Group Title: Citrus Station mimeo report - Florida Citrus Experiment Station ; CES 65-2
Title: Evaluation of Hunter Citrus Colorimeter for measuring the color of orange juices
CITATION PAGE IMAGE ZOOMABLE
Full Citation
STANDARD VIEW MARC VIEW
Permanent Link: http://ufdc.ufl.edu/UF00072424/00001
 Material Information
Title: Evaluation of Hunter Citrus Colorimeter for measuring the color of orange juices
Series Title: Citrus Station mimeo report
Physical Description: 5 leaves : ; 28 cm.
Language: English
Creator: Huggart, R. L
Barron, R. W
Wenzel, F. W
Citrus Experiment Station (Lake Alfred, Fla.)
Florida Citrus Commission
Publisher: Florida Citrus Experiment Station :
Florida Citrus Commission
Place of Publication: Lake Alfred FL
Publication Date: 1964
 Subjects
Subject: Orange juice -- Color -- Florida   ( lcsh )
Orange juice -- Quality -- Florida   ( lcsh )
Genre: government publication (state, provincial, terriorial, dependent)   ( marcgt )
bibliography   ( marcgt )
non-fiction   ( marcgt )
 Notes
Bibliography: Includes bibliographical references (leaves 4-5).
Statement of Responsibility: R.L Huggart and R.W.Barron and F.W.Wenzel.
General Note: Caption title.
General Note: "October 6, 1964."
 Record Information
Bibliographic ID: UF00072424
Volume ID: VID00001
Source Institution: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: oclc - 75956879

Full Text

1 -


Citrus Station Mimeo Report CES 65-2
October 6, 1964


Evaluation of Hunter Citrus Colorimeter for Measuring the Color
of Orange Juices

R. L. Huggart and R. W. Barron
Florida Citrus Commission
and
F. W. Wenzel
Florida Citrus Experiment Station
Lake Alfred, Florida


Investigations on the development and evaluation of a low-cost, $800 to
$1,000, colorimeter for determining the color of orange and other citrus juices
have been in progress since 1963. The Committee on Citrus Products at the 1963
annual meeting of the Institute of Food Technologists expressed the need for
such a colorimeter. Five thousand dollars were provided by the Florida Citrus
Commission for this purpose and an order was placed with Hunter Associates
Laboratory, Inc., for the development of a citrus colorimeter. A prototype
instrument was demonstrated and a paper presented by Hunter (1) at the 14th
Annual Citrus Processors' Meeting held on October 8, 1963, at this Station.
The theoretical basis for the design of the Hunter, Model E45, Citrus Colori-
meter was discussed in this paper, together with consideration of some of the
characteristics of orange juice that effect its color. Some results were also
presented which had been obtained by using this instrument for measuring the
color of orange juices.

The purpose of this paper is to present and discuss data that have been
obtained during the past year by the authors in evaluating the Hunter Citrus
Colorimeter. The instrument was returned for modifications to the Hunter
Associates Laboratory on two occasions during the past year. Results, per-
taining to the Hunter E45 colorimeter, given in this report were obtained
after its latest modification.


Experimental Results and Discussion

Hunter E45 values and visual color scores for reconstituted frozen concen-
trated orange juices are reported in Table 1. The authors, together with
either Mr. M. D. Maraulja or Mr. R. W. Olsen scored the juices using USDA
plastic color comparator tubes and procedures. Also, given in Table 1 are the
Hunter Color Difference Meter (HCDM) "Rd", "a" and "b" values for both the con-
centrates and the reconstituted juices. These 16 samples of frozen orange
concentrates were prepared by mixing various commercial Florida concentrates in
such a manner that wide ranges of HCDM values were obtained in the composite
samples. The range for the Hunter E45 values was from 62 to 313. The HCDM
"Rd" values for the reconstituted juices ranged from 20.3 to 27.1; the "a"

Florida Citrus Commission and
Florida Citrus Experiment Station,
Lake Alfred, Florida. 400-10/6/64-RLH











values from -7.5 to -3.0; and the "b" values from 28.1 to 29.7. HCDM values
within these ranges: would be found in reconstituted juices from at
least 95% of the commercial 420 Brix frozen concentrated orange juice packed
in Florida. These samples contained juices from early, mid- and late season
varieties of oranges.

