Document 8Q1ZwRJb499dQvvvkD5e95Yk

cDMirom DUPONT MARSHALL LABORATORY LIBRARY REPOST NO* PEH5$?1 COPY #7 s. i, m mm m w o ir s c o mpan y , me. FABRICS AND EINtSHBS BBPAKBWX r esear c h d iv is io n DiTisscm pr o c es s en g im-er in c , g r o u p COLOR OPERATIONS SECTION REFERENCE COW wm tot ciicumw m 3 1989 qiAMCTERIZATIQN OF COLORANTS FOR METALLIC FINISHES By: W, S. Armstrong Date Issued: December 1968 Project: 4I-0029, Instrimental Color Technology RETURN TO MARSHALL LAB. LIBRARY ABSTRACT This report describes the development of a practical system for character!%lug colorants for metallic products so that dual-angle color vectors can be SN3 SO W !- 4>t n CM TABLE OF CONTENTS i. smim Introduction ................. Results and Occlusions Continuing Work ................................... * . ... ................... . , Aclmov;ledgnent ..................................... . . . , . II. DISCUSSION Background ................. . . 6 General Metied o Characterisation .......... Relationship of Elat and High Angle Color Measurements Directional Scattering of Aluminum Method ....... Dual Angle Characterization Method .......... Correlation with QoXorimeters Behavior of High Scatter Pigments Production Shading Experience ............................................. . Conclusions o .................... . III. INFERENCES ...... ...................... XYo APPENDIX A. Colorimeter Viewing Geometry . . . . ......................... .... B. Comparison. of "Directional Scatter5* Vectors .......... C. Comparison of 867"Line Computed f? Manual Vectors ...... . D. Comparison of 94Saline Computed Manual Vectors ....... . E. Mathematical Proof of Common Intersection of Spectral Curves . . F. Statistical Correlation of Tristimlus Values . H O H i -Mt f-s Cf5 ^4 CO Sn5O tM >** 09 NJ K> K> 3 ~*3 3 C?' CM Ssi f DUP030002685 i - mwm INTRODUCTION Vector shading f metallic colors has been successfully carried out by C.O.S. for some four years by using dual-angle color measurements to account for the marked directional behavior of metallics. tMortunately existing color theory did not cover the metallic field, so it was necessary to laboriously prepare and measure dual-angle vector panels for each colorant in a tom&a before it could be cospiter-shaded. This routine is tin<*consuming (requir ing about erne man-day per formula) and often inaccurate because of the weighing of very small' quantities of colorant. Manual vector preparation then effectively limited instrumental shading to a few large volume promts whose repetitive manufacture could justify the cost of vector preparation. To make routine instrumental shading of metallics practical it was necessary to develop a technique whereby reliable dual-angle vectors could be rapidly computed while retaining compatibility with the existing methods of measuring and shading. The objective of this study was the development of technology and equipment necessary to rapidly predict dual-angle vectors tor metallic colors from formula composition. The bulk of the development work was carried out between October 1964 and June 1966. RESULTS AND CONCLUSIONS Over 30 colorants in each of two mill base lines (901 and 913) were characterized in metallic products using different techniques. These characterization data were then used to generate computed dual-angle color vectors on a variety of S67 and 945 line formulas covering a wide color range. Correlation of these computed vectors with manually derived ones is good with the final system. A comparison of computed and manual vectors is given in Appendices B, C and D. The soundest overall approach was determined to be: 1.. Characterize the colorants in aluminum at too separate illumination and viewing conditions approximating those used on the "Colormaster" equipped with an Angular Viewing Device. 2. Determine the statistical relationship between tristimulus readings developed from the dual-angle spectral curves and those on the "Colormaster." 3. Compute formula and vector tristimulus values (using . characterization data from step 1) for the "Flat" case and convert to equivalent "Colomaster" tristimulus values (using relationships from step 2) before converting to aLj &a, Ab. 4. Repeat step 3 for the "High Angle" case. - 1- .Actual pimt 3*d3S5g sxpsrianca o b MS-ilse .at. :.%rlvvi using upp*3$St ss&th exfaxim&zaZ caemetsrissfioa fata showed that tl cots',ited 7sotors sts Sspal to or better than msimi actors ia -required Mis/Imtcb, Tlss cfc&mtarAzatioa sad vmxwt mo^atm programs to been edified to handle tfes system outlined above sad are fully .operational. <xmmm mm: Bxp&rimce gained during the experimental work with the prototype Reflectjmce Attachment lias been utilized is the design. and fabrication of m inprored unit for permanent use. The improved mit is .being used for tbs routine dual-angle characterization f ell metallic lines. The- 847.,. 86?., 9279 .and MS lines are shaded using confuted sectors, (This .system' is being extended into the "Pulax? metallic lines and data will be available for essentially all metallic products within tbs next year.. ) Technical wrik is continuing on proving the absolute'accuracy of characterization data. The second phase of the metallic program will be to utilize the .dual-angle characterization-data in the computer formulation of metallic colors. AGvMCWU3DffiNT The Parlin Plant Laboratory greatly helped the 913-lin.e characterization work by letting do all the required bases to -944-liae single pigment quality. XSA:IL II ~ DISCUSSION BACKGROUND Metallic colors have been successfully vector-shaded m a comexcial basis by C.Q,,S since August 1964 (Ref* 1). Hie tedwique tliat made metallic vector shading possible was the dual angle color difference measurement. For nonmetallic (solid) colors the color difference between batch and standard is measured on the colorimeter at a single viewing angle normal to the panel surface. Bor metallic colors the color difference between batch and standard is measured at the normal (flat) angle and also at a fixed grazing angle (high-angle) to account for the highly directional effect of the metallic pignent. Therefore two sets of vectors, flat and high-angle, aye required to shade a metallic color to match the standard at all angles. The principal deterrent to generating metallic vectors via computer was the inability to predict the high angle effects. Flat vectors for metallic colors could be calculated with fair accuracy by characterizing the colorants in the same manner as solid colors, substituting aluminum in place of white as the reference colorant, but there liras no established technology for calculating the high, angle data. References 4, 5, and 6 indicate tliere are no satisfactory solutions to the metallic color problems in the literature to date. In fact reference 6 makes the following statement in a preface: ''There is no current theory whatsoever for the color of metalized paint films; any progress here represents a step into the unknown." The work described in this report was aimed at a practical solution to the problem by adapting existing theory to the metallic case rather than attempting to evolve a new theory. One of the restrictions on the use of the Kubelka-Hunk equation is that it assumes essentially perfect diffusion of light flux within the colorant layer. This has been interpreted to mean that the pigment particles must be randomly oriented and that therefore metallic flakes (which would tend to be mostly horizontal) would fall outside the scope of this equation (Ref, 3), Some limited experiments showed tliat the Kubelka-Munk equation seemed to work fairly well with simple metallic colors. Therefore it was logical to assume that the Kubelka4imk equation would be adequate at least for vectors where we aye interested in relatively small composition and color differences. Two approaches were taken to the problem of high angle characterization. Hie first was to make the flat characterization as with solid colors and then calculate what hypotlietical K and $ values would be required for the aluminum to give the measured high angle effects. This method was tried first because it required no equipment modifications. The second approach was to characterize the colorants at two different viewing angles. Tills metliod required design and fabrication of a special vievdng attachment for the spectrophotometer. -3 DUP030002688 Gi WMM OF OiMACTElZATI The following is a brief description of the mathematics involved to determine the color characterisation data for a mill base. Nomenclature used in the equations: K ** Absorption Coefficient . S * Scattering Coefficient R Reflectance at Complete Hiding W * Weight Fraction of a Component in a Mixture Subscript c - refers tocolorant (base) being characterised '* x refers toreferencecolorant " n - refers tomixture n " m - refers to mixture m At complete hiding the KubelhaMsmk equation gives the relationship (Ref. 2, p, 389): | * $%* (1). The absorption coefficient (IQ and the scattering coefficient (S) of a mixture are the weighted sums of the component coefficients. Therefore the K/S ratio of a mixture of a reference colorant and another colorant can be written: K - % Kr* Wc Kc 5 B^Sp-T^-Sj (2). For a two-compement mixture we can eliminate one of the