United States
Environmental Protection
Agency
Industrial Environmental Research ~^
Laboratory
Cincinnati OH 45268 '/ i \
Research and Development
EPA-600/S2-82-059 Sept. 1982
Project Summary
Tomato Cleaning and
Water Recycle
Walter W. Rose
This summary discusses the findings
of a study designed to develop a mud-
removal/water recycle system for a
tomato dump tank operation employing
rotating soft rubber discs.
Over a three-year period (1974,
1975, and 1976), modifications were
made to a mud-removal-water recycle
system. A false bottom-ejector trans-
port system was developed to prevent
soil from settling in the dump tank.
Also, a physical/chemical treatment
system was developed to remove soil
from water being recycled back into
the dump tank. Colloidal particles
were removed from the thickener
overflow tank before being recycled
back to the dump tank or discharged
to the sewer. A vacuum belt dewatering
unit was constructed and evaluated
for dewatering mud from the thickener
tank prior to final disposal.
Two different designs of rubber disc
machines were evaluated and found to
be equally effective in cleaning the
surfaces of tomatoes. Minimal quanti-
ties of water were required in the
cleaning of the tomatoes. Energy in
the form of spinning soft rubber discs
removed surface soil from the toma-
toes. It was also observed that stems
were removed from the tomatoes by
the disc cleaners.
This Project Summary was developed
by EPA's Industrial Environmental
Research Laboratory, Cincinnati, OH,
to announce key findings of the
research project that is fully docu-
mented in a separate report of the
same title (see Project Report ordering
information at back).
Introduction
Mechanical harvesting of raw products
has increased the amounts of soil being
transported to food processing plants.
The quantity of such soil for tomatoes is
about two percent of the product
weight. The soil is characterized in two
forms: (1) as clods of dirt and (2) as
smear soil on the surface of the product.
The increase in soil on the product has
caused food processors to use more
water to clean the product.
When the velocity of water is insuffi-
cient to scour the bottom surface of the
tank, soil solids accumulate in the
bottom of dump tanks or initial flumes.
After several hours, the soil accumulates
to the extent that product flow is
impaired and process downtime is
required for the removal of the soil.
The food-processing industry has
used high water-scouring velocities to
prevent soil from settling, or allowed the
soil to settle. In the latter case,
processing was stopped periodically to
remove the accumulated soil. The use of
high water-scouring velocities results in
the discharge of excessive amounts of
water, which adds to the hydraulic
portion of surcharges. The discharge of
inert solids (soil) to a municipal waste
treatment system may result in opera-
tional and maintenance problems. The
intermittent shutdown bythe processing
plant reduces plant productivity and
requires additional water for cleanup
purposes.
This project was initiated to seek
solutions to problems related to the
processing of machine-harvested toma-
-------
toes. The specific objectives of the
project were:
• To demonstrate on a commercial
scale the substantial reduction in
the volume of water required to
clean tomatoes by using a flexible,
spinning rubber disc machine.
• To develop a water recycle system
for tomato dump tank water.
• To compare the cleanliness (based
on bacteriological and physical
criteria) of tomatoes processed in a
conventional washing system with
that of a low-water cleaning/
washing system.
• To compare the plant effluent (with
regard to volume, BOD, and suspend-
ed solids) of conventional versus
low-water cleaning systems.
• To make economic comparisons
between (a) conventional washing
systems versus disc cleaner; (b)
conventional washing system with
dump tank recycle (no chemical
flocculation) versus disc cleaner
with dump tank recycle (no chemi-
cal flocculation); (c) disc cleaner
with dump tank recycle (no chemi-
cal flocculation) versus disc cleaner
without dump tank recycle; and (d)
disc cleaner with dumptank recycle
(with chemical flocculation) versus
disc cleaner with dump tank recycle
(no chemical flocculation).