The'Hunter E45 values for the reconstituted orange juices (Table 1)
gradually increased as the visual color scores increased from 32 to 38 and
the HCDM "a" values did likewise, except in the case of sample C-8 which had
a color score of 32, a Hunter E45 value of 76 and a HODM "a" value of -6.7.
Both of these Hunter values for sample C-8 are greater than those for sample
C-9. This was caused by a greater degree of lightness (whiteness) as indi-
cated by the larger HCDM "Rd" values of'29.3 and 27.1 for this concentrate
and.reconstituted juice, respectively. The Hunter E45 and HCDM "a" values
show the same trend relative to the visual color scores because both instru-
ments can be used to measure the degree of redness in the color of the orange
juices.

Color measurements were also made with both the Hunter E45 and the HCDM
on three samples of 650 Brix California frozen orange concentrates, submitted
by Mr. Gordon Biesel, Orange Products Division, Sunkist Growers. Visual
color scores for these products were also determined. The color of the re-
constituted juices, as shown in Table 2, was better than that of the OJ3
plastic comparator tube, which corresponds to a color score of 38.

Minimum and maximum color scores and HCDM values are listed in Table 3,
which were obtained from the examination of 527 samples of frozen concen-
trated orange juice. These products were packed in Florida commercial plants
during the past three seasons. Since the Hunter Citrus Colorimeter was not
available until 1963, the only Hunter E45 values given are for samples ob-
tained during the 1963-64 citrus season. The percentages of the total number
of samples of commercial frozen orange concentrate, whose color characteris-
tics are indicated by the minimum and maximum Hunter E45 and HCDM values, are
small. This is indicated by the frequency distribution tables of HCDM values
for reconstituted commercial frozen concentrated orange juices, which are
included in Citrus Station Mimeo Report CES 65-10. This report is also in-
cluded along with all of the papers presented at this meeting, copies of which
were given to all persons attending.

The frequency distribution of color scores for samples of commercial
frozen concentrated orange juice are presented in Table 4. These samples
were collected during the past three citrus seasons and the 1963-64 samples
were used to determine the correlation between the color scores and the
color values of the reconstituted juices, as indicated by either the Hunter
E45 values or the HCDM "a" values. Since a major difference between the
color of the 1963-64 midseason samples and that of the samples collected
during the two previous years was indicated, this opportunity is taken to
call attention to it.

From the frequency distribution data in Table 4, it is evident that the
percentage of midseason samples given a color score of less than 34 was 0.0%,

Florida Citrus Commission and
Florida Citrus Experiment Station,
Lake Alfred, Florida. 400-10/6/64-RLH







Table 1. USDA color scores, Hunter Citrus Colorimeter values and Hunter
Color Difference Meter values for samples of commercial frozen concentrated
orange juice


USDA Hunter
color E45
scores1,2 values
32. 76
33 62*
96
34 110
118
110
138
170

35 193
198
36 203
37 232
276
38 298
I" 309
313*


Rd
29.3*
26.0
25.9
26.0
27.1
25.2
25.1
23.5
23.3
23.8
22.2
21.8
21.2
20.3
20.9
19.0*


Hunter Color Difference Meter values


Concentrates


a
3.2
1.7*
2.2
3.3
3.2
2.8
3.6
4.0
5.5
5.0
5.4
5.9
6.9
7.1
8.7*
7.9


b
34.3*
32.7
32.6
33.2
33.4
32.7
33.1
32.5


Reconstituted


a
-6.7
-7.4
-7.5*
-6.7
-6.9
-7.1
-6.9
-6.5


Rd
27.1*
26.2
25.5
25.5
25.4
25.0
25.0
23.6


32.7 23.8 -5.4
32.5 24.2 -5.8


32.0 22.8


-5.4


32.2 22.8 -4.6
31.9 21.8 -4.0


31.2
31.8
31.1*


20.9
21.9
20.3*


-3.8
-3.0*
-3.1


1


USDA color scores determined visually using plastic comparator tubes. These
tubes designated as OJ3, 0J4, OJ5, and 0J6, corresponding to 38, 36, 34, and
32 color score points, respectively, were first used during the 1963-64
citrus season.


Indicates either minimum or maximum values.
2 The color scores were determined by personnel of the Florida Citrus Commission
and the Florida Citrus Experiment Station and not by processed foods inspectors
of the Agricultural Marketing Service of the United States Department of
Agriculture.