weight fractions, since Wr * 1~WC. If we have two mixtures of the two colorants we can use equation (2), to solve explicitly for Kc and Sc in terms of Kr and S*. If % and flfo are the weight fractions of a colorant in two different mixtures with the reference, then the equations for Sc and Kc become: $r - [(1-%)/%) (Sr EK/S]n - Kr)l - [{l-lfo) /%0 (Sr [K/Sj* - Kr)J --imtrnmir'--w--~~~^TM"~-- fsi. Kc * (hW Wm) (Sr [K/S] - Kr) * Sc [K/S] (4). If we make two different mixtures of a colorant and reference and measure their reflectances (% and %), we can calculate [K/S)B and [K/SJjn from equation (1). Assuming we know Kr and $r we can then calculate Sc and Kc from equations (3) and (4). R is a function of wavelength, so we may have a different value of Kc and Sq at each wavelength across the spectrum. We have found that it is sufficiently accurate to pick reflectances every 10 nanometers across the visible spectrum from 400 to 700 nanometers inclusive. This gives a total of 31 points at which Sc and Kc ate calculated for each colorant. *4 DUP030002689 the -rallies ft* md Sy for the 'reference may be determined absolutely by methods outlined by j M (fef, 2..). Ikwever, all these methods ara depsnctet upon the acesuyacy of film thickness measurements on thin films. If m are intenstwd only in color, and. mot hiding power, m da not md the absolute values for % and Sr, as shown by Edwards d Yining in 1961, If m as values for %, the corresponding values of Xy are determined fro the spectral carve of the reference and equation (l). The Sc and Kc values Which w calculate for each colorant using equation (4) are then actually relative to the reference values. This is a mathematically sound approach so long as all the colorant K and S values used in a particular cwqmitim are relative to the sane reference, A white dispersion is used as reference for solid colors because white has a fairly uniform spectral curve which is markedly affected by any colorant which is added and because white is comtion to most formulas, The value of % for the white reference was assumed equal to 1.0 at all wavelengths. Theoretically we could use any two compositions of colorant and reference to determine Sc and Kc, but there are several practical limitations to consider. Since equations (3) and (4) are dealing with differences between spectral curves (in terms of K/S) then accuracy is improved by making these differences as large as possible. Also we find that Kc and Sc values calculated from two different pairs of mixtures do not agree exactly. This disparity is probably due to the combined effects of experimental error in make-up and measuring and to limitations in the theory. To get good conereojnise values for Kc and Sc across the foil composition range for solid colors, we standardized on three compositions 10/90 and 40/60 colorant/reference, and masstone (100%) colorant. Sc and Kc are calculated for the 10/90 - masstone pair and then from the 40/60 - masstone pair and averaged. The desirability of using masstone -whenever possible was shown by Edwards and Vining in 1961 and demonstrated mathematically by Nichols and Orchard (Ref, 3.). Because of the difference in dispersion and flocculation properties of different vehicle systems, individual pigments must be characterized as mill bases in each particular system where they are used, for example W-224 would be characterized separately as 42-224 for use in alkyds, as 901-224 for use in acrylic lacquers, etc. Once the K and S values for each dispersion required for a particular formula have been calculated, these data may then be used to compute the color vectors for the formula from its composition. The principal appearance difference between metallic and nan-metallic colors is the effect of viewing angle. The appearance of metallics can change markedly with viewing angle, while most solid colors appear essentially unchanged at different angles. The constancy of a solid color with viewing angle can be explained, at least for a perfectly diffuse surface, by Lambert's laws Ia *= In Cos(A) where la intensity of reflection at an angle A to the normal and In intensity of reflection normal to the surface. -5- DUP030002690 If we replace Ia Ijy the tA^s&at&ja vtiUies QJ> % sad % of the surface measured m. the normal, m cm mites Qa - % Cos {A} % Ka Cos (A) Ba Sft Cos(a) Sufestl'totiag in th cube-root equation for modified Adsms coordinates md solving in teams of J.n, a.n and bj, we get: La * (CosA}*/* On) - (InCosA)*/2) (12. SS) as (CbsA)i/2 ba " (CosA)1/2 (an) (bn) If we can put any credence in the second tern of the L equation, it indicates that L would be affected mors by viewing angle than a or b. There will be a change in saturation with viewing angle, but since both a and b change proportionately, there will be no change in hue. The coordinates change slowly with viewing angle as can be seen by the following values of (CosA)1/2, Angle A (CosA)*/2 0 1,0 45 0.3 83 0*5 89 0,25 The diagram in Appendix A shows the illumination aid viewing angles involved in the Angular Slewing Device for the "Colomasisf' colorimeter* The 45, 0 illminating5 rdewing angles of the 5'flat" position are the same as those used for solid colors. The "Hi^angle" position is obtained by rotating the panel 70 about the center line of the illuminated area. Note that this changes both the illuminating and viewing angles ,, since they are measured from the normal to the panel. We can calculate what the "high-angle" reflectance of a Lambertian (perfect diffusing) surface should be in relation to its "flat'* reflectance on this device* The change in illuminating angle will have no effect on the results as long as the photo cell "sees55 the entire illuminated area in both cases. Therefore for the Lambertian surface: % * G Co s (70j = 0.342 Gf , % - gf Cos(70) - 0.342 gf Bh * B os(70ft) 0*342 Bf Thus the ratio of the high-angle to flat tristimulus values should be 0,342 for a perfect diffuser on tlw? Angular viewing Device,' Table I below $hnm actual measured values of the high-angle/flat tri-stimulus ratios and the flat and high-angle L, a, b values for a series of solid colors.. These panels were actually .gloss enamels and therefor did not approach the perfect diffusing surface'. However s 'we can see that the high- angle/flat tri-stimulus ratio is fairly constant xmm panel to panel and averages 0,376 - 6 DUP030002691 f&r .all readings* The ratios falling furthest from the .average are for the bilack, .dark Mae, and dark gr-ees where extranely low reflectances had to fee Pleasured .for the high-angle case* This deviation may be .clue to minor inaccuracies in these low reflectance measurements. These data Indicate that -the I.a^ranLe/flat relectsn.ce .ratio is constant for solid color panels of she same gloss level regardless of hue or saturation* Tie last column in Table I marked s<Hue. Angle1 is arctan (b/a) expressed in degrees with the *a axis taken as zero degrees. For example, a pure xxd would have a hue angle of 0% a pure yellow 90% .a pure green 180, etc. . %f we Ignore the first four panels, which are adsrcmstic, m see that for solid color panels with significant saturation the hue angle is .independent of viewing angle, which is what was predicted for the perfect diffusing surface. Table II shows the same measurements made on a series of metallic panels, most of them current automotive colors. We see that fox the metallic panels the high-angle/flat ratios are much lower and more variable than the solid panel ratios. The average variation in flat to high-angle line angle is six times as great for the metallic colors as for the solid colors. There is also a much larger variation among the metallic colors. The Copper shows a shift of only 0,1 from flat to high*angle hue, but the Bright Blue shifts 24,8 with the other colors falling between these two extremes. These results, tabulated on the next two pages, show that there is no sinple geometrical reXationslnp between flat and high-angle reflectance for metallic colors as there is for solid colors. In fact the relation-* ship appears to be composition dependent. This means that the high-*angle reflectance cannot be calculated from the flat data for metallic colors, and that some new technology would be required to predict the high-angle behavior of a metallic composition. Two separate approaches were taken to the high-angle problem. The first was to find some property of the' alirniinum flake alone which would account for the angular effects of all metallic colors. The second was to treat the flat and high-angle properties of a colorant mixture as two separate mixtures. The investigations were done in the chronological order given here because the first could be started with existing instrumentation, while the second required instrument design modifications. - 7 - DUP030002692 i r r- r-- : ' O i H so <*> CM H US r-i .c SO O co tn CM CM MS* CA * -? VO .CS m M CM : C3N .o * <J m CO CM CO Is** o S>* ftp F"* CM o Cv ftQ oo gs CM Sf ft .ft o ft ft CO Mf a < ?? CS! H ii CM CM ti r**J- CM ii ~s t ? "" '"'SET^T **~-,i2rw "sg'Sjc H MS h* tn t-J -s- r-S O 6 * e 9 9 a HO f 1 O 4* oo * 3 ? P". <0 VO H ft ft sf s gs v0 CM {V. *r n0 OfS*. vO S'. OS <n us * 00 o ft# PI US (JS us o . o r*. ! I 1*.^ {'M V0 vo CM <M DO .