Dump Tank Water Recycle
The initial approach used in the
development of a water recycle system
was to maintain a high water velocity in
the dump tank and to withdraw, treat,
and return the water. The withdrawn
water was treated, then returned to the
dump tank at various locations. It was
postulated that the water would be of
adequate volume and have sufficient
head to maintain a scouring effect a long
the entire bottom of the dump tank.
The primary objective of the 1974 test
program was to determine if it was
feasible to maintain a closed-loop
system of water usage around the
tomato bin dump. The second objective
was to evaluate the technique of
utilizing high scouring velocities at the
base of the bin dump to prevent
accumulations of settleable solids
within the bin dump itself. The third and
final objective was to determine what
benefits, if any, would accrue from a
coagulation and flocculation system
used in conjunction with the closed-
loop water recycle system. The findings
of the 1974 water recycle project
indicated the following problems:
• Insufficient data coverage.
• Unsatisfactory mud removal pro-
cedure.
• Clogging of recycle system me-
chanical elements.
• Lack of soil solids concentration.
• Lack of liquid level control in the
bin dump.
• No automatic control of coagulant
dosage.
Major modifications were required in
subsequent operations of the recycle
system to assure its ability to meet the
general project objectives. The four
objectives set for the 1975 study were
as follows:
• Modify and improve the water
recycle system for the tomato
dump tank.
• Compare wastewater quantity and
quality under different operating
modes.
• Compare the quality of final dis-
charge waters from the system
with and without the addition of
coagulation/flocculation.
• Economically compare the se-
quences of operation of the initial
bin dump with and without the
water recycle system, and with the
water recycle system operating in
different modes.
Model Dump Tank Study
In January 1975, a model bin tank
study was developed. The proposed grit
collection system consisted of a false
bottom near the point where tomatoes
were dumped into water. Grit would
pass through the false bottom and be
transported by means of an ejector to a
solids separation system. Two alternative
methods were used to introduce recir-
culated water into the bin dump: (1) jet
flow and (2) weir overflow. The jet flow
system sustains high velocities for
greater distances through the bin dump
tank than does the weir overflow
method. These methods were evaluated
to determine if soil deposition could be
concentrated near the point of introduc-
tion.
Soils selected for evaluation of
deposition were Yolo Loam and Hanford
Sandy Loam. Handford Sandy Loam is
predominantly a sandy material, whereas
Yolo Loam contains a significant amount
of clay (Table 1). The major objective of
the model study was to develop quanti-
tative design criteria for scaling the
solids collection and transport system
for a full-scale bin dump operation.
Table 1. Study Soil Compositions,
Average Weight Percent
Component
Sand
Silt
Clay
Yolo Loam
22.5
50.0
27.5
Hanford
Sandy
Loam
74.5
16.0
9.5
Test results showed no significant
differences between the jet flow and
weir overflow methods in the distribution
patterns of soil deposits. However, the
Hanford Sandy Loam did not develop the
concentrated, settled condition experi-
enced with Yolo Loam. With Hanford
Sandy Loam, characteristic "ripples" or
"humps" were observed on the down-
stream side of the deposited solids.
Also tests were conducted using both
low- and high-scour channel velocities.
Using low-scour velociti-es, only 30
percent of the loaded soil was effectively
collected and transported from the bin
dump to the solids separation system.
However, excessive scour velocities in
the channel resulted in a reduction in
overall collection efficiency. Using high-
scour velocities, patterns of soil deposi-
tion showed an accumulation of relatively
fine soil particles immediately down-
stream of the hopper and false bottom.
Observation of the flow pattern in the bin
dumptank indicated that this accumula-
tion of material was primarily due to the
transport of soil particles by turbulence
generated by the high-scour channel
velocities.
Full-Scale Water Recycle
System
The design for the full-scale recycle
system included the false bottom,
ejector solid transport, screen and solid
discharge hopper, swirl concentrator,
sludge thickener, and chemical floccu-
lation systems. The four modes analyzed
are discussed below.
