Falling Film Evaporation in Vertical Tubes

ESDU 98010
Issued April 1998
Falling film evaporation

ESDU by Accuris Copyright material. For current status, contact ESDU.
in vertical tubes
Endorsed by
The Institution of Chemical Engineers
The Institution of Mechanical Engineers
ESDU 98010
ESDU DATA ITEMS
Data Items provide validated information in engineering design and analysis for use by, or under the supervision
of, professionally qualified engineers. The data are founded on an evaluation of all the relevant information, both
published and unpublished, and are invariably supported by original work of ESDU staff engineers or consultants.
The whole process is subject to independent review for which crucial support is provided by industrial companies,
government research laboratories, universities and others from around the world through the participation of some
of their leading experts on ESDU Technical Committees. This process ensures that the results of much valuable
work (theoretical, experimental and operational), which may not be widely available or in a readily usable form, can
be communicated concisely and accurately to the engineering community.
We are constantly striving to develop new work and review data already issued. Any comments arising out of your
use of our data, or any suggestions for new topics or information that might lead to improvements, will help us to
provide a better service.
THE PREPARATION OF THIS DATA ITEM
The work on this particular Data Item was monitored and guided by the following Working Party on behalf of the
Heat Transfer Steering Group:
Mr C.R. Brecknell — Wellman Process Engineering Ltd

Mr P. Martin — AMEC Process Engineering Ltd
Mr M.J.C. Moore — Kvaerner John Brown
Dr V.V. Wadekar — HTFS, Hyprotech
Dr D.R. Webb — UMIST
The Heat Transfer Steering Group first met in 1966 and now has the following membership:
Chairman
Prof. P.J. Heggs — UMIST
Vice-Chairman
Dr R. Cheesewright — Brunel University
Members
Dr T.R. Bott — Birmingham University
Eur. Ing. R.J. Brogan — HTFS, Hyprotech
Eur. Ing. T.J. Loraine — Stone & Webster Engineering Ltd
Dr A.C. Pauls* — Independent
Mr R.E. Sacks — Air Products Ltd
Dr J. Taborek* — Independent
Eur. Ing. M.A. Taylor — Independent
Mr G.H. Walter — Independent
Dr D. Wilkie — Independent.
*
Corresponding Member
The Technical work involved in the assessment of the available information and the construction and subsequent
development of the Data Item was undertaken by construction and subsequent development of the Data Item was
undertaken by
Mr G.H. Walter — Independent.
The person with overall responsibility for the work in this subject area is Mr S.J. Pugh, Head of Thermofluids Group.
ESDU 98010
FALLING FILM EVAPORATION IN VERTICAL TUBES
CONTENTS
Page
1. NOTATION AND UNITS 1
2. INTRODUCTION 4
3. PURPOSE AND SCOPE 5

4. LEADING CHARACTERISTICS OF FALLING FILM EVAPORATORS 6

4.1 Tube Dimensions 6
4.2 Thermal Characteristics 6
5. ANNULAR FILM COEFFICIENT 6

5.1 Flow Regimes in Vertical Tube-side Falling Films 6
5.2 Film Coefficients 7
5.2.1 Average coefficient in sub-cooled zone (tube entry) 7
5.2.2 Local film coefficient in evaporation zones 8
5.2.2.1 Evaporation from a laminar film, Ref, x < 30 8
5.2.2.2 Evaporation from a gravity controlled wavy-laminar film 8
5.2.2.3 Evaporation from a gravity controlled turbulent film 8
5.2.2.4 Combination of wavy-laminar and turbulent film expressions 9
5.2.3 Evaporation from a film with vapour shear 10
5.3 Evaporation in the Presence of a Non-condensable Gas 10
5.4 Evaluation of Film Transport Properties 11
5.5 Onset of Nucleate Boiling 11
6. DATA BANK 12
7. DESIGN DETAILS 13
7.1 Liquid Distribution in Inlet Channel 13
7.2 Minimum Tube Wetting Rate 15
8. INFORMATION AND DATA REQUIREMENTS 16
9. CALCULATION PROCEDURE 17
9.1 Calculation of Film Coefficient 17
9.2 Calculation of Tube Length 18
9.3 Check for Onset Of Nucleate Boiling In Evaporation Zone 19
i
ESDU 98010
10. WORKED EXAMPLE 20
11. REFERENCES AND DERIVATION 25
FIGURE 27
APPENDIX A VAPOUR SHEAR ENHANCEMENT 28
A1. Discussion 28
APPENDIX B EVAPORATION FROM GRAVITY-CONTROLLED TURBULENT FILMS AT HIGH

PRANDTL NUMBERS 29
B1. Discussion 29
APPENDIX C LIQUID FILM BREAKDOWN 31
C1. Discussion 31
ii
ESDU 98010
FALLING FILM EVAPORATION IN VERTICAL TUBES
1. NOTATION AND UNITS
Units
SI British
A heat transfer area of tube internal surface, πDL m2 ft2
C 0, C constants in Equations (5.2) and (5.3) respectively – –

cp specific heat capacity J/kg K Btu/lb °F
D internal diameter of tube m ft
g acceleration due to gravity m/s2 ft/h2

8
(9.81) ( 4.17 × 10 )
gc constant of proportionality in Newton’s second law of – lb ft/lbf h2
8
motion (1.0) ( 4.17 × 10 )
h fg specific enthalpy of vaporisation (latent heat) J/kg Btu/lb
K min , K T constants in Equations (C1.1) and (C1.2) – –