-si .- a| CO SN| Psi Nj CMj _ ___i 00 IS Fs 00 p oo co j c s <s Jl CM aa \ Sf 3} CN S> ON COa S'*** to CO ir r~t <& *?* co rM <X> CM CM a CO H V Mi* OS 00 ! * -4 CSS SO *4 fsl oo AC ! rM tH {M tH CO CO VO 00 co eft ap c n tn r-^ rM t3 o in CA CN so 0*> -h + On aa O US i ;.?9- oa CM H ?A C\ ri pH n & CO Hs- HO S'. US 11 p*^` Es. os s CM MS PS P*5 !1 m cm H M* O9 <t ON |x* ! i Oa -4. oO so co H >*1 Oa USa 00 o .'St- -*s M M "? DO a OS O co ev US PS V0 US sr SO *<s* 13 us o CM co ft e* o o- PS I1 r*i P* 0V !*- 9* vO V0 <A PS %<j* f>* C0a CM ON CO jt SV ON ON ea mh ii ' c m **8US * l/N rl *n *? 4* *r o us so us aa 4* f SO T-i Os H a PS PS ! M3- to sia- .* cs H CO VO H tf} Xi '<32*a tf'1 CO COa CM o m CM CO so 0v O vO co PS o H , cn A** <0 ro cn a 05 oo PS r*. vfi C,O n ON na JQ M-s r> ftJ ON >* CO CM <0 <n 9 O05s na CO tn 00 Q CA CO a ON CO CO us l^p CMa .CO co CO0 cn o* JNa, 9**! cn<9 N ON CO i--x ..........i". o* N* -C 6 PS . Sia- 00 us CO. OS CO PS a ?* *H m ON .CMa CO <* ON CO a <ft r* CM O CO ON cn a CM C?v PSa ON H n a O o sy a . .. . .. 0S r-il V CO O 0 f^v rM ON ^ m O PS? PS M3- . o o| ft ft Sf' K)f MS' H O Hi vj* 0 co usf us r*-- 3SN O O H CO CO 00 CO PS] <! H H | rv> c m <3* rM | 1 p jj | J ST OM? s3* 0^3 at o ms o\ cn| CM 0N ri cn \o *,? h r*. vof h os * *? co r^l r*} S *4- c m! ps h i 4 af co | o r-*. us a c m oo CNS oo <r CM SS* *a vo CM of 1 MO 3 0 1- "1 3 | A V0i | | KJ FSj un Cs CO PS \(5 M? H i OS vol CS csl i a S co p*.? rS S ! US <3s| p'5 o s a as o o? :1 .1 1 00PS US H w cn ps PS r~; n m co! H ml si ^ in5 H| be h-'i po W 55 > <5 ! I j | i ri u rfsj >t w H *H1 tn SSl ! ! i <a 1 M O s& , i > j -u ,aj d ra H *r<? fri Wj ! -sj 4Si cs a& H "ri fo S3 t --^ j IT .. j s co *a| - a .fcj $ f-'i vc*j < --. AS ti.1 *q r'l {i--* p) ............. ! r i 1 ,3 44 .} q1 3 f ,ft-*r*>l ii W 42 mM H vl fe sa; | 1! T-; i KS i ! { . W%- 2 44 43 CS CSC r*. 'vi Pu aa; ii : *0 C) CO vOf %oj . .a; ri Of ] H *Sfl US r*i! | O O'l I ^ c m| MS' rM| a| O O! | H *rt 5fci rfl .0 a .1 o\ w oo s- NO CM in c m O IA pv Is" ae CM CM OH co CA CM a 00 ON <f H on m .r*. o n 0 OH en H OS W| PS US| ft 9 HO j 00 *4 in os a CM Mi* <rt H Pi! **3- H| oof 1 o us PS PS aw UO O CM H . OS c m ! VD Mi * O i j PS H i**. i~i a ft US O CM cH PS CM 00 o O VO S CS o OS US CS aa " o P" PS 43 .sS M rM *i"i &> at, M-S rp>si i?dC | 1 Si fs i a) I .a? a | s*< !*< f O w! s !i H *ri| te Wf 1 CS 1 i J 8) J J-j | cs I J 44 | i s-s HH Cm !SJ 1 r*1. r*5 O !>*) f) %l :sv. r ; i > ;C0 CA aa tn i/n CO CO i*v a ** m on ^st 4` 4*0 ^ PS H aa * eo vo f 4- bs US US a *<r n o 3*^ 00 WA vo '<5* \\ \6 Os a ha V0 !P *?* CM -cs* cn co 9 <0 US S'* CO US . VO **s- ft .ON O CO CM a CM CO h* ^ .. to PS o to ON to CO PS a0 sof -X \ ii>,V0' ,vr?-n'f]> i i o ol Hi .9 W W.CMi .-i-i'l' cn VO m .*> -a SSl s*H CHi 4* 4*. oo o': CM H a. PS CM P*i H. s vo 1 PS 1 i | s? ] on i cn 1 1 .ha CN 3 CO !i CA | n PS i . o r*. PS PS co n I H MS-1 US S"'3 PS of US CMJ 1 VS US E WS PS} O 1 \ p s ca a to co < 00 tO *4- I OS PS| 9 OJ co r*| vo CM ! -si } Q| H *v*il JK3| Sv CO H CO a9 Mfr Ht V0 CM H US O CM aa r*-0 oo S5* H 0M in h .s O ss- | H1 . l ha vn| *cr r^r a a3 H| j j H -ft t8 ^ H. *H U S3 W .jaj ss m H H1? ::S^ ' 1 5? I ^ 1 1 5 ! Hi >> 1 1 ! I i 4 1^1 . TJ 4 e 1 S3 B> I *H 5 <y t 1 e i B4 DUP030002693 o of o 1 11' '*He3!' *H:r vCOsf r-`i <iP'"'3S oO o Wf 4 mH<?\ tv? l*jf O* CM? in sH 41 v<.| .<?sM> n-5I| >| . 4; ' ; Si'; cMn Si-8f DUP030002694 mmmwm, s c at t er in g m mmw im me h b d Since the only difference between. m>lM sad .metallic colors is the aluminum* it seems reasonable to assume that the variation in the high .angle reflectance .of -the metallics described above could be accounted for by some property of the aluimum alone, We evaluated the hypothesis that the aluminum flakes would .have different effective scattering coefficients at different angles and that-the dual angle effects .of metallic colors could be calculated 'by merely using one .set of aluminum scattering coefficients for the flat case .and .a different set for the liigb-angXe case. Characterization data were calculated on a series of 901line colorants to test out this hypothesis. The characterization for the flat case was handled the same as solid colors except aluminum was used as the reference with an assumed S ** 1.0 at all wavelengths. The characterization panels were then measured on the Dual Angle '"Colormaster5 and the high--angle/flat reflectance ratio calcu lated for .each one. The flat K and S data were then used on the Model II Color Formulating Computer (Ref. 7,,) to determine what the quasi high-angle S value for aluminum would have to be to yield the measured Mgh-angle/flat reflectance ratio for each characterization panel. The Color Formulating Computer has the capability of computing spectral curves and tristiwlus values for any composition of up to six colorants. Each of the six input stations accepts the 31 pairs of K and S values for an individual colorant. A concentration dial connected to each input station makes it possible to independently vary the concentration of any colorant in the mixture. The technique used on the computer was to enter the colorant K mid S data in the normal manner, but to enter the aluminum K and the aluminum S values as two separate colorants on different input stations. This made it possible to vary the effective S value of the aluminum independently. The concentration dials were initially set to the composition of the panel being analyzed and the flat G, 11, and B values were determined on the computer. Then the concen tration dial for the aluminum S values was adjusted until the largest of the tristimulus values was equal to its flat value multiplied by the measured high/flat ratio fox that panel. For example, for the 10/90 868-750/867-200 panel 'die largest tristinailus reading was B, which had. a measured high/flat ratio c 0,125 on the ''Gplonaaster." The concentration dial for 868-7S0 was set at 10, the 867-100 K and the 867-100 S concentration dials ivere both set at 90 on the Color Fomsiating Computer, and the computed B value of 40,1 was recorded. Next the 867-100 S dial was decreased until the computer B value was equal to 0,125 x 40.1 or 5.01. At this point the .5 concentration (3.97 in this case) was recorded. The high-angle quasi S of .aluminum can then be calculated: (3,97/90) X 1.0 * 0.0441.' The lugh-angle aluminum S value calculated fox 867-100 alone was 0,0434 and the average of the values calculated .from all 20 characterization panels was O.0433, The 10* colorant panels averaged .0439 and the 40* colorant panels averaged .0426. Vectors were calculated for nine 1965 867-iise products and compared with the manually prepared vectors for these products. The results are tabulated in Appendix B, The majority of the computed vectors compared fairly well with the manual ones* Dark saturated colors such as 867-97213 (Maroon) and 867-37224 (Dark Sreen) -shewed the poorest correlation between, asmptvted. and manual results, Suitable K and S data were not available -for white at the time. - 10 ~ DUP030002695 ttee# pggawsitsi W-&S3 end. W-fe-S ttWmd the largest rdgb 'grgie devi&tiws* tea weakness of tbs directional S approach was that the quasi 'high- angle B value had $0 be a compress figure, The three lugh/flst trlstimalus ratios for a panel were not equal*, so cm had to be selected to calculate the S- value. The S value chosen would not give the correct Mgh/fXat ratio for the otter two fcrfetisa&us values This difference is snail for uasatuxsfced colors s but cm be qjuite significant as saturation is Increased. A possible cause steras femt the 'Ijamping of two --affects Into the high angle S value, 1;te .mans! towering of reflectance at a grtping angle (which would be experienced even with ..solid colors) and the ad^lttonal lowering of reflectance at the grassing angle due to the directional affect of the altsdaum alone. St might he possible to inwove the range and accuracy of fills method by handling the 'two effects separately. Itefter work along these lines was stopped at this point because the prototype dual-angle reflectance head for the spsetrephetometer was completed. We decided to evaluate it and pursue the more basic two-angle characterization approach because It was less speculative In nature* QIAMCTB'Rl2ATIQM MfIBOD ; The -west direct approach to the dual angle vector problem is to characterize all the necessary colorants in. atosuinum at the same two sets of iX!iinaiion and viewing conditions that are used on the Colorimeter Angular Viewing Device, The flat and high-angle colors can then be treated as two separate entities, eliminating the need to find any .rimthematical relationship between tbera. The standard spectroplK)toaster reflectance attachment utilizes an integrating sphere to collect the reflected light coming off the panel in all directions, whereas the colorimeter .measures the light reflected at essentially cm angle only. There is no commercially available equipment which could satisfactorily match the colorimeter illumination and viewing conditions on the spectro photometer, so a special unit had to fee designed and built. This initial unit was built as an experimental prototype to determine the feasibility of such a system and replaces the standard ..integrating