Energy consumption during Refractance Window evaporation of selected berry juices

INTERNATIONAL JOURNAL OF ENERGY RESEARCH
Int. J. Energy Res. 2004; 28:1089–1100 (DOI: 10.1002/er.1017)
Energy consumption during Refractance Window1
evaporation of selected berry juices
C. I. Nindo1, J. Tang1,n,y, J. R. Powers2 and K. Bolland3
1 Department of Biological Systems Engineering, Washington State University, Pullman, WA 99164-6120, U.S.A.
2 Department of Food Science & Human Nutrition, Washington State University, Pullman, WA 99164-6376, U.S.A.
3 MCD Technologies, 2515 S. Tacoma Way, Tacoma, WA 98409-7527, U.S.A.
SUMMARY
The Refractance Window1 evaporator represents a novel concept in the design of evaporation systems for
small food processing plants. In this system thermal energy from circulating hot water is transmitted
through a plastic sheet to evaporate water from a liquid product ?owing concurrently on the top surface of
the plastic. The objectives of this study were to investigate the heat transfer characteristics of this
evaporator, determine its energy consumption, and capacity at di?erent tilt angles and product ?ow rates.
The system performance was evaluated with tap water, raspberry juice, and blueberry juice and puree as
feed. With a direct steam injection heating method, the steam economy ranged from 0.64 to 0.84, while the
overall heat transfer coe?cient (U) was 666 W mÀ2 8CÀ1. Under this condition, the highest evaporation
capacity was 27.1 kg hÀ1 mÀ2 for blueberry juice and 31.8 kg hÀ1 mÀ2 for blueberry puree. The energy
consumption was 2492–2719 kJ kgÀ1 of water evaporated. Installation of a shell and tube heat exchanger
with better temperature control minimized incidences of boiling and frequent discharge of condensate. The
steam economy, highest evaporation rate and overall heat transfer coe?cient increased to 0.99,
36.0 kg hÀ1 mÀ2 and 733 W mÀ2 8CÀ1, respectively. Copyright # 2004 John Wiley & Sons, Ltd.
KEY WORDS:
falling liquid ?lm; steam economy; energy consumption; blueberry; raspberry; juice
1. INTRODUCTION
Intensive use of thermal energy and loss of product quality of heat sensitive liquid foods are two
major concerns in the design and operation of evaporators used in the food industry. Multi-
stage evaporators have been developed that operate at sub-atmospheric pressures to increase
energy e?ciency and reduce thermal degradation of products (APV CREPACO, 1992). Those
systems are, however, very expensive and operate well only within a narrow range of ?uid
viscosity. The Refractance Window1 (RW) evaporator, developed by MCD Technologies Inc.,
is relatively simple, inexpensive and can be used with a diverse range of products, including
those with high sugar and pulp content. It basically uses hot water at normal atmospheric
nCorrespondence to: J. Tang, Department of Biological Systems Engineering, Washington State University, Pullman,
WA 99164-6120, U.S.A.
y E-mail: jtang@mail.wsu.edu
Contract/grant sponsor: Washington Technology Center
Published online 16 July 2004
Received 17 October 2003
Copyright # 2004 John Wiley & Sons, Ltd.
Accepted 19 January 2004