Conventional Washing
System
The conventional washing mode
consisted of four washing steps: dump
tank, inside washer flume, distribution
flume, and final rinse. The wash water
was countercurrent to the direction of
the tomatoes (Figure 1). In this system, a
significant savings in water consumption
was realized. The conventional washing
system adopted for this study used
water recycling, and was not the same
-------
. -C-,
v f
Dump
Tank
Inside
Flume
Distribution
Flume
Ordinary
Pump^
Legend
Tomato Product
Process Water
"• Make Up Water
Water Meter
Water Quality
Sample Station
Final
Inspection
\Final Rinse
Sewer
(Further Processing)
figure f. Flow diagram of conventional cleaning system.
as the "conventional" washing systems
commonly employed in food processing.
Conventional Washing with
Water Recycling System
Tomatoes coming into the plant were
processed in the manner previously
described. The only difference in this
first mode of operation was the incor-
poration of a swirl concentrator into the
water recycle system (Figure 2).
Disc Cleaner with Water
Recycling System
In this mode of operation, tomatoes
were transported directly from the
dump tank to the rubber disc cleaner,
and subsequently transported into a
distribution flume, and then to a final
inspection stage (Figure 3). The rubber
disc provided an alternative method of
cleaning. It replaced the flumes and
sprays in the usual cleaning process.
Disc Cleaner with Water
Recycle and Chemical
Flocculation
For this mode of operation, the flow
diagram is similar to the previous flow
scheme. Here, however, an internal
chemical coagulation and flocculation
step was added prior to the sludge
thickener process (Figure 4). The
chemical coagulation-flocculation sub-
system consisted of four components:
(l)recirculationpump, (2)tubefloccula-
tor, (3) slip-stream turbidimeter, and (4)
chemical feed mechanism.
Analysis of Water
Consumption
In a typical tomato processing plant,
water can be used for (1) filling, (2)
operation, or <3) cleanup. The water
used for filling purposes represents that
portion of water used to fill up a dump
tank, an inside washer, or a distribution
flume. The water used for operational
purposes represents that portion which
is continuously utilized, such as bin
washing, cleaning of the trash belt, and
the final rinse of tomatoes. Cleanup
water is that portion of water used for
floor cleaning, tank and flume washing
during a shutdown.
Table 2 presents the average amount
of water consumption for various water
usages. The data indicate that for the
same usage purposes, there was no
significant variation in the total daily
usage at the various modes of operation.
Approximately 7 percent was utilized for
filling purposes, 55 percent for opera-
tional purposes, and 39 percent for
cleanup purposes. The following obser-
vations and conclusions may be drawn
from these data:
• The average daily tonnage of
tomatoes processed increased
substantially when water recycling
measures were employed.
• A decrease of 26 percent in the
average total daily water usage
was realized when the disc cleaner
with water recycling and chemical
flocculation was applied. A decrease
in water usage was also noted for
filling and operational purposes.
-------
Screen & Swirl
Concentrator
Sludge
Thickener
(gw
@
6W
Final Rinse
Legend
Final
Inspection
0
Solids
@v
Tomato Product
Process Water
—"- Make Up Water
I Solids Slurry
A*J Water Meter Sewer
W\ Water Quality
Sample Station
LO Bin Dump Level Control
Figure 2. Flow diagram of conventional cleaning with water recycle system.
• A decrease in the average unit of
water consumption (gallons of
water used per ton of tomatoes)
occurred when water was recycled.
Soil Solids Loading
Analyses of soil solids distributions at
various modes of operation are presented
in Table 3. In a conventional washing
system, 31 percent of the soil is
removed from the dump tank, and the
remaining 69 percent is discharged into
the sewer. Utilization of more sophisti-
cated removal measures (i.e., disc
cleaner, and chemical flocculation)
significantly increases the unit weight
of soil solids removed from the sludge
thickener. At the same time, the
percentage of soil solids discharged into
the sewer decreases. The amount of soil
solids carried to the processing plant is a
function of soil type, moisture content of
the soil when the tomatoes are har-
vested, and method of harvesting.