Florida Citrus Commission and
Florida Citrus Experiment Station,
Lake Alfred, Florida.
400-10/6/64 RLH


juices
b
28.5
28.3
28.1"
28.9
28.3
28.2
29.2
28.9
29.3
29.0
29.0
29.4
29.3
28.9
29.7*
29.4


Sample
code
C-8
C-9
C-10
C-6
C-7
C-11
C-5
C-12
C-13
C-4
C-3
C-14
C-15
C-2
C-16
C-1


1










Table 2. Color measurements on samples of California frozen concentrated orange juice
Hunter Color Difference Meter values Visual Visual
Description 650Brix 42oBrix1 12Brix1 120Brix1 color color
of samples Rd a b Rd a b Rd a b Hunter E45 scores2 scores

Early season 11.2 + 8.8 25.8 19.8 + 7.4 32.2 24.1 -0.9 33.0 390 41 40
Midseason 10.8 +13.5- 25.6 18.2 +13.0 31.7 21.9 +3.6 32.9 562 43 43
Late season 10.4 +16.1 25.2 16.7 +15.9 30.6 19.5 +5.9 31.4 669 45 45

Prepared from 650 Brix sample.
2
Hypothetical score determined by extension of regression line for relationship between Hunter Color
Difference Meter "a" values for reconstituted juices and visual color scores, using comparator tubes
(glass filled with plastic). This scatter diagram and regression line were published in Proc. Florida
State Hort. Soc. 75: 331-336. 1962.

Hypothetical score determined by extension of regression line for relationship between Hunter E45 Citrus
Colorimeter values and visual color scores, using plastic comparator tubes (See footnote, Table 1).







Florida Citrus Commission and
Florida Citrus Experiment Station,
Lake Alfred, Florida.
400-10/6/64 RLH











Table 3. Minimum and maximum visual color scores, Hunter Color Difference
Meter and Hunter Citrus Colorimeter values for samples of commercial frozen
concentrated orange juices collected semi-monthly during three citrus seasons


Visual color scores

HCDM "Rd" values

it "b" if
u "aE5
t11 IEb,5 ,

Hunter E45 values


Reconstituted juices
Season 1961-62 1962-63 1963-64
Samples 198 165 164
Min. Max. Min. Max. Min. -Max.
34 39 33 37 32 38


19.4
-8.0


28.2
-3.2


21.9
-7.3


29.8
-1.9


18.4
-7.9


26.2
-1.6


25.5 31.6 28.9 35.0 26.2 31.9


78 378


Florida Citrus Commission and
Florida Citrus Experiment Station,
Lake Alfred, Florida.
400-10/6/64 RLH








Table 4. Comparison of frequency distribution tables of visual color scores for samples from midseason
and late season packs of commercial frozen concentrated orange juices collected semi-monthly during three
citrus seasons

Reconstituted juices
Midseason Late season Total
Visual samples-% samples-% samples-%
color Season 1961-62 1962-63 1963-64 1961-62 1962-63 1963-64 1961-62 1962-63 1963-64
scores1 Samples 101 78 76 97 87 88 198 165 164
32 -- 1.3 -- -- -- 0.6
33 -10.3 40.8 -2.3 - 6.1 18.9
34 10.9 43.6 38.1 1.1 2.3 5.6 21.2 18.9
35 40.6 41.0 13.2 56.3 3.4 20.7 49.1 7.9
36 35.6 5.1 6.6 1.0 32.2 27.3 18.7 19.4 17.7
37 11.9 - 20.7 8.1 56.8 16.1 4.2 30.5
38 1.0 -- 68.0 -10.2 33.8 -5.5
39 -- -- 10.3 -- 5.1 -
40


Uolor scores determined visually in 61oi-o6 and
1963-64 using plastic tubes (See footnote Table


1962-63 using plastic in glass comparator tubes; in
1). The plastic in glass tubes were numbered 1, 2, 3,


and 4 and corresponded to 38, 36, 34, and 32 color score points, respectively.


Florida Citrus Commission and
Florida Citrus Experiment Station,
Lake Alfred, Florida.
400-10/6/64 RLH











10.3% and 42.1% for the 1961-62, 1962-63 and 1963-64 seasons, respectively.
This indicates that the color of frozen orange concentrates, packed during
the midseason period, has become progressively poorer during the past three
citrus seasons. Late season packs of frozen concentrated orange juice almost
always will be given higher color scores than those products packed during
the midseason period. This is also evident from the distribution data in
Table 4. Color scores less than 34 were not given to any of the 1961-62 or
1963-64 late season samples and to only 2.3% of the 87 late season samples
that were examined during the 1962-63 season, when the severe freeze of
December 1962 occurred.

To evaluate the Hunter E45 Citrus Colorimeter, it was necessary to deter-
mine correlation coefficients between the visual color scores for reconsti-
tuted frozen concentrated orange juices and the redness attribute of the color
of these juices, as indicated by the Hunter E45 values. These correlation
coefficients are listed in Table 5, together with those between the visual
color scores and the HCDM "a" values, which also indicate the redness of the
color of the juices.


