wd
`
`Effect of Textile Properties on the Bidirectional
`Solar-Optical Properties of Shading Fabrics
`
`MAUREEN M. GRASSO AND BRUCE D. HUNN
`Centerfor Energy Studies, The University of Texas at Austin, Austin, Texas 78758, U.S.A.
`
`ABSTRACT
`This paper presents an experimental approach to and preliminary results of mea-
`surements of bidirectional solar-optical properties of shading fabrics. Transmittance
`and reflectance profiles for three woven and three knitted fabrics are presented as a
`function ofaltitude and azimuth anglefor selected incidence angles ofsimulated direct
`solar radiation. The wovenfabrics exhibit transmittance profiles that are sharply peaked
`in line with the incident beam. At normal incidence, woven fabrics with transparent
`yarnsresult in high peak transmittance ofapproximately 41%, whereas opaque yarns
`and dense structures result in low peaks of 0.5-2%. All of the wovens tested exhibit
`low reflectances of 1-2%. Knitted fabrics also show sharply peaked transmittance
`profiles in line with the incident beam. However, the open knits have peak values of
`up to 63% at normal incidence, whereas opaque knits show peaksofonly 1-4%. Peak
`reflectance values for all the knits were low at approximately 1-2%, even with metal-
`lized yarns.
`
`-
`
`
`
`Conventional interior shading devices control solar
`heat gain and provide a variety of other benefits in-
`cludinglight and glare control, privacy, reduced cooling
`loads, and preservationoffabrics and art objects subject
`to ultraviolet degradation from exposure to sunlight.
`A report by the Center for Energy Studies at The Uni-
`versity of Texas at Austin has shown the importance
`of window shading to control solar heat gain in resi-
`dential and commercial buildings [15]. A major con-
`clusionofthat study is the potential advantage ofshad-
`ing devices that combine good shading coefficient and
`U-value (insulating) properties as an overall energy
`Strategy, even for a cooling-dominated climate. For
`certain applications, interior shading devices that can
`be managedovera diurnalcycle, such as draperies and
`fabric shades, have an advantage over shading devices
`that cannot be diurnally managed (screens, awnings),
`because they have the potential to combine added
`shading properties for the cooling season with good
`thermal insulation for the heating season.
`Previous research on the performance of shading
`devices has focused on the heating load reduction (in-
`sulating) properties of fabric shades and draperies, or
`on their daylighting characteristics [2-14, 17, 18, 20,
`23]. Little work has focused on the shading properties
`of fabrics, particularly in response to daily and seasonal
`variationin incidentsolar angles. None ofthe literature
`
`reviewed has examined the design of interior fabric
`shading devices with respect to the influence of the
`textile components on solar load control.
`Furthermore, the variability of shading coefficients
`with incidence angle and with direct/diffuse radiation
`mix is not well known for cases in which the direc-
`tionality of reflected and transmitted radiation is im-
`portant. An innovative approachto this problem is to
`use carefully selected combinations of shading device
`configurations—fabric, yarn and fiber structure, as well
`as finishes—to achieve the desired solar load control
`characteristics (directional reflectance and visibility
`characteristics). For example, the reflectance ofa fabric
`shading device can be maximizedby selecting the ap-
`propriate fiber length, yarn structure, and fabric struc-
`ture.
`Fabrics used as shading devices can create significant
`opportunities for solar load and glare control, but di-
`rectional
`solar-optical properties (reflectance and
`transmittance) of fabrics are needed to design draperies
`and shades that are effective for this purpose. These
`directional characteristics are neither well known nor
`well understood, and there has beenlittle research on
`how best to incorporate desired directional reflectance/
`transmittance characteristics into shading fabric de-
`signs. Quantitative design methods are needed to design
`improved fabric systems.
`
`Textile RADF247!Js7 (1992)
`Page 1 of 11
`
`247
`
`DBR Finance, Inc., &x4o40)4/42.00
`DBR Finance, Inc., Ex. 1011
`
`