Vacuum Belt Dewatering
Device
In September 1976, the vacuum belt
dewatering unit was evaluated for cake
solids production rate, dewatering
efficiency, drying factor, sludge volume
reduction, and solids recovery efficien-
cies. The unit was analyzed with and
without the addition of a chemical
coagulant. The following observations
were made in comparing the two sets of
data:
• A slight increase in the thickness of
the feed sludge and cake with
chemical coagulation at the clarifi-
cation-thickening stage.
• Increases in the sludge loading
rate and sludge cake production
rate with chemical coagulation.
• A significant increase in the total
solids concentration for the sludge
feed and cake with chemical
coagulation.
In comparing the performance of the
vacuum belt dewatering unit with other
sludge dewatering devices, two param-
eters are worth noting, cake solids
production rate, and the cost of sludge
dewatering. The cake solids production
rate of the vacuum belt ranged from 10
to 117 Ib/hr/ft2, while a cake solids rate
of 1 to 8 Ib/hr/ft2 was typical for other
types of dewatering devices (i.e.,
vacuum filter or filter press). Also, it
should be noted that the cake production
rates for the vacuum filter and filter
press used in this comparison were
based on dewatering municipal sludges;
cake production rates would be different
for sludge from a tomato processing
plant.
The second significant feature of the
vacuum belt dewatering system was its
low unit cost, estimated to be approxi-
mately $2 per ton. Unit costs of other
dewatering devices ranged from $12 to
$29 per ton of dry solids.
The data showed significantly larger
sludge solids loading rates and cake
production rates when chemical coagu-
lation was employed. The average
sludge solids loading rate with chemical
coagulation was 76.2 Ib/hr/ft2 versus
38.9 Ib/hr/ft2 without chemical appli-
cation. Similarly, the average cake
production rate with chemical applica-
tion was 75.2 Ib/hr/ft2 versus 38.3
Ib/hr/ft2 without the application of any
4
-------
Screen & Swirl
Concentrator
Sludge
Thickener
Submers-
ible
Kf/na/
Rinse Legend
V
Final
Inspection
Tomato Product
f Process Water
--»- Make Up Water
* Solids Slurry
1W
Solids
i!lH Water Meter
(W) Water Quality
Sample Station
Sewer
\±/ Bin Dump Level Control
{Further Processing/
Figure 3. Flow diagram of disc cleaner with water recycle system.
chemicals. Results for the other param-
eters were approximately the same.
Tomato Cleaning with
Mag nu washer
In September 1976, a Magnuwasher,
significantly different from the flat bed
disc unit, was evaluated. This machine
contained parallel rows of soft rubber
discs arranged in a circle. By means of a
drive mechanism, a tumbling move-
ment was imparted to the product, and
the spinning of the discs provided the
cleaning. The Magnuscrubber was
analyzed in the same manner as the flat
bed disc cleaner.
The data indicated that variations in
the volume of water used by this device
influenced the characteristics of the
effluent. As more water was used, the
concentration of measured parameters
decreased. The results suggest that
tomatoes after Magnuscrubber treat-
ment were as clean as those found after
the final rinse. A summary of the
effluent from the Magnuscrubber is
provided in Table 4. The BOD results
suggest that variations in the volume of
water had no effect on the amount of
organic matter being discharged. The
COD and TOC results were more
variable with flow. The TOC increased as
the volume of water increased, whereas
COD was maximum at 7 gpm. Apparent-
ly, this device can clean tomatoes with
minimal amounts of water.
Economic Analysis
Because of the wide variation in
wastewater treatment costs, a summary
comparison of alternative recycle
systems is presented in Table Sforthree
wastewater agencies. Industrial waste-
water charges among San Jose, East
Bay Municipal Utility District (EBMUD),
and Sacramento, California, were
evaluated for cost-effectiveness under
the various cleaning methods. The total
cost of processing tomatoes involved
the following:
Cost of water.