---

1090
C. I. NINDO ET AL.
pressure to evaporate water from liquid foods. Small fruit and vegetable processors who cannot
a?ord expensive capital investments in multi-e?ect vacuum evaporators may ?nd this
equipment very useful. In addition to concentrating food products, the RW evaporator is
versatile and has potential for use in processing nutraceutical, pharmaceutical, biotechnology,
and chemical products, as well as in waste recovery and remediation for a diverse range of
industries.
Before a new type of food processing equipment is commercialized, it is important to conduct
adequate experiments to evaluate its performance. For an evaporator, the measures of
performance usually include steam economy (energy e?ciency), amount of water it can
evaporate per hour (i.e. its capacity), operating temperature range, and nature or type of
products it can handle. The procedures for determining energy use and other performance
indices for vacuum evaporators are outlined by Minton (1986), Rumsey (1986), and Chen and
Hernandez (1997). Similar performance characteristics are needed for the RW evaporator in
which the energy for evaporation is obtained from hot water.
In the present con?guration of RW evaporator (Figure 1), the water that acts as the heat
source is ?rst heated by directly injecting steam into a water tank. The water is heated to the
temperature required for the process, usually a few degrees below boiling point (96–988C), and
maintained at that level to avoid formation of bubbles that would reduce heat transfer to the
product. The hot water is then circulated beneath a transparent plastic with the product ?owing
concurrently on the upper part of the inclined ?at surface of the plastic. The energy from the hot
water is transmitted through the plastic sheet for heating and evaporation of water from a liquid
product that makes a number of passes through the evaporator until the desired concentration is
reached. Compared to conventional falling ?lm evaporators, the design of RW evaporator
makes it possible to attain higher solids concentration without fouling of the evaporation
surface. The plastic surface only needs cleaning at the end of the process or whenever a di?erent
product is to be processed.
The objective of this study was to determine the performance characteristics of the RW
evaporator, and particularly to document its energy consumption per unit weight of water
evaporated, heat transfer per unit of surface area, evaporation capacity at di?erent tilt angles,
and ways of improving the system.
Figure 1. Layout of RW evaporator showing temperature measuring points.
Copyright # 2004 John Wiley & Sons, Ltd.
Int. J. Energy Res. 2004; 28:1089–1100

---

ENERGY CONSUMPTION DURING REFRACTANCE WINDOW1 EVAPORATION
1091
2. THEORETICAL CONSIDERATIONS ON HEAT AND MASS BALANCES
IN RW EVAPORATOR
The evaporator capacity, which is a measure of the amount of water evaporated per hour, can
be estimated from the overall mass balance and dry matter balance:
mp;in ¼ mp;out þ mw
ð1Þ
mp;in  xin ¼ mp;out  xout
ð2Þ
where mp;in and mp;out are the mass ?ow rates (kg sÀ1) of ?uid product at the inlet and outlet of
evaporator, respectively; mw is evaporation rate (kg sÀ1); xin and xout are the solids
concentration (decimal) at the inlet and outlet points, respectively.
Combining Equations (1) and (2) yields:


xin
mw ¼ mp;in
1 À
ð3Þ
xout
By measuring the solids content xout during the evaporation process, the evaporation capacity
ðmwÞ at di?erent elevations and product ?ow rates can be calculated (Figure 2).
The rate of heat transfer per unit surface area is an important parameter that is necessary for
evaluating the performance of evaporators. To determine the heating rate, the overall heat
transfer coe?cient (U), the e?ective evaporation surface (A) and the temperature di?erence
(LMTD) between the hot and cold ?uids must be known. The relevant equation is
DTin À DTout
Q ¼
ð
1
UAðLMTDÞ
and
LMTD ¼
4Þ
lnðDTin=DToutÞ
where Q1 is the energy for evaporation and sensible heating of product (kW); DTin the
temperature di?erence between circulating hot water and liquid product at the inlet (8C); DTout
the temperature di?erence between circulating hot water and liquid product at the outlet (8C).
Apart from pre-heating of juice to evaporation temperature, thermal energy from steam
condensing in the circulating hot water or within the heat exchanger is used mostly for
evaporating water from the product. The energy supplied by saturated steam is the same as the
amount released when the steam condenses. Therefore, for steady state conditions, the latent
heat of condensation ðQ Þ
Þ
s
is utilized for heating the product to evaporation temperature ðQp
and evaporating water from the product ðQ Þ;
w
with the remainder constituting thermal losses to
the surrounding ðQ Þ:
L
Most of these losses occur through the bottom stainless-steel plate
supporting the plastic. If the condensate that normally circulates back to the boiler hot well is
Figure 2. Representation of heat and mass transfer within the evaporator.
Copyright # 2004 John Wiley & Sons, Ltd.
Int. J. Energy Res. 2004; 28:1089–1100