`

`248
`
`TEXTILE RESEARCH JOURNAL
`
`Our goal in this study is to apply textile theory to
`the developmentoffabric interior shading devices for
`commercial andresidential buildings, primarily for use
`in cooling-dominated climates. Our specific objective
`is to determinethe directional shading performance of
`fabric shadesso as to identify innovative textile designs
`(fiber / yarn/ fabric) and their incorporation into shad-
`ing devices that are mosteffective for cooling load re-
`duction while maintaining daylighting (with reduced
`glare), visual, insulating, and aesthetic qualities.
`This paper presents the experimental basis for the
`developmentof these innovative shades. We describe
`a bidirectional solar-optical property apparatus and
`present
`transmittance/reflectance data for several
`conventional woven and knitted fabrics. Preliminary
`results show theeffect of fabric physical properties on
`the range of bidirectional properties to be expected
`from conventional shading fabrics.
`
`Experimental Approach
`
`The solar-optical properties normally reported for
`shading devices are hemispherical transmittance and
`reflectance [1]. Thus, single values of transmittance
`and reflectance are given that define the fraction of all
`incident (diffuse and direct) radiation transmitted
`through the shade or reflected in the forward hemi-
`sphere. These properties are integrated overall incom-
`ing and outgoing angles.
`Whenthedirection ofeither the incident or outgoing
`radiation is important, however, directional solar-op-
`tical properties are specified. Thus, directional-hemi-
`spherical properties integrate the transmittance orre-
`flectance over the outgoing hemisphere for a beam of
`radiation at an arbitrary incidence angle. Similarly, the
`bidirectional solar-optical properties are specified for
`all possible incidence angles and all possible transmitted
`and reflected angles.
`The angular transmittanceis defined as the fraction
`of the incident radiant energy that is scattered into a
`solid angle at position (a, ¢), where a is the altitude
`and ¢ is the azimuthal angle in the transmitted hemi-
`sphere (see Figure 1). Thus, the bidirectional trans-
`mittance (specific transmittance ) is expressed by
`
`1(a, 0:3 %, %) = eee
`
`di,
`
`is
`
`iy
`
`3
`
`(1)
`
`where a; and ¢; are the incident altitude and azimuth
`angles, a, and @, are the transmitted altitude and azi-
`muth angles, di, is the differential radiance at the po-
`sition (a,, ¢,), and de; is the differential incident radiant
`enerpage (sesfntiqily collimated) per unit of fabric
`Page 2 of 11
`surface area [22]. Thus, the bidirectional transmittance
`
`is defined as the ratio of the exitant radiance (radiant
`flux per unit solid angle per unit area normal to the
`direction considered) to the incident irradiance ‘(ra-
`diantflux per unit area of sample). However, because
`the incidentradiantflux (de; ) is measured in the direct
`normal orientation, for non-normal incidence,
`
`T( aj, Pi; &1, 1) =
`
`3
`
`(2)
`
`di,(a;, Oi; 1, pr)
`de; , (a;, $;) COs 9;
`where 9; is the angle of incidence. The integration of
`(a; , ;; &;, o;) overall solid angles in the transmitted
`hemisphereyields the directional-hemispherical trans-
`mittance. Similarly,
`the bidirectional reflectance is
`given by
`
`pay, Pi; O84, br) =
`
`di,( aj, 0:3 G1, Pr)
`de}, (ai, i) cos UF
`
`(3)
`
`where the subscript 7 indicates the reflected direction.
`
`Incident hemisphere
`—j—
`
`Transmitted hemisphere
`—--
`
`Transmitted ray
`
`Incoming ray
`
`
`Surface normal
`
`Plane of sample
`
`a =Altitude angle
`o =Azimuth angle
`6 =Incidence angle
`
`ee
`
`FIGURE 1. Conceptofbidirectional transmittance.
`
`For the apparatus used in this study, the angle of
`incidence of a radiant beam incoming to the fabric
`sample is defined by its azimuth angle (i.e., the angle
`between the surface normal andthe horizontal projec-
`tion of the incident beam) andits altitude angle (1.¢.
`the angle between the incident beam and the horizon-
`tal). The outgoing radiant energy in the transmitted
`or reflected hemisphere is similarly characterized by
`azimuth ang) RBRUaMaes "FARESEe Be pense”
`DBR Finance, Inc., Ex. 1011
`coordinates.
`
`

