Group Title: Citrus Station mimeo report - Florida Citrus Experiment Station ; CES 66-8
Title: Problems in selecting juice extractor for state test rooms
Full Citation
Permanent Link:
 Material Information
Title: Problems in selecting juice extractor for state test rooms
Series Title: Citrus Station mimeo report
Physical Description: 7 leaves : ill. ; 28 cm.
Language: English
Creator: Blair, James G
Florida Citrus Commission
Citrus Experiment Station (Lake Alfred, Fla.)
Publisher: Florida Citrus Commission :
Citrus Experiment Station
Place of Publication: Lake Alfred FL
Publication Date: 1965
Subject: Oranges -- Processing -- Florida   ( lcsh )
Genre: government publication (state, provincial, terriorial, dependent)   ( marcgt )
non-fiction   ( marcgt )
Statement of Responsibility: James G. Blair.
General Note: Caption title.
General Note: "400-10/12/65-JGB."
 Record Information
Bibliographic ID: UF00072440
Volume ID: VID00001
Source Institution: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: oclc - 76756570

Full Text

Florida Citrus Commission and
Citrus Experiment Station CES 66-8
Lake Alfred, Florida- 400-10/12/65-JGB

Problems in Selecting Juice Extractor for State Test Rooms

James G. Blair
Florida Citrus Commission
Lake Alfred, Florida

Back in 1961, the Florida Citrus Commission together with other interested
groups requested the USDA to undertake a thorough review of the official in-
spection procedure used in Florida for sampling and testing oranges for pro-
cessing to determine pounds-solids per box. In accordance with this request, a
special survey committee was formed composed of personnel from the USDA and
Florida Agricultural Experiment Station.

Their report was submitted in March of 1961 and stated that the major
problem areas at that time were 1) the selection of a representative sample
from a load of oranges and 2) the determination of juice content in the sample.
Consequently, they studied these problems in detail and made the following ob-
servations and recommendations with regard to extractors.
Juice Determination
"Acknowledgment was made at the start of the survey that the function of
the Inspection Service is to provide accurate, uniform, standardized determi-
nations of pounds-solids at all inspection stations without regard to vari-
ations in commercial operating procedures which result in different rates of
recovery between plants. Present methods of juice determination do not meet
the above requirements. Following are some of the problems noted by the

1. Two methods of juice extraction* are used without any positive
method of assuring agreement either among methods or between methods.

2. Adjustments are made on test equipment which are arbitrary in
nature and are designed to balance test results with industry output.

3. Examination of peels from the extractors indicate that size of
the oranges, peel thickness, variety and fruit condition affect effici-
ency of juice extraction.

4. A considerable amount of juice is discarded from the finisher
and is not reflected in total juice content nor is there any assurance
that the discard is constant from machine to machine. It is also
suspected that other variables may affect the amount of juice

5. The machines do not completely clean out at the end of each

6. There are no precise methods for determining machine adjust-
ments and machine tolerances which often affect extraction efficiency.

7. Machine adjustments are complicated in nature and are not under
complete control of the Inspection Service.

8. No inherent check is provided in present methods to detect
errors in weighing.

One standard method should be used by the Inspection Service for determining
juice content of oranges. The amount of juice determined should have a fixed
correlation to "total" juice content as determined by a standard laboratory pro-
cedure to be devised. Following are the specifications that should be met by
the method of juice determination:

1. Juice determination must be independent of variety, size, peel
thickness or condition of fruit within tolerances to be established.

2. Juice determination must be consistent among similar samples
of fruit over measurable and allowable ranges of machine tolerance.

3. The determination method must be completely objective.

4. Extraction machines must maintain critical settings and any
variation from these settings must be obvious.

5. Machine adjustments should be such that they can be locked in
place with a seal that can be broken only by authorized Inspection

6. Composition of the juice from the extractor should be such that
it would facilitate subsequent analyses.

7. Juice extracted from the samples should be satisfactory for
subsequent plant use. (Further processing may be necessary).

8. The machine should "clean out" between samples.

9. The machine should be durable and reasonably simple to

10. All components of the sample should be available after ex-
traction to see that their total weight equals the initial sample
weight as a check on weighing. (The need for this requirement would
be partly eliminated through use of recording scales).