L length of zone m ft
·
M mass flow rate per tube kg/s lb/h
m, n exponents in Equations (5.2) and (5.3) respectively – –

2 1/3
α ηf
Nu f modified film Nusselt number = ---- --------- – –
λf 2
ρf g
ηf cp
Pr f film Prandtl number = -------------f – –
λf
·
Q rate of heat transfer W Btu/h
q· heat flux W/m2 Btu/h ft2
4Γ 4M· f – –
Re f film Reynolds number = ------ = -------------
η f πDηf
S solubility, see Section 8 kg/kg lb/lb
T temperature K or °C °R or °F
U overall heat transfer coefficient referred to inside W/m2 K Btu/ft2h°F

diameter of tube
U′ combined heat transfer coefficient for heating medium, W/m2 K Btu/ft2h °F
tube wall and any fouling, referred to inside diameter of
tube
· · ·
xg vapour mass fraction = Mg / ( Mg + Mf ) – –
Issued April 1998 – 31 pages

1
ESDU 98010
α heat transfer coefficient W/m2 K Btu/ft2h °F
Γ liquid loading or liquid flow rate per wetted perimeter kg/s m lb/h ft
= M· f /πD
δ liquid film thickness m ft
η dynamic viscosity kg/s m lb/h ft

(Pa s)
θ eff effective mean temperature difference K °F
Λ dimensionless group defined in Equation (A1.4) – –

µ relative error in prediction of Nusselt number defined in – –

Equation (6.1)
λ thermal conductivity W/m K Btu/h ft °F
ρ fluid density kg/m3 lb/ft3
σ surface tension N/m lbf/ft
τi interfacial shear stress N/m2 (Pa) lbf/ft2
τ *i dimensionless interfacial shear stress defined in – –

Equation (A1.5)
χ concentration, see Section 8 kg/kg lb/lb

Extra Subscripts
b at boiling or saturation temperature

evap in evaporating zone
exp experimentally measured value
f liquid film
g gas or vapour phase
h heating fluid
i/f at liquid-vapour interface
in at inlet
l liquid
lm logarithmic mean value
min minimum value
mixt mixture
n-v non-vaporising component
onb at onset of nucleate boiling
2
ESDU 98010
pred predicted value
sub in sub-cooled zone
w refers to inside tube wall surface or any dirt film present, as appropriate
x at distance x along tube from onset of evaporation
Superscript
mean or average value
Notation for temperature distribution

Tube wall
Evaporating liquid film
Liquid-vapour interface
Ti/f Direction of
Th Tw Tg heat flow
Heating Vapour or
medium vapour/gas mixture
Corresponding thermal resistances and heat transfer coefficients are as follows.
Th Tw Ti/ f Tg
U′ αf αg
He Liq Va
po
tub ating uid
( ur/
e w me e va ga s
all d p ora ph
and ium, tin ase
a ny g ) fi
fou lm
lin
g
A vapour/gas phase resistance is present when a two or more component mixture is being evaporated or
when a liquid is being evaporated in the presence of a non-condensing gas.
When the vapour phase is an almost pure component, Ti/f = Tb , Tg = Tb and the resistance diagram
becomes
3
ESDU 98010
Th Tw Tb
U′ αx
2. INTRODUCTION
The heat transfer mechanism for falling-film evaporation is one of conduction or convection through a
liquid film from a heated surface to the vapour-liquid interface where evaporation takes place. Designs are
such that no nucleation (nucleate boiling) occurs at the heated surface. Typically, liquor is fed to the tops
of tubes arranged as a bundle and flows down the walls as a film under gravity. Vapour-liquid separation
occurs usually at the bottom of the tubes, but some designs do have counter-current flow of liquid and
vapour. A typical arrangement of a falling film evaporator is shown in Sketch 2.1.
Liquor may flow within or outside tubes, but structural and maintenance considerations, together with
performance predictability and accepted practice, almost invariably dictate that evaporation occurs on the
tube-side with the heating medium on the shell-side.
With evaporation inside tubes the liquid flows down the tubes as an annular film. Resistance to heat transfer
in the case of evaporation of a single component liquid is concentrated entirely in the liquid film. There is
an additional mass transfer or diffusional resistance in the liquid phase, resulting in reduced heat transfer,
when evaporation from a liquid mixture takes place. One physical explanation is that preferential
evaporation of the more volatile or light component(s) concentrates the less volatile or heavy component(s)
at the interface, increasing the equilibrium temperature at the interface. The temperature driving force across
the film, ( Tw – Ti/f ) , and hence the effective evaporation rate is reduced. When the vapour phase is an
almost pure component (such as water in a concentration process), the vapour phase mass transfer resistance
will be negligible. When the vapour phase is also a mixture, a vapour phase diffusional resistance may be
present and the effective evaporation rate will be reduced further.
4
ESDU 98010
Liquor in
Liquor distribution
device, see Section 7.1
Tubes set proud of tubesheet

to provide circular flow weirs,
see Section 7.1
Heating fluid in
Diagonal cut to assist

