Effect of Fiber Weave Structure in Printed Circuit Boards on Signal Transmission Characteristics

applied sciences

Article Effect of Fiber Weave Structure in Printed Circuit Boards on Signal Transmission Characteristics

Bei Chen 1,2, Ruohe Yao 1,*, Hongfei Wang 3, Kuiwei Geng 1 and Juan Li 2

1 School of Electronics Information Engineering, South China University of Technology, Guangzhou 510640, China; [email protected] (B.C.); [email protected] (K.G.) 2 China Fastprint Circuit Corporation, Guangzhou 510663, China; [email protected] 3 Dongguan ITEQ Corporation, Dongguan 523000, China; [email protected] * Correspondence: [email protected]; Tel.: +86-20-8711-0449

Received: 26 November 2018; Accepted: 11 January 2019; Published: 21 January 2019

Abstract: In this paper, we characterized and compared signal transmission performances of traces with different specifications of fiber weave. Measurements demonstrated that the dielectric constant, impedance fluctuation, and differential skew were all affected by fiber weave style. For flattened fiber weaves, the dielectric constant fluctuation reached 0.18, the impedance fluctuation amplitude was 1.0 Ω, and the differential skew was 2 ps/inch. For conventional fiber weaves, the three parameters were 0.44, 2.5 Ω, and 4 ps/inch respectively. Flattened fiber weave was more favorable for high-speed signal control. We also discussed the other methods to improve the fiber weave effect. It turned out that NE-glass (new electronic glass) fiber weave also had better performance in reducing impedance fluctuation and differential skew. Furthermore, made the signal traces and fiber weave bundles with an angle or designing the long signal line parallel to the weft direction both are simple and effective methods to solve this problem.

Keywords: ﬁber weave effect; high-speed; impedance ﬂuctuation; differential skew

1. Introduction It is known that in today’s printed circuit board laminates, there is a plate-like material made of electronic glass fiber cloth that is 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 relatively dielectric constant (Dk) of ~6 and ~3, respectively, presenting a nonhomogeneous medium for signal transmission. The performance of differential transmission traces is dependent upon the underlying material properties of the laminates used [1–7]. At high edge rates, especially transmission rates exceeding 25 Gbps, the impact of the fiber weave effect on high-speed signal transmission becomes non-negligible. For 25 Gbps single channel transmission rate or coming 56 Gbps, the fiber weave effect continues to challenge the PCB (short for Printed circuit board) industry to provide novel solutions. Research on the impact of fiber weave effect on ≥25 Gbps high-speed materials has become more meaningful. In this work, the dielectric constant fluctuations of different types of fiber weaves are studied using a vector network analyzer. Then the impact of the dielectric constant fluctuation on the impedance of single-ended lines and the signal propagation delay of differential lines are discussed. Furthermore, we also discuss 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 in-depth assessment of the fiber weave effect.

Appl. Sci. 2019, 9, 353; doi:10.3390/app9020353 www.mdpi.com/journal/applsci Appl. Sci. 2019, 9, 353 2 of 10

2. Materials and Methods

2.1. Materials and Equipment We focused on the ﬁber weave effect only for high-speed PCBs, so the research must be based on high-speed PCBs. Here, high-speed laminates and prepreg based on PPO (short for Polyphenylene Oxide) resins that were supplied by Panasonic were used to produce multilayer PCB. A vector network analyzer (VNA, E5071C) from Agilent was used to measure TDR (short for time domain reﬂectometry), the S-parameter, and phase difference. To ensure the accuracy of the measurements, the tested frequency bandwidth was 20 GHz, and the rise time was 22.3 ps.

2.2. Methods In the design of the tested multilayer board, the thicknesses of the laminate cores were 0.08 mm to 0.20 mm, and different styles of prepreg were used. The style of the ﬁber weaves contained prepreg 1035, prepreg 1080, prepreg 3313, prepreg 2116, and prepreg 1078.

The graphic design for the Dk test show as Figure1: X was the width of the fiber weave bundle, Articleand x 0 was the transmission line width on the circuit board. Most transmission line were designed <10 mil, significantly less than most of the sizes of the fiber weaves (see Table1, X2 and Y2). This would Effectcause the transmissionof Fiber lines Weave directly above Structure the warp or weft in fiber Printed weave or in Circuit the middle of two fiber weaves (see Figure1). Dk vary with the location of the transmission line. From Table1, we can see Boardsthat most spaces on ofSignal the fiber weaves Transmission (the value of (X3 – X 2)Characteristics or (Y3 – Y2) in Table1) <10 mil (0.4 mm), so we could design the spaces of single-ended microstrip lines to X + 0.4 mil to obtain the single-ended 1,2 1, 3 1 2 Beiimpedance Chen , Ruohe across differentYao *, Hongfei positions Wang of the , fiberKuiwei weaves. Geng and Juan Li