Table 5. Correlations between visual color scores and the Hunter
Citrus Colorimeter values and also the Hunter Color Difference Meter
"a" values for reconstituted commercial frozen concentrated orange
juices

1963-64 Season
Midseason Late season Total
Visual samples-76 samples-88 samples-164
color scores
and Correlation coefficients (r)1
Hunter E45 values 0.80 0.80 0.94

HCDM "a" values 0.63 0.53 0.832

All of the correlation coefficients in this table are statistically
significant at the 1% level.

2 Corresponding coefficient of 0.91 was reported (3) after examination
of 215 samples packed during 1954-55 season.


Kramer and Twigg (2) consider that a correlation coefficient of 0.90 or
better is an excellent indicator of human evaluation; also, that a correlation
of 0.80 or better is satisfactory for use, although a higher correlation is
desirable. This is predicated on experience in their laboratory that corre-
lation coefficients rarely exceed 0.94, when determined from data obtained by
two different sensory panels in evaluating such characteristics as the color
or flavor of a food product. Therefore, any correlation coefficient, listed
in Table 5, that is equal to or exceeds 0.80 is considered by the authors to
be of satisfactory practical importance. Since the correlation coefficients

Florida Citrus Commission and
Florida Citrus Experiment Station,
Lake Alfred, Florida. 400-10/6/64 RLH











between the visual color scores and the Hunter E45 values for both the midseason
and late season reconstituted juices are 0.80, the possibility of using this
instrument in the future to measure the color quality of orange juices in quality
control laboratories of citrus processing plants is considered to be good at
this time.

It should also be pointed out that the correlation between the subjective
visual color scores of orange juices and the objective Hunter E45 values is
much better than that when the HCDM values are used, especially when the mid-
season and late season samples are considered separately.

There is a problem connected with the measurement of color by a single-
reading instrument which Mr. Hunter pointed out recently. 'Mr, Hunter wrote
"I would like to comment on the fundamental question you asked that concerns
a problem which is common to all one-dimensional color scales. Color is
three dimensional, no one-dimensional scale can eliminate the problem of mis-
grading odd-ball colors.

We have the same problem with the Tomato Colorimeter even though the
Tomato Colorimeter obtains signals for "L", "a", and "b" separately. It re-
duces the three signals by a specific equation so as to obtain values for
tomato color on a single scale. In the process, it becomes possible to
misgrade dark samples discolored by mold. They are graded as tomato juice of
the very best color. The instrument measures according to a formula and the
formula is applied to any colors, not just colors that look like tomato juice.

The Tomato Colorimeter is a $2725 instrument because it measures all
three color quantities. If the users are willing to measure these quantities
separately and use graphs, they could obtain fool proof ratings of juice. So
long as they measure by equation, however, they run the risk of misgrading
odd-ball samples.

The same situation exists with the E45 where you are measuring orange
juice color by the equation: Color = k (A/Y -C). There is no possibility
that I can see for getting fool proof instruments which properly grade juices
on a one-number scale. The user will have to be told to reject samples that
do not look like orange juice."

The authors have been advised by Mr. Hunter that better correlation may
be possible if some other modifications are made to the Hunter Citrus Colori-
meter. Since this was desirable, the instrument has been shipped to him for
such modifications. When it is returned to the authors, it will be evaluated
again during this citrus season, using not only frozen concentrated orange
juices but also other citrus juices, such as white or red grapefruit, lime
and lemon.

References

1. Hunter, Richard S. 1963. A colorimeter for grading citrus juices.
Citrus Station Mimeo Report CES 64-3. Fourteenth Annual Citrus Processors'
Meeting, Florida Citrus Experiment Station, Lake Alfred, Florida.

Florida Citrus Commission and
Florida Citrus Experiment Station,
Lake Alfred, Florida. 400-10/6/64-RLH











2. Kramer, Amihud and Bernard A. Twigg. 1962. Fundamentals of quality
control for the food industry. 512 pp. The Avi Publishing Company, Inc.,
Westport, Connecticut.

3. Wenzel, F. W., and R. L. Huggart. 1962. Relation between Hunter
Color Difference Meter values and visual color of commercial frozen concen-
trated orange juice. Proc. Florida State Hort. Soc., 75: 331-336.


Florida Citrus Commission and
Florida Citrus Experiment Station,
Lake Alfred, Florida.
400-10/6/64 RLH




University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs