Group Title: Citrus Station mimeo report - Florida Citrus Experiment Station ; CES 68-8C
Title: Juice extraction for pounds-solids determination
CITATION PAGE IMAGE ZOOMABLE
Full Citation
STANDARD VIEW MARC VIEW
Permanent Link: http://ufdc.ufl.edu/UF00072451/00001
 Material Information
Title: Juice extraction for pounds-solids determination
Series Title: Citrus Station mimeo report
Physical Description: 5 leaves : ; 28 cm.
Language: English
Creator: Blair, James G
Citrus Experiment Station (Lake Alfred, Fla.)
Florida Citrus Commission
Publisher: Citrus Experiment Station :
Florida Citrus Commission
Place of Publication: Lake Alfred FL
Publication Date: 1967
 Subjects
Subject: Citrus -- Processing -- Florida   ( lcsh )
Citrus -- Yields -- Florida   ( lcsh )
Genre: government publication (state, provincial, terriorial, dependent)   ( marcgt )
non-fiction   ( marcgt )
 Notes
Statement of Responsibility: J.G. Blair.
General Note: Caption title.
 Record Information
Bibliographic ID: UF00072451
Volume ID: VID00001
Source Institution: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: oclc - 76803621

Full Text




Florida Citrus Commission and
Citrus Experiment Station CES 68-8C
Lake Alfred, Florida. 450-10/12/67-JGB


JUICE EXTRACTION FOR POUNDS-SOLIDS DETERMINATION '

J. G. Blair
Florida Citrus Commission
Lake Alfred, Florida


Pounds-solids has been a measurement of quality for about 20 years, but
there still are many people engaged in various citrus activities who do not
understand what it means. Although this measurement is not new the significance
of its importance is relatively new. In 1961 the USDA was requested to outline
the problems associated with obtaining pounds-solids measurements and submit
recommendations for their solution. The industry accepted the USDA recommend-
ations, and so, the Florida Citrus Commission established the Pounds-Solids
Project. The objectives of the project were covered in detail by the USDA re-
port, but for easy reference they are given below in a condensed form:

1. To design a sampler that can be installed in the fruit
receiving system of all processing plants, sample
selection to be absolutely nonbiased and controllable
for size, regardless of the size of load or rate of
unloading.

2. To encourage the development of an extractor that will
be dependable, simple to operate with minimum adjustments,
and easy to clean; juice determination to be independent
of variety, size, peel thickness, or condition of fruit
within tolerances to be established, Consistent results
must be obtained from similar samples of fruit over
measureable and allowable ranges of machine tolerance. It
would be desirable for the machine so developed to have no
commercial application.

3. To develop a method of automatically determining Brix and
acid having a complete print out system.


Priority of accomplishment followed the same order as that listed above. A
sampler has been developed that completely satisfies the first objective. The
use of this sampler is now mandatory for all plants requesting State Certification.
Some additional experimentation has been done on ultimate sample size and the re-
sults of this preliminary work were presented at the Florida State Horticultural
Society meeting in November 1965.

Research has been continuous toward accomplishing the objectives outlined
for the development of a juice extractor. A quick solution to the problem was
of course desirable but it takes time to develop and prove an extractor that
will conform to the exacting specifications. The industry was understandably
impatient during the last 5 years despite the progress made. During this
period, a new single-head and several models of multiple-head machines were
built by Brown Machinery and tested as part of the pounds-solids project. In
fact, the Model 2701 machine was recommended for fresh fruit house use, but so
far, this recommendation has not been implemented. FMC Corporation was, also,










engaged in developing experimental extractors although their approach was some-
what more conservative than Brown Machinery Company. Even so, engineering and
manufacturing time was necessary to make new orifice tubes and design and re-
design the modified principles of operation.

The results of the 1965-66 pounds-solids testing program were disappoint-
ing in that they did not completely meet the requirements established by the
industry. Our report of these findings was closely followed by a meeting of
representatives of key industry organizations at which time it was decided that
a final decision on a test room extractor could not be delayed beyond March 1967.
This decision was predicated on the paramount importance of having a single
machine as soon as possible. This strong desire for quick answers necessitated
a program designed for limited goals instead of a continuing program for the
accomplishment of the objectives established by the USDA. The modified Brown
2701 machine was selected as the control to be used as the base for all analyti-
cal comparisons. All results obtained previously indicated that the experimental
extractors operated most efficiently at a theoretical 100% level. Despite this,
the proponents of reduced yield setting were successful in having the program
designed so that the extractors would hopefully, produce something like 5% less.
This added variable only cluttered the program which in turn made the analysis
more complex.