`

`May 1992
`
`To relate these definitions to quantities measured
`with our apparatus, we note that the radiant flux emit-
`ted from a differential element of the sample surface
`that is intercepted by a differential element of the re-
`ceiving surface (the sensor) is given by Siegel and
`Howell [21] as
`d*@ = di dA, cos 0 dAcensor/R?
`
`(4)
`
`
`
`FIGURE 2.Bidirectional solar-optical properties apparatus.
`
`where di is the radiance(flux per unit area of emitting
`surface projected normal to the direction considered,
`per unit solid angle) of the sample surface element d4,,
`cos @ is the cosine of the angle of exitance, dAsensor iS
`the differential sensor area, and R is the distance from
`the sample to the sensor. But the flux received by the
`sensorareais its irradiance, given by
`d?b/dAsensor = di dAs COS 0).R? =i Gesscr
`which is the sensor reading [19].
`emittance spectrum of the sun. Another lenscollects
`Now the incidentirradiance e; is measureddirectly,
`the light to produce a collimated incident (incoming)
`without the sample in place. Combining Equations 2
`beam. A masking device is placed between the light
`and 5 and noting that cos @ = cos a cos ¢, the bidirec-
`source and the sample to exclude extraneous light
`tional transmittance becomes
`(Figure 2). The source lampis situated 1 m (3.28 ft)
`R
`tla é ) — sensor
`from the fabric sample.
`(6)
`ts Yrs
`
`Twosilicon photodiode light detectors with active
`areas of 5.1 mm? (0.008 in.*), are used to measure
`angular transmittance and reflectance. The detector
`includes a built-in operational amplifier that amplifies
`the output signal to a more readily measurable range.
`Although the sensors are not ideally suited to the solar
`spectrum, their responseis fairly flat in the 0.45-1.0
`micron range.
`Each sensor is mounted inside a 9.5 mm (0.374 in.)
`diameter black tube to block stray light from reaching
`the sensor by highly restricting its acceptance angle to
`10° (5° divergence oneitherside) (see Figure 3). This
`ensuresthat the sensor sees only the central portion of
`the sample andthatonly energy transmitted orreflected
`into a small solid angle is sensed. As shown in Figure
`3, at an incidence angle of 0° (beam normalto the
`sample) the sensor views a 17 mm (0.7 in.) diameter
`area at the center of the 50 mm (2.0 in.) diameter beam.
`At the most oblique position of the sensor (65° sample
`rotation), it views an area having a width of 43 mm
`(1.7 in.), which is entirely within the illuminated por-
`tion of the rotated sample. Thus the sensor averages
`the light transmitted through, or reflected from, ap-
`proximately circular areas ranging in diameter from
`17-43 mm (0.7-1.7 in.). The characteristic length of
`the fabric pattern is always smaller than this diameter
`and is usually at least an order of en smaller.
`However, for
`that eeflHea
`oan
`knit orwoven‘BBRinanceelts
`Xndteh spa-
`DBR Finance, Inc., Ex. 1011
`
`te
`
`|
`
`(3)
`
`2
`
`e,}cosa;,cos @,dA, ~°
`
`where dA, is the sample area viewed by the sensor; the
`birdirectional reflectance is obtained similarly. The
`measured results presented below are reported as rel-
`ative intensity readings, @sensor/@;, and are related to
`the transmittance (or reflectance) at each point by
`Equation 6. Thus therelative intensities must be mul-
`plied by a scaling factor, which is essentially an area
`ratio, to obtain the transmittance or reflectance. These
`values then must be integrated overall possible solid
`angles in the receiving hemisphere to obtain hemi-
`spherical properties.
`
`INSTRUMENTATION
`
`, Wedesigned an apparatus to measure the bidirec-
`tional reflectance and transmittance of direct (beam)
`radiation for small, 102 x 102 mm (4 X 4 in.) fabric
`samples. This apparatus is based on the large scanning
`radiometer developed at Lawrence Berkeley Labora-
`tory to measure bidirectional solar-optical properties
`of samples of fenestration system layers [19]. The ap-
`paratus is composed of a power source that emits direct
`(beam) radiation to the small square of fabric being
`tested and twolight-sensitive silicon photodiodes that
`measure the beam radiation reflected from, or trans-
`
`ans through, the fabric (see Figure 2).
`apage36squaresiqis a 75 W Xenonarc lampwith a35 m in.Oodrture lens, in order to simulate the
`
`
`Page 3 of 11
`
`