---

1092
C. I. NINDO ET AL.
collected and weighed, then QL also includes the enthalpy in discharged condensate. Therefore,
Q ¼
þ
þ
ð
s
Qp
Qw
QL
5aÞ
where
Q ¼
¼
p
mpcpDTp; Qw
mwlw; and DTp ¼ Tp;out À Tp;in
ð5bÞ
For fruit juices, the speci?c heat capacity ðcp, kJ/kg8C) and latent heat of evaporation
(lw, J/kg) at a known temperature Tpð8CÞ are given by the following equations (Rao and Vitali,
1999; Chen and Hernandez, 1997):
cp ¼ 4:187f1 À xinð0:57 À 0:0018ðTp À 20ÞÞg
ð5cÞ
lw ¼ 2499 Expð À 0:001016TpÞ
ð5dÞ
After determining Q
;
p and Qw it is possible to calculate the heat transfer coe?cient, U. The
value of the product side heat transfer coe?cient hp is a measure of how e?ciently the latent
heat from steam is transferred to the product being concentrated. Given that the temperature of
the plastic surface on the hot water side is nearly the same as the bulk water temperature, the
overall resistance to heat transfer between the circulating hot water and the vapor bulk is then
provided by the plastic sheet and ?uid food being concentrated. The value of hp depends on the
super?cial velocity of the liquid product and its physical and thermal properties such as viscosity
and thermal conductivity.
1
1
d
1
¼
þ
þ
ð6Þ
UA
hwA
kA
hpA
The Mylar1 plastic sheet used has a thermal conductivity of 0.155 W mKÀ1, a thickness of
0.2 mm, and covered a surface area of 6.64 m2. The resistance to heat transfer by conduction
across the plastic sheet is ?xed. Therefore, the magnitude of U depends on the combined
radiation and convective heat transfer on both sides of the plastic sheet. From the above
relationships, it is possible to calculate the ratio of energy used for evaporating water from
product ðQ Þ
Þ:
1
to the net thermal energy supplied by condensing steam ðQs The evaporator
performance can be expressed in terms of steam economy, i.e. the amount of water evaporated
per kilogram of steam consumed, namely:
m
ðm
Steam economy ¼ w ¼
p;in À mp;outÞ
ð7Þ
ms
ms
3. MATERIALS AND METHODS
Experiments were conducted with water (to establish baseline parameters), raspberry juice,
blueberry juice and blueberry puree after diluting the commercially available concentrates to
about 10% solids content. The raspberry juice concentrate was supplied by Milne Fruit
Products Inc. (Yakima, WA) while blueberry products were obtained from Valley Processing
Inc. (Sunnyside, WA) courtesy of Overlake Foods Corp. (Olympia, WA). The berry juice
concentrates and puree were shipped overnight to MCD Technologies Inc. (Tacoma, WA),
allowed to thaw, then mixed and diluted as required for the RW evaporation tests.
Steam for heating the circulating water was supplied from a boiler (Steam generator model E-40,
Clayton Industries, CA) and either directly injected into water contained in an insulated 200-gallon
Copyright # 2004 John Wiley & Sons, Ltd.
Int. J. Energy Res. 2004; 28:1089–1100