sphere reflectance attactenent on the Bausch Lomb Spectronic SOS spectrophotometer. The Dual Angle Reflectance Attachment duplicated as well as possible the illuminating* Viewing geometry of the colorimeter equipped with the Angular Viewing Device. (See Appendix. A.) Exact duplication could not b attained because the spectrophotometer uses a single phototube to view both reference and sample while the colorimeter uses two separate phototubes.. The reference and sample .in the spectral Dual. Angle Reflectance attachment -are situated so that they are iltoadnated with 45 incident light as in the colorimeter# The phototub is .located at the .intersection, of the normals to the reference' and sample (flat position) and the axis of the tube is along the bisector o.f the angle between the normals. We can say that the geometry of the two instruments is. similar although not congruent. Bane difficulties encountered during the Directional Scattering work indicated that tte standard 19/90 40/66 -nasstone "mixtures used for .solid colors -in white mrs not adeq^Sste for metallic cteracteriratios, One of the problems was that the iO/SO'reflectance curve was often so close to the aluminum curve ll DUP030002696 tlsat slight measuring .errors caused the equations tc yield negative values fox S. In sows .cases with the very transparent pigments it was almost impossible to get .complete hiding with the mssstone and still have a usable panel, Tliere was also some evidence-, that since true siasstone was never reached in metallics, using raassteme data could distort the results., A new set o standard concentrations was used: 20/80, S0/S0, 80/20, and 33/2 color/almlnum, Only three of those wore used for any particular colorant 5 but the selection of which three to use depended on 'die properties of the particular colorant, Masstor.es were also used where they could he attained with a reasonable number of coats and where there was a good chance the colorant would be used in saturated colors approaching the masstone of the colorant. The digital cenputer characterization program was modified to mate it completely general so it could handle any concentration. All panels used were half blade, and half white primed so that hiding could be checked mstrmsentally on the spectrophotometer. Two colorants, HHL24 and W-630, siiowsd anomalous behavior in aluminum aid are discussed in a separate section. All other colorants characterised behaved normally in the flat position. In the high .angle position many of the reds and yellows showed reflectances higher tlian the aluminum in the region of their dominant wavelength. This is no problem as long as the data are consistent; i,e,, the reflectance curve for a mixture of two colorants should fall between the curves of the two colorants themselves. It can be shown mathematically (Appendix E) that if the spectral curves of two colorants intersect, the curves for all mixtures of the tiro must intersect at the sane point. In most cases due to small measuring errors the spectral curves did not all intersect at a common, point, so minor adjustments Iiad to be made to mate the data consistent. Figure I shows flat and .high angle curves typical of blue and green organic pigments. Increasing the colorant concentration lowers all points on the spectral curve of the mixture. Figure n shows typical behavior of reds and yellows which have higher reflectance than aluminum in their dominant region in the high-angle position but not in the flat. OQ'dRBLAlTQH WITH COIORMeiERS During the early stages of characterization work we found float there was a marked offset between the spectral curve of a panel run on the standard Integrating sphere reflectance attachment and the curve of the same panel run on the Dual Angie Reflectance Attachment in the flat position. There had been fairly good correlation between colorimeter readings and standard spectrophotometer readings (using the integrating sphere) on solid colors, A* reasonably good correlation between the two instruments is necessary for accurate color work. A statistical correlation was made between the spectral curves of all panels run on the Dual Angle Reflectance Attachment .and these same panels measured m the Dual Angle "Colormaster" colorimeter, A linear correlation was computed with the following results: Colorimeter Value ?= = (3) (Spectral Value) . Correilation Coeff Flat High Anglo '0,02 *0.40 1.11 0.531 ,98 .97 12 REFLECTANCE (2) W . REFLECTANCE IRON OXIDE RED METALLIC MIXTURES NORMAL VIEWING REFLECTANCE m REFLECTANCE (D 12- h ig h -a n g l e v iew in g ' + + + + 4 4 4 4 4 4. + + 4 + + + " + . + + 4 8. a 4-A a a a a a a A A a AAA A -A A A 9. . 4. o A ' tf <> A A A <> . A A A 4 <> O 4 A. A A. 4 9'- * *V V V V V V V V V V V V V V V .400 500 BOO WAVELENGTH (NM) \ - 34 - v- xr V A $ A A A A A + '4> : + 4- 4 '700' DUP030002699 .correlation constants wc?rs thea used to adjust each point or. the Dual Angle spectral curves so that their computed ttistimlus values w o k id then coincide with the values measured on 'the colorimeter. Then the K and values were computed .for the colorants using the adjusted spectral curves-;. This technique was used to characterise 39 ** SOI-line colorants and the results used to' compute flat and high angle vectors for 26 ~ 867~line formulas. A comparison of the computed and manual vectors for these formulas is given In Appendix C, Tlie statistical correlation raaaticsied above was a composite of Gr R, aid- B values and therefore gave a compromise adjustment of the reflectance curve across the entire spectrum, If G, R, and B values are correlated separately, there is a significant difference between the .constants obtained. This indicated that it might be more accurate to mhs individual adjustments to the computed tristimulus values during the vector calculations' than to use a bulbed correction factor on the reflectance curves. This -second technique was used in the dual-angle characterisation of 36 913-XIne colorants. Reflectance curves for the metallic characterisation panels were run on the Dual Angle Reflectance Attachment and a separate statistical correlation was made for G, R, and B between the spectral data and the colorimeter. K and S values were calculated for the colorants from the original reflectance data. The correlation constants for each tristimulus value were built into the vector program so that after the tristimulus values of a formula were confuted, they were individually adjusted to coincide with the colorimeter data. Vectors have been computed for approximately 125 - 945-line formulas using the 913-lino metallic characterization data. A comparison of computed and manual vectors on those formulas wliich had manual vectors prepared is shown jn Appendix D, In most cases where the computed vectors do not compare well with the manual ones, we know why. For the two silvers 97397 and 97786 tile manual 913-1121 Flat AL movement is much larger than the computed. This is because 'die formula purposely loads light for these colors. If the vectors were computed on the final composition, the AL would be larger. The manual vectors for 97473, 97547, 97797 and 97814 show *&L Flat movements for white. White will darken metallic colors in this lightness range, which the computed vectors show. The computed data for 97788 is inaccurate because the characterization data for 913-2664 did not include masstone data and there is very little aluminum (0,52%) in this formula. The computed and manual vectors for 97805 do not compare well, but shading experience using the .confuted data Iras been good. BEHAVIOR OF HIGH SCATTER PIGMENTS SUCH AS W-3.24 AND 1-650 Certain pigments behave abnormally enough in metafiles to make them extremely difficult to handle mathematically. Fortunately there .are very few of these pigments which are used in metallic formulas. Write and If 630 are two such pigments presently being used in metallic formulas which show, extreme deviation. Figure III shews spectral curves of several white/aluminum mixtures in the flat and high angle positions. In the flat position the reflectance initially decreases as the white conceixtiation increases, then the trend reverses and reflectance .increases mth increasing white concentration. In the Mghwangle position white in all proportions increases the reflectance, Figure IV shows reflectance versus white content at a single wavelength. * 15 DUP030002700 I : . - -' . -v-- " ' FIGURE III. WHITE METALLIC MIXTURES REFLECTANCE (2) REFLECTANCE <2> ' : FIGURE IV EFFECT OF ADDING WHITE TO ALUMINUM (WAVELENGTH - 550 NM) NORMAL VIEWING' ' DUP030002702 REFLECTANCE <%> FERRITE YELLOW METALLIC NORMAL VIEWING .* 867-0100 *<>*. 20, 40, 100% 868-0630 BO 40_ j + + + + + + ** + .