`

`250
`
`TEXTILE RESEARCH JOURNAL
`
`
`
`Control Arm
`
` ae
`
`
`Samplea
`
`65 fe,max. rotation
`Incident Beam
`if
` “,deg7
`=67mm =
`
`
`
`
`
`
`
`
` HELE —|gReuyt ®&:as.
`
`FIGURE 3. Photodiode configuration.
`
`tially averaged values and considerable data scatteris
`likely.
`Because this project involves working with small
`squaresof fabric, the apparatus is approximately 0.46
`m (18 in.) high and is manually operated. The sensor-
`holding armscan bepositioned at every 5° of azimuth
`while readings are taken with the data acquisition sys-
`tem; the data are imported to a spreadsheet for later
`manipulation and graphical presentation in three-di-
`mensional surface plots. Similarly, the sensor tubes lock
`into place at every 5° ofaltitude on the arcs of rotatable
`sensor-holding arms (see Figure 2).
`The sample-holding plane, with its two rotational
`degrees of freedom,consists of an inner ring revolving
`within an outer ring. With a fixed source, the degrees
`of freedom of the sample holder providefor a full range
`ofazimuth and altitude anglesto be setfor the incident
`beam. However, the sensor tube contacts the holderat
`rotational angles above about 65°, Thus data can be
`taken at angles up to 80° only at selected locations
`where clearance is adequate. Because of blockage of
`the incident beam by the reflected sensor arm, reflec-
`tance data cannot be taken within 30° of the source
`beam.
`
`CALIBRATION AND OPERATION
`
`At each angularposition ofinterest in the transmit-
`ted or reflected hemisphere, the bidirectional property
`distribution is determined by dividing the measured
`intensity at that position by the measured intensity of
`the incident (unattenuated, with no samplein place)
`beam,corrected for its angle of incidence. This incident
`beam intensity is measured just before and just after
`each set of readings is taken, and the average of these
`two is used. Readings at the angular positionsare taken
`approximately every second as the rotating arm 1s
`manually swept through the 5° stops in the azimuthal
`andaltitudinalarcs.
`Transmittance and reflectance measurements were
`taken at incidence angles representing a rangeofalti-
`tude and azimuth angles typical of solar conditionsat
`east-, south-, and west-facing windows during winter
`and summer conditions. Transmittance measurements
`were taken for incidence angles of(0, 0) and (45, —45).
`but only the results for the incidence angle of(0, 0)
`are presented. Similarly,
`reflectance measurements
`were taken for incidence angles of (45, —45) and (75.
`—30), with results presented only for the (75, —30)
`incidence angle. Theresults are the average offive re-
`peated runs unless otherwise noted.
`
`Results and Discussion
`
`Before collecting solar-optical property data for the
`fabric samples, we calibrated the instrumentation and
`madep
`mbipty ymeasurements.
`Information re-
`Page 4 of 11
`garding these tests can be found in reference 16.
`
`PHYSICAL AND OPTICAL PROPERTIES
`Physical PYBRUPEee!dESRIPEOF peach
`DBR Finance, Inc., Ex. 1011
`shadingfabric tested are given in Table I. Two general
`
`