Heating fluid out concentrate drainage
Vapour out
Concentrated fluid out
Sketch 2.1 Sectional arrangement of falling film evaporator with co-current flow
3. PURPOSE AND SCOPE
This Data Item presents the available data and provides correlations for predicting the local film coefficient
for flow inside plain vertical tubes of circular cross-section and hence the performance of falling film
evaporators under steady state conditions. Data available for assessing liquid mixture effects are insufficient
for developing reliable design procedures*. Consequently this aspect of falling film evaporation is not
considered further in this Data Item.
*
A method accommodating mass transfer effects in falling film evaporation, based on film theory, has been presented by Palen et al.13
The paper references most relevant work undertaken since 1980. It should be consulted for guidance where diffusional resistance in the
liquid film is deemed to be significant.
5
ESDU 98010
4. LEADING CHARACTERISTICS OF FALLING FILM EVAPORATORS
4.1 Tube Dimensions
Length typically 6 m to 11 m (20 ft to 36 ft) but may be as short as 1.5 m to 3 m (5 ft to 10 ft), particularly
in deep vacuum applications. Diameters are typically 25 mm to 64 mm (1 in to 2½ in).
4.2 Thermal Characteristics
(i) Liquid retention times are short, implying liquid hold-up and overall inventory are low.
(ii) Suppression of nucleate boiling implies low liquid film temperature differences, ( T w – Ti/f ) . As a
rule the overall temperature difference ( Th – Tg ) will be less than 15 K (27°F) and frequently less
than 10 K (18°F).
(iii) Film heat transfer coefficients are high.
(iv) Pressure drop through the tubes is often negligible.
(v) Absence of nucleate boiling and low temperature differences reduce possible fouling tendencies.
(vi) The flow pattern is that of a liquid annulus with a vapour core. Because a two-phase mixture is
absent in the vapour core there will be no hydrostatic head effect*. Absence of an imposed
hydrostatic head allows a falling film evaporator to be used at absolute pressures approaching the
hydrostatic head of the film thickness.
The combination of short liquid retention times and the ability to operate at low overall temperature
differences makes falling film evaporation particularly suitable for concentrating most heat sensitive
liquids, whilst the absence of any significant hydrostatic head effect can make falling film evaporators a
superior alternative to thermosyphon reboilers in vacuum service (also in terms of operational stability).
5. ANNULAR FILM COEFFICIENT
5.1 Flow Regimes in Vertical Tube-side Falling Films
Flow within an annular film when evaporation is taking place at the free surface may be either laminar or
turbulent, depending upon the value of the liquid film Reynolds number, Ref = 4Γ/ηf . At the lowest
Reynolds numbers ( Ref < 30 ) the flow will be fully laminar. At higher values of Reynolds number the flow
becomes firstly rippled or wavy whilst remaining essentially laminar. At even higher Reynolds numbers
the flow becomes turbulent.
In laminar films, the heat transfer mechanism is one of conduction through the film so, as evaporation
proceeds and the film becomes thinner, the film coefficient may be expected to increase. However, this
effect may be partially or even wholly offset by the increase in viscosity as the liquor becomes progressively
more concentrated. (In laminar flow the film thickness is proportional to the cube root of the viscosity.)
With turbulent films the mechanism, outside the laminar sub-layer, is one of convection and the film
*
The hydrostatic head effect may best be understood by reference to a liquid to be evaporated and introduced at the bottom of a vertical
tube. The boiling temperature will be higher at the bottom of the tube than towards the top on account of the static pressure associated
with the head of liquid or two-phase mixture above. The mean overall temperature difference, which is the driving force for heat transfer,
will be reduced. This effect is particularly significant where overall temperature differences are small.
6
ESDU 98010
coefficient will decrease as evaporation proceeds and the film becomes thinner.
If sufficient vapour is generated and the vapour velocity down the tubes is high enough, gravity flow will
be enhanced by vapour shear at the vapour-liquid interface.
The mechanisms of heat transfer through an annular film are similar, whether the process is one of
condensation of a vapour or evaporation of a liquid in the absence of nucleation at the heated surface.
Correlations for the prediction of local liquid film coefficients may therefore be expected to be of similar
form in both cases. This has been assumed in much of the literature. Here, available data specific to
evaporation are presented.
Transitions between different flow regimes are not abrupt; they occur smoothly over ranges of the
controlling parameters and a finite length of the tube.
ESDU 9102412 may be consulted for further information on the hydromechanics of annular films in vertical
tubes.
5.2 Film Coefficients
If a sub-cooled liquid zone exists the average or mean coefficient in that zone may be estimated from the
expressions presented in Section 5.2.1. In the evaporative zone variations in the liquid physical properties
as evaporation proceeds are such that local or point values of the film coefficient should be used (see
Section 5.2.2). The total heat exchanger area will be
·
Q ·
∆Q
A = ∑ ----------------------------- .
U ( Th – Tg )
(5.1)
0
5.2.1 Average coefficient in sub-cooled zone (tube entry)
At tube entry the liquid may be sub-cooled. In these circumstances all the heat transferred from the tube
wall goes to raise the bulk temperature of the liquid to its saturation or equilibrium temperature. No heat
is transmitted across the free surface.
The appropriate expression depends on whether flow within the film is laminar, transitional or turbulent as
defined by the value of the film Reynolds number, 4Γ / η f . The expressions are those of Wilke1, derived
from data over the range 5 ≥ Prf ≤ 210 , and as presented by GaniÄ7 and Seban5.
αδ 4Γ m 0.344
-------- = C 0  ------- Pr f , (5.2)
λf η 
f
where δ is the film thickness given by
0.5
ρ f δ ( gδ ) 4Γ n
- = C  ------- .
---------------------------- (5.3)
ηf η 
f
7
ESDU 98010
TABLE 5.1 Constants for Equations (5.2) and (5.3)