+ + * + + A . U .9 AA A4 AA < ' V 20 J * < A A A ^ . A. A 9*9 VVAAV vvy /So soo.So WAVELENGTH (NM) V VV > ?' + + + $ &&0 TOO / HIGH-ANGLE VIEWING ' REFLECTANCE <%) DUP030002703 lids is a real phenomaMi md -act s-ym quirk m a. particular mz^reioig instrm&i tin effect is 4$uite obvious visually and on fee 'Xtotemmstwi" and ft variety of ^ctta^tty^txs* Toe ph^ai^'ica can M explained, tmt casitat -It Ma.d3.e4 mfes&atie^ly by fee 5ubelM-4%?ik option, ife formulating rm$o for nikiM in nstsiUcs is in the deqress^ refitstc rsgib., life higher owicmtrftticfts fee colots take m ft milky look. This is fertunate m that we do set Mrs to mthemtl-cally sisifeste fee reversal of fee rsfiectaaee tread Ife is r vector catoilatiens. it appears then feat we could .caloiiste m effective; set of 1C and 8 vaims for white from two sets of reflectance curves in fee -clecreasiaf refieefasce regie* K. and 8 values caIo.ilated this way gwe good vector iweisehts -.in the light ceiors, but v&ty inacairate ones in dark colors. Emmxi&tmu of the manual vector data stowed. feet fee flat white vmtms darkened colors wife L values above 40 and HghtBned mtuis with L values below 30* the white behaved is the fist position as though it had a masstone wife an t value somewhere between 30' end 40. For the 9Ql*3ine characterisation this quasi raasstone reflectance curve was assumed .parallel to fee a&eiinu reflectance Curve and ..having an l va3.ua of 35. These quasi-masstone data ware then .used to calculate X and S values for 901-124 in alwiimsa. Mute vectors computed using these data correlated much batter with the manual ones* When the '913-line bases were characterised another method was tried to determine the quasi masstone o the white* A -series of mixtures composed of bleck sad aluminum ia fee proper lightness range was prepared and a small amount .of white added to each. t%ea a point was reached where fee white did not change the lightness (i 32) , this was taken as the quasi masstpn of the white. Figure V shows the spectral curves of several mixtures of 868-0630 and 867-0100, The curves are normal in fee blue end .of the spectrum, but behave abnormally in -fee red end, 'We would expect fee mixture curves to fall between the `two mass tone curves, but fee flat curves of ail mixtures below 501 868-0630 fall below both, raasstone curves, Fortunately we can get around ibis serious problem by .characterizing 'W-630 -at concentrations below .501 868-0630 where we have psuedo-nomal behavior* Any -formula with higher than 0/50 868-0630/867*0100 ratio is closer to a .solid color than to a metallic. Shading performance at Parian on fee S^S-liab using computed dual-angle vectors has been good. Analysis of the data on 41 batches shaded by computed vectors and 48 batches shaded wife manual vectors showed an average of 3,3 hits/batch, using computed -vectors and 3.6 hits/batch using manual vectors. On eight colors which war shaded by both manual and computed vectors, fee camputed data averaged 3.1 hits/batch while the manual vectors averaged 4,1 hits/batch. Therefore, the confuted vectors appear to be slightly better than the rngisuai ones. - 2,t - DUP030002704 Gmpited vectors were generated for all new 1957 model year 857-line colors starting in June 1966 and have been used for shading all of the 867 and 927 line products ever since that time. Overall shading performance on the automotive lacquers has been equal to or better than with the manually prepared vectors* omiissoNS The relatively good comparison between the computed and manual vectors for most colors and the good production experience with the confuted vectors indicate that the <hialangle characterization approach is sound. The prototype reflectance head was built to provide sufficient data to determine the feasibility of tMs technique and to determine what refinements would be necessary for a permanent working model. Experience gained with the pro'* totype has been used in the design of a precisely fabricated working model. Characterisation panels for all 901 and 913 line bases have been remeasured on the new model as have the majority of the 42-line and 330-line mill bases. The spectral data punch system, K,S calculation program, and color vector calculation program have all been modified to handle the dual-angle characterization system. The initial phase of the metallic characterization program has been completed. We now have a working system by which we can routinely characterize colorants in any line for the generation of shading vectors. The second phase, which has been started and will continue indefinitely, is the development of techniques to improve the accuracy of the characteriza tion data, A third phase is the application of the dual-angle characterization data to pigment identification and color formulation. The Mark II Color Formulating Computer has been modified to accept two reference curves (flat and high-angle) for this purpose. - 20 - DUP030002705 i. Viudng* R !1,, .Process EsgrUeerpng Report HB-GS^S* Shading of Mstallics,i! 2, Judd, '0, B, aJid Wysxeckl* 0, wColot on Biislr.&s-s, %fer.ce., .aid . Industry/* 2<L & 063# Wiley, 3,, Nichols, 0, $,, end Orchard., St ,, J 73fc, &?& M,,, J55, If2 (Feb, X96S3, ' ...`.. 4* Bapi<isas Jl. Ri GQlaj^'^imving, 3* .22,. paa,,-Foh, jJSS)., So J5avids<ai, It, R, and Heffimdiugei% M,. Color Bn^i^eeting* 4, 13 {idaysk 1366)., ." ................. Billseysr,, el: el, ;0sin/ Tech, 44M 1.43 (X36S)-, 7.o Fai&habsr., M, 2r> &g.ineen,Eg Department Report EPL-65-7^ J'sMsifc H Cblorarit Ibismlaticn CaKootst Insttvctiou MaimI!i (June 19653-, DUP030002706 APPENDIX A COLORIMETER VIEWING GEOMETRY NORMAL VIEWING (45,0) A-l Panel HIGH-ANGLE VIEWING (25, 70) DUP030002707 mmtx b ammsm m m?~um amsm mb mw q a l v b s t o s . (Directional Scatter of Muniinua Method) Manual vector mvmmts are shown la red eluted '* w 83 s* blue 23 DUP030002708 867-9721S SamLe 867-97216 Fmm 867-97217 Ti 867-97220 BaSOK .100 458 855 253 its 100 458 253 855 *.2 ..86 *3.58 -1.22 `*1.52- -1.50 -2,38 -1.19 ' 0.50' *2*02 -r.002 1,71 2.01 -0.80 . -142 1,10 1.21 0* 52 042 -1.84 -3.31 -1.23 -io07 ' 0o7$ 1.26 -0.24 -048 6-1.04 . -044 +2.23 -9.16 -1.78 -1.49 -2.34 -2.95 -0.97 -2.36 -140 -1.95 847 0,39 2.42 2,70 .-3.00 -4.30 -0.83 -0. 03 -0,72 -0.83 "2oS2 : -2.13 2o53 2.69 347 0.56 105 1,28 1,96 100 1.27 2.60 750 -142 -0.88 557: -1.80 . -1.25 253 -1 o69 -2.0S 2.31 3.03 0.54 -0.24 146 2.36 0.24 -0.19 -2.00 -1.29 -0.97 -0.52 0,,4S 0.55 -1.86 *1.58 1.06 0.62 1.66. 0.84 0.80 ' 0,30 -0.75 *1.80 *0,56 -0.60 0,59 0.03 -040 0,39 1.15 1,29 -1.57 *2.28 -1.17 -1.70 1.03 0.47 0,72 0,81 -0.52 *0.74 -046 -0.61 -0.40 -0.03 -0.55 0.86 -1.44 -2,79 -0,80 -0.35 *0.14 0,60 1.43 3.44 ~0.88 *1.37 0,80 0,37 0,06 0.67 -1.62 -1.83 -140 4.39 -0,47 -0.91 1.38 1.04 -0.82 -1.26 2.62 l.Sl 1.56 0,70 0.82 0.85 -0.50 -0.21 "1.19 -0.53 "'0.97 -1,29 1,16 0.60 0.S8 049 -0,89 -0,27 -0.15 0.36 1,29 1.39 0.49 0.43 045 0,35 0,23 -0,01 -1.06 -0,94 +1.09 0.71 582 -2.83 -.0.4! 3.02 ' -2.30 +1046 0.60 -0,33 0.36 588 -0.40 2.02 -0.34 0.05 1,80 -0.58 0.78 1.36 253 -0.32 -0.69 1,74 -0.41 -0o 50 1.91 -041 . -0,29 100 >0.86 1.00 -0.08 -0.44 -1.42 -1.10 y iT'tt""an iinfill mri-TtnirniH nm 0.05 -0,04 .-1.07 -1.48 0.88 -0.08 24 DUP030002709 PRODUCT 867-97221 4B!MA<SSMEnHI : 100 10S 2S3 658 559 867-97223 105 750 253 458 867-97224 Dark Gmm 750 253 105 612 * *<* . FLAT * -s%-twrr M +3.43 +2.98 -1.00 "2.43 +2*52 +2*25 =4,7.0 -I, 79 -1*37 to2fl26 -0.79. -2,20 -1.66 -2O60 +3,34 +4.26 -2.00 -2.48 1.44 +1.38 4b. +0.06 +0.26 +0.3.2 +0,22 +0.63 *1,55 +2.84 +2,97 "2.54 -3,30 +2.70 +3.80 -0,77 -1.08 -1,00 -1.97 -1.24 -1.03 .+0.94 +1,35 -.2,37 -2.11 0.46 +0,88 -0,10 +0.44 -0.79 "1.07 -0.17 -0,20 +0.08 +0.07 +2.72 +1,91 -0.47 +0,13 -1.33 -0.31 -0.94 +1.91 -0.57 +0.45 +2.64 -1.63 +3.39 -1.93 +0.33 +2.37 -1,55 -4.01 +0.68 -0,98 -0,18 +0.03 +0.31 1.07 +1.52 +3.69 +0.80 +0.80 HIGH ANGLE 7 Aft ' '' ~SBUBIUIW~.IU' -0.63 -1.33 +0, M -0,56 +1,48 0.59 -1.19 +0,17 -0.62 "0o 89 -1,41 -0.67 -1.34 +0.40 +0.94 -0,90 +1.45 +1.95 +5,72 +2.54 +3,40 -1,44 -1.20 +0.70 +0.09 -4,46 1,57 +1,78 +1,45 "0.57 -0.17 -0,70 -0.94 -0,20 -0,06 +.0,67 -0.17 -1.54 -0,51 +0.25 +0.46 "rVt.* t&*i8. +0.48 -0.17 +0,07 -0.29 -0.57 +0.11 +0.04 +1.65 . +0,83 -0,1.8 +0.64 -0.39 -0,30 +0.52 +0.19 +0,26 +3,21 -0.28. -0,12 +0.36 +0,14 -0.28 -0.12 -0.51 -5.23 '-0.58 -0.93 +0.15 +0.22 +0001 +0.09 +0.44 +4,37 - z$ - DUP030002710 AHBMIX C COMPARISON CP 867-liffi GOMHJIfiD AND MANUAL IfSCOTS --^---"issrapr^itsi--Manual vector moveiasnts are shown .in red. Counted, vector movements are ..shorn in blue. The cube root color coordinates measured from the standard panel and those ecsspited by the vector program are shown at the bottom of the data for each formula. 26 DUP030002711 PEODUtT 867-97211 Oiiffisr 867-97212 MueHlray 867-97213 .aBwAiWSS*1 100 664 6253 552 588 COORD m-w ,w-Ei--"Ltr Kirriaw aa TMT5 *1.65 *1,49 s \^f*.*n7*/. .- 0.64 -0,30 -0,54 *GO10 *1,60 -0.76 -0.43 *1.10 *1.09 -0.7B 4.44 -1.4S 1.77 *0.54 *0.78 .,,n as *1.92 *9.74 1.46 -2,10 *0.97 -3.22 '-0.70 -3.16 ' *0.o 76 "4,57 -3.55 24.16 27.05 -2.37 -5.79 -5.00 6551 -2,05 -1.46 -2.69 , "1.95 1804 -2.26 -1.96 *1.77 *2.64 6253 "2*83 , -2.83 -0.09 *0.25 .105 1.76 *0*8.3 1.73 0.77 124 . -1.18 -1.26 0 *0,24 COORD :5632 58.53 *0.44 -3.47 -2.61 -2.84 -0,60 -1.12 1.52 *1,85. <1,39 1.81 *0,74 *0,44 -8,17 -7,30 ` 4804 820 355 100 6253 124 COORD 0.62 0.07 *0.84 *0.45 -13.55 -0.13 -1.23 -1.16 -2.85 -0.98 -1.46 -0,95 1.76 0.69 *4.17 *1.08 *2.59 1.85 1.10 *1.75 *1.65 *0. 36 <0.23 ' 0,31 -0*56 -0.23 -1.87 - -0,89 -0.71 "0.26 *1.11 *1.72 22.06 . J.<- 81 22.2:5 . *22.26 *0.11 *s,m *5.05 ' yy* 0.57 HIGH ANGLE *--* -0.26 sow * \r-, . ;< `j *0.02 *0,10 *0.49 -0.42 1.4:5 *0.14 *1.99 -0,32 -0,19 -0,22 -1,17. -0.42 -1,80 -O.SO *0.15 -1.01 *1.07 *1.39 * j> #0cm t\ rzt? -1.69 -,*yvj c &*,y*&.(. -2,50 1S.38 12,26 -2.06 -2,20 -5,75 . -0,77 .