---

ENERGY CONSUMPTION DURING REFRACTANCE WINDOW1 EVAPORATION
1093
tank or passed through a shell and tube heat exchanger. The latter had a length of 0.95 m with shell
diameter of 0.21 m. A PID controller that responded to the set water temperature in the tank was
used to regulate the steam supply to both heating systems. For the direct steam injection, the water
over?owing from an outlet pipe near the top of the tank was collected and weighed to determine
the quantity of steam supplied. The condensate discharged from the shell and tube heat exchanger
was measured similarly. The circulating water was heated to a temperature of about 978C before
starting product circulation. For easy adjustment of product ?ow rates, an electronic speed control
was connected to a 3HP positive displacement rotary lobe pump (Waukesha Cherry-Burrell,
Delavan, WI). During calibration, with speed settings at 184, 169 and 150 rpm, the pump
respectively discharged 160, 147, and 130 kg of diluted product per minute. Besides the setting of
feed pumping rates, experiments were done with the evaporator tray tilted at 24, 30, and 378 from
the horizontal position. The rate of evaporation was measured by recording the decrease in the
weight of product contained in a 200 kg capacity barrel, which was placed on a platform scale.
Another small compressed air pump actuated by a ?oat device in the evaporator header tank
intermittently pumped more of the dilute feed product into the header tank to maintain the
product level. To measure the evaporation rate, the decrease in volume in the supply tank was
recorded until all the product was emptied into the header tank. At this point, the constant level in
the header tank could no longer be sustained.
The total solids content of juice was measured using automatic temperature compensating
type hand refractometers (Atago ATC-1E and ATC-2E for brix ranges 0$32% and 28$62%,
respectively). For higher total solids contents, concentrates were diluted before the measurement
of brix. The brix number of the concentrates was then calculated from the dilution factor and
measured brix of the diluted samples. For temperature measurements, type-T thermocouples
were connected to a data logger (Model 21X, Campbell Scienti?c Inc., Logan, UT) and their
output relayed to a computer to display the temperature pro?les of product and circulating hot
water at various points (Figure 1). The overall heat transfer coe?cient, energy consumption and
evaporator capacity were determined from the recorded temperature and ?ow rate data.
4. RESULTS AND DISCUSSION
4.1. Baseline study with tap water as feedstock
Table I shows the results of evaporation tests that were conducted on a prototype production
machine using tap water as feedstock at a ?xed evaporator tilt of 378. Each of those experiments
lasted 1.0 h. The quantity of water evaporated from the 6.64 m2 surface ranged from 131 to
158 kg hÀ1 (19.7–23.8 kg hÀ1 mÀ2), while the energy consumption was from 2533–2787 kJ kgÀ1 of
water evaporated. Modi?cation of the air handling system and increase in circulating hot water
?ow rate increased the evaporation rate from 143 to 173 kg hÀ1 (21.5–26.0 kg hÀ1 mÀ2) (Tables I
& II(a)). Further adjustment of the air?ow pattern within the evaporator using a ba?e reduced
the splashing that occurred on the top cover and the inner walls. The amount of evaporated
water condensing on the top cover and falling back onto the feedstock was also minimized.
Since the ?ow direction of air is opposite to that of ?uid, residence time of ?uid product
increased at higher air?ow, leading to more evaporation. This countercurrent ?ow of air against
the inclined evaporation surface created some bene?cial turbulence and ripples on the ?uid
product. Zheng and Worek (1996) showed the positive in?uence of such ripples on heat and
mass transfer in thin ?lm evaporation by using equally spaced agitated glass rods on an inclined
Copyright # 2004 John Wiley & Sons, Ltd.
Int. J. Energy Res. 2004; 28:1089–1100