At this point, the committee wishes to emphasize that the above specifi-
cations are for an extractor for use by the Inspection Service on small
samples and the suitability of a machine for that purpose has no relationship
at all to the suitability of that or any other machine for commercial operations.
At the present time, there are two types of extractors used for inspection pur-
poses, but it is possible that neither of the presently used machines will meet
the above specifications.


Long Range Project

A research project should be initiated to solve the problems of sample
selection and juice determination on samples of oranges drawn for inspection
purposes at processing plants. Since the major problem is in the design and
testing of machines, an engineer competent and experienced in this field
should head up the project. Considerable emphasis also must be placed on

Florida Citrus Experiment Station
and Florida Citrus Commission,
Lake Alfred, Florida. -400-10/12/65-JGB

the statistical side of the problem and precautions must be taken that no
machine is tested and approved without consulting the statistician for a de-
tailed plan for obtaining test data. Such data must then be subjected to a
thorough and competent statistical analysis.

In order to expedite the development of the equipment, a project leader
should be designated who would be able to devote full time to this project
and who would have no other responsibilities. Moreover, it is recommended that
this research project not be terminated immediately upon the solution of the
presently pressing problems. Consideration should be given to a critical eval-
uation of other inspection problems that may now exist or develop in the rapidly
changing and expanding citrus industry.

Following is a list of work which should be under taken through the proposed
research project as soon as possible:

1. A standard laboratory method for determining juice content of oranges
should be developed.

2. Existing machines for extracting orange juice should be evaluated
according to standard laboratory method and the specifications set
forth in this report.

3. If necessary, work should be started on the improvement of an existing
extractor or the development of an entirely new inspection extractor
for juice determination.

The committee recognizes that one may be mislead by the sometimes erroneous
belief that the modification of existing systems and machines is the best and
most logical approach to the solution of a problem. We would encourage the pro-
posed research project to go beyond this limited concept and explore all possible
approaches to the determination of orange quality which may employ entirely new
techniques and principles. Recent developments in instrumentation make feasible
many measurements that were thought to be impractical or impossible a short time

The committee suggests that the Citrus Experiment Station at Lake Alfred,
Florida would be the logical place to locate the research project.

In addition to the work to be done under the above research project, it is
recommended that other interested individuals or concerns be encouraged to de-
sign machines that would meet the specifications which have been presented for
the sampler and extractor. However, any such machines developed should be
subjected to the same impartial and objective evaluation as those developed under
the project. The data from these tests would then be turned over to the
Inspection Service and the Florida Citrus Commission which would approve or re-
ject the machines on the basis of those data and any other tests which the
Inspection Service might wish to make to determine the suitability of the
machines for inspection purposes.

Supplementary recommendations and suggestions of a more detailed nature will
be furnished the project leader of the above research project when he has been
designated. Several ideas for the solution of the above problem have been

Florida Citrus Commission and
Citrus Experiment Station,
Lake Alfred, Florida. 400-10/12/65 JGB

brought forth by the committee which can best be furnished the project leader
through consultation."

That just about covers the portion of the USDA report on extraction, and
one can readily see that it represents a sizeable challenge to this industry.
The challenge and the recommendations of the report were accepted, and renewed
efforts have been made toward solving the specific problems outlined. Despite
the freeze of 1962-63, a great deal has been accomplished on this complex pro-
ject in such a short period of time.

Some of you may still remember the electronic pounds-solids determinator
that was tested in 1962. Then in 1963-64, there was the Anderson extractor
which was tested in conjunction with many methods of separating the juice from
the heavy pulp and rag. There have, also, been other devices tested, some like
those just mentioned that had to be abandoned for ones that showed more immedi-
ate promise.

Results of some early tests made at the Citrus Experiment Station on ex-
tractors like those used by the Inspection Service for official testing purposes
are shown in Table I. By simple statistical analysis, an effort has been made
to display the consistency in the juice yields of the two machines being com-
pared. Since the year was 1963, the only unfrozen fruit available was from the
experimental groves of the Indian River Research Laboratory. Effect of sample
variation was reduced to a minimum through actual sizing, counting and weighing
of the fruit so that the samples tested would be as near alike as it was physi-
cally practical to make them. Approximately 40 individual tests were run on
each machine and the deviations were figured on this basis. The averages, as
well as ranges, for the complete program are shown with an indication that the
results obtained from the FMC machine were more uniform. Since this was based
on only one limited variety of fruit, further performance testing was necessary.