-0,78 -1.80 . -2.02 -0.S2 -1.22 *1.39 *2.91 -0,74 -1.89 -0,26 -2,39 *1.40" 0.66 *1.45 *0,83 *1,70 *2,03 0,61 *4.03 27.21 26,00 -0.26 -6,70. . -l,4f * -1.73 -0.24 *0,40 *1.17 *3.18 *0.80 *1,34 -0.80 -0.84 -5.34 -9,73 *0,09 *0.28 *0.72 -Oc'50 -0.15 -0.06 -0.52 -0,88 -1,76 -0.09 -0.68 -0,65 *1.12 *0.62 *3,01 *2,16 *1.63 *0o86 0.13 *0.22 *0.24 *0,12 -0,02 -0,20 -0,27 -0.11 -1.66 -0.32 -8.55 -0,13 *0,75 ' *1.34 *0.15 16.74- '*.32 l.SS 17,07 *8,68 3,86 - 17 - DUP030002712 --PiR-vOmDnfU'aifCiur-Tvs 867-07215 saajs"^ 86? "97216 m'twm TWquoise TMS-0*Si;' B 100 458 6253 834 124 0001 4i. v2,54 5,50 *0. 8.3 -2,03 *2,08 *2,44 -1.02 ->0,85 -0,20 -1,06 52,44 55,53 Buar ;Aa 46 =1,64 0.64 0.86 1,45 0.42 :KMi 0.98 0.44 "0,38 7.18 3,54 *2,59 0*^ 1.96 2.36 1,77 0,02 0.3S . -0,61 1.30 17.08 16,47 ' 855 6253 -.2.37 -3,49 -2.00 ' -SaSQ 2.50 3-64 -0.45 0,15 2.94' 2,14 -0.75 -1,39 100 1.06 0.48 -0.94 2,29 <0.36 -2.72 458 -0.86 0,23 2.03 *%<% 0p83 3.31 6750 -0,18 -0,38 -1.15 1.02 -0.62 0.25 105 GOOF OoSS *0,85 5S.S4 61,19 -0,04 "0,50 1.52 -0,67 2,15 -3,93 8.81 11.11 - 6750 6253 6557 105 100 124 COOKS 1,34 -0,76 -1.59 1.61 1.43 -1.63 1.20 1.05 <1,85 -1.58 "0,61 -1,16 1.10 +2.21 n.i6 3,20 1,08 2.19 1,02 2,26 -1.42 0.07 -3.7,5 ' +2.97 -vi &V :j! .*V* "'A* 59.07 -13.37 0.40 0.59 0.7S 0.47 -1,60 -1.95 04 55 0.17 0,26 0,12 -0,60 0,16 '6,47 -3,09 til AM3L& &r-a.-- ,*m **?. '.'0,7? 1,57 ft & n -0,21 -1.12 -1.53 -1,45 0,, 0.51 li * 0.11 -lls? }\ x*i<r* 0.60 io4i 0,02 0,S0 *j*0 ^ 1.20 0.62 1,02 1.38 3.98 -0.55 xXXi-*l-<1 "' 0,01 .27.39 3.95 27,35-. 3.18 * 1vi5? <t $\# 9,28 0,42 *2.26 1,29 2,21 *1,76 1.46 -1.30 <2.38 ,34 0.30 -0.80 -1,54 0.45 1,12 0,47 0,.20 -:0.40 -1.71 0.19 -1.20 0.29 0,49 0,86 lrv46 Cl,-40 0.38 -1.32 -0.94 0.S2 0,43 - 0.65 0.66 0.12 0,7$ -0.8$ -1.7,1 28.76 30.15 0,22 . 4.79 -1.56 5.84' 0 0,46 ';0.6? -0,91 " 0,7:9 *0.95 1,12 0.41 -0,13 -0,94 0,19 -0.19 1,38 0,64 0,80 9,71 0.97 1.51 0.23 2,30 3.92 4,44 >a <s v? v 1.75 26.01 5.69 27,34 -11.14 6.09 0,37 0.42 0,34- 1,00 =1.38 0.37 -0.06' 6.15 0.35 0,02 "0.64 "1.68 -4.18 DUP030002713 PRODUCT base 867=9721,8 6750 KHRaqS>ise 6253 6552 124 100 COORD &L 0.08 -1.24- FLAT A,*+*a*ra =5,56 =2,44 *1.80 +2.43 -X.02 =1.99 +0-.45 +1.62 +1.40 *1.78 -1.20 *0,6? =2.08 . *1.19 -1,62 =1.96 +1.24 +0.59 *0,82 +0.20 =1.26 -0.26 +1.92 +3.10 +0.20 =1,23 =0.48 28.17 - -4,86 -8.49 28.8? -5,93 . =7.14 867=97219 mrm~ 6804 =1.96 =2,27 +1.97 ' =0,31 +3.54 =0,90 6551 -1.80 =2.48 -Go 2$ =1.04 -1.S8 -2.14 6253 -1,66 -2,08 *0.06 +0,55 *2.43 +2,09 loo +2,05 -0ol9 *1.26 +3.80 +1.13 +2.45 124 =0.87 +0.39 =0.29 =1.99 +0.88 +1.46 COORD 57.85 55.86 -1.78 . -14.10 -6.67 =12.36 867=97221 Orasr-* 6253 -1,50 =1.33 =1.30 =0.76 *0.64 +0,69 653 =2,31 2.53 *2.82 =1.57 2,27 +1,71 100 +1,90 -0.54 +0.18 +3,06 -1,33 +0,41 124 -0.91 -0.12 +0.04 =1.29 =0,88 0.14 559 . -2.10 +0.89 =2,37 =1.69 +0.83 -2.12 105 - COORB +1*7.2 +2.36 S6.02 55.18. -1,46 =2.27 .+7,83 +8.08 +O.S2 *0.69 =3.97 =4.86 HIGH ANGLE AL Aa' -&12 ' e,43 =0,54 -1.86 +0.87 =0*28 -0,34 *,46 +0,37 *CI054 +1,01 -0.26 -1.22 *0.57 *1.00 =0,61 =1.85 *1.93 +5,35 -0.65 -2.05 -1,89 -0,63 *0.44 +1,23 -0,10 =0,44 =0,33 +0,17 16.52 +0,77 12.51 '=2.89 -4,20 -7.64 =0.33 *1.47 +1.24 . +3.13 -0,32 ' *0,07 =1.49 =0.82 -0.74 -0,12 -1.27 -0,79 *0.03 +0,14 =0.97 =1.37 *1.92 +2.21 *1.00 -0.10 +2.70 +1,92 *1.79 3.18 =0,50 +2,10 25.29 25.70 r.0,59 -5,07 +0,57 +2,57 -1.38 -1,05 -.9,16 =10.81 -1.19 =0,99 =1,22 =0.83 *0.08 +1,54 *2,21 +2,06 -1.08 =1,14 *0.64 +0.98 27.26 27.21 -1.32 =1.17 +1.18 1.23 -0.51 +0.02 +1.65 +1.40 *0.12 -0,16 -0,75 - =1.,44 +3,72 *3,11 *0.44 0.26 *2,14 +1.57 +0.14 +0,22 -0.70 =0.28 -1.29 =1,36 +0.16 *0,06 -2.14 =3,42 29 - DUP030002714 PKOmcr 867-97223 inair EASE iiw*w<r I0G RAT . aJU ' ' aa 2,89 2.59 =0,0.2 1.18 '' S MJ.29 -0,81 458 -0.82 -0.25 2.37 "1.48 0.42 3.70 105 2,21 0.56 -0.58 1.39 1.54 -1.19 6750 -1,11 -1.13 -2.28 "2c 60 -0.2? 0,04 6253 -1*28 -1,94 0,56 0,60 -0.O4 0,03 124 -1,01 -0.13 -0.31 -4,41 1.57 -1,52 OQORP - 54,81 -4.20 59.52 -6o04 0.86 2,18 c-$ W--ftW 1.09 2.03 -.22 -0.70 112(31 ANGLE ` 16 8.37 3,21 . *1 *,-0; a p..v 5 -0.15 0.6$ 0.70 1,06 2.27 1.S7 0,88 0,15 0.85 -0.X0 -0.55 -0,37 -1.07 -1.36 -2.33 v" -fl -0.39 -0.62 =1,53 0.16 -.0.09 -0,23 -0.35 3.61 -2 .S3 6.66 . 44.28 0*14 0,26 26,45 28.23 -2.97 -8.46 -0,08 -0.18 867-97224 isreen 105 6253 6750 124 612 6557 ODO&D 2.85 3,41 -2.09 -2.24 0.85 0.8? , =0,83 -0,94 1.58 0,84 -0.25 -0.07 -0.82 0.11 -0.26 -0.86 -1.23 -1,32 2,41 4.91 -X.69- -0.55 -2,55 -1.12 0,44 0,67 =0.84 a. 22 0.88 1.32 -0.03 0,66 -0.81 -0*40 0,92 -1,10 21.53 '-2.59 19,91 -2,07 -2.08 -1.00 0.5* 0.68 -0.40 =1.28 0.01 . 0.13 -0.1? -0.09 0,20 0.51 0,05 0,06 -0,07 -0.12 -0,69 -1.78 -0.50 ' -1,18 1.66 10.99 -1.50 -0.30 -4.19 - 2,63 0.27 0,92 -0.41 2,05 S, 23 1.29' 8 -0.24 0.74 0.20 "Q-3& -0.05 15.38 0.55 11,62 -3.51 =1,98 -3,50 * 30 - DUP030002715 FiffiJM 867.-97S43 SSt o k 6253. 355 2.00 860 3804 1393 ar" MM i&& -0.54 =0*34 +0.96 -0.33' +2,95 +2*94 +4,84 =0.65 -1.41 -0,92 -2.17 -0.79 +2.23 +0.67 +2,62 0.65 +0*63. -2.07 =0*07 +2.42 -2.20. -0*39 1.54 +G..S8 +1*19 0.75 -0,43 -0.90 -2.19 -0,66 0301 23.18 +20.65 +7.78 HI msuB- . f M, ' - Aw--a fife. *0.33 -1.65 *0,14 -0.4? +0..84 +0.81 +0,40 . +0,55 +2.2? +1,98 +0.14 0.19 -0,03 -1.26 "0.20 +1,06 =0*70 -0,73 3-3*00 +3.52 4.3$ -2.32 .*0.95 +0,17 26.92 +20.34 15.89 +11.09 '0*S -0*27 +1.14 +0,46 +0.26 -0,37 40.21 2.52 -0.36 -0.48 -0.49 -1.59 +2*04 6*86 867*97344 6253 Wssm'Wv 6557 6551 im 124 COORD =0*42 *lo09- +0*08 +0.46 +1.22 +2*70 -0.03 *1.68 -1.22 =1*62 *2.09 +3.62 -0.32 =1*00 +0,20 =0*21 +0.79 +0,62 =0*22 =0*26 +1.12 +0.84 -0.92 =0.78 -1.08 -0*66 +0*54 +1.51 -0.1$ +2.73 43,59 -5.34 =24.72 44*15 -11.70 =22*39 =0.24 . -0.49 =0*32 -0,04 =0.42 -0*57 +0.64 +1*22 1.63 +2.03 20.87 20*75 0.16 -0.03 +0.9S +0*34 +1.00 +0,47 -0*25 :0.73 =6*63 -0.90 0.-64 -2,69 +0.97 +1.27 -0.43 -0,37 -0.56 -0,19 +0.2.2 +0.65 ' -2.10 "2,22 =13*37 =15*36 - 32. - DUP030002716 w m7~mmi 105 g 124 6557 6253 6864 6750 100 c m) 3.48 4.34 ' -2,33 '-0-.56 40,5$ *0M 40,42 -1,24 >0,94 |,90 0,91 'X .-44 '1,35 -0.67 AQ*64 41,86' 2,98 Lti 42.01 -1.18 -0,67 -0,45 '01,37 42.04 -00 66 "1,39 26.50 '-9.58 27c 23 *11,64 0,67 jL&l 0.07 -14 -2.20 0.9S 40,83 =0.12 0,36 4,1,35 1.85 "0.33 4 ,23 -8,28 "9.88 ^ga"nn iiiii)Hi,iii.iiii,,:,jii ;i. mm mSOKSf'.'t** .; c- 1,05 1C*2 -2,08 1.00 3.06 -rxv,.*3P .u>** 0 -4UI2 "1,87 1.06 41.36. -0.89 -0,45 =0.20 40.30 0,90 . 40.77 4e.?f 40.38 -0,17 -0,47 -0.06 -0,7.8 40.26 -0,08 0.45 =0.23 17.00 14.61 -0,70 =6.72 *'0,S8 0,19 * $,* iM -0,71 -0,46 -1.15 0.54 0,53 * $ tWm:-; 0.73 #. 40 0,78 -9.2$ 0 -4,97 -7,56 6551 100 6253 6557 124 6804 GQQRD =2.88 '"2.-81 4#. 22 44.00 41.76 41,06 ~*JSU *? 40.21 "2,0.0 "1.72 40,46 40,90 "2.21 -1,99 40.14 ."0,33 "0.94 -1*68 1.58 41.53 -1.18 -1.84 2,84 3.49 43,75 -6.02 47.07" "12,62 "1.62- - "0.40 -1.23 *1,43 40,72 0,88 43.15 1,89 41.53 -1.21 1,96 -0,73 43,08 42.35 -0.30 *0,85 "0.3 40,17 "1.67 =1,21 -0,41 * rO.86 0,42 0.44 -0.09 42,25 2.67 48.37 -1.18 *>1.33 40.48 "0.34 =0,12 -0,72 0,84 42,15 -21,98 -19,48 21.05 21.36 6,99 "3,79 -0,81 -0,97 . 41.42 1,17 1,76 1.71 "1,00 *0.74 ' -2.71 "1,18 0,64 40.63 -2.19 "13.88 DUP030002717 mouucr 867-97590 Sum BASS 105 6353 m 6551 4804 458 COORD Mi' mr M +2.4S *2,35 *1.17 -0.77 ~0,8? -1,01 -1.1? -1.70 +8.57 +0.39 +0.72 '-056 "0.44 -0.55 -0.49 -1.57 *8.55 "'0.15 +1.38 *2.04 +0.08 +0.68 0,17 0.41 21.08 +S.26 23.33 13,12 m -8.65 -0.95 *0.21 + 0.28 "0,68 -0.61 -0.83 -0.95 0 -0.23 *1.20 *2.87 -1.53 0.18 mmPMGw M/ v to +0,45 *0.19 0.63 -0.18 -0.35 -0.29 -0,48 -0,33 -0,60 *11,04 0*08 0.45 0.58 +0,78 +0.00 -0.63 -0,96 -0.08 -0.32 .-0.06 -0.60 -0.51 t O.OS *0,26 *0.76 ' *0,51 0 -0.05 +0,15 +0,13 *0.46 +0.60 0.23 0,97 15,48 17.99 *1.25 1.80' -0,31 1.14 867-57591 Stow 613 , +0.02 -0.43 -0.23 . -0,56 *1.35 1.69 105 +2.02 -1.11 -3,10 +1,87 -1.42 -4.51 1483 -i;.2s a,,64 *0,45 +1.08 +1.90 1,94 6253 -1.66 -1.90 -0.33 -0.29 -1.04 -1.38 834 "1.8? *2.76 +0.72 . -2.00 *2.29 -0,75 100 2.80 -1.18 -1.44 +2.99 0.90 -2.98 COORD 56.18 57.85 7,52 12.85 +0,86 16,92 *0.02 0.13 1.00 1,14 -0,63 -1,06 -1.05 "0,99 -0.34 -0,92 0.93 1.62 28.17 27.22 -0.26 -.0,08 --0.05 "1.87 +0.29 *0.76 -0.82 +0.11 +1,65 +2.25 "0,S3 -0.69 +1.48 +0,34 +0.S4 +0,73 -1.89 -1.35 0.53 *0,02 - -1,02 -0.92 *0.70 +1.05 -0.60 -1.16 6.92 *8.62 - S3 DUP030002718 09 MDUQT mh m<"anv 867-97592 Wmm BASE 1816. 460 1393 6253 105 100 COORB- 3T HAT &a -1.61 1*89 *0O5 1.1? -1.83 1*30 =2.33 -2.14 -2,48 ~3,07 2.16 *0,03 -dc-52 -1.95 "1,14 +2.43 I, 21 0.71 "0,09 2.14 1.20 1.11 *0,49 29.60 *23.46 35.01 . *19*14 Ad ` BPW'r' .* "2,42 *0.57 -0.75 -1.77 -2.30 "0.19 "1,23 -0,63 *1.43 *0.29 *1.50 *0,58 13.14 *8,79. At w* (< "0,31 -1.29 HIGH ME ^la MM. w-- sir**TM, 5 *0,15 *5.04 -.0,52 "0.42 "1.13 -1.03 -0,49 "fiU.SS 4.21 -0,40 +0.20 "0.41 "0,20 "2,01 "1,99 -X. 06 . "0,70 -0,43 -1,89 "0.72 "0,49 *0,54 +0,40 +0.55 -0,66 0.39 -0.34 *0.46 *0.59 *0.77 -0.10 *0,38 0,15 17.72 . +9.-91 12.06 +19,02 +3.44 '+6,56 ui-mm MS1FW m .96 .7? 