Figure 1. The graphic design of the Dk test. Figure 1. The graphic design of the Dk test. We also designed differential microstrip lines (the other type of impedance line, other than a We also designed differential microstrip lines (the other type of impedance line, other than a single-ended line) with different spaces to discuss impedance fluctuations influenced by the fiber single-ended line) with different spaces to discuss impedance fluctuations influenced by the fiber weave effect. weave effect. The tested boards were produced by the following procedure: CCL (short for copper-clad The tested boards were produced by the following procedure: CCL (short for copper-clad laminate) cutting → inner layer process → laminating → drilling → PTH (short for plating through laminate) cutting → inner layer process → laminating → drilling → PTH（short for plating through hole) → electronic plating → back drilling → outer layer process → solder mask → component mark hole)→ surface → electronic finishing plating→ test. → back drilling → outer layer process → solder mask → component mark → surface finishing → test. 3. Results and Discussion 3. Results and Discussion 3.1. Dielectric Constant Fluctuations of Different Types of Fiber Weaves 3.1. Dielectric Constant Fluctuations of Different Types of Fiber Weaves Impedance fluctuations can be attributed to local variations in the Dk and the structural integrity of theImpedance stack-up, fluctuations both of which can dependbe attributed on the to laminate local variations fiber weave. in the TheDk and dielectric the structural constant integrity (Dk) is a ofkey the parameterstack-up, both of impedance, of which depend and its on fluctuation the laminate significantly fiber weave. affects The impedance. dielectric constant Different (Dk) types is a of keyfiber parameter weaves showof impedance, different characteristics.and its fluctuation significantly affects impedance. Different types of fiber weavesFor any show particular different stack-up characteristics. configuration, there are two worst-case effective Dk extremes that occur,For theany highest particular and stack-up the lowest configuration, Dk. When all there yarn are strands two worst-case align with eacheffective other, Dk and extremes the trace that lies occur,directly the overhighest the and center the of lowest a low Dk. resin, When a “glass all yarn peak” strands resin, align the highest with each Dk isother, gotten. and When the trace the lies trace directlylies directly over the over cen the middle of a high resin, a “glass valley” region, the lowest Dk is gotten. Due to the

Appl. Sci. 2019, 9, Firstpage-Lastpage; doi: FOR PEER REVIEW www.mdpi.com/journal/applsci Appl. Sci. 2019, 9 FOR PEER REVIEW 2 Appl. Sci. 2019, 9, 353 3 of 10

ter of a low resin, a “glass peak” resin, the highest Dk is gotten. When the trace lies directly over thetrace middle passing of acrossa high fiberresin, weave a “glass bundles, valley” like region, the feelingthe lowest of crossing Dk is gotten. railroad Due tracks to the diagonally trace passing in a acrosscar (see fiber Figure weave2)[ 2bundles,], the trace like exhibits the feeling a cyclical of crossing variation railroad in Dk. tracks For differentdiagonally styles in a ofcar fiber (see weaves, Figure 2)fiber [2], weavethe trace bundle exhibits dimensions a cyclical are variation different, in includingDk. For different a fiber weavestyles of bundle’s fiber weaves, width fiber (X2 and weaveY2) bundleand pitch dimensions size (X3 and are Ydifferent,3) (see Figure including3 and a Tablefiber1 weave). X represents bundle’s width warp direction,(X2 and Y and2) andY representspitch size (weftX3 and direction. Y3) (see Figure 3 and Table 1). X represents warp direction, and Y represents weft direction.

Appl. Sci. 2019, 9 FOR PEER REVIEW 3

FigureFigure 2. 2. TheThe Dk Dk variation variation due due to to fiber ﬁber weave. weave.

FigureFigure 3. 3. PhotosPhotos to to illustrate illustrate fibe ﬁberr weave weave bundle bundle parameters. parameters.

Table 1. Measurement results of ﬁber weave bundle parameters. Table 1. Measurement results of fiber weave bundle parameters.

MeasurementMeasurement Results Results (mil) Style Style X1 X1 X2X2 X3 Y1Y 1 Y2 Y2Y3 Y3 10351035 0.82 0.82 8.88.8 14.2 0.78 0.78 12.4 12.413.7 13.7 10801080 1.6 1.6 8.2 8.2 17.0 1.1 1.1 12.1 12.122.4 22.4 1078 1.4 14.2 16.2 1.0 17.6 17.8 1078 1.4 14.2 16.2 1.0 17.6 17.8 3313 1.9 13.1 16.2 1.5 11.0 16.3 21163313 2.2 1.9 14.113.1 16.2 17.2 1.5 2.0 11.0 15.516.3 17.3 2116 2.2 14.1 17.2 2.0 15.5 17.3

Appl. Sci.With 2019 the, 9 FOR parameters PEER REVIEW of X and Y, we could design different graphs to test Dk fluctuation amplitude4 With the parameters of X and Y, we could design different graphs to test Dk fluctuation for different styles of fiber weave. The fluctuation amplitude of Dk is shown in Figures4 and5. It was amplitude amplitudes for different of 1080, styles 3313, of and fiber 2116 weave. were The much fluctuation larger than amplitude those of of 1035 Dk andis shown 1078. inComparing Figures 4 found that the Dk fluctuation amplitudes of 1080, 3313, and 2116 were much larger than those of 1035 andwarp 5. direction It was found to weft that direction, the Dk fluctuation it was found that the Dk fluctuation amplitude of weft direction was and 1078. Comparing warp direction to weft direction, it was found that the Dk fluctuation amplitude relatively lower than that of warp direction. of weft direction was relatively lower than that of warp direction.