---

1094
C. I. NINDO ET AL.
Table I. Capacity and energy consumption of a prototype RW evaporator using water as feedstock
(evaporator tilt angle: 378, hot water ?ow: 0.9 kg sÀ1).
Heat transfer (kW)
Feed circulation Evaporation Condensate
Water
Product
Steam
Evaporation
Steam
rate (rpm)
rate (kg hÀ1) ?ow (kg hÀ1) evaporation
heating
supply heat (kJ kgÀ1)
economy
184
145
195
94.5
12.1
112
2778
0.74
184
131
181
85.1
10.8
101
2787
0.72
184
137
186
89.0
11.1
105
2771
0.73
184
144
181
94.1
10.7
101
2533
0.79
184
158
200
103.3
12.3
112
2538
0.79
Mean
143 Æ 7
189 Æ 7
93 Æ 5
11.4 Æ 0.6 106 Æ 5 2681 Æ 117
0.76 Æ 0.03
Table II. Capacity and energy consumption of RW evaporator after modi?cation of air?ow system
(evaporator tilt angle: 378, hot water ?ow: 6.8 kg sÀ1).
Heat transfer (kW)
Feed circulation Evaporation Condensate
Water
Product
Steam
Evaporation
Steam
rate (rpm)
rate (kg hÀ1) ?ow (kg hÀ1) evaporation
heating
supply
heat (kJ kgÀ1)
economy
(a) Tap water as feedstock
184
169
225
109
20
147
2778
0.75
184
170
250
111
18
163
2727
0.68
184
174
234
114
22
152
2811
0.75
169
177
230
116
18
150
2828
0.77
Mean @184 rpm
173 Æ 4
235 Æ 8
113 Æ 3
20 Æ 2
153 Æ 5
2746 Æ 49
0.74 Æ 0.03
(b) Blueberry juice and blueberry puree as feedstock
184z
139
194
90
15
127
2571
0.72
184z
148
192
96
14
124
2557
0.77
169z
135
185
88
13
121
2685
0.73
150z
126
183
82
14
120
2729
0.69
172 Æ 12z
137 Æ 7
189 Æ 5
89 Æ 4
14 Æ 1
123 Æ 3
2636 Æ 72
0.73 Æ 0.03
184}
211
274
137
19
179
2492
0.77
(c) Raspberry juice as feedstock
184
132
169
85.8
18.4
95
2601
0.78
184
124
171
81.2
18.0
97
2796
0.73
184
154
212
100.7
15.7
118
2761
0.73
Mean
137 Æ 12
184 Æ 19
89.2 Æ 7.6 17.4 Æ 1.1 103 Æ 10
2719 Æ 79
0.75 Æ 0.02
Feed type: zblueberry juice; }blueberry puree.
stainless-steel evaporation surface. The rods produced waves or eddies that increased the heat
and mass transfer several times. For the RW evaporator, a balance is needed between ?uid
turbulence and the amount of splashing that is acceptable. Despite the good results of air?ow
adjustments, the direct steam injection system used during these baseline studies was not very
Copyright # 2004 John Wiley & Sons, Ltd.
Int. J. Energy Res. 2004; 28:1089–1100

---

ENERGY CONSUMPTION DURING REFRACTANCE WINDOW1 EVAPORATION
1095
e?cient. The water temperature was di?cult to control leading to supply of more steam than
was necessary. The energy consumption therefore increased, causing a slight reduction in steam
economy (Table II(a)). In later experiments, a new controller was installed and the direct steam
injection system replaced with the shell and tube heat exchanger.
4.2. Evaporation of blueberry and raspberry juices
Blueberry and raspberry juices are usually concentrated to 658Brix in commercial operations,
while their puree counterparts are limited to about 288Brix. The change in total solids content
and viscosity of fruit juices and purees has an in?uence on the energy usage during evaporation.
For the RW evaporator, the tilt angle, temperature and ?ow of both ?uid product and
circulating hot water, may also in?uence the evaporator performance. To evaluate the
performance of the evaporator, temperatures at di?erent points (Figures 3 and 4), evaporation
rates (Figure 5), and steam economy (Figure 6) were plotted for di?erent tray elevations and
product ?ow rates.
Figures 3 and 4 are typical temperature pro?les showing the e?ect of circulating water ?ow
rate on ?uid temperatures during evaporation of blueberry juice. At a water ?ow rate of
0.9 kg sÀ1, about 18–208C temperature di?erence between hot water at inlet and outlet points
was observed (Figure 3). However, when the water ?ow rate was increased to 6.8 kg sÀ1, the
temperature of circulating hot water dropped only by an average of 3.28C. For both cases of low
and high water ?ow rates, the product temperature at the inlet and outlet di?ered by less than
1.28C. As water in the product changes into vapor, it causes product cooling which prevents any
signi?cant product temperature rise between the inlet and outlet points. Tsay and Lin (1995)
observed that liquid vaporization tends to be high at higher product inlet temperature. Since the
product temperature in an evaporator operating at atmospheric conditions mainly depends on
the heating medium temperature, it is advantageous to have the highest possible product
temperature that does not lead to quality degradation.
Figure 3. Typical temperature pro?les during evaporation of blueberry juice under direct steam injection
heating with water circulating at 0.9 kg sÀ1.
Copyright # 2004 John Wiley & Sons, Ltd.
Int. J. Energy Res. 2004; 28:1089–1100