6S51 -1.5S *2.14 6557 ' -1.29 "2.76 6253 -2.12 -2.67 100 *1.36 *1.42 124 "0,48 -.3.38 COORD $9.86 61,68 * .34 +0.59 * .25 +0,60 '-0.90 "1,93 "2,37 2.56 "1,73 *0,24 -2.19 "334 *0.26 *0,79 "0.02 +1,00 0.03 +0,39 *0,50 +0.65 "0,05 *0,13 ' 0,22 ' *0,08. *0.70. -1,45 -3.89 -3.25 *0.39 *0.14 0.30 +0,26 -1.20 -3,02 -0,54 "3,60 "1,07 -1.12 -2,72 "3,98 "1,34 -3,62 +9.01 "4.2? +Q063 *0.23 +0,72 +0.71 *2.13 +6,48 "0.30 +7,27 30.47 31.88 "0.72 *2,26 >0.51 +1.03 "1.07 -3,92 -1.52 -3.71 *0,50 +1,09 *,'22 *0.79 -0.51 *0.98 "2,70 "4,89 - 34 - DUP030002719 C"19 MODOCT "-"^W-W*******^ BASE RAT &L - "ig 867-97692 Ctei|>a.gn 6S4 100 6253 m 460 124 COORD -0.66 '"0.58 *1.71 *1.53 -1.61 -1,85 *1*31 -1,21 -0.75 -0,58 -0.91 -4,43 60,33 62,48 *0,09 *2.39 1.50 *0,62 -0,48 -1.54 "1.90 *0.10 *0,03 -1.13 -0o 91 *1.84 *1.33 *0,71 *0,17. *0.87 0,53 *1.62 *1.06 *0.21 *0.11 -0.14 -3.53 *1.59" *9.81 -0,08 40,43 RICH m&M h p ~ f*iwar 40 0.31 *0,30 -0,13 *0.04 *1.24 *2.08 0.46 0,52 -9,27 -0.46 -0.49 -2,64 -1,04 -1.15 *0,03 *0.01 -0,81 -0,76 *0.41 -0,60 *1.13 *1,30 *0.65 0.88 .-0.47 -0,62 *0.55 *0,64 *0.09 1.42 *4,30 *3,69 -0.86 1.79 *0.65 ' -1,89 30.43 -0.22 31,47' . "0,30 *5,51 +11,40 m-mmi G5CT~" 458 6253 612 100 834 124 0QP8P ygagftwcwwaw.iiiiir .miTW -1,10 -4.02 *1.82 -1048 *0,32 *0,09 1.67 *2,12 *0.87 -0.47 *00.94 -1,79 54,00 55.51 .2o09 *2,43 *2.54 *2,79 -0O3S -0,10 -1.54 -1.17 *0.03 -0,25 *0.71 *0,75 -1.03 -0,93 -1.00 -1,02 *1,15 *0.61 -0,02 -0,28 *0,2.5 -0,41 *4.08 2,00 -00 90 *2.30 17.68 18,56 -0,12 *0.-S4 -1,60 *1151 -0,84 -0,85 -0.50 0,13 *0.20 *0,11 -0154 -0,04 *0.002 0,93 -0,92 -0, 73 -0.0S -0,26 <*0,04 0,67 *2.75 2,22 25.96 27,13 -0.05 ' -1,03 .*1,86 *1,67 +0.91 +1,04 -1.11 *0.76 +0.33 *0,47 -0.00 -0,71 *0.24 +0.29 +1.99 +0,07 *7.41 +8.76 - 35 - DUP030002720 HOOUCt. iBV>ArSwiHr,r. HAT Sit ~ J_' *67-97842 6293 **0.72 0,43. -0..22 498 -DJI 0.34 -1,96 0. $0 6750 -0*43 "0,80 -1.63 -2.13 100 *105S *0.80 2,42 0,57 105 6,86' 1,02 +0.79 0.72 .124 -0.21 =0.20 -3.31 0,60 COORD 60.72 -0.66 61,59 "3o05 *3.43 *4,21 -0.39 -0,21 -1.02 -1.61 -1.50 -2,32 *0.07 -1,57 1.87 *4,72 . -0.55 HIGH MM ' &a *0.03 njsfh*iS.S> -0,38 -1,23 -0.42 -0,81 *0.10 0.43 *0.03 *0,01 *2.90 *2,84 30.27 32,51 *0.17 *0.02 -1.58 -1,90 0.11 1,01 *0.32 *0.45 -0.40 *1.01 --2..GS -3,36 +2.02 +2..S3 -8,45 -0.39 -0.60 -1,75 -0.93 -1.27 *0.06 -0,60 *1,23 +3.29 867~97844 mrtffir 124 6750 100 6253 65S7 COOED . . .70. ,70 -0.-06 0,11 3.98 3.40 -1,02 -0,64 42 -0,21 20,05 18,55 0.09 =0.13 -0000 -0.48 -2,47 -0,06 0.20 -0.56 1.18 *0.83 *0.63 *2,17 -0,94 =0,74 *0,59 *0,1:9 -1.73 -2,09 *1.86 *1,67 -1.35 -1.41 -6,47 -5.85 0.52 *2.09 *0,10 -0.18 *0,65 *1,10 -0JL7 -0.04 0 -0.48 14.95 11.12 +0.06 -1.06 *0.34 1.09 -0.20 -0,56 +0.02 *0,14 +0.35 +0,78 1,82 =0,98 *0,89 -0,36 0.02 +0,40 -0,59 0,08 +0.79 +1.07 -0.72 -1.40 -2.7* -7,33 " 'p~ii n n"urn n~i 1111111 > t t ii 11 ii mi urn ii m>>i *i............. lin h i m rij --iminrrY''- iiu mh .......................................................................................... njniMturiMiin 11 mi 36 DUP030002721 867*97846 m 6750 ess? 6557 100 124 COORD 867*97847 GrsaaBfHW ...fWl'WI'Wl.1** 6750 2612 105 6253 1482 124 COORD p ***^-' " mnn'nw.wfc.w 867-98106 mr-- 100 62.53 124 6551 .318 COORD *2.33 1.54 -0.49 "0.45 *0. .**5 0,16 -1,55 -1.52 -1.54 -1.44 *0,68 -0,15 "lo85 -2.76 *1,40 *0,56 2.14 ' Ml. 38 *2,41 *0.03 -0.33 -1.78 -0.13 -0,05 56.17 S8.16 -2,73 *0.64 *1,6? +1.80 +0.37 +0.44 1.27 1.14 -2.77 -3,0? *1,36 1,20 -0,61 +0,78 -8,20 -8,28 *0.67' *0.77 -0.75 0.49 -0,97 -1.35 "1,20 -1.51 *0.9? 2,15 *2,54 *4.29 38,08 28,00 0.39 +0.78 -1,00 1.59 *0,23 -1.19 -0.72 -1.44 +0, ss *3.26 -0,95 *3.78 -2.28 -7.71 *1,16 +0,9$ +0.37 0.72 *6.00 +1,71 -1.73 -1,79 *0.$5 +2,27 -9,98 +1,60 -5.7S -8,28 -1.65 -1.36 -0.27 -006 *1.2? *0.74 -0.82 -Q< 96 -0.58 -0.52 -0.75 -2.93 57.66 61,52. *.2o66 *3.56 -0.74 -1.43 -0,18 -0.16 -4.05 -5,25 -2 28 -0,99 31,90 34,26 -3.91 -3.23 -1.43 -0,67 1,67 *1,62 *0.60 *0,23 *0,33 0,05 -0,42 *0,88 -3,40 ..-5,49 -1.54 -1,38 "0, 53 *0.14 -0,95 -2,28 -1,59- "3,31 +3,46. 1,87 *0.18 +0,23 -0.84 +0,58 *1.08 -1,07 *0.68 *0,92 +1.39 +1.66 rO.ll *0.07 -0.70 -0,97 +0.58 -0.25 *1.66 - -0.24 *1,60 -0,51 0.20 +0.41 -0.17 = 0.79 *3,56 5,45 -1.94 *5,07 *0,17 *1,34 28.13 28.89 Tjfi -rfrmini MTiii rii r urn -3.14 -9.43 *0,26 -0.35 *0.65 *1,87 -0.02 -0,99 -0.63 -0.67 *1,84 +1.76 "0,60 -0*66 -.0,18 -0,32 0.88 +1.28 *0.21 0.88 0-0.20 0 +0.34 -0.78 *0.12 +3.17 3,23 -0,24 -0.51 "0.60 ' - +2.06 -0,83 +2,61 *0.84 0 -0.12 +0.90 -+0.84 2,14 +0,32 -0,65 -1,86 -1,27 *1.65 *2.06 "6.42 *1,35 -0.84 -2,87 >0*82 +1,72 -0,15 . -0.58 *1.87 *1,35 *3.82 0.62 -0.45 "0ol6 *0,23 +0,27 2.22 +0.72 +2.85 -26.03 -4,8? -21.38 17,26 +4.72 14,04 . +0.86 ,-12.18 *17,19 57 DUP030002722 d CCMBftRISCH OF M54JMB jOCMPCtBP AND MANUAL VECTORS (Dual itagle System) Manual vector swoamts shewi in red Owpited " w w bin Tte cubs toot color coordinates assured fro the standard pel and those, computed fey the vector program are shorn at the bottom of the data for each fomila.o - 38 - DUP030002723 9*9-9903 Sssa Base m m 750 no 12% COOED, CGIBUTSD TO MfiSUAL WSOT8 m m fibf - 2.25 '+ '2,21 - lo%9 2.62 - 0,68 - 0.03 -- 0,68 - 0.37 - 0.1% - 1,68 - o;ai 1.85 + 2.96 -- 2.68 4- 2,88 - 2.98 * 2,20 - 0,93 1,13 - l.%6 81.59 0.28 21.73 4- 1,28 - 1.96 * 1.12 + a.%5 4 2.8% 4 2.00 4-2,73 2,35 - 2.5% - 2,50 4- 0,90 -12.99 -23.91 m ... .... ........... . - 0.37 4 1.22 -0,95 o%% 4 0.96 - 0.9? 0.22 - 0*27 4 2.36 - 0.27 - 0,57 4 2,39 0.0% - 0.09 4 0.29 - 0,05 - 0.80 4 1,32 4 0,%2 4 0.35 - 0.78 4 0.50 - 0.13 -1,08 4 1.32 - 0,07 - 2,%T 4 2.32 -- 0.& -3.8% 15.1% 4 0.75 - 7,0% 25.96 4 l.g% r 6%% 9%59S60% 33.m no 552 253 12% COOED, 3.68 0.25 * 3,77 4- 0.25 - 3.32 0.0% - 3%1 4* 0,23 - 2.32 4* 0,15 - 2.28 4 0.32 - 0.62 4- 0.50 - 2.38 4- 0.82 %8.%8 - %.0% %8o 20 - 7.82 + 2,25 4- 2.3% - 3.37 3.38 4 3.68 4* 2.12 - 0.6Q 4* 2.%% -18.82 -17.99 4 1.37 4 1,0% 1.60 . - 1.59 - 0.92 - 0.79 + 2.33 4 2,85 2%.07 25.10 - 0.06 4 1.30 - 0.36 4 2.02 4 0.71 - 1.T6 4 1.0% - 2,13 4 0.36 4 2,6% - o.n 4 2.16 - 0.%5 - 1,82 -- 0.68 -1.6% -1.95 -22.52 - 2.07 -12.08 Silver 121 551 83% COOED o 2.52 4 0.21 - 1.10 + 0.56 4- 0.23 - 0.86 0 4 0.21 - 0.50 - 0,2% 4 0,10 - 0,2? - 1.35 - 1.05 - 2.2% - 2.85 - 1.75 - 2,72 - 1.25 - 0.88 - 2.62 - 2.02 - 0.73 - 1.63 - 0,83 4* 0,69 4- 2.08 - 0.56 4 0.5% Ik* - 1.01 + 0.66 4- 2.69 . 0.2% 4 0.35 62.57 4- 1.36 - 0.1% 6%.0@ 1,08 > o,6o 31.32 32.33 to.a * 0.11 - 39 - DUP030002724 &-3 m.mm........ Bas@ ifo* Mlm m 63k m 63B 12* CO0SB. ALT M? Aby 4 4ol6 - .03 - 3.97 4 Oo99 '<? 0.14 - 2.55 4 0.10 - 0CM3 4 2.29 - 0.52 0.19 4 3,43 1.02 4 0.34 * 3t * 2.43 4 0.8O -- 0,48 do69 ' 3.54 - 0.05 - 1.12 4- 2,25 4 0,16 4. 2*66 - 0.82 * 1.90 - 3.33 .4 0.74 -1.68 59,3? 4 1.39 4 9,46 6144 - 1.23 t 9o 6s MjH lag Abg 4 0.46 -- 0,06 -- .2*0$ 4 0.20 - 0.09 -- 0.99 4 0.25 -- 0.11 4 1.28 4 0.30 - 0,13 4 1.3$ * 2.78 4 0,36 - 2*87 . < 1,60 4 0*30 - 0.2? - 0.90 4 2*56 0.02 - 0,35 4 1,31 4 0,44 4 3.76 - 0,11 4 0,05 4 3.93 - 0*34 - 0*39 31.18 ~ 1,51 4 4.46 30,28 - 0.20 4 3.55 9k$~$T5k7 Bing , 110 551 253 124 aid 750 103 000KD, 4 3.01 4 0.22 4 2.32 4- 2,64 4 0.13 0.87 I068 - 0 *66 <** 2X2 - o,99 - 0.32 io3$ -- X036 4 0.18 1.56 = 0o38 4 0.10 4 0.93 041 4 0.33 4 0.28 0o4l 4 0.25 4 0.59 2.12 4 2.92 0ci9 - 0.89 4 lol8 0oS6 < 0.06 * S6l 41,69 * 0.39 - IpSX 4 0173 * 2*67 4 0,70 4 2.97 4 2.28 4 0.21 * 0.94 51,37 - 2.28 13.72 47.35 6.09 -13.29 4 1.09 4 0.09 4 1.64 4 0.6? -- OoOS 4 0.78 - 0,70 4 0.13 : 0$ - 0,50 - 0,0? - 0,57 4 o.or 4 1.27 - 0.24 4 0.01 4 0,91 4 1.74 - 0.25 - o4t 4 0,94 - 0,19 0*48 - 1.25 4 2.05 4 0.3,4 - 0.42 4 0,76' -- 0,07 -- 0.23 - 1,90 4 1.11 - 0*18 " -- O.69 4 o.4l 4 1,15 4 0,21 4 2.n 4 0* 97 - 0.11 4 o.l 25.26 - 0.28 25,31 - 2,03 fljggMjggj - 40 DUP030002725 m&wm Dark Orgy 253 750 559 105 110 124 m^wrn WSiiilrv^eewr com 110 253 124 BIS coor d. 945^97774 Bill 576 no 253 819 124 COORD, 1B 24 0 - 2,38, 0,03 3,35 * 0.24 - 0.92 - 0,07 - 0,37 .=*. 2,26 * 0,30 <= 2.04 4 2,35 @.98 ~ i.20 1.22 - 1.21 3.90 0,07 .0,85 4.47 0,37 * 1.39 2.75 * 0.I7 * 0,60 + 3.34 0.32 - 1,05 0.27 * 0.18 0.48 0.60 - 0.09 0.52 27.52 2.93 * 6.19 26.69 1,67 -.5,94 * 1.56 * 0.O4 0.31 * 0,82 0,03 * 0,15 - 2.75 ' 0,05 - 1,93 9.03 0,41 + 0.10 - 1.17 0.07 8,75 0.57 0,05 0,04 2.99 * 2.38 2.77 * 2.59 2,22 * 2,22 62.32 0,95 - 1,99 63,73 ~ 0.39 0,76 1.36 * 0,88 - 2.28 - 1.52 0.94 0.93 3.77 - 1.29 0.45 5,40 - 1.98 - 0,42 * 0.94 - 0,14 * 2.85 - 0,96 0.27 2.75 0.93 2.18 0.55 1,50 2.92 1.29 0.26 * 0.22 0.74 - - 0.35 - 0.39 3,13 35.72 r 2.87 "23.01 33.49 3,59 -22.94 - 8,30 ' 0, 21 * 0,25 - 8.6 0.32 * 0,92 - 0.02 * 0.18 - 0.22 8,09 - 0.27 - 0.04 - 0.25 * 0.50 0.20 - 0.24 0.27 - 0,27 * 0.92 0.62 0,44 1.28 * 0,25 <* 0.70 - 0.42 + 0.42 - 0.29 0,67 0.14 * 0.39 + 0.85 .41 - 0.92 2.37 * 0,46 - 1.81 25.66 2,52 - 4.00 16,61 0.42 - 2,30 0.45 <* 0.04 * 0.20 * 0.01 0,81 - 0.07 1.82 - 0.13 * 0.46 - 1,26 t 0,85 0.36 * 1*42 0.17 - O.10 1,18 - 0,07 0,12 - 2.55 * 1.94 2.00 - 1.33 1,88 - 1,33 29,01 0.28 * 1,58 31.41 0.36 * 0,36 0,42 0,72 0,66 - 0,37 + 8.38 0,27 + 1.36 - 0.70 0,30 1,46 0,66 * 1.37 0.32 9,20 1.76 * 0,24 0,33 * 1.47 - 0.20 * 0.69 * 0.42 - 0.17 0.67 0,67 1,70 - 0.70 - 2.54 3,12 - 0.49 * 5,53 17.64 * 2,73 -11.25 17,38 0.62 * 8,48 - 41 - ' DUP030002726 Csde Base tibp tap Abp '.Lt, Mm no 654 334 25$ 124 105 0301. 945-97783 Rose no 807 253 124 doom 9IS-87786 Silver 121 S57 818 834 124 253 COORD. 0o48 - 0.06 - 0.92 0.35 > 0.18 - 0.93 - 0.05 - 0.09 * 0.57 -'0.35 ' - 0.35 * 2.04 0.74 + 1.22 * 0.98 i 1.38 + 1.43 ' 2,58 - 4.02 0,33 ' - 1.63 * 2.04 0.12 - 0,42 - 0.70 * 0.25 0.12 * 3.70 0.11 - 1.82 0.30 * ' 0.28 - 1.55 + 6.05 0.26 - 1.51 60.94 <2.63 < 9.15 63.20 0.21 8.29 * 0.64 * 0.43 - 0.19 * 2.20 1.18 - 1.02 0.91 1.09 0.45 * 3.57 4.07 1.38 - 0.60 ` 1.25 - 0.63 - 3.10 *> 0.28 - 0.34 + .03 - 1.02 + 0,18 0,01 0.12 - 0.95 59.13 * 6.14 2,51 57.56 + 4.33 3.51 * 4.35 e 0,10 0.34 1.78 0.21 0.05 1.92 - 2.75 3.32 - 2.80 * ,3.60 - 3.94 * 2.50 4 2.57 2,43 - 2.77 * 2.50 - 2,27 9 -1.54 2,22 9 0.15 - 1,26 * 1.14 1.41 * 0.03 + 0.11 * 4043 * 6,07 * 0.03 0.02 - 2.32 * 0.34 * 1.00 - 2.10 0.08 0,71 61.61 1.46 - 3.79 62.33 - 0,85 - 1.57 42 - % tan *% * Op 05 * 0,12 * 0,48 0.05 * 0.11 n 0.49 - 0,07 - 0,09 + 0.12 + 0,15 - 0,10 0.77 - 0.58 * 0.62 * 0.58 0.30 0.76 1,68 *> 2.84 - 0,88 * 1,73 -1.35 0,11 - (US 2,64 3,97 * 0. 