In the course of our testing program, the preliminary results obtained from
tests on the experimental Brown 2701 extractor were most encouraging. It was
decided that during the 1963-64 season a comprehensive field test would be con-
ducted on this new principle of extraction. Since the hand-ream method of de-
termining juice content is the only presently approved method, the program was
based on comparing tests from machines using the new extraction principle versus
the hand-ream method. Five machines were built and placed in five different
locations and the results of many tests were collected and tabulated. A com-
plete report including statistical analyses on these tests was distributed last
year. While this report covered only the work on grapefruit, some tests were
conducted on oranges, as well, and the results of these tests are shown in Table
II. The primary comparison was the hand-ream versus the 2701 machine, but re-
sults of tests run on the presently used testroom extractors were included for
additional information. The use of sized fruit in the medium size ranges only,
reduced somewhat the extraction variation caused by fruit size, but this variable
will be covered in subsequent data.

While it is recognized the determination of internal quality is considerably
affected by individual fruit sizes, there are only limited data available on the
effect these sizes have on different extractors. Therefore, yield comparison by
sizes has become the standard for almost all testing at the Citrus Experiment

Florida Citrus Commission and
Florida Citrus Experiment Station,
Lake Alfred, Florida. 400-10/12/65 JGB

Table I
Comparison Brown and FMC Extractors
East Coast Valencias--1963


Brown 400 FMC 091

Percent Lbs.-Sol. Percent Lbs.-Sol.
Juice per Box Juice per Box


Avg. Yield
Avg. Lb.-Sol.








Table II

Extraction Comparison
Samples Randomized from Packinghouse Fruit--1963-64

Fruit Hand Brown Brown FMC Hand Brown
Size Variety Area Ream 2701 400 091 Ream 2701

252 PA Ridge 59.4 60.8 58.0 58.0 2.14 1.50
200 63.5 62.5 59.2 61.0 2.19 1.23
200 Seeded 55.7 56.8 57.3 55.3 1.54 .97
Mixed Ham. 52.8 53.6 53.1 44.4 3.03 1.25
252 PA E.C. 57.4 59.8 56.5 58.0 1.78 .72
200 55.8 57.5 55.8 58.9 2.40 1.66
200 57.1 59.2 56.7 58.0 2.35 1.03
163 57.7 59.2 56.4 57.9 1.46 1.35
163 58.7 59.1 56.7 58.5 1.73 .72
163 Val. 62.6 61.6 59.1 60.4 .93 1.02






.561 .300


To illustrate better the effect of size, Fig. 1 shows the results of tests
conducted in which the 2701 and the FMC were compared. The period was late in
the 1963-64 season and the fruit used was Ridge Valencias. The FMC loses some
yield on the small fruit while on the large sizes the loss is quite significant.
Conclusions cannot be made from such a limited experiment but there seems to be
an indication the new Brown principle of extraction is less biased for size than
the FMC.

In addition to the effects caused by size, there are, also, those caused by
different varieties. To obtain information about resulting effects of varieties
more tests were run during the 1964-65 season comparing the FMC machine with the
Brown. Fig. 2 has been prepared showing the plus and minus differences using
the FMC machine as the base. On seedless fruit, all tests except one showed the
Brown obtaining a significantly higher yield, while on the seedy fruit the Brown
had a slightly less yield. The average percent difference was plus 3.05% on the
seedless and minus .5% on the seedy. The conclusions here are that the FMC is
slightly biased for varieties.

The FMC Corporation have been engineering a modification to their orifice
tubes to eliminate this bias but, initially, did not meet with much success. In
addition, several pneumatic type orifice restrictors have been tried with some
being excluded from further testing. The principle, however, appeared to have
considerable merit and FMC is sufficiently encouraged to continue with their
engineering development program.