---

1096
C. I. NINDO ET AL.
Figure 4. Typical temperature pro?les during evaporation of blueberry juice under direct steam injection
heating with water circulating at 6.8 kg sÀ1.
Figure 5. Evaporation rate for blueberry juice at di?erent evaporator tilt angles and juice ?ow
rates under direct steam injection heating.
With stable thermal input conditions, the log mean temperature di?erence (LMTD), de?ned in
Equation (3) and plotted in Figures 3 and 4, can be used to monitor the performance of the
evaporator. The temperature pro?les indicate that LMTD increased from 258C to about 308C
when the circulating hot water ?ow rate was increased from 0.9 to 6.8 kg sÀ1, respectively. This
increase in LMTD as a result of increase in water circulation rate might have contributed to the
high evaporation (Table I and II(a)). To maintain the high evaporation rate, the rest of the
Copyright # 2004 John Wiley & Sons, Ltd.
Int. J. Energy Res. 2004; 28:1089–1100

---

ENERGY CONSUMPTION DURING REFRACTANCE WINDOW1 EVAPORATION
1097
Figure 6. Quantity of water evaporated from blueberry juice per kilogram of steam used
under direct steam injection heating.
experiments with berry products were conducted with hot water circulating at 6.8 kg sÀ1. Yan and
Soong (1995) investigated heat and mass transfer along an inclined heated plate with ?lm
evaporation and reported that a reduction in the inclined angle causes an increase in the air-liquid
interfacial temperature, which in turn leads to a larger latent heat exchange. In the present study,
the very high ratios of feed ?ow to evaporation rates might have masked the in?uence of tilt angle
on the interfacial temperature. Apart from inclined angle, the viscosity and total solids content of
?uid product, and the heat capacity rates of the ?uids ?owing on both sides of the plastic sheet
might also in?uence the heat and mass transfer. For the three evaporator bed tilt angles used in this
study, evaporation rates were higher at higher product circulating rates (Figure 5). The foregoing
observations suggest that higher circulating water ?ow rate is preferable because it results in higher
interfacial air–liquid temperature which increases the latent heat ?ux. The magnitude of the
evaporative latent heat ?ux we obtained is several times that of the sensible heat ?ux (Table IV).
This indicates e?cient conversion of latent heat from condensing steam into heat ?ux for
evaporation and agrees with conclusions made by Yan and Soong (1995).
When the evaporator tray was tilted at 378 from the horizontal with hot water circulating at
6.8 kg sÀ1, the average energy required to evaporate 1 kg of water from blueberry juice, blueberry puree
and raspberry juices were 2636, 2492 and 2719 kJ, respectively (Table II(b) and (c)). Commercial fruit
juice concentrates are usually at higher brix than their puree counterparts, so the high amount of
dissolved solids binds the water molecules more tightly resulting in a rise in boiling temperature. At a
higher circulating water ?ow rate, the water temperature is nearly invariable and since the boiling point
rise of the higher brix juice is more than that of puree, slightly more energy was used in the evaporation
of blueberry juice to the higher brix than in the puree. With the direct steam injection heating method,
the highest evaporation rates recorded during the concentration of blueberry juice and blueberry puree
were 180 and 211 kg hÀ1 (i.e. 27.1 and 31.8 kg hÀ1 mÀ2), respectively (Tables II(b) and III(a)). After the
direct steam injection heating system was replaced with shell and tube heat exchanger, the highest
evaporation rate for blueberry juice was 239 kg hÀ1 (38.0 kg hÀ1 mÀ2) (Table III(b)).
Copyright # 2004 John Wiley & Sons, Ltd.
Int. J. Energy Res. 2004; 28:1089–1100