Laminar Transitional Turbulent
Flow
4Γ/ηf ≤ 1600 1600 < 4Γ/η f ≤ 3200 4Γ/η f > 3200
C0 0.029 0.212 × 10–3 0.181 × 10–2
m 0.533 1.2 0.933
C 0.866 0.073* 0.137

*
n 0.5 0.84 0.75
0.5
*
Two sets of values for C and n (0.095, 0.80) and (0.051, 0.87) giving values of ρ f δ ( gδ ) /η f within 10%
of one another over a wide range of film Reynolds numbers are presented by Seban5. The values in
Table 5.1 are means of these values.
5.2.2 Local film coefficient in evaporation zones
For evaporation at the free surface to take place the whole of the heat taken up from the tube wall is
transmitted across the film, implying a temperature gradient. Therefore, before evaporation can take place,
all the heat transferred from the tube wall must go into superheating the film to the extent that the liquid at
the liquid-vapour interface is at its equilibrium (saturation) temperature. Data are few for heat transfer in
this thermally-developing region and no substantiated calculation methods are available. However, this
zone is usually short compared with the overall lengths of tube required in most falling film evaporator
applications and may be accommodated by an additional length of tubing.
5.2.2.1 Evaporation from a laminar film, Ref, x < 30
The Nusselt solution for heat transfer through a laminar gravity controlled film is
2 1 /3
αx ηf – 1/4
Nu x = ------ --------- = 1.10 Re f, x . (5.4)
λf ρ2 g
f
5.2.2.2 Evaporation from a gravity controlled wavy-laminar film
– 0.22
Nu x = 0.82 Re f, x . (5.5)
This is Equation (5.4) multiplied by an enhancement factor suggested by Chun and Seban4.
5.2.2.3 Evaporation from a gravity controlled turbulent film*
0.29 0.63
Nu x = 0.0097 Re f, x Pr f . (5.6)
* 0.4 0.65
The correlation of Chun and Seban4, Nu x = 0.0038Re f Pr f , is frequently cited. Åsblad and Berntsson17 point out the experimental
procedure resulted in a systematic variation of maximum Reynolds number with evaporation temperature and thus with Prandtl number,
and questioned the suitability of extrapolation beyond the following experimental limits.
Prandtl number 5.7 2.9 1.8
Maximum Reynolds number 8000 14 500 21 000
8
ESDU 98010
The constant and exponents were obtained from a regression analysis in the form of Equation (5.6) of the
available data (see Section 6).
5.2.2.4 Combination of wavy-laminar and turbulent film expressions
The gravity controlled wavy-laminar and turbulent components, Equations (5.5) and (5.6), may be
combined into one expression in a form proposed by Chen et al.10. The correlation is
1 /6
 Re f, x Pr f 
1.73 3.78
Nu x =  ---------------- + ---------------------------------
0.31
. (5.7)
 1.32 12 
 Re f, x 1.2 × 10 
Wavy-laminar Turbulent
gravity gravity
component component
The maximum Prandtl number of the available data from which Equation (5.6) and hence Equation (5.7)
was developed was 5.7 (see Section 6). The correlations for wavy-laminar and fully laminar flows,
Equations (5.5) and (5.4), are Prandtl number independent so extrapolation to much higher Prandtl number
flows is allowable. However, this is not the case with gravity controlled turbulent flows.
Predictions from Equation (5.6) and hence Equation (5.7) at Prandtl numbers in excess of about 7, when
the flow or part of the flow down the tube is turbulent, should be treated with extreme caution.
A correlation for high Prandtl number fluids and a discussion of its limitations are presented in Appendix B.
The form of relationship between modified Nusselt number, Nu x , and liquid film Reynolds number, Re f, x ,
expressed by Equation (5.7) is illustrated in Sketch 5.1.
Equation (5.7) applies for wavy-laminar and turbulent film flows and to the transition between them.
9
ESDU 98010
2
8
0
=1
7 Pr f
6
Nux 5 =5
Pr f
3
=2
Pr f
2
=1
Pr f
0.1
100 2 3 4 5 6 7 8 1000 2 3 4 5 6 7 10000 2 3 4 5 6 7 100000
Ref,x
Sketch 5.1 Relationship between modified Nusselt number and liquid film Reynolds number
5.2.3 Evaporation from a film with vapour shear
Vapour shear enhanced flows are infrequently encountered in falling film evaporator applications except
when operation is under vacuum. Should vapour shear enhancement be considered operative, an extended
form of Equation (5.7) may be used, as presented in Appendix A.
Under vacuum conditions the pressure drop may well be of such magnitude as to reduce significantly the
vapour saturation temperature and hence the overall temperature difference driving force for heat transfer.
The magnitude of the frictional component of any pressure drop may be calculated from the interfacial
shear stress, τ i , given by Equation (A1.5) of Appendix A.
5.3 Evaporation in the Presence of a Non-condensable Gas
If operation below the saturation temperature of an evaporating liquid is required, a ‘non-condensable’ gas
may be introduced into the vapour stream. The partial pressure and hence the corresponding evaporation
temperature of the liquid to be evaporated will be reduced.
If the flow of gas and vapour are co-current with the liquid film the resultant vapour shear may be sufficient
to increase significantly the local film coefficient. However, there will also be an additional resistance to
heat transfer in the vapour phase due to the presence of the gas. This resistance will increase with the mole
fraction of the non-condensable component.
10
ESDU 98010
If the gas/vapour flow is upwards, that is counter-current to the liquid film, then that flow will be limited
by flooding. The flooding velocity is the maximum vapour velocity above which liquid will be entrained
in the vapour phase or is otherwise carried upwards rather than draining to the bottom. Guidance for the
calculation of vapour flooding velocity is given in ESDU 8903811.
5.4 Evaluation of Film Transport Properties
The expressions for the film coefficients given in Section 5.2.2 are based on the assumption of constant
fluid properties across the film and were derived at a mean film temperature given by
T f, m = T w – 0.75 ( T w – T b ) . (5.8)
Evaluation of properties at this temperature is considered appropriate for laminar films. In the absence of
data at higher film Reynolds numbers, this expression should be used for wavy-laminar and turbulent film
conditions.
In practice, variations in density, specific heat capacity and, usually, thermal conductivity across the film
are small. These properties may be evaluated at the vapour saturation temperature without introducing