Parallel to warp direction 3.7 3.6

3.5 1035 1080 3.4 3313 k

3.3 2116

D 1078 3.2 3.1 3.0 2.9 0 5 10 15 20 25 30 35 X direction /mil

Figure 4. The Dk ﬂuctuation amplitude parallel to warp direction. Figure 4. The Dk fluctuation amplitude parallel to warp direction.

Appl. Sci. 2019, 9 FOR PEER REVIEW 5

Appl. Sci. 2019, 9, 353 4 of 10

Parallel to weft direction 3.7 3.6 3.5 1035 1080

k 3.4 3313 D 3.3 2116 1078 3.2 3.1 3.0 2.9 Appl. Sci. 2019, 9 FOR PEER REVIEW 0 5 10 15 20 25 30 35 6 Y direction /mil

Figure 5. The Dk ﬂuctuation amplitude parallel to weft direction. Figure 5. The Dk fluctuation amplitude parallel to weft direction. 3.2.3.2. Impact Impact of of the the Fiber Fiber Weave Weave Effect Effect on on Impedance Impedance Fluctuations Fluctuations

ThereThere are are two two main main types types of oftransmission transmission lines, lines, differential differential lines lines and andsingle-ended single-ended lines. lines.The influenceThe inﬂuence mechanism mechanism of the of fiber the ﬁberweave weave effect effect on different on different impedance impedance types types is different. is different.

3.2.1.3.2.1. Effect Effect of of the the Fiber Fiber Weave Weave Ef Effectfect on on Single-Ended Single-Ended Impedance Impedance Lines Single-endedSingle-ended impedance impedance is is affected affected by by location location distribution. distribution. Thus, Thus, when when the the Dk Dk exhibited exhibited a a cyclicalcyclical variation variation with with the the location location distribution, distribution, it it was was reflected reflected on on the the single-ended single-ended impedance impedance TDR TDR curves.curves. The The effect effect of of fiber fiber weave weave style style on on single-en single-endedded impedance impedance fluctuation fluctuation is shown is shown in Figure in Figure 6. It6. wasIt was found found that that 1035 1035 and and1078 1078 had hadrelatively relatively lower lower impedance impedance fluctuation fluctuation amplitudes, amplitudes, whereas whereas 1080, 3313,1080, and 3313, 2116 and had 2116 relatively had relatively higher higherimpedance impedance fluctuation fluctuation amplitudes. amplitudes. Here, 1078 Here, and 1078 1035 and had 1035 the lowesthad the impedance lowest impedance fluctuation fluctuation amplitudes, amplitudes, below 1.0 below Ω for 1.0 50Ω Ωfor designed 50 Ω designed traces, traces,whereas whereas for other for stylesother stylesof fiber of fiberweave, weave, impedance impedance fluctuation fluctuation amplitudes amplitudes could could reach reach 2.0–2.5 2.0–2.5 ΩΩ. .After After flattening flattening treatment,treatment, the the medium medium homogeneity homogeneity of of 1078 1078 and and 1035 1035 was was better better than than the the traditional traditional style style of of fiber fiber weave.weave. The The test test results results were were consis consistenttent with with theoretical theoretical calculations. calculations.

FigureFigure 6. 6. EffectEffect of of fiber ﬁber weave weave style style on on impedance impedance fluctuation. ﬂuctuation.

3.2.2.3.2.2. Effect Effect of of the the Fiber Fiber Weav Weavee Effect Effect on on Differential Differential Lines DifferentialDifferential lines lines are are shown shown in in Figure Figure 7 and were mademade upup ofof twotwo halveshalves ofof differentialdifferential pairspairs (designated(designated “D+”“D+” and and “D −“D”).−”). They They ran overran differentover different media (resin media vs fiber(resin weave) vs (seefiber correspondingly weave) (see correspondinglydifferent Dk values) different and had Dk differentvalues) and propagation had differe properties,nt propagation especially properties, velocity. especially It can be concluded velocity. Itthat can the be signalconcluded propagation that the velocities signal propagation of the two halvesvelocities of the of differentialthe two halves pairs of in the Figure differential7 were different, pairs in Figurewhich 7 resulted were different, in impedance which fluctuationsresulted in impedance of differential fluctuations lines. We of candifferential also call lines. these We impedance can also callfluctuations these impedance differential fluctuations skew. differential skew.

Appl. Sci. 2019, 9 FOR PEER REVIEW 7 Appl. Sci. 2019, 9, 353 5 of 10

FigureFigure 7. 7. PhotosPhotos to to illustrate illustrate differential differential lines.