Since the new Brown principle of extraction seemed to come closer to ful-
filling our requirements than anything else available at the time, arrangements
were made with Brown Citrus Machinery Company to build three automated prototype
machines capable of handling the load of a processing plant test room. They
agreed to build these machines and have them ready for the 1964-65 season. The
program for the field testing was set up so that one machine would be at a plant
using Brown extractors and one would be at a plant using FMC extractors. Adams
and Minute Maid in Auburndale were chosen because they had the room for the
additional equipment and personnel and had the capacility of obtaining a double
sample. We are most grateful for the cooperation extended by the management of
each of these plants and hope the results were mutually beneficial. Additional
Inspection Service personnel were assigned each of these machines and they were
operated on a continuous basis.

Results of approximately 40 days of testing between March 5 and May 5 have
been analyzed and are shown on Table III. The average yield and deviation for
each day for each plant were calculated and the figures at the bottom of the
table are the averages. The results between plants cannot be directly compared
due to different fruit and sampling techniques, but the in-plant comparisons
shows the Brown 2900 yield being higher than the official machine in one plant
and lower in the other. The pounds-solids does not follow this pattern which is
not at all unusual. In both cases, the standard deviation indicates more uniform
yield from the Brown 2900.

The third machine was located at the Citrus Experiment Station where con-
siderable test work was done on adjustments, measurement of juice losses,
comparison with the 2701, etc.

Meanwhile, FMC Corporation had progressed with their development of a
pneumatic restrictor on the orifice tube and began to get satisfactory results
about the middle of the 1964-65 season. Beginning October 10, 1964, and
Florida Citrus Commission and
Florida Citrus Experiment Station,
Lake Alfred, Florida. 400-10/12/65-JGB

Field Run Ridge Fruit- Period 4/7 thru 6/10/64
-- BROWN 2701
--FMC 091




Fig. I
324 252 200 163 125


-- / \ SEE[
S/ \

\ ,

o /
\ /



Note: I.FMC Mach-
ine used as the base
Samples of approx-
imately 45 Ibs. each




J.G.BLAIR 9/10/65

0- 0-
\ /
\ /






Table III

Comparison Brown and FMC Extractors
Canning Plant Fruit--1965

Minute Maid Adams
Brown 2900 FMC 091 Brown 2900 Brown 400

Percent Lbs.-Sol. Percent Lbs.-Sol. Percent Lbs.-Sol. Percent Lbs.-Sol.
Juice per Box Juice per Box Juice per Box Juice per Box


Average 51.122 6.006 52.116 6.013 52.634 5.732 51.879 5.612


Average 1.20 -2.11 -1.67 -2.22 --
Range 1.93 2.33 1.74 2.35 -

Table IV
Comparison FMC Ext
Model 091

Air Control

Fruit- FCCC
Dev. Yield
- 51.751
.4173 50.193'
.3930 51.218
.0336 51.960
.0632 50.502
- 50.775

Fruit CES
Dev. Yiel
1.539 -
-" 1.662 52.31
'1.711 47.61
1.'875 52.5!
1.572 54.5i
2.325 x -
2.732 \- -

d D


ruit FCCC
ev. Yield
- 51.770
1991 50.957
1917 50.977
2972 51.652
2520 50.104
- 51.055
- 50.324

- .2267
-- .3837


- \ .2350
- .1055

- 1.868







continuing through March 4, 1965, approximately twenty series of tests were con-
ducted comparing the current "standard" setting on the FMC Model 091 state test
extractor to various and numerous pneumatic orifice restrictor arrangements
using both seedy and seedless oranges.

One setting or arrangement of the pneumatic orifice restrictor was found to
give very encouraging results on both seedy and seedless varieties. This
arrangement was encouraging because slight changes in the operating air pressure
had a definite effect on juice yields with a comparable relationship on both
fruit varieties and also functioned very effectively retaining the rag, peel and
seeds in the orifice tube.

On March 22, 1965 a meeting to review the above test data was attended by
representatives of the Florida Citrus Commission, Florida Department of Agricul-
ture and FMC personnel. It was decided at this meeting to install the air
controlled components on a state test extractor in a test room of a commercial
plant where duplicate fruit samples from each load of fruit could be obtained
to compare the standard FMC state test setup with the most promising air con-
trolled restrictor arrangement. It was, also, decided that State Inspection
personnel would "man" this installation, conduct tests and keep records for th,
remainder of the Valencia orange season.