The pounds-solids extractor program during the 1966-67 season had more
effort concentrated into a short period of time than any single project ever
conducted at the Citrus Experiment Station. The carrying out of this accele-
rated program was facilitated through the cooperation of Florida Department of
Agriculture, Florida Citrus Commission, Florida Agricultural Experiment Stations,
and the machinery manufacturers

Dr. Frank G. Martin of the Statistics Department, Agricultural Experiment
Station, Gainesville, who has been providing statistical assistance for the
Pounds-Solids Project, prepared a detail plan for obtaining necessary data for
sound statistical evaluation. The three main variables he had to consider were
machine precision (repeatability), fruit size, and variety biases. Within the
capability of the Station facilities, personnel available and limited time,
these variables were satisfactorily measured.

The decision date of March 1967 was too early to permit testing of Valencia
oranges and even necessitated an unusual early start to assure sufficient data
for sound statistical analysis of the other citrus varieties. The fruit for all
tests was preselected from trees that most nearly represented the variety de-
sired; these trees were marked for harvesting later, as needed.

The seedless variety used was Hamlin from rough lemon rootstock; the seedy
varieties were Parson Brown, initially, followed by Pineapple. Tests were begun
in late September 1966, although the fruit was extremely immature and not
generally acceptable as canning plant fruit. The last tests were run in early
February at which time the seedless (Hamlin) variety was over-mature but typical
of some of the fruit processed at this time. Pertinent information from each of
approximately 1000 tests per month was recorded for immediate, as well as,
future engineering and statistical analyses.

Florida Citrus Commission and
Florida Citrus Experiment Station,
Lake Alfred, Florida. 450-10/12/67-JGB








The amount of data gathered is mountainous but has been carefully evalu-
ated. The basic conclusions reached are sound and are supported statistically.
One of the most important was machine repeatability. Table 1 shows estimates
of tolerance limits which state the range in percent about the mean within
which (plus or minus) it is expected 95% of the yields fall. For example, if
100 samples of Large fruit were extracted on the Brown 3300 machine, the juice
yield of 95 samples would be within plus or minus 1.11 percent of the average.


Extractor


3300

091B


3300

091B


Table 1. Estimates of machine repeatability
Seedy Seedless Small Large

Size


1.14

.72


1.22

.88


1.00

.66


1.11

.73


Constructed

1.16 -

.82 -


Average


1.05

.70


1.19


The conclusions that can be drawn from above test data is that the FMC
machine (091B) was consistently about 0.35 percent better on precision.

Table 2 includes the results of the tests relative to the size variable,
The information contained in this table may be summarized as follows: the
Brown 3300 extractor always obtains higher yield on the Large sizes of fruit.
This is clearly illustrated by comparing the average differences on the
Seedless type:


Brown 3300
FMC 091B
Hand Ream


Small
+1.54
-1.23
+2.87


Large Difference
+1.43 0.09


-2.30
+1.59


1.07
1.28


The comparison on the Seedy type is approximately the same, therefore, it can
be concluded that the Brown extractor is biased for size.








Florida Citrus Commission and
Florida Citrus Experiment Station,
Lake Alfred, Florida. 450-10/12/67-JGB









Table 2. Variance from control for size


Extractor




2701 (control)
3300
091B
091
Hand ream


2701
3300
091B
091
Hand ream


2701
3300
091B
091
Hand ream


2701
3300
091B
091
Hand ream


Series
Sl




54.46%
+ 0.76
2.77
4.29
+ 2.51


54.75%
- 0.37
2.91
3.90
- 1.98


Parson
56.80%
+ 1.60
+ 2.70
+ 1.64
+ 1.99


57.11%
+ 1.75
+ 2.98
+ 2.19
+ 1.07


Series Avg. Series
S2 SI+2 S3

Small (252's)
Hamlin Seedless
58.08% 56.27% 54.92%
+ 1.38 + 1.07 + 2.53
0.76 1.76 0.25
2.43 3.36 1.72
+ 2.40 + 2.45 + 3.07


Large (163's)
Hamlin Seedless
58.97% 56.86% 56.08%
+ 1.54 + 0.95 + 2.38
- 1.50 2.20 1.94
- 2.44 3.17 3.03
+ 0.92 + 1.45 + 1.46


Small (252's)
Brown-Seedy
58.04% 54.42% 5
+ 0.97 + 1.28 +
+ 2.47 + 2.58 +
+ 1.26 + 1.45 -
+ 1.96 + 1.97 +


Large (163's)
59.15% 58.13% 57.10%
- 1.31 + 1.53 + 1.13
- 1.99 + 2.48 + 0.61
- 1.59 + 1.89 0.88
- 0.42 + 0.74 0.72