`

`May 1992
`
`Le
`Identification
`
`Wovens
`B9
`D3
`
`B19
`"
`Knits
`Bll
`
`TABLE I. Physical properties of shading fabrics.
`Yam
`twist, TPL,
`W XF
`
`Yarn
`count,
`WXF
`
`Yarn
`size, tex,
`WXF
`
`Fiber content
`
`251
`
`Fabric
`weight, g/m?
`(oz/yd")
`
`Cover
`factor K,
`WX F
`
`Optical
`description
`
`polyester (semidull)
`polyester (bright)/rayon
`(striation-bright)
`Polyester (bright)/
`cotton
`
`17 x17
`17 * 12
`
`6x6
`
`83x 74
`98x69
`
`9.8 x 9.7
`817.8 x 28.9
`
`56.3 (1.66)
`155.7 (4.60)
`
`10.7% 9.5
`17% 15.3
`
`29 x 19
`
`196 x 174
`
`333.2 (9.84)
`
`16.7% 10.3
`
`transparent, sheer
`dense, essentially
`opaque
`dense, opaque,
`lustrous
`
`Structure
`
`plain weave
`satin weave
`
`Dobby weave
`
`raschel knit
`
`polyester (dull)
`
`NA
`
`24 x 11
`
`NA
`
`75.9 (2.24)
`
`NA
`
`open, rectangular
`grid
`14 x 25
`NA
`241.3 (7.13)
`NA
`dense, essentially
`
`opaque
`nonwoven
`(striation-bright)
`backing
`dense, opaque
`NA
`67.4 (1.99)
`NA
`36 x 56
`NA
`polyester (semidull)
`raschel knit
`D2
`metallized face
`eea sS——_
`
`
`
`A2 raschel knit with—polyester (bright)/rayon NA
`
`fabric categories are represented: wovens and knits.
`Within each category, we tested samples of both uni-
`form andirregular structures. Fabrics in each category
`Tanged from open to opaque or nearly opaque with
`respect
`to light
`transmission, We measured yarn
`twist (turns per inch), yarn count (ASTM D3775),
`yarn size (tex), cover factor K (yarns per inch/
`\cottoncount), and fabric weight (option C, ASTM
`D3776) for each fabric sample where possible. Fiber
`content was determined by the manufacturer.
`
`Woven Fabrics
`
`The three woven fabrics tested have dissimilar textile
`characteristics. The sheer, plain weave fabric (B9) is
`characterized by a high, balanced yarn count, fine yarns
`(9.8 tex, warp and weft), medium twist, and light
`Weight. A significant aspect of this fabric is its trans-
`lucent, semidull (moderate delustering) filament yarns;
`therefore, light can be transmitted through the open
`Spaces as well as through the yarns themselves (Figure
`4a). On the other hand, sample D3, the 4 X 1 satin
`
`
`
`FIGUR
`@v5)
`Gifriqs4(a) plain weave (sample B9), (b) satin
`weaveea 3OF sal and (c) Dobby weave (sample B19).
`Page 5 of 11
`
`weave (sateen), is a dense, essentially opaque fabric
`(Figure 4b). This high-count fabric has a dull luster
`even though it is composed ofthin staple yarns (17.8
`tex warp, 28.9 tex weft) of medium twist that are made
`of bright(little or no delustering) polyester fibers. This
`satin weave is heavier than the plain weave sample
`(B9) and has a higher coverfactor.
`The Dobby weave (B19) is the heaviest fabric, with
`large yarns ( 196.2 tex warp, 173.9 tex weft), a low yarn
`count, and a high cover factor comparable to the satin
`weave. Since there is the potential for yarn slippage,
`somelight can be transmitted through the open spaces
`of the weave. The staple yarns have a high luster re-
`sulting from the low twist and bright polyester fibers
`present in the yarn structure (Figure 4c). These textile
`characteristics have a major influence on the trans-
`mittance andreflectance ofthis fabric.
`
`Knitted Fabrics
`Thethree knitted fabrics are similar in that they are
`all raschel warp knits, but the similarity ends there.
`Oneof the knitted fabrics has an openstructure, while
`the other two are more opaque, with one having a met-
`allized coating on the face and the other a nonwoven
`backing. Sample B!1 is a lightweight, open, uniform
`rectangular-grid raschel knit constructed with dull
`polyester fibers (Figure 5d). In contrast, sample A2
`could be characterized as an open casement with a
`nonwoven, random-webbacking, makingit essentially
`dense, opaque, and of medium weight(Figure Se). The
`fibers used in the face are bright polyester and bright,
`striated rayon. Finally, sample D2 is a more “tradi-
`tional” raschel knit, dense and opaque with a metallic
`finish or coating on the face (Figure Sf); the polyester
`fibers used are semidull.
`The warp I
`ture ofthese fabrics made their
`characterizati BieFINACeINGHeXardAtthe
`DBR Finance, Inc., Ex. 1011
`
`