The calculation method of differential skew is shown in Equation (1) [8]. It is the two lines’ The calculation method of differential skew is shown in Equation (1) [8]. It is the two lines’ propagation delay (tpd1 is the propagation delay of one line of the differential pairs, and tpd2 is propagation delay (tpd1 is the propagation delay of one line of the differential pairs, and tpd2 is another line): another line): Skew = |tpd1 − tpd2|. (1) Skew = |tpd1 − tpd2|. (1) For a microstrip differential line, the propagation delay (tdp) can be calculated using EquationFor a (2)microstrip [2]: differential line, the propagation delay (tdp) can be calculated using Equation (2) [2]: tpd = 85*(0.475Dk + 0.67)1/2. (2)

In Equation (2), Dk is the mediumtpd dielectric = 85*(0.475Dk permittivity, + 0.67)1/2. and tpd is strongly related to Dk. (2) Appl. InSci.Because Equation 2019, 9 theFOR (2), laminatePEER Dk REVIEW is the was medium not a homogeneous dielectric permittivity, medium, whenand tpd a is transmission strongly related line walkedto Dk. in a8 differentBecause place, the the laminate effective was dielectric not a permittivityhomogeneous of themedium, medium when was a different. transmission The medium line walked dielectric in a rent. The medium dielectric permittivity (Dk) for a specific transmission line (see Figure 8) could differentpermittivity place, (Dk) the for effective a specific dielectric transmission permittivity line (see of Figurethe medium8) could was be differ calculated using Equation (3), be calculated using Equation (3), where εG is the dielectric permittivity of the fiber weave, Dk is the where εG is the dielectric permittivity of the fiber weave, Dk is the dielectric permittivity of the resin, dielectric permittivity of the resin, and H is the total thickness of the fiber weave and resin: and H is the total thickness of the fiber weave and resin:

ε = {0.5*εG*W*(H1 + H2)*0.6 + Dk*W*[H − 0.3*(H1 + H2)]}/W/H ε = {0.5*ε *W*(H1 + H2)*0.6 + Dk*W*[H − 0.3*(H1 + H2)]}/W/H G (3)(3) == {0.3* {0.3*εGεG*(H1*(H1 + + H2)H2) ++ Dk*[HDk*[H −− 0.3*(H10.3*(H1 + H2)]}/H. H2)]}/H.

Figure 8. Photos to illustrate the Dk calculation.

According to Equations (1)–(3), we could calculate the maximal differential skew for differentdifferent styles of single fiberfiber weaves with various differentialdifferential transmissiontransmission lines (see Table2 2).). As As can can be be seen seen fromfrom thethe table,table, forfor some designed transmissiontransmission lines,lines, thethe lineline waswas bigbig enough to have a fatal influence influence on thethe high-speedhigh-speed signal signal transmission. transmission. In In addition, addition skew, skew values values were were influenced influenced by the by designed the designed line’s widthline’s width and space and(W/S). space (W/S). It was foundIt wasthat found theoretical that theore linetical values line were values close were to zero close in to special zero situations.in special Thissituations. happened This whenhappened (W+S) when equal (W+S) to X3 equal or Y3. to ThisX3 or is Y3. attributed This is attributed to the fact to that the the fact two that lines the oftwo a differentiallines of a differential pair had thepair same had the relative same location relative tolocation fiber weave to fiber bundles weave underbundles this under situation, this situation, and the effectiveand the effective Dk of the Dk two of linesthe two was lines the was same. the Since same. it wasSince difficult it was difficult for designers for designers to obtain to detail obtain data detail of thedata fiber of the weave fiber and weave other and functional other functional design requirements, design requirements, this method this wouldmethod be would difficult be todifficult realize. to realize. Table 2. Maximal differential skew for different styles of fiber weave.

Table 2. Maximal differentialDifferential skew for di Skewfferent (ps/in) styles of fiber weave. W/S (mil) 1035 1080Differential 3313 Skew (ps/in) 2116 1078 W/S (mil)Weft Warp1035 Weft1080 Warp Weft3313 Warp Weft2116 Warp Weft1078 Warp 4/4 4.7Weft 3.5 Warp 7.2 Weft 5.1 Warp 6.3 Weft 3.1Warp Weft6.4 Warp 4.6 Weft 2.8 Warp 1.9 4/84/4 0.9 4.7 0.43.5 6.57.2 5.15.1 6.56.3 1.33.1 2.66.4 3.64.6 2.22.8 1.9 1.6 4/8 0.9 0.4 6.5 5.1 6.5 1.3 2.6 3.6 2.2 1.6 5/5 4.3 2.3 6.6 4.7 7.3 5.6 5.1 4.8 2.7 2.0 5/9 0.4 0.3 5.0 4.7 3.0 1.8 1.0 1.2 1.2 1.1 6/6 0.9 0.4 5.2 4.6 5.7 4.1 3.4 3.3 2.3 1.5 6/8 0.0 0.0 2.8 4.3 3.5 2.2 1.2 1.4 1.5 0.8 7/7 0.0 0.0 1.7 4.0 2.7 2.6 0.6 1.1 1.0 0.7 7/9 0.4 0.3 0.8 3.9 0.8 0.8 0.1 0.1 0.2 0.1 8/8 0.7 0.4 0.4 3.1 0.4 0.9 0.0 0.3 0.1 0.3 8/10 1.87 1.47 0.3 1.9 0.0 0.5 0.8 0.0 0.1 0.0