Wastewater treatment costs.
Energy cost.
Equipment capital and deprecia-
tion costs.
Solid waste disposal cost.
Chemical cost.
Labor and maintenance costs.
Indirect costs.
A summary of the total costs for
tomato cleaning for the alternative
recycle systems considered is presented
in Table 6, which summarizes all
identifiable costs for the alternative
systems where the costs were not
equivalent for all systems considered.
The costs range from a high of approx-
imately $1.50 per ton for the modified
conventional washing system with
recycle to a low of $0.92 per ton for the
disc cleaning system with water recycle
and solids concentration. The advan-
tages of utilizing water recycle systems
in terms of overall cost savings are
clearly demonstrated in this table. In
this analysis, the overall cost, including
capital costs of the required equipment,
is reduced by as much as $0.59 per ton
of tomatoes processed. Hence, there is a
potential net annual savings of approxi-
mately $20,000 per year for a 36,000-
-------
-[/w
@
Washer ,_c
A/7
l"^
/0$ *r
Screen & Swirl
Concentrator
. , . Tfl.fl
75W) •— J
i
Tube
Flocculator
W
Sludge
Thickener
Solids
@W
^ \ ^
« /T
S3 ^
M^
Tiw.
^V"S
©
- ^XH
\7W
v r
Dump
Tank
f K
- Jj
Disc
Cleaner
f
I
€
, , Disc Rinse
~-\M\
••{M]-»-
UM\
•rMMb
gw^
(F
0
Distribution
Flume
' i f /na/ Rinse -
f \.
JW 1 I
Final
Inspection
H
urther Processing)
Legei
VTon
f /Vo
-•*- /Wa
f So/
0 Wa
@ l/l/a
Sa
® Bin
r^
Tan/
W
.x
[/
£>N
i\- ....
-'r — T
3
r\\
i fr\tf
ra/7*-3| ^'
,(m
x^^
7
-------
Table 2. Summary
Mode of Operation
Conventional
Conventional
(with swirl, with-
out flocculation)
Disc Cleaner
(with swirl, with-
out flocculation)
Disc Cleaner
(with swirl, with
flocculation)
of Average Water
A verage
Tomato
Processed
(ton/day)
481
445
594
605
Consumption at Various Modes
of Operation
Average Water Consumption (gal/ day)
Filling
Purpose
10.700
7,800
7,000
6,400
Operational
Purpose
81,300
59.800
44,300
51,100
Cleanup
Purpose*
41.500
41.500
41.500
41,500
Total
133.500
109,100
92,800
99,000
Average Unit
Water Consumption
Rate
(gal/ton)
278
245
156
164
* Averaged value.
Table3. Average Total Soil Solids Loadings at Various Operational Modes
Mode of Operation
Total
Tomatoes
Processed
(ton/
day)
Soil Solids Removal
from Dump Tank
(Ib/day) (Ib/ton)
Soil Solids Lost
to Sewer
(Ib/day) (Ib/ton)
Soil Solids Removal
from Thickener
(Ib/day) (Ib/ton)
Total Soil
Solids to
Plant
(Ib/day)
Soil Per
Unit of
Tomato
(Ib/ton)
Conventional
481 2.757 5.7 6,029 12.5
8,786
18
Conventional
(with swirl.
without floc-
culation)
Disc Cleaner
(with swirl.
without floc-
culation)
Disc Cleaner
(with swirl.
with floccu-
lation)
445 941 2.1 2.332 5.2 1,453 3.3 4,726 11
594 684 1.2 1.609 2.7 3.353 5.7 5,646 10
605 856 1.4 2.920 4.8 4.568 7.6 8,344 14
Table 4. Characteristics of Effluent
from Magnuscrubber
Flow(GPM) COD BOD TOC
4
7
13
337
440
420
• (Ib/day) •
281
283
282
119
158
172
The daily average tonnage of
tomatoes processed increased
substantially during modes of
operation with water recycling.