significant error. If temperature differences are large, greater than about 20 K, the liquid viscosity, ηf , must
be evaluated at the mean film temperature, Tf, m . Usually in falling film evaporators temperature
differences are small so viscosity may also be evaluated at the vapour saturation temperature.
5.5 Onset of Nucleate Boiling
In falling film evaporator applications where nucleate boiling cannot be tolerated it is necessary to know
the wall superheat, ( T w – T b ) , required to initiate nucleation. Frost and Dzakowic3 have derived the
following expression, relating wall superheat to heat flux at the onset of nucleate boiling.
λ l h fg ρ g 2 1
q· onb = ------------------- ( T w – T b ) --------- . (5.9)
8σ T b onb 2
Pr f
The equation was developed on the premise of a wide range of active nucleation cavity sites being present
on the heating surface. This is not usually the case and consequently higher values of heat flux or wall
temperature than those predicted by Equation (5.9) may mark the onset of nucleate boiling. This feature
was confirmed by comparison with experimental data for a variety of fluids, including a number of organics,
water, refrigerant-12 and liquefied gases*.
*
Hydrogen, nitrogen, oxygen and neon.
11
ESDU 98010
6. DATA BANK
A data bank of local or point data was prepared from information given in Derivations 14, 15, 16 and 17.
The scope of the data bank and the fit of the data are given respectively in the following table and Figure 1.
Åsblad and
Data source Struve14 Berntsson16 Chun15
Berntsson17
Liquid R.11 R.12 R.11 R.12 R.114 Water
Tube diameter mm 32 o.d. 25 i.d. 41.8 i.d 29 o.d.
Operating pressure MPa 0.101 0.25 – 0.85 0.100 0.502 0.154 0.004 – 0.1
Reduced pressure 0.023 0.061 – 0.206 0.023 0.121 0.046 0.00018 – 0.0045
Liquid Prandtl No., Pr f 4.4 3.0 4.3 3.1 5.7 1.8 – 5.7
Film Reynolds No., Re f 115 – 8280 2030 – 18 400 50 00 – 27 00000 320 – 21 000
Heating method Condensing steam Electric resistance Liquid refrigerant Electric resistance
heating flowing counter- heating
current to
evaporating stream
The basis of comparison with predictions in all cases was the relative error, µ , defined as
Nu pred – Nu exp
µ = ---------------------------------------------- . (6.1)
Nu exp
12
ESDU 98010
7. DESIGN DETAILS
7.1 Liquid Distribution in Inlet Channel
Uniform distribution of liquid to all tubes in the inlet (top) channel is essential and individual tubes must
be wetted evenly around their circumference at entry if design performance is to be achieved.
There are a number of methods for achieving uniform distribution, of which some are outlined in this
section. Selection depends very much on the experience and practice of the equipment designer and
manufacturer.
A common means of obtaining uniform circumferential distribution in individual tubes is to allow them to
protrude above the tubesheet. Vertical slots, whose roots are all the same height above the face of the
tubesheet, are cut as shown in Sketch 7.1. Liquid is thus introduced into each tube over a series of weirs.
An alternative method is to replace the castellated tube entries by inserts designed to impart an initial swirl
to the incipient film.
Tubesheet
Sketch 7.1 Castellated tube entry
In order to maintain uniform distribution to all tubes it is essential for the liquid to be able to flow across
the tube sheet without significant loss of head.
One arrangement is shown in Sketch 7.2. The combination of an impingement plate and the weir
circumscribing the tube bundle ensures an even distribution around the bundle circumference. This
arrangement is suitable for small bundles with tubes placed on a large pitch.
For larger bundles distribution across the tubesheet may be facilitated by the provision of flow lanes and/or
a peripheral distribution weir provided with V-notches on its upper edge. The form of distributor will also
serve to counteract any maldistribution arising from non-verticality.
An alternative method is the multiple orifice distributor plate installed in a trough situated above the top
tubesheet as shown in Sketches 7.3 and 7.4.
Other arrangements include the introduction of the feed as a spray through a nozzle or group of nozzles,
suitably situated in the inlet (top) channel, and the insertion of restrictor orifices in each tube entry,
dimensioned to maintain an even head of liquid across the tubesheet.
13
ESDU 98010
A A
Impingement plate
Weir
Section on AA
Sketch 7.2 Circumferential weir distributor – Smith9
In all designs great care must be taken during the design stage and during fabrication and installation to
ensure the tubes are truly vertical and the tubesheet truly horizontal. For more detailed information
Reference 9 should be consulted.
Liquid entry
Impingement plate
Orifice plate
(distributor)
Top tubesheet
Sketch 7.3 Multiple orifice distributor – Minton8
14
ESDU 98010
orifice
tube
Sketch 7.4 Detail of orifice distributor plate – Minton8
7.2 Minimum Tube Wetting Rate
There is a minimum wetting rate that enables a continuous liquid film to be formed and maintained around
the tube circumference at tube entry. If the minimum wetting rate is not attained, rivulets separated by dry
areas will be formed, resulting in reduced area for evaporation. Hot surfaces are more difficult to rewet
than cool surfaces so the inadvertant formation of dry areas under running conditions can present operational
difficulties. If the quantity of feed liquor per tube is insufficient for the minimum wetting rate at tube entry
to be attained or the ratio of feed to evaporation is low, provision should be made for recirculation of a
proportion of the product liquor.
Knowledge of the precise nature of film breakdown is still incomplete and definitive data for the generalised
prediction of minimum wetting rates are sparse. A procedure of sufficient generality cannot therefore be
recommended. If information specific to particular combinations of fluid/process conditions is available,
it should be used.
Further discussion of film breakdown will be found in Appendix C.
15
ESDU 98010
8. INFORMATION AND DATA REQUIREMENTS
The following data are required.
System Variables
*Local
• film boiling temperature, Tb .
• Total mass flow rate per tube, M· f .
• Tube inside diameter, D .
In the case of a concentration process, additionally:

mass of solids
• Local solubility, S , defined as -------------------------------------- , or
mass of mixture
mass of non-vaporising component
• local concentration, χ , defined as ------------------------------------------------------------------------------------ .
mass of mixture
With vapour shear enhancement, additionally:
• Local vapour mass fraction, xg .
If a full design or performance calculation is required, additionally:
• Local heating medium temperature, Th .
• Combined heat transfer coefficient, referred to the inside diameter of the tube, for heating medium,
tube wall and any fouling, U ′ .
Physical properties
• Density, specific heat capacity, thermal conductivity and dynamic viscosity of the liquid phase at
the local boiling temperature, ( ρ f , cp , λ f , η f ) .
f
With vapour shear enhancement, additionally:
• Density and dynamic viscosity of the vapour phase at the boiling temperature, ( ρ g, η g ) .
If a full design or performance calculation is required, additionally:
• Latent heat of vaporisation of the vaporising component, h fg .
If a check for onset of nucleate boiling is required, additionally:
• Surface tension at the local boiling temperature, σ .
• Density of the vapour phase at the boiling temperature, ρ g .
*
In a concentration process account must be taken of boiling point elevation (BPE) in calculating the temperature driving force for heat
transfer. As evaporation proceeds the boiling point of a solution rises.
Boiling point elevation (BPE) is defined as the difference between the boiling point of a solution and the boiling point of the solvent
(usually water) at the same pressure.
16
ESDU 98010
9. CALCULATION PROCEDURE
All calculations are per tube.

9.1 Calculation of Film Coefficient
(i) Calculate, as appropriate, the mass flow rate of solids (or non-vaporising component), M· n-v , from
· ·
M n-v = S ( or χ )M f (9.1)
or the mass flow rate of mixture from
·
· M n-v
M f = ------------------- . (9.2)
S ( or χ )
(ii) Calculate the liquid loading, Γ , from
·
Mf
Γ = -------
- (9.3)
πD
and hence the film Reynolds number, Re f , from
4Γ
Re f = ------- . (9.4)
ηf
(iii) Calculate the liquid film Prandtl number, Pr f , from
ηf cp
f
Pr f = -------------- . (9.5)
λf
(iv) In the sub-cooled zone, if present, calculate the liquid film thickness, δ , from Equation (5.3)
re-arranged as
2/3
 Cη ( Re ) n
δ =  ----------------------------
f f
(9.6)
 0.5 
 ρf g 
using mean values of the physical properties and Reynolds number in the zone and the appropriate
values of C and n from Table 5.1.
(v) In the sub-cooled zone calculate the average film coefficient, α , for the zone from Equation (5.2)
rearranged as
λf m 0.344
α = ----- C 0 ( Re f ) Pr f (9.7)
δ
using mean values of Ref and Prf for the zone and the appropriate values of C 0 and m from
Table 5.1.
17
ESDU 98010
(vi) For each increment in the evaporation zone:
If Re x ≥ 30 calculate from Equation (5.7) the local Nusselt number, Nux , from
1/6
 1.73 3.78 
 0.31 Re f, x Pr f 
Nu x =  --------------- + -------------------------------  . (9.8)
1.32 12
 Re f, x 1.2 × 10 
 