3.3. Effects of Trace Direction on Impedance Fluctuation Figures 9 and 10 show the effects of trace direction on impedance fluctuation for the 1035 fiber weave. As can be seen from the tested TDR response, the impedance fluctuation amplitude of the warp direction was relatively higher than that of the weft direction. This can be explained with the data in Tables 1 and 3. In the weft direction, the fiber weave bundles were thinner and better flattened, and the medium homogeneity was better. Thus, for high-speed PCB designing and manufacturing, especially 25-Gbps backplanes, long signal traces should be parallel to the weft direction [8,9]. Appl. Sci. 2019, 9, 353 6 of 10

Table 2. Cont.

Differential Skew (ps/in) W/S (mil) 1035 1080 3313 2116 1078 Weft Warp Weft Warp Weft Warp Weft Warp Weft Warp 5/5 4.3 2.3 6.6 4.7 7.3 5.6 5.1 4.8 2.7 2.0 5/9 0.4 0.3 5.0 4.7 3.0 1.8 1.0 1.2 1.2 1.1 6/6 0.9 0.4 5.2 4.6 5.7 4.1 3.4 3.3 2.3 1.5 6/8 0.0 0.0 2.8 4.3 3.5 2.2 1.2 1.4 1.5 0.8 7/7 0.0 0.0 1.7 4.0 2.7 2.6 0.6 1.1 1.0 0.7 7/9 0.4 0.3 0.8 3.9 0.8 0.8 0.1 0.1 0.2 0.1 8/8 0.7 0.4 0.4 3.1 0.4 0.9 0.0 0.3 0.1 0.3 8/10 1.87 1.47 0.3 1.9 0.0 0.5 0.8 0.0 0.1 0.0

3.3. Effects of Trace Direction on Impedance Fluctuation Figures9 and 10 show the effects of trace direction on impedance fluctuation for the 1035 fiber weave. As can be seen from the tested TDR response, the impedance fluctuation amplitude of the warp direction was relatively higher than that of the weft direction. This can be explained with the Appl. Sci. 2019, 9 FOR PEER REVIEW 9 data in Tables1 and3. In the weft direction, the fiber weave bundles were thinner and better flattened, andAppl. the Sci. medium 2019, 9 FOR homogeneity PEER REVIEW was better. Thus, for high-speed PCB designing and manufacturing,9 especially 25-Gbps backplanes, long signal traces should be parallel to the weft direction [8,9]. 62 parallel to 经向平行走线the warp direction Impedance fluctuation(Z0)/ 60 62 parallel to 经向平行走线the warp direction Impedance fluctuation(Z0)/ 58 60 56 58

/ohm B 0 54

Z 56 C D /ohm 52 B 0 54 E Z CF 50 DG 52 E F Ω 48 50 G

0.0 500.0p 1.0n 1.5n 2.0n Ω 48 Time/s

0.0 500.0p 1.0n 1.5n 2.0n

Time/s Figure 9. Impedance fluctuation of traces parallel to the warp direction. Figure 9. Impedance ﬂuctuation of traces parallel to the warp direction. Figure 9. Impedance fluctuation of traces parallel to the warp direction. parallel to the weft direction Impedance fluctuation(Z0)/ 62 纬向平行走线 60 parallel to the weft direction Impedance fluctuation(Z0)/ 62 纬向平行走线 6058 56 58

/ohm B 0 54

Z 56 C D /ohm 52 B 0 54 E Z CF 50 DG Ω 52 E F

48 50 G Ω 0.0 500.0p 1.0n 1.5n 2.0n

48 Time/s 0.0 500.0p 1.0n 1.5n 2.0n

Figure 10. Impedance ﬂuctuation of tracesTime/s parallel to the weft direction. Figure 10. Impedance fluctuation of traces parallel to the weft direction.

Figure 10. Impedance fluctuation of traces parallel to the weft direction.

Table 3. Variations in warp and weft directio ns of different fiber weave types.