A 41 percent decrease in the
average unit water consumption
rate (gallons of water per ton of
tomatoes processed) was realized
when disc cleaner with water
recycling and chemical floccula-
tion were applied.
When water conservation and
reuse measures were implemen-
ted, the soil solids removed from
the dump tank per ton of tomatoes
processed decreased, soil solids
lost to the sewer decreased, and
the soil solids removed from the
thickener tank per ton of tomatoes
processed increased.
Estimated incoming soil solids
ranged from 10 to 20 Ib/ton of raw
tomatoes (13 Ib/ton average).
Estimates were based on the sum
of soil solids collected from the
dump tank, lost to the sewer, and
removed from the sludge thicken-
er.
The amount of soil solids varies
considerably and depends upon
the type of soil in which the
tomatoes were grown, the moisture
content of the soil when the
tomatoes were harvested, and the
method of tomato harvesting.
Recommendations
Methods for the removal of soil from
water, the recycling of water from dump
tank operations, and the cleaning of
tomatoes using a low volume of water
have been demonstrated, and should be
placed into commercial use. There is a
need to find an effective method to
eliminate vines, stems, grass, and other
materials that accumulate at product
transfer points, as well as to periodically
clean the false bottom of spacings
which become clogged with soil clods,
rocks, metal trash, etc.
-------
Table 5. Summary Comparison of Wastewater Treatment Costs for Alternative Recycle Systems
Agency
San Jose
EBMUD
Sacramento
Average
Modified
Conventional
Washing Plus
Recycle
($/ton)
1.104
1.237
1.324
1.222
Modified Conven-
tional Washing
Plus Recycle
and Solids
Concentration
($/ton)
0.469
0.495
0.531
0.498
Disc Cleaning
Plus Recycle
and Concentra-
tion
($/ton)
0.312
0.438
0.467
0.406
Disc Cleaning
Plus Recycle,
Flocculatioir,
and Concentra-
tion
($/ton)
0.255
0.437
0.357
0.350
Table 6. Summary of Costs for Tomato Cleaning
Overall Cost
Water Use
Wastewater
Treatment
Energy
Equipment
Capital
Cost
Solid Waste
Disposal
Flocculant
TOTAL
Modified
Conventional
Washing with
Recycle
($/ton)
0.074
1.222
0.016
0.186
0.005
—
1.503
Modified
Conventional
Washing with
Recycle and
Solids Con-
centration
($/ton)
0.066
0.498
0.038
0.605
0.009
—
1.216
Disc Cleaning
with Water
Recycle and
Solids Con-
centration
($/ton)
0.042
0.406
0.034
0.424
0.012
—
0.918
Disc Cleaning
with Water
Recycle. Solids
Concentration.
and Flocculation
($/ton)
O.044
0.350
0.035
0.474
0.013
0.017
0.933
Potential net annual savings = (1.503 - 0.933) (36.000 ton/year) = $20.520.
Walter W. Rose is with National Food Processors Association, Berkeley, CA
94710.
Kenneth A. Dostal and Harold W. Thompson are the EPA Project Officers (see
below).
The complete report, entitled "Tomato Cleaning and Water Recycle," (Order No.
PB 82-255381; Cost: $12.00, subject to change) wilt be available only from:
National Technical Information Service
5285 Port Royal Road
Springfield, VA 22161
Telephone: 703-487-4650
The EPA, Project Officer can be contacted at:
Industrial Environmental Research Laboratory
U.S. Environmental Protection Agency
Cincinnati, OH 45268
.S. GOVERNMENT PRINTING OFFICE: 1982/559 -092/0517
-------
> =• o>
8 la
V, U)
5 w
m O
m O
a
a
CO <0
O
W
ru
aotrx
oxxc
IT
m
2
O
m > T)
m
W n n o
w-< s 3
Oi o g
I
------- |