If Re x < 30 calculate from Equation (5.4) the local Nusselt number, Nux , from
–1 / 3
Nu x = 1.10Re f, x . (9.9)
(vii) Calculate from the definition of the Nusselt number, Nux , the film coefficient, αx , from
1/3
 ρ 2 g
α x =  ---------
f
λ f Nu x . (9.10)
 2
 ηf 
9.2 Calculation of Tube Length
In sub-cooled zone
·
(viii) Calculate the heat load, Q , from
· ·
Q = M f c p ( T b – T in ) . (9.11)
f
(ix) Calculate the mean overall heat transfer coefficient, U , from
1 1 1
---- = ------ + ---- rearranged as
U U′ α
U′ α
U = ------------------ . (9.12)
U′ + α
(x) Calculate the effective mean temperature difference, θeff , between the heating medium and liquid
film from
θ eff = logarithmic mean temperature difference. (9.13)
(xi) Calculate length of tube required to raise the bulk temperature to its saturation value, Tb , from
Equation (5.1) rearranged as
·
Q
L sub = ---------------------- . (9.14)
πDUθ eff
18
ESDU 98010
For each increment in the evaporation zone
(xii) Calculate heat load, Q· , from
· ·
Q = M evap h fg . (9.15)
(xiii) Calculate as before (step (ix)) the mean overall heat transfer coefficient, U , from*
U′α x
U = -------------------- . (9.16)
U′ + α x
(xiv) Calculate the effective mean temperature differences, θeff , between the heating medium and the
vapour from
θ eff = arithmetic temperature difference. (9.17)
(Note: if the heating medium is a liquid undergoing significant temperature change in the zone, the
logarithmic mean temperature difference should be used.)
(xv) Calculate the incremental length of tube required from Equation (5.1) rearranged as
·
Q
L evap = ---------------------- . (9.18)
πDUθ eff
9.3 Check for Onset Of Nucleate Boiling In Evaporation Zone
(xvi) Calculate the temperature difference across the liquid film, ( T w – T b ) , from*
U′ – U
( T w – T b ) = ------------------ ( T h – T b ) . (9.19)
U′
(xvii) Calculate the incremental heat flux, q· , from
·
Q
·q = -----------
- . (9.20)
πDL
(xviii) Calculate the temperature difference across the liquid film for the onset of nucleate boiling at the
heat flux calculated in step (xvii) from Equation (5.9) rearranged as*
2· ½
 8σ T b ( Pr f ) q
( T w – T b ) onb =  ------------------------------------ . (9.21)
 λ f h fg ρ g 
(xix) Compare ( T w – T b ) onb from step (xviii) with ( T w – T b ) from step (xvi).
If ( T w – T b ) onb < ( T w – T b ) , nucleate boiling may occur.
*
The expressions presented in steps (xiii), (xvi) and (xviii) are based on the assumption of the evaporation of a single component where
the vapour phase is an almost pure component. See second resistance diagram in Notation.
19
ESDU 98010
10. WORKED EXAMPLE
The total flow per tube of saline water entering a falling film evaporator at atmospheric pressure is
M· f = 0.012 kg/s. The water enters at 85°C, 0.04 kg/kg salinity and is to leave at 0.20 kg/kg salinity. The
tube inside diameter is D = 0.035 m. The combined heat transfer coefficient for the heating medium
(condensing steam), tube wall and fouling referred to the inside tube diameter is 4000 W/m2 K and the
temperature of the heating medium is constant at 110°C. It is required to check that the design will be
satisfactory within the constraints of a maximum allowable tube length of 11 m and absence of nucleate
boiling.
Note. For illustrative purposes a constant temperature condensing heating medium has been specified. In
practice the condensation coefficient may vary along the length of the tube. If the heating medium is a
liquid its temperature will vary and the shell-side will be baffled. Accommodating varying service fluid
film coefficients or baffled shell-side single-phase flow considerably complicates procedures.

ESDU 9102412 should be consulted for information on the calculation of the condensate film coefficient
in vertical downflow outside tubes* and ESDU 830386 for the calculation of the baffled shell-side
coefficient for single phase flow.
Physical properties are as follows.
Sub-cooled Evaporation zone

zone
(Mean zone (Properties are those at zone outlets*)

properties)
S 4 0.04 0.08 0.12 0.16 0.20 kg/kg

ηf 0.000365 0.000308 0.000338 0.000374 0.000417 0.000459 kg/m s
λf 0.673 0.679 0.679 0.679 0.679 0.679 W/m K
cp 4010 4024 3847 3682 3526 3379 J/kg K

f
ρf 990 982 1011 1040 1068 1102 kg/m3
ρg kg/m3
0.595
σ 0.0627 0.0598 0.0605 0.0618 0.0630 0.0666 N/m

BPE – 0.61 1.54 2.65 3.89 5.23 K
Boiling –
100.6 101.5 102.7 103.9 105.2 °C
Temp
h fg – 2257 × 103 J/kg
*
Mean zone properties in the evaporation zone are tabulated in Table 10.1.
*
Although ESDU 9102412 is specific to condensation inside tubes, the procedures are equally applicable to condensation outside vertical
tubes.
20
ESDU 98010
For the purposes of this example the tube is divided into five increments with the film coefficients calculated
for each increment. The increments are the sub-cooled zone, S = 0.04 to 0.08, 0.08 to 0.12, 0.12 to 0.16
and 0.16 to 0.20. They are numbered 1 to 5 respectively. Interpolated values of physical properties are
used as appropriate.
Calculations are presented in detail for the sub-cooled and first evaporative zones. Calculation for all zones
are summarised in the final tabulation.
All calculations are per tube*.
Sub-cooled zone
(i) The liquid loading (Equation (9.3)) and average film Reynolds number (Equation (9.4)) are
respectively
0.012
Γ = ------------------------ = 0.109 kg/s m
π × 0.035
4 × 0.109
Re f = ------------------------ = 1294.
0.000337
(ii) The film Prandtl number (Equation (9.5)) is
0.000337 × 4016
Pr f = ------------------------------------------ = 2.00 .
0.676
(iii) The film thickness, δ , is obtained from Equation (9.6) as
2/3
 C × 0.000337 × 1305 n
δ =  -------------------------------------------------------- .
 986 × 9.81
0.5

From Table 5.1 C = 0.866 and n = 0.5 since flow is wavy-laminar ( Ref ≤ 1600 ) , so
–4
δ = 2.27 × 10 m.
(v) The average heat transfer coefficient is given by Equation (9.7) as
0.676 0.533 0.344