Warp Direction Weft Direction Table 3. VariationsFiber Weave in warp Style and weft directions of different fiber weave types. ∆Dk ∆Z0/Ω ∆Dk ∆Z0/Ω Warp Direction Weft Direction ConventionalFiber Weave fiber Style weave 0.58 3.80 0.44 2.73 ∆Dk ∆Z0/Ω ∆Dk ∆Z0/Ω Flattened fiber weave 0.25 1.56 0.18 1.11 Conventional fiber weave 0.58 3.80 0.44 2.73 3.4. Methods to ImproveFlattened Fiber Weave fiber Effects weave 0.25 1.56 0.18 1.11

3.4.3.4.1. Methods Flattened to Improve Fiber Weaves Fiber Weave Effects

3.4.1. FlattenedFlattened Fiberfiber Weavesweave is a type of fiber that is treated by open-filament or flatten-weave technology. Figure 11 shows the differences in the two styles of fiber weave: Compared to regular glassFlattened fiber cloth, fiber flattened weave fiber is a weave type ofis morefiber dispthatersed, is treated the dielectric by open-filament uniformity oris better,flatten-weave and the technology.difference in Figure the dielectric 11 shows constant the differences between thein th differentiale two styles lines of fiberbecomes weave: smaller. Compared to regular glass Thefiber properties cloth, flattened of fiber fiber weaves weave could is more be expressed dispersed, by the a phase dielectric differential uniformity (∆Phase), is better, which and is thethe differencepropagation in thedelay dielectric at different constant transmission between frequencthe differentialies (see lines Equation becomes (4)), smaller.and we could use a vector networkThe propertiesanalyzer to of detect fiber ∆weavesPhase. could be expressed by a phase differential (∆Phase), which is the propagationThe ∆Phase delay of at 1078 different (a type transmission of flattened frequencfiber weave)ies (see was Equation much lower (4)), thanand wethat could of 1080 use (a a typevector of networknormal fiber analyzer weave) to detect(Figure ∆ Phase.12). This means that a flattened fiber weave can improve a PCB’s signal The ∆Phase of 1078 (a type of flattened fiber weave) was much lower than that of 1080 (a type of normal fiber weave) (Figure 12). This means that a flattened fiber weave can improve a PCB’s signal Appl. Sci. 2019, 9, 353 7 of 10

Table 3. Variations in warp and weft directions of different ﬁber weave types.

Warp Direction Weft Direction Fiber Weave Style ∆Dk ∆Z0/Ω ∆Dk ∆Z0/Ω Conventional ﬁber weave 0.58 3.80 0.44 2.73 Flattened ﬁber weave 0.25 1.56 0.18 1.11

3.4. Methods to Improve Fiber Weave Effects

3.4.1. Flattened Fiber Weaves Flattened fiber weave is a type of fiber that is treated by open-filament or flatten-weave technology. Figure 11 shows the differences in the two styles of fiber weave: Compared to regular glass fiber cloth, flattened fiber weave is more dispersed, the dielectric uniformity is better, and the difference in the dielectric constant between the differential lines becomes smaller. The properties of fiber weaves could be expressed by a phase differential (∆Phase), which is the propagation delay at different transmission frequencies (see Equation (4)), and we could use a vector Appl. Sci. 2019, 9 FOR PEER REVIEW 10 network analyzer to detect ∆Phase. The ∆Phase of 1078 (a type of flattened fiber weave) was much lower than that of 1080 (a type of transmission quality. According to Equation (4), the skew values of the 1080 and 1078 fiber weaves normal fiber weave) (Figure 12). This means that a flattened fiber weave can improve a PCB’s signal can be obtained (4 ps/in. and 2 ps/in., respectively). Thus, for a 25-Gbps PCB, especially a backplane, transmission quality. According to Equation (4), the skew values of the 1080 and 1078 fiber weaves a flattened fiber weave is often used for stack-up. In recent years, flattened fiber weaves have been can be obtained (4 ps/in. and 2 ps/in., respectively). Thus, for a 25-Gbps PCB, especially a backplane, widely used in high-speed laminates, and traditional 106 and 1080 fiber weaves have been abandoned a flattened fiber weave is often used for stack-up. In recent years, flattened fiber weaves have been in the Megtron 7. widely used in high-speed laminates, and traditional 106 and 1080 fiber weaves have been abandoned in the Megtron 7. ∆∆PhasePhase == Skew* Frequency *360. (4) (4)

Appl. Sci. 2019, 9 FOR PEER REVIEW 11

Figure 11. The structure of regular weaves and ﬂattenedflattened ﬁberfiber weaves.

Figure 12. ∆PhasePhase of of regular regular and and flattened ﬂattened fiber ﬁber weaves.

3.4.2. NE-Glass Fiber Weaves NE-glass (new electronic glass), a product of Nittobo, shows improved electrical and mechanical performance over traditional E-glass and is widely used in some low-loss laminate products (for example, it was used in Nelco's N4000-13 SI and Meterowave 2000, TUC's TU872 SLK SP, and Panasonic's Megtron 6 NE and Megtron 7NE). NE-glass shows better temperature stability, lower Dk, and lower loss than the equivalent E-glass. The reduced Dk of NE-glass yields important benefits for high-performance signaling applications [10]. Materials that using NE-glass have less variation between the Dk of resin. With a ratio of glass/PPO resin as 46:54, we could calculate the nominal laminate Dk to be 3.6 with E-glass, but with NE-glass would be 3.35. Figure 13 shows the impedance fluctuation (Z0) of traces routed on NE-glass. When compared to Figures 7 and 8, it was found that traces routed on NE-glass showed much lower impedance fluctuations (Z0). Figure 14 shows phase differentials for differential pairs produced with NE-glass and E-glass. The results indicated that the ∆Phase of NE-glass was much lower than that of E-glass. Due to its excellent performance, NE-glass has been widely used in low-loss, very low-loss, and ultralow-loss laminates.

