect. This paper allows high-speed digital designers to have a more-depth assessment of fiber weave effect.
4.Conclusions
The dielectric constant fluctuation of different types of fiber weave are inconsistent, 1080, 3313 and 2116 are much larger than that of 1035 and 1078 (1080 is 0.44 and 1078 is 0.18). This fluctuation characteristic will fed back to the effect on impedance and signal propagation delay. 1035 and 1078 have relatively lower impedance fluctuation which is below 1.0 ohm for 50 ohm designed trace(1080, 3313 and 2116 can reach 2.0-2.5 ohm) , and lower differential skew with various differential transmission lines(1078 is 2 ps/inch and 1080 can reach 4.0 ps/inch).
For the methods to improve fiber weave effect, we advised high-speed digital designers to use flatted fiber weave (like 1078 and 1035) and NE-glass fiber weave（like 2116NE) . We also suggest to design the long signal line paralleled to the weft direction or rotates the image with an proper angle( if don't know the fiber weave pitch, 5o is recommend).
Point 2: The outcomes which appear in Figures 7 and 8 should be explained much better in the text. Additionally, the Figures 7, 8 and 14 should be redrawn in order to improve their visibility.
Response 2: We accept comments from reviewers, please refer to the following pictures, And we added table 3 to explain the variation between weft and warp direction. Thank you!
Point 3: The theoretical part of the manuscript is very poor. The authors should explain more detailed the methods which are used and their novelty, comparing with previous works.
Response 3: We accept comments from reviewers, we adjusted the structure of "Results and discussion", and added some more detail explain for the results. The following is the catalogue of the part of "Results and discussion", some more detail explain please refer to the revision 1.0 of the article. Thank you!
"3. Results and discussion
3.1. Dielectric constant Fluctuation of different type of fiber weave
3.2. Impact of fiber weave effect on impedance fluctuation
3.2.1. Effect of fiber weave effect on single-ended impedance line
3.2.2 Effect of fiber weave effect on differential line
3.3. Effect of trace direction on impedance fluctuation
3.4. Methods to Improve fiber weave effect
3.4.1 Flatted fiber weave
3.4.2 NE-glass fiber weave
3.4.3 Making trace and fiber weave bundle with an Angle"
Point 4: The authors could add a few more recent and significant works in the section “References”.
Response 4: We accept comments from reviewers, and added 1 recent reference. Thank you!
13 Bucur, D. Fiber Weave Effect - a performance-limiting factor, 10th International Conference on Communications (COMM), 2014.
Author Response File: Author Response.docx
Round 2
Reviewer 1 Report
I acknowledge the changes made to the manuscript by the authors and can recommend its publication.