09 0.31 + 0,39 6.55 - 0.20 0.19 + 0.70 - 0.13 - 0,77 * 0.77 30,79 + 0.78 * 5.51 31,00 + 0,27 + 4.19 + 0,45 0.43 - 0.10 + 0,53 0,46 0,41 * 0.76 0.47 * 0.25 - 1.42 2.16 0.82 - 0.32 - 0.59 > 0.01 - 0.84 - 0.19 T 042 + 1,34 + 0.20 * 0.01 + 4,37 ? 0,14 - 0,02 28.55 + 3.66 2,95 29.04 + 1.94 2,42 1.59 + O.X3 + 0.38 '* 0,58 0.02 0.12 * 1,26 1.48 - 2.32 - 1,54 1.54- 2,46 - 1.46 2.02 * 1.61 ~ 1,27 i.n > 1.22 0,22 * 0.65 + 3.00 0,43 * 0.87 3.40 * 1.22 - 0,21 0*22 + 1.34 - 0.14 -.0,17 - 1.55 0.01 + 0.78 - 1.23 * 0.09 * 0.43 28.13 * 0.95 - 2.74 30.00 - 0.22 * 0.67 DUP030002727 945-47788 .k$m SSI 464 - 0.45 -urn * 0.75 * 4.77 0.85 0.76 * 0.46 - urn - 1.31 1.60 - 2,29 8.87 > .8,1.2 ** 0.19 0,33 1.14 9.59 0,42 - 0.37 -ZM - 0.45 - 0.34 4 0.# 4 lP56 - 0.61 1.19 * 0.77 ' - 0.14 4 0.01 4 0.27 253 - 0.46 4 0.76 * 0.14 - 0.17 4 0.3S > 0.24 3.41 * 2.12. - 0.91 0,56 - 0.10 ~ 0.47 121 1.47 * 0,63 r 3.49 0.41 - 0.08 - 0.73 * 1.58 . 0.97 1.21 1.14 - 0.39 - 1.21 124 * 0.16 0.30 - 0.97 1.73 - 0.73 - 2.95 0.29 4 0.42 - 0.74 - 0.06 0.32 - 0.36 557 * 0,61 * 0.02 1.11 - 0,12 0.22 - 0;23 coom. 21,99 2.10 - 7.41 30.18 - 8.55 10.77 15.48 4.09 a 5.46 16,74 - 3,92 4 0.07 945': 01789 *S!7f#l5oHne(p, 121 253 0.67 - 0.05 0,83 1.22 * 0.01 0.64 -r 1,00 + 0.07 1.93 - 1.07 0.01 1,35 4 5.09 - 0.28 4 1.04 4 0. 53 - 0,08 * 0.16 - 1.02 0.14 4 1.41 - 9,43 - 0.06 0.78 - 0.34 - 0.17 - 0.45 2,00 - 0.08 - 0,94 - 124 - 1,23 * 0,24 1.43 4 2,31 * 0.26 - 1,12 * 2,36 - 1.56 - 3.32 - 1.20 - 0.26 - 1.88 551 - 3.09 * 1.58 - 3.83 - 1,05 '* 0,19 - 1.59 2,86 4 3.42 - 1,32 - 0,91 4 2.35 - 0,74 818 - 2.54 2.64 v 0.98 - 0.72 * 1.26 - 6.29 CODED. 49.39 * 0,57 -11.24 51.19 * 2.62 -12.S4 23.71 4 1.24 - 8,01 23.45 - 0.24 - 5.56 945-97790 Bwaiiywiiwi 576 819 - 1.30 2,08 '* 0.27 - 0.65 4 0,28 - 0.72 4 1.42 4 2.10 - 2,40 - 2.73 4 0.03 0.47 - 0,41 - 0,67 - 0.03 4 0,01 4 0.72 - 0,22 4 0.80 + 0,80 - 0,84 - 0,97 - 0,12 4 0,27 0,75 - 0.32 4 2.39 - 0,24 4 0.03 4 1,21 253 - 0,83 - 0,20 4 1.71 - 0.24 0 18 4 0.80 121 1.63 1,48 4 0.04 < 0.28 -y 0.52 0.53 4 0.26 4 0.19 * 0.30 4 0,31 - 0,06 - 0,26 4 0.10 - 0.44 - 0,02 0.02 - 0.20 4 0.10 750 0,23 - 0.88 4 0,51 - 0,06 v 0.25 4 0,11 124 mm,- 0.90 4 0.41 25,06 29,19 4 1.40 0,28 4 3.22 4 1.18 - 1.71 4 0,93 - 101.,3426 4 1.16 2.52 16.03 17.41 0.27 - 0.34 t ft# - 1.97 - 2,77 : 5:18 -* DUP030002728 945**97795 r rfn3***nft9&'<** no m 253 124 asm mmw Sana Gold m m 253 124 834 - oGcm 945-97798 Caballes Tan no 253 654 ' 834 124 362 CCOROo mm * 4.34 - 1.72 * 048 * 2.S6 - 1.79- 1,06 3.02 * 2.22 1.70 1.87 1.77 - 1.13 2.99 0.82 1,00 - 1.59 ^'0,62 * 1.02 - 0.88 * 0.22 * 0.09 - 1.33 * 0.60 * 0,43 54,72 * 4,96 ^ 5.40 57.86 5.48 - 5,61 2.13 - 0.23 0.63 1.81 .* 0.82 * 154 *5* 0.00 0.10 0,6.5 * 0^,00 047 * 1.44 - 2.50 - 040 " 1.08 4 1.02 0,09 0,44 4* 0*26 - 0,08 * 0.21 1,26 ' . 0.04 - 1,05 0.41 * 0.70 9.69 0,54 0.74 0.87 53.44 2.44 *10.20 53.41 -9* 2.09 13.27 2,21 0.39 - 041 2,00 - 145 * 0,86 ~ 2.78 1.43 - 2.3$ - 1.40 * 0,28 - 1.08 * 0.09 042 * 0.70 * 0,91 0.72 * 241 0.58 * 0.89 * 0.46 0,48 0,75 0.55 0,24 - 0.03 - 0.32 - 0,77 * 0.47 - 1,03 a 0,94 * 1.28 - 0*23 - 0.27 6,05 * 1,23 45.36 . 5.53 *14.36 47.44 6.56 *17.33 0.90 - 0.15 * o.si * 147 - 247 * 0,46 - 2.00 1.77 6*59 r 0,76 1.2$ - 0.53 '* 2,43 ? 0.32 * 0.87 - 0,90 r 0*52 8.69 + 1.24 * 0.31 * 0.48 * 1,86 * 0.06 - 0,45 25,93 * 3*78 - 3.63 27.39 * 3.73 - 2.79 * 0.57 - 0.30 - 0*45 ' * 0.01 - 0.24 - 0.86 * 040 * 0.17 * 047 * 0.15 - 0,04 * 0.57 - 141 o 0.46 - 1.08 - 0.87 * 0,03 - 0.46 * 1.44 - 0*23 * 0.27 2,04 *? 0,33 * 0.09 0.13 * 0.47 * 0.32 ?. 0.15 * 0.48 * 0,73 28,93 * 2.11 5.23 30.81 * 0*66 6.66 * 0,66 -- 0.06 * 0*08 0,43 - 0,43 * 0,29 - 1.45 - 0,93 + 0.34 + 0.60 - 0,06 * 0,06 * 1.27 * 1.83 * 0.07 * 0,62 24.57 26.26 - 1,11 * 0,31 * 0.30 - 043 * 0*58 * 0.57 * 0*44 0,34 * 2*29 * 2.21 * 345 * 3.38 * 2,66 - 6,95 * 0.21 * 0,93 * 0,30 * 0,46 * 0,60 * 0.59 * 0.38 0.66 * 5.26 * 8.60 44 - DUP030002729 Jtoutete,-. Bas Ms> . &m 6bp 94SH07WO Bit, Blue 253 0,46 - 0.41 2,63 0*45 0.52 2.20 ~ 0.62 1.48 0.80 559 - 1.14 + 1.55 0.85 m . * 1,53 .0,0ft - 1.80 1.52 0.00 - 2,26 m * 0.43 * 0.61 @.86 0.31 - 0,12 - 0.13 m^wmi m&mm JdL. m 4819,23 MJSi-- * 22,,2225 1,24 - 9.67 - 7.36 2.02 2.89 4lfi m - 0.11 0.54 *> 0.64 - 0.10 -0,13 0,57 o 0,48 0,44 - 0,18 0.2$ 0.32 0,23 + 0.S4 - 0.79 0,21 0.01 - 0,46 6.35 0,90 * 1.97 0,60 0.08 - 1.48 m * 0.87 0.06 : tU TOTTWt w t SSI SIS 253 124 cam. W-Mos fttatireim 110 f* 3.72 4.61 3,02 - 3.2ft - 1,79 1.57 - 1.59 * 2,65 2.59 0.42 0.45 : ill - 1.17 - 0.89 2.71 2.00 - 1,30 - 2,31 m * 1,73 0.49 - 0.06 - 0.38 1.07 2,02 zlkil- 2,04 :* .6,46.. 1,09 - 0.13 : 1:81 : m : 8:88 .m - 0.89 - 0.91 1.11 - 0,96 3.49 3.14 247 1.53 0.S0 0,19 ? 0.84 - 0,56 0.57 0,42 240 1.39 - 1,75 * 1.02 . m : 1*8 4:8 : m :fcS- 0.16 0.23 1.43 654 J* 0,33 + 0,11 2.13 : ta :Wtoil750 - 1.30 - 0,36 ;fc :k -m 630 : oil : 8:11 i1:SJ 18:81 t HI : B B mo:ll55? : fcfi : ::: fc&oil124 * f:S : : m "4-J 18:B 030. ftft ;.1;S a* -S-.lt iMt 45 DUP030002730 3tx W M$-mm Gold BSS m 484 m emm* Kft-97814 Pale Gold 121 674 484 255 124 ( C005D, 945-97816 Charcoal 253 121 124 818 759 o xsd. . 6%' 4ap % 8% Sib$ 2,32 f + 0.60 1.49 -0.41 * OJ85 * 1.84 - 1*61 1.41 * 0.79 .1.94 - 0.6$ 1.62 58.02 0.23 58.79 - 1.46 * 2.43 *= 0#80 1.94 - 0.84 2*46 * '2.72 1*55 1*8$ 1.17 - 0*91 ` *12.99 15*52 - 2.47 - 3.06 " o. 0.20 - 0.78 - 0*56 0.51 0*44 30,06 30*01 0.15 0.62 0.00 - 0.10 0,92 o.n l.SO - 1*10 1.92 0.97 044 t- 0,52 - 1,24 - 1*37 0*82' 0,87 * 0*80 - 0,59 6*56 8*1$ 1.45 ' -.1*39 * 049 - 0.06 2.14, - 0.33 0*21 2*57 - 0*14 * 0.03 0.75 - 0.45 0*24 1,12 2.14 * 2.83 3,99 - 4.01 2*58 3*94 1.61 2.18 1*52 - 1.19 * 1*52 1.44 - 3.54 * 2.40 0.16 2.21 58.57 58.37 2.61 > 2.70 * 2.71 4 3.38 - 0.48 0.13 0.15 - 0*31 * 2.03 0.71 0. 24 0*01 * 0.07 - 0.02 *> 2*00 - 1*41 0.12 - 1*98 12,81 14*66 0.19 - 0,09 0*02 0.03 2.90 - 1*47 2,13 2*93 30.88 28*67 - 0.75 - 0*72 0*69 1*00 - 0.55 - 0*11 * 0.13 - 0,26 1*13 0*48 0.66 0*02 0.55 0,03 - 2*68 - 1*23 0.86 0,89 * 7.28 7,88 - 0.10 - 0.09 0.07 0,10 0.38 - 0.37 - 0.55 - 0.19 0.04 - 0*05 1.87 - 0.09 - 0.26 4.19 <*. 0.15 0,15 - 0.98 0.70 - 1.12 r 1.26 1*28 - 1.35 - 0.16 0,58 - 0.42 - 0.17 0*61 - 0.27 : * 0.40 1.71 * 0.39 " 0.79 - 2.10 - 0.47 - 0.12 * 0.66 - 0.24 - 0*19 - 0.51 a Oo 22 35.85 2.16 - 0*44 36.25 - 0.13 2,28 18*01 0,67 a 2.60 18,36 * 0*01 0,70 DUP030002731 APPESWX B Assua a mixture of two pigments, a colorant aid a reference,, than: 1. If the masstme spectral curves o the two pigments.do mot intersect*, is it possible for the spectral curve of any mixture of"tITe two to Intersect liter masstone .curve? At any wavelength we can find the ratio of the two absorption and scattering coefficients: Let jjj and "n Than at a specific wavelength [K] 32 Kr [(l-c)/c * m] C^mix (2)o Sr E(l"C)/c*n] At the point of Intersection of a mixture with a masstone curve, either: From Equation 2 we can see that either of these conditions can occur only if n h i. But, if n * m, then , which means the two masstone curves intersect* Therefore, we have shown that mixture curves will intersect masstone curves only if the masstone curves themselves intersect. 47 DUP030002732 2 If the masstone spectral curves of two pigments intersect, do curves of ail mixtures of the two pigments intersect at the sane point? At point of intersection of masstones: Therefore lbms m and Sc m Sr HjSy Therefore all mixtures regardless of concentration (c) will have a K/$ * K/S of the reference and hence all mixture spectral curves will intersect at the point of intersection of the masstone curves ,, 48 *> DUP030002733 STATISTICAL CQRRSLAXXCgf OF TRISTIMUUUS VALBES Linear Regression: Y % sK 867 Line data i "Colomaster," Y * Spectrophotometer 945 Line data X 83 Spectrophotometer* Y * "Colonaaster'' DATA 8 867 Plat 6* R, B 80/20 H to 50/50 ts to N iif '* + A1 ir to to 20/80 H to n " + A1 to n G All ti to G 80/20 to to G 80/20 50/50 to High Angle G,R,B u to If " 80/20 SO/SO to to to " 20/80 0.09 -0.130 -0,970 0.80 "0.03 -1,51 -0.12 -0,700 3,87 3.93 4,21 1.0299 1.1243 1,1768 1.1387 1.1726 1.2292 1,0614 1.1541 0.5806 0.5933 0.5748 20/80 + A1 4.08 0.5986 r3 , .976 ,974 ,976 .974 .974 ,985 ,968 .974 ,960 .970 .923 ,926 N 57 60 63 69 63 60 19 39 57 60 60 63 945 Flat G It to R ft to B to tt R8 to High Angle G n ft " R to to " B to to Ri 2.82 2.98 2.10 3.29 -2.37 "2,33 *1,74 *2,21 Note; r3 * (correlation coefficient)2 is! * Number of samples 0.6619 0.6983 0.6889 0.6779 1.8635 1.6604 1.8338 1.8011 .912 .941 .973 ,956 .970 .903 .958 ,904 67 67 67 67 67 67 67 67 -i. 49 * DUP030002734 DISTRIBUTION Co W,, THEOBALD WlLMo FF J. Oo GRAVES Si T Co Wo STAHL Ro Jo KNAKE SI 9? * *1 :i P. Jo GRAHAM EXP. STAoo F, F. Oo Ho BULLITT, JR. - MARSHALL LAB. Go Eo LEWIS ' ,-n Go I. MUU10LLAND - FLINT DEVo LAB. No PAPPAS MARSflALL LAB. Qo Eo LENGEL MEQJELEN F, Wo TROMBLEY AJAX R, II. VINING W.'il. EDWARDS .PHI LA* 49 Wo So ARMSTRONG Po lit (C.O.S.) FILE Copy No. l 2 3 4 5 6 7 a 9 10 u 12 IS 14 15-16 .>U DUP030002735