Appl. Sci. 2019, 9, 353 8 of 10

3.4.2. NE-Glass Fiber Weaves NE-glass (new electronic glass), a product of Nittobo, shows improved electrical and mechanical performance over traditional E-glass and is widely used in some low-loss laminate products (for example, it was used in Nelco’s N4000-13 SI and Meterowave 2000, TUC’s TU872 SLK SP, and Panasonic’s Megtron 6 NE and Megtron 7NE). NE-glass shows better temperature stability, lower Dk, and lower loss than the equivalent E-glass. The reduced Dk of NE-glass yields important benefits for high-performance signaling applications [10]. Materials that using NE-glass have less variation between the Dk of resin. With a ratio of glass/PPO resin as 46:54, we could calculate the nominal laminate Dk to be 3.6 with E-glass, but with NE-glass would be 3.35. Figure 13 shows the impedance fluctuation (Z0) of traces routed on NE-glass. When compared to Figures7 and8, it was found that traces routed on NE-glass showed much lower impedance fluctuations (Z0). Appl. FigureSci. 2019 14, 9 FORshows PEER phase REVIEW differentials for differential pairs produced with NE-glass and E-glass. The12 results indicated that the ∆Phase of NE-glass was much lower than that of E-glass. Due to its excellent performance, NE-glass has been widely used in low-loss, very low-loss, and ultralow-loss laminates.

NE-Glass（2116）

Impedance fluctuation(Z0)/Ω Impedance 52 NE-Glass(2116)

46 /ohm

0 B

Z 44 C 42 D E F 40 G 38 Appl. Sci. 2019, 9 FOR PEER REVIEW 0.0 500.0p 1.0n 1.5n 2.0n 13 Time/s

Figure 13. Impedance ﬂuctuation of traces on NE-glass. Figure 13. Impedance fluctuation of traces on NE-glass. 20 20 E-Glass(2116) NE-Glass(2116)

10 10

0 0 Diff. Phase Diff. Phase Diff

-10 -10

-20 -20 1.990 GHz 3.990 GHz 1.990 GHz 3.990 GHz Frequency Frequency FigureFigure 14. 14.∆ ∆PhasePhase ofof Electronic-glassElectronic-glass versusversus NE-glass.NE-glass.

3.4.3. Making Traces and Fiber Weave Bundles with an Angle 3.4.3. Making Traces and Fiber Weave Bundles with an Angle Making signal traces and fiber weave bundles with an angle is a relatively simple and effective Making signal traces and fiber weave bundles with an angle is a relatively simple and effective method to solve the differential skew problem [11,12]. There are two practical methods to achieve this: method to solve the differential skew problem [11,12]. There are two practical methods to achieve (1) The PCB manufacturer rotates the image. This reduces the utilization rate of laminates and this: increases production costs. Usually, a 10 degrees rotation angle makes production costs increase 8–10%. For customers,(1) The PCB this manufacturer is a big increase; rotates the image. This reduces the utilization rate of laminates and increases(2) The production designer costs. rotates Usually, the trace. a 10 This degrees approach rotation requires angle makes designers production to rotate costs the increase signal trace 8%– during10%. For layout. customers, This would this is increase a big increase; the difficulty of layout and affects layout efficiency. (2) The designer rotates the trace. This approach requires designers to rotate the signal trace during layout. This would increase the difficulty of layout and affects layout efficiency. Another question is what is the right angle. Here, we tested the effects of three kinds of rotation angles (5 degrees, 10 degrees, and 15 degrees). As can be seen from Figure 15, the ∆Phase results of the three kinds of rotation angles were basically identical. This means it did not take too much of an angle between trace and fiber weave bundle to resolve the fiber weave problem. Theoretically, a trace merely has to cross two weave bundles along its length, so that the effect on the two adjacent traces is equalized (see Figure 16 and Equation (5)) [1]. According to the pitch size of the fiber weave bundles and trace lengths, the rotation angle could be calculated.

Appl. Sci. 2019, 9, 353 9 of 10

Another question is what is the right angle. Here, we tested the effects of three kinds of rotation angles (5 degrees, 10 degrees, and 15 degrees). As can be seen from Figure 15, the ∆Phase results of the three kinds of rotation angles were basically identical. This means it did not take too much of an angle between trace and ﬁber weave bundle to resolve the ﬁber weave problem. Theoretically, a trace

Appl.merely Sci. has2019, to 9 FOR cross PEER two REVIEW weave bundles along its length, so that the effect on the two adjacent traces14 is equalized (see Figure 16 and Equation (5)) [1]. According to the pitch size of the ﬁber weave bundles and trace lengths, the rotation angle could be calculated.