`

`252
`
`
`
`FIGURE5. Raschel knit fabrics: (d) open rectangulargrid (sample
`B11), (e) nonwoven, random-web backing (sample A2), and (f)
`metallic coating on the face (sample D2).
`
`raschel knit structure, we were unable to accurately
`determine the yarn twist and yarn size (tex) ofall sam-
`ples and thus were unable to calculate the cover factor
`K. The yarn count or gauge is perhaps most meaningful
`for the metallized sample (D2) because of its more
`traditional knitted structure. However, because the op-
`tical descriptions given in Table I are meaningful for
`all samples, we used them in characterizing these fab-
`rics.
`
`TRANSMITTANCE MEASUREMENTS
`
`Woven Fabrics
`
`The bidirectional transmittanceresults for the plain
`weave are shownin Figure 6 for an incidence angle of
`(0, 0). This sort of three-dimensional surface plot for-
`
`TEXTILE RESEARCH JOURNAL
`
`matis used to display all of the solar-optical property
`data. The angular transmittance is plotted against a
`grid representing every 5° in altitude and azimuth (up
`to 65°) in the transmitted hemisphere. The point-by-
`point average offive repeated runsis plotted exceptas
`otherwise noted.
`The plain weave (B9) has a peak transmittance of
`0.42, which occurs directly in line with the incident
`beam. The peak is sharp (base spread of only +10°),
`indicating that this fabric does not disperse thelight
`either vertically or horizontally. The light is transmitted
`betweenthe fine yarns of the fabric structure as well
`as through the transparent semidull fibers. The presence
`of the delustering agents in the polyester filaments does
`not disperse the transmitted light in any noticeable
`pattern.
`Because ofits higher coverfactor ( 1.66 greater), the
`satin weave (sateen) (D3) has a much lower trans-
`mittance than the plain weave; at normal incidence,it
`has a transmittance peak of 0.005 or 0.5%, compared
`to 42% for the plain weave with translucent yarns of
`semidull fibers (B9) (Figure 7). In addition, there ts
`greater horizontal dispersion of the transmitted light
`with the satin weave (base spread of +20°), yet the
`shape of the peak is symmetrical. Outside the base cone
`of +20° there is evidence of a low signal, but unlike
`the plain weave, the drop-off is gradual, indicating a
`broad, low-level dispersion of the incident beam. Even
`thoughthe fibers in this satin weave are characterized
`as bright, the dense fabric structure prohibits light from
`being transmitted and contributes to the greater dis-
`persion oflight that is transmitted; some of the light
`scattering may result from thestriations presentin the
`rayon fibers.
`
`
`
` Multiple Run Average
`
`07
`
`08
`
`Ae
`2
`“4
`03 g
`>
`02 s
`
`Or
`
`oe
`
`
`
`AUISNALNIAALLW13
`
`INTENSITY
`
`Ny
`Pt
`Ors
`ries Somes
`jreteteseeaten
`‘
`peg eee eat
`BE ee ee
`MEET
`OGIEfA LLALRD
`CROCS rae
`ir
`COEDSOYi
`
`RELATIVE
`
`07
`
`08
`
`Df
`
`Z
`5
`Z os
`g
`g o2
`
`a}
`
`=
`
`Page 6cofeb1Piain weave sheerfabric,
`Page 6 of 11
`transmittance at (0, 0) incidence.
`
`DBR Finance, Inc., Ex. 1011
`DBR Finance, Inc., Ex. 1011
`FIGURE 7. Dense satin weave, transmittance at (0, 0) incidence.
`
`

`

`May 1992
`
`The Dobby weave (B19), the heaviest fabric, resem-
`bles a herringbonetwill (Figure 4c). It is constructed
`of bright, polyester staple fibers of low twist and ex-
`tremely thick yarns relative to the other samples. The
`cover factor is comparableto the satin weave; however,
`the yarn count is much lowerin this sample. Conse-
`quently, more light is transmitted through this fabric
`(peak of 0.022) than through the satin weave (D3) by
`a factor of 4 (Figure 8). The peak is sharper and better
`defined (with very little horizontal dispersion ) than that
`of either the plain (B9) or satin (D3) weave samples,
`indicating the importance of fabric structure to the
`transmittance distribution observed. The plot flattens
`out, but not to the extent observed with the plain weave.
`At this normal incidence angle, only 2%of the lightis
`transmitted, whereas 42% of the light is transmitted
`through the plain weave. The light transmitted in the
`Dobby weave is only through the interstices of the
`loosely woven structure, whereas in the plain weave
`sample, light is transmitted not only through the in-
`terstices, but also through the transparent semidull
`polyester fibers. Evidently little light is being trans-
`mitted through the bright white polyester and cotton
`fibers in the Dobby weave. In later examplesofreflec-
`tance, one can see slightly more reflectance with the
`Dobby weave (B19) than with the plain weave (B9),
`hence the light that is not transmitted is reflected in a
`diffused pattern.
`
`
`
`MMISNALNIBALLTae
`
`RELATIVEInTENSITY
`
`FiGurReE8. Dense Dobby weave, transmittance at (0, 0) incidence.
`
`Knitted Fabrics
`As shownin Figure 9, the bidirectional transmittance
`splar-grid raschel knit fabric (B11)
`of thhas Page|eotttoccurring at angle (0, 0) directly
`Page 7 of 11
`
`253
`
`behind the incident beam. This sharply shaped peak
`(base spread of only +10°) indicates virtually no dis-
`persion of the incident beam.This occursfor the open
`knit structure because little incident light is reflected
`or refracted either vertically or horizontally from the
`path of the incident beam; most of the light passes
`directly through the open spacesin the knit.
`
`o7
`
`a8
`
`05
`
`ny
`
`fn
`
`Multiple Run Average
`O7
`
`08
`
`05
`
`E
`
`5
`a3 x
`
`Q2 5
`
`oy ©
`
`=—
`
`z 03
`
`5 02
`
`of
`
`=
`
`FiGuRE 9. Rectangular grid raschel knit,
`transmittance at (0, 0) incidence.
`
`At normal incidence (0, 0), the transmittance dis-
`tribution for the raschel knit with nonwoven random-
`web backing (A2) appears as in Figure 10. This dense
`fabric structure results in a transmittance of only about
`1.4%, compared to values of more than 60% for the
`
`O.014 aot
`
`aot
`
`xae
`Lo
`Z ows
`j ee
`~
`=
`
`SES
`ees SoeSee
`Ss SE ssSse
`
`Muitiple Run Average
`
`=
`
`wo E
`2
`ae
`1.004 5
`Ooo
`ey
`
`FIGURE “OBREMahteyt
`Oy x Ta
`DBR Finance, Inc., Ex. 1011
`
`