Series Avg.
S4 S3+4


53.63%
+ 1.50
- 1.18
- 1.86
+ 3.53




54.72%
+ 1.44
- 2.89
- 2.56
+ 2.00


54.28%
+ 2.01
- 0.71
- 1.79
+ 3.30




55.40%
+ 1.91
- 2.41
- 2.79
+ 1.73


Pineapple Seedy
5.79% 54.16% 54.97%
1.16 + 0.41 + .78
1.41 + 0.95 + 1.18
0.27 0.66 0.46
0.95 + 1.42 + 1.18


54.95%
+ 0.63
+ 0.13
- 0.37
- 0.08


55.97%
+ 0.88
+ 0.37
- 0.62
- 0.40


Overall
Average




55.27%
+ 1.54
1.23
2.57
+ 2.87


56.13%
- 1.43
- 2.30
- 2.98
- 1.59


56.19%
+ 1.03
+ 1.88
+ 0.49
+ 1.57


57.05%
+ 1.20
+ 1.42
+ 0.63
+ 0.17


Table 3 includes the results of the tests relative to the variety variable.
The information in this table can be summarized as follows: the FMC 091B ex-
tractor always obtains less yield from the Hamlin variety. This is clearly
illustrated by comparing the average differences:


Brown 3300
FMC 091B
Hand ream


Hamlin
+0.88
-2.03
+1.94


Parson Brown
and
Pineapple
+0.64
+0.98
+0.70


Difference
0.24
3.01
1.24


Therefore, it can be concluded that the FMC extractor is biased for variety.

Florida Citrus Commission and
Florida Citrus Experiment Station,
Lake Alfred, Florida. 10/12/67-450-JGB









Table 3. Variance from control for variety
Constructed (324's through 163's) Series
Series Series Avg. Series Series Avg. Overall
Extractor Cl C4 C1+4 C5 C7 C5+7 Average


Hamlin Seedless
2701 (control) 54.71% 58.10% 56.40% 59.46% 54.61% 57.04% 56.72%
3300 2.28 + 0.57 0.85 + 0.89 + 0.98 + 0.92 + 0.88
091B 2.78- 1.79 1.98 1.10 3.07 2.09 2.03
091 4.09 3.40 3.74 2.19 3.35 2.78 3.26
Hand ream + 2.33 + 1.94 + 2.13 + 1.10 + 2.42 + 1.76 + 1.94

Parson Brown-Seedy Pineapple Seedy
2701 (control) 53.72% 57.43% 55.58% 57.75% 55.24% 56.50% 56.04%
3300 + 0.54 + 1.39 + 0.96 + 0.66 + 0.01 + 0.33 + 0.64
091B + 1.55 + 3.52 + 2.53 + 1.30 0.17 0.56 *+ 0.98
091 + 0.62 + 1.90 + 1.26 + 0.06 0.92 0.43 + 0.41
Hand ream + 0.76 + 2.28 + 1.52 0.74 + 0.53 0.11 + 0.70



No particular analyses were made for condition of the fruit since this was
not part of the program, but observation of the results indicates that both
machines perform best on fruit at its prime stage of maturity.

Juice analysis. The primary purpose of making a juice analysis is to find
the Brix value. This is now being accomplished in the State Test Rooms by the
use of a hydrometer. The accuracy of a hydrometer under carefully controlled
conditions is + 0.1 degree Brix. Improvement of this accuracy together with the
elimination of the human element in the actual measurement is desired. Therefore,
research is being conducted toward accomplishing both of these objectives. At
present, a unit built by Electron Machine Company is being tested to determine
how closely it meets the USDA requirements. The principle of operation is based
on refractive index which can be related to Brix value. Desired accuracies of,
at least + 0.02 degrees Brix amounts to + 0.00008 on the refractive index scale.
It seems impractical to expect repeated accuracy like this from single measure-
ments; therefore, multiple measurements should be taken and the results inte-
grated to provide an average figure that, if repeatable within the desired
accuracy, would be satisfactory. These results will have to be corrected for
acid, after which both the acid and the corrected Brix figures would be printed,
punched, recorded, or whatever was desired. The use of a system just described
would necessitate an automatic acid determinator with an output signal that
could be coupled with the output signal from the Brix determinator.

Other principles of measuring Brix are being considered. Density is one of
the most promising. This can be measured through the use of an accurate balance
or by determining mass through ultrasonics. There is, also, a good possibility
that both the sugar and acid content could be measured colorimetrically with an
auto-analyzer. Extended research in any of these fields would require funds
outside of those presently budgeted for this project.


Florida Citrus Commission and
Florida Citrus Experiment Station,
Lake Alffed, Florida. 450-10/12/67-JGB




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