Appl. Sci. 2019, 9 FOR PEER REVIEW 15

Figure 15. ∆PhasePhase before before and after rotating.

Figure 16. Calculation of the rotation angle.

4. Conclusions The dielectricdielectric constant fluctuationfluctuation of different typestypes ofof fiberfiber weavesweaves werewere inconsistent:inconsistent: 1080, 3313, and 21162116 areare muchmuch largerlarger thanthan 10351035 andand 10781078 (1080(1080 isis 0.440.44 andand 10781078 isis 0.18).0.18). ThisThis fluctuationfluctuation characteristiccharacteristic fed back to the effect on impedance and signal propagation delay. Here, 1035 and 1078 had relatively lower impedance fluctuations, which were below 1.0 Ω for 50 Ω designed traces (1080, 3313, and 2116 could reach 2.0–2.5 Ω), and lower differential skews with various differential transmission lines (1078 was 2 ps/inch. and 1080 could reach 4.0 ps/inch.). For methods to improve the fiber weave effect, we advise high-speed digital designers to use flattened fiber weaves (such as 1078 and 1035) and NE-glass fiber weaves (such as 2116NE). We also suggest designing long signal lines parallel to the weft direction or rotating the image with a proper angle (if the fiber weave pitch is not known, 5 degree is recommended).

Author Contributions: B.C. makes substantial contributions to conception and design, acquisition of data, analysis and interpretation of data; R.Y. gives final approval of the version to be submitted and any revised version; H.W. participates in drafting the article or revising it critically for important intellectual content; K.G. participates in drafting the article or revising it critically for important intellectual content; J.L. submit this article. Funding: This work was supported by the Science and Technology Project of Guangdong (2014B090901017) and the Science and Technology Project of Guangzhou (201604010086). Conflicts of Interest: The authors declare no conflicts of interest. Appl. Sci. 2019, 9, 353 10 of 10

References

1. Dudek, R.; Goldman, P.; Kuhn, J. Advanced Glass Reinforcement Technology for Improved Signal Integrity. Printed Circuit Design & Fab. 2008, pp. 40–42. Available online: https: //www.pcdandf.com/pcdesign/index.php/articles-by-subject/210-signal-integrity/3378-advanced- glass-reinforcement-technology-for-improved-signal-integrity (accessed on 20 January 2018). 2. Scott, M.; Chris, H. The impact of PCB laminate weave on the electrical performance of differential signaling at multi-Gigabit data rates. In Proceedings of the Design Conference, Santa Clara, CA, USA, 31 January 2005. 3. Cai, J.Q. PCB used layer insulation film and high function glass cloth Printed Circuit Information. Circuit Inf. 2011, 12, 26–31. 4. Avendano, V.; Champac, V.; Figureas, J. Signal integrity verification using high speed monitors. In Proceedings of the European Test Symposium, Corsica, France, 23–26 May 2004; pp. 114–119. 5. Loyer, J.; Kuuze, R.K.; Ye, X.N. Fiber Wave Effect: Fiber Weave Effect: Practical Impact Analysis and Mitigation Strategies. In Proceedings of the Design Conference, Santa Clara, CA, USA, 29 January–1 February 2007. 6. Sharawi, M.S. Practical Issues in High Speed PCB Design. Potentials IEEE 2004, 23, 24–27. [CrossRef] 7. Jiang, L.H.; Zhu, Y.H.; Zhang, N.; Zhuo, N.Z.; Xing, H.D.; Wang, H.B. The Research of the Low Dielectric Properties of Fiberglass and the Application of Fiberglass in High Frequency Printed circuit Board. Fiber Glass 2016, 3, 1–6. 8. Bucur, D. Fiber Weave Effect—A performance-limiting factor. In Proceedings of the International Conference on Communications, Bucharest, Romania, 29–31 May 2014; pp. 1–4. 9. Zhang, T.; Chen, X.; Schutt-Aine, J.E.; Cangellaris, A.C. Statistical analysis of fiber weave effect over differential microstrips on printed circuit boards. Signal Power Integr. 2014, 40, 1–4. 10. Luevano, G.R.; Shin, J.; Michalka, T. Practical Investigations of Fiber Weave Effects on High-speed Interfaces. In Proceedings of the IEEE Electronic Components & Technology Conference, Las Vegas, NV, USA, 28–31 May 2013; pp. 2041–2045. 11. Shlepnev, Y.; Nwachukwu, C. Modelling Jitter Induced by Fibre Weave Effect in PCB Dielectrics. In Proceedings of the IEEE International Symposium on Electromagnetic Compatibility, Raleigh, NC, USA, 4–8 August 2014; pp. 803–808. 12. Cao, Y.; Li, J.; Hu, J.; Zheng, M. Design Consideration for 10 Gbps Signal Transmission Channel in Copper Backplane System. In Proceedings of the International Conference on Computational Science & Its Applications, Santander, Spain, 20–23 June 2011; pp. 215–218.

© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).