This installation was made at Florida Citrus Canners Cooperative, Lake
Wales; completed on March 29, and operated through June 8, 1965. During the 2
days of actual operation, duplicate samples were compared from 340 truck loads
of Valencia oranges between the FMC standard state test setting and the pneu-
matic orifice restrictor arrangement. Various other comparisons were also mad
including triplicate samples from 8 different lots of randomized oranges pre-
pared at the Citrus Experiment Station.

Due to the lack of operating experience and test data on the 1-1/4" air
controlled orifice tube, both pneumatic orifice tubes (1" to be on 3" cups and
1-1/4" tube on 4" cups) were initially operated at the same air pressure. Dup-
licate samples were compared on 41 loads of fruit at 10 psi, 39 loads at 12-1/2
psi and 83 loads at 15 psi air applied to both tubes.

As the Valencia season progressed and the larger fruit (over 3-1/4" dia.)
became available, some preliminary tests were made comparing various air press-
ures on the 1-1/4" pneumatic restrictor to the standard setting. These tests
indicated that the air pressure should be approximately doubled on the 1-1/4"
air controlled tube to yield about the same juice yield as the 1-1/4" standard
setting. Beginning May 14 and continuing through the remainder of the season,
the operating air pressure of the 1-1/4" orifice tube was increased to 30 psi,
and the 1" tube continued to operate at 15 psi air. Duplicate samples were
tested in this manner on 177 truck loads of fruit. Results of some of these
tests are shown on Table IV. The data chosen for comparison were selected first
because it was run at the optimum air pressure and included fruit samples from
the Station as well as those from the canning plant receiving system.

It may be noted that up to 4% variation existed in some individual tests
but the average juice yields from the 340 duplicate samples were nearly the
same. It may also be noted that generally the juice yields from the pneumatic
Florida Citrus Commission and
Florida Citrus Experiment Station,
Lake Alfred, Florida. 400-10/12/65-JGB

restrictors are more uniform than the standard state test setting. To us this
demonstrates the theory that variations in the number of seeds, peel thickness
and fruit size all have an effect on the pressure developed in the "fixed"
restrictor of the standard machine, while the pneumatic restrictor applies a
uniform pressure regardless of these variables.

Our plans are to continue similar tests on the pneumatic restrictor
arrangement, using both seedy and seedless varieties, during the coming early
and midseason orange operation in the state test facilities of a commercial
plant. Also, a second pair of prototype pneumatic restrictors will be installed
on the machine at the Citrus Experiment Station for evaluation during the coming
1965-66 season.

While these field tests were being conducted, experimentation was being
done at the Station to determine the effect of many relationships such as yield
versus size, high yield-low Brix and visa versa and loss of yield due to hand-
ling. Results of some of these comparisons are explained by the following:
For example, Figure 3 shows a comparison of the yield from both the FMC and 2701
extractors versus individual fruit sizes. The results shown here are similar
to those of Figure 1 on the same comparison except that they cover all fruit
varieties instead of just the Valencias. Again, it may be noted that the curve
for the Brown 2701 machine has a more uniform slope indicating less effect on
results from large and small fruit.

On seven different tests during the period of March 25 through May 26
where the fruit was recirculated through our sample testing system, the
losses resulting from handling amounted to from 0.11 to 2.12 percent. There
are several factors that contribute to this loss but the most significant seems
to be degree of maturity. The fruit receiving system at the Station as used
for sampling tests, is gentle in comparison to most of those used commercially.
So under normal operating conditions, the results mentioned above could be con-
sidered minimum. A load of fruit which was tested at the Station this last
season had to be rehandled several times, commercially, and the loss in yield
as recorded by official tests amounted to 2.202 pounds of juice per box.

The information reported here may not have a direct bearing on the deter-
mination of pounds-solids but it was obtained as a result of it and is being
passed along for what it may be worth to you.

In this report, extraction is considered to include the finishing operation.

Florida Citrus Commission and
Florida Citrus Experiment Station,
Lake Alfred, Florida. 400-10/12/65-JGB

x BROWN 2701
o--- o --FMC 091

252 200 163 125

- o

55 -

53 m

51 7

49 6

47 5

45 4


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