---

1098
C. I. NINDO ET AL.
The average mass of water evaporated per unit mass of steam used (steam economy) was
0.64–0.84 with direct steam injection heating method (Figure 6). These ?gures improved to 0.80–
0.99 after the shell and tube heat exchanger was installed (Table III(b)). The direct steam
injection is by design a more e?cient heating method, but during this study it was di?cult to
control the temperature to avoid boiling of circulating water. The boiling led to enthalpy loss in
discharged hot water that could have been used for the evaporation process. Therefore,
replacement of the direct steam injection with the shell and tube heat exchanger reduced the
wastage of circulating hot water and improved steam economy. The results of steam economy
for both heating systems are very representative of the overall system performance since the data
for each combination of tilt angle and product pumping speed were obtained for residence times
of more than 1 h after reaching steady state operation condition. Aboabboud et al. (1996)
analysed an atmospheric evaporator that included thermal energy recycling and obtained a
steam economy of 2.83. Budin et al. (1998) reported a steam economy of 0.91 for a single e?ect
vacuum evaporation process, while the 2- and 3-e?ect tomato paste evaporators investigated by
Rumsey (1986) had economies between 1.38 and 2.60. Fellows (1988) reported steam economy
values between 1.67 and 3.33 for one- to three-e?ect vacuum evaporators with vapor
recompression, and 0.91 to 2.5 for those without vapor recompression. The steam economy
obtained for the RW evaporator with shell and tube heat exchanger was from 0.80–0.99.
Considering that the RW evaporator is operated at atmospheric conditions, its steam economy
is comparable to 0.7–0.9 reported by Aboabboud et al. (1996) for evaporators without thermal
energy recycling. Though operating it at higher inclines facilitates gravity ?ow of ?uid product
on the evaporation surface, steam economy values at 37 and 308 tilt angles did not appear
di?erent (Figure 6).
Table III.
Feed
Evaporation
Condensate
Evaporation
Tilt
circulation
rate
?ow
heat
Steam
LMTD
Coe?. U
angle
rate (rpm)
(kg hÀ1)
(kg hÀ1)
(kJ kgÀ1)
economy
(8C)
(W mÀ2 8CÀ1)
(a) Steam economy and energy for evaporation of blueberry juice at di?erent tilt angles and feed pumping rates
(direct steam injection system)z
378
184
180
230
2668
0.78
30.1
666
378
169
136
185
2685
0.73
28.7
530
378
150
126
183
2729
0.69
29.6
485
308
184
156
186
2646
0.84
29.3
590
308
169
148
184
2612
0.81
29.1
555
308
150
131
190
2549
0.69
29.1
479
248
184
144
163
2606
0.89
28.9
543
248
169
114
161
2609
0.71
28.8
430
248
150
97
151
2611
0.64
28.8
367
(b) Performance of the evaporator after replacing steam injection system with shell and tube heat exchangerz
378
184
196
246
2718
0.80
36.0
619
378
150
223
245
2587
0.91
35.3
685
308
184
239
253
2598
0.95
35.4
733
308
150
225
249
2618
0.91
35.3
699
248
184
234
253
2624
0.92
35.2
728
248
150
228
229
2624
0.99
34.8
722
z Blueberry juice used as feedstock.
Copyright # 2004 John Wiley & Sons, Ltd.
Int. J. Energy Res. 2004; 28:1089–1100

---