`

`254
`
`TEXTILE RESEARCH JOURNAL
`
`open fabric structures. Again, the transmittance is
`sharply peakedin line with the incident beam, but the
`intensity falls off the shoulder of the peak less abruptly
`than with the open fabrics. This is because nolightis
`directly transmitted through open spaces in the knit,
`but is dispersed through the random-web nonwoven
`backing. The transmitted light is gently dispersed over
`a region +15° from the peak, falling smoothly out
`to 65°.
`The transmittance distribution for the raschel knit
`with the metallized face (D2) is sharply peaked and
`well defined, even with the knitted fabric structure. Its
`reflective properties result from the metallized coating
`on the fabric face and the semidull polyester fibers in
`the structure (Figure 11). As expected,
`less light is
`transmitted through this fabric (13.5%) than for the
`open raschel knits, where the light is transmitted di-
`rectly through the interstices. However, morelight is
`transmitted than for the raschel knit with the nonwoven
`backing (A2), in which case the backing diffuses the
`light. The base spread of +15° indicates a slight dis-
`persion of incident light at the base, which mayresult
`from the delustering agents in the polyesterfibers.
`
`AMSNELLNI
`
`BAL138
`
`FIGURE 11. Metallized raschel knit fabric,
`transmittance at (0, 0) incidence.
`
`REFLECTANCE MEASUREMENTS
`
`Woven Fabrics
`
`Because ofits translucent, semidull filament yarns
`and the structure of the fabric, the sheer plain weave
`fabric (B9) transmits most of the incident beam
`througe Hrsfsti¢es but some through the translu-
`Page 8 of 11
`cent fibers. This results in a relatively low peakreflec-
`
`tance of 1.5% at an incidence angle (75, —30) (Figure
`12), which is representative of conditions near solar
`noon in the summerat manylatitudes. Ofthelight
`that is reflected, there is a sharp peak at (—75, 30),
`indicating forward (specular) reflection; the rolling,
`curved shape of the reflectance profile indicates a
`weaker back and lateral reflection. The delustering
`agents in the semidull polyester fibers may be influ-
`encing this lateral and backward reflection.
`
`INTENSITY
`
`aLn
`FO
`SSeS
`BERS,
`Sete?
`SOC HITS
`
`RELATIVE
`
`FIGURE 12. Plain weave sheerfabric, reflectance
`at (75, —30) incidence.
`
`The opaquestaple yarns, bright polyester and bright,
`striated rayon fibers, and weft floats present in the satin
`weave sample (D3) cause it to be more reflective than
`the plain weave sample (B9) of translucent filament
`yarns. Thus, as Figure 13 shows, at the (75, —30) in-
`
`RALISNALNI
`
`aniv1s
`
`
`
`RELATIVEINTENSITY
`
`DBR Finance, Inc., Ex. 1011
`DBR Finance, Inc., Ex. 1011
`FIGURE 13. Dense satin weave, reflectance at (75, —30) incidence.
`
`

`

`May 1992
`
`cidence angle, the satin weave has a peak reflectance
`of 2.3% compared to 1.5% for the plain weave at the
`same incidence angle. Thereflection profile at this angle
`shows a strong forward (generally specular) reflection,
`with no back orlateral scattering. The presence of
`bright polyester fibers contributes significantly to the
`reflection distribution observed. Anotherreason for the
`forward reflection distributionis that less yarn surface
`is exposed to the incident beam,resultingin little lateral
`scattering. Note the sharp peak at (—75, 30), in line
`with the beam.
`Because the Dobby weave (B19), like the satin weave
`(D3), is very dense,resulting in no exposed yarn sides,
`its reflectance profiles are quite similar to those for the
`satin weave. We saw nobackscattering andlittle lateral
`scattering, and the reflection was nearly specular(Fig-
`ures 13 and 14). Both samples contain bright polyester
`and either bright striated rayon or cotton. The low twist
`in the Dobby sample contributes to the forward (spec-
`ular) reflection, whereas the floats in the satin weave
`sample are significant in influencingits reflectance dis-
`tribution.
`
`
`
`AUSNALNIBALLS
`
`INTENSITY
`
`RELATIVE
`
`FIGURE |4, Dense Dobby weave,reflectance at (75, —30) incidence.
`
`Knitted Fabrics
`At an incidence angle of (75, —30), the resulting
`reflectance pattern for the open rectangular-grid raschel
`knit (B11) is shown in Figure 15. Forthis situation,
`the incident beam is reflected fairly sharply in the plane
`of the beam in both the forward and backward direc-
`tion, forming the ridge at azimuth = 30°. The backward
`scatter results from light penetrating the fiber and re-
`nectingangpring agents.In thiscase, the peak
`Page 9 of 11
`is at an
`allitude of —75°, indicating specular reflectance
`
`255
`
`at these high angles. A low peakreflection of1.1% in-
`dicates that most of the incident beam passes directly
`through and the remaining light is reflected back in
`the direction of the beam. Even though the white yarns
`appearfairly reflective, the presence of dull polyester
`fibers (heavily delustered ) and the openstructure pre-
`sent relatively little reflecting surface.
`
`
`
`AMISNALNIANIL138
`
`INTENSITY
`
`RELATIVE
`
`FIGURE 15. Open, rectangulargrid raschel knit,
`reflectance at (75, —30) incidence.
`
`The dense, essentially opaque raschel knit with a
`random-web nonwoven backing (A2) has a reflectance
`pattern similar to that for the open knitted fabrics, as
`illustrated in Figure 16 for an incidence angle of (75,
`—30). Light is reflected offthe dense structure oflight-
`
`
`
`MMISNALNIBAL1S8
`
`Multiple Run Average
`
`INTENSITY
`
`RELATIVE
`
`Fic
`
`iets ‘DBRFI at ariee heeex Ot
`‘
`se wi
`“
`a
`DBR Finance, Inc., Ex. 1011
`
`

`

`256
`
`colored (beige) bright polyester and bright striated
`rayon fibers, giving a peak reflectance of 2.2%. Again
`the reflectance surface is entirely in the forward direc-
`tion. Onereasonis that this sample, unlike the open
`rectangular-grid raschel knit (Figure 15), does not ex-
`hibit backward reflection. In addition, the absence of
`delustering agents in the fibers reduces the effect of
`backward scattering of light. Even though the rayon
`fibers have striations, they do notdiffuse thereflection
`in thelateral direction.
`The reflection pattern of the raschel knit with the
`metallized coating on the face of the fabric (D2) was
`sensitive to the fabric’s mounting in the sample holder
`and to specular reflection from the yarns in the knit
`structure. Wrinkling of the fabric appears to change
`the reflection pattern observed. At an incidenceangle
`of (75, —30), there is forwardreflection with a peak
`at 2.2%, as shownin Figure 17. The secondary lateral
`peaks occur from yarn reflection within the knit as well
`as from thepresenceofdelustering agents in the semi-
`dull polyester fibers.
`
` 4
`
`\e Saas SS Sot
`pa. V\Wneersoesses SocSeS
`aoe oe
`SSS Se
`
`
`
`
`
`RELATIVEINTENSITY
`
`FIGURE 17. Metallized raschel knit, reflectance
`at (75, —30) incidence,
`
`At this high solar angle, the reflectance distribution
`for the raschel knit with the metallized coating (D2)
`is similar to those for the raschel knit with nonwoven
`backing (A2) and for the open rectangular-grid raschel
`knit (B11) in that the distribution observed is in the
`forward direction; however, the raschel knit with the
`metallized coating (D2) exhibits a greater backward
`reflection distribution. The reason is that the structure
`
`has dpugertyqgfnaduill) polyesterfilaments and the
`
`Page 10 of 11
`coating
`Creates greaterreflection off the yarns.
`
`TEXTILE RESEARCH JOURNAL
`
`Conclusions
`
`The research reported in this paper showsthat con-
`ventional fabrics exhibit directional solar-optical prop-
`erties that depend strongly onthetextile structure. The
`solar-optical property measurement methodologyde-
`veloped in t