α = ----------------------------- × 0.029 ( 1294 ) ( 2.00 ) = 4995 W/m2 K .
–4
2.27 × 10
(viii) Heat transferred in the zone is given by Equation (9.11) as
·
Q = 0.012 × 4016 ( 100.6 – 85 ) = 752 W.
(ix) The overall heat transfer coefficient is given by Equation (9.12) as
4000 × 4995
U = ------------------------------- = 2221 W/m2 K.
4000 + 4995
*
This calculation procedure follows the order of Section 9.
21
ESDU 98010
(x) The effective overall temperature difference (see Equation (9.13)) is
( 110 – 85 ) – ( 110 – 100.6 )

θ eff = θ lm = ---------------------------------------------------------------------- = 15.9°C .
( 110 – 85 )
loge -----------------------------------
( 110 – 100.6 )
(xi) The length of tube required is given by Equation (9.14) as
752
L sub = -------------------------------------------------------------- = 0.19 m.
π × 0.035 × 2221 × 15.9
Evaporation zone 1
(i) From the definition of solubility (salinity) the mass of solids in the liquid at zone entry is given by
Equation (9.1) as
·
M n-v = 0.04 × 0.012 = 0.00048 kg/s
and remains constant throughout the evaporation regime.
At zone exit the mass of the liquid mixture is given by Equation (9.2) as
· 0.00048
M f = ------------------- = 0.006 kg/s.
0.08
The mean liquid mass flow in the zone is
· 0.012 + 0.006
M f = ---------------------------------- = 0.009 kg/s.
2
(ii) The liquid loading and film Reynolds number are given by Equations (9.3) and (9.4) respectively as
0.009
Γ = ------------------------ = 0.0819 kg/s m
π × 0.035
4 × 0.0819
Re f = --------------------------- = 1014 .
0.000323
(iii) The film Prandtl number is given by Equation (9.5) as
0.000323 × 3936
Pr f = ------------------------------------------ = 1.87 .
0.679
(vi) The local Nusselt number is given by Equation (9.8) as
1.73 3.78 1 / 6
 0.31 ( 1014 ) × ( 1.87 ) 
Nu f =  ---------------------------- + ----------------------------------------------------------  = 0.181 .
– 1.32 12
 ( 1014 ) 1.2 × 10 
22
ESDU 98010
(vii) The local evaporation film coefficient is given by Equation (9.10) as
997 × 997 × 9.81 1 / 3

α =  ---------------------------------------------------- × 0.679 × 0.181 = 5577 W/m2 K.
 0.000323 × 0.000323
(xii) Heat transferred in zone (heat load) is given by Equation (9.15) as
· 3 3
Q = ( 0.012 – 0.006 )2257 × 10 = 13.54 × 10 W.
(xiii) The overall heat transfer coefficient is given by Equation (9.16) as
4000 × 5577
U = ------------------------------- = 2329 W/m2 K.
4000 + 5577
(xiv) The effective overall temperature difference will be the arithmetic temperature difference (see
Equation (9.17)) and is
100.6 + 101.5
θ eff = 110 – ---------------------------------- = 9.0 K and Tb = 101.1 ºC .
2
(xv) The length of tube required is given by Equation (9.18) as
3
13.54 × 10
L evap = ----------------------------------------------------------- = 5.87m .
π × 0.035 × 2329 × 9.0
Check for onset of nucleate boiling in evaporation zone 1
(xvi) The temperature difference across the liquid film is given by Equation (9.19) as
4000 – 2329
( T w – T i/f ) = ------------------------------- ( 110 – 101.1 ) = 3.7 K .
4000
(xvii) The zonal heat flux is given by Equation (9.20) as
3
13.54 × 10
q· = ------------------------------------------ = 20 980 W/m2.
π × 0.035 × 5.87
(xviii) The temperature difference across the liquid film for the onset of nucleate boiling is given by
Equation (9.21) as
· ½
 8 × 0.0602 × ( 101.1 + 273 ) × ( 1.87 ) 2 × 20 980
( T w – T b ) onb =  -------------------------------------------------------------------------------------------------------------------------- = 3.8 K.
 0.679 × 2257 × 10 × 0.595
3

Study of the performance summary in Table 10.1 shows that a tube length of 10.34 m is required. The
likelihood of nucleate boiling is greatest in Zone 2 where the calculated temperature difference across the
film (step xvi) is close to the temperature difference calculated by Equation (5.9) for the onset of nucleate
boiling (step xviii). The possibility of nucleate boiling should be removed by increasing the tube length
by an additional amount, accompanied by a corresponding reduction in steam temperature.
23
TABLE 10.1 Performance Summary
Sub-cooled Zone Evaporation Zones
Zone 1 2 3 4 5
Mean Physical Properties
ηf kg/s m 0.000337 0.000323 0.000356 0.000396 0.000438
λf W/m K 0.676 0.679 0.679 0.679 0.679
cp J/kg K 4016 3936 3765 3604 3453

f
ρf kg/m3 986 997 1026 1054 1085

ρg kg/m3 – 0.595 0.595 0.595 0.595
σ N/m 0.0613 0.0602 0.0612 0.0617 0.0648
Boiling temperature, T b °C – 101.1 102.1 103.3 104.6
Calculated Quantities
M·
24
at zone exit
mixt kg/s 0.012 0.006 0.004 0.003 0.0024
Mean M· mixt in zone kg/s 0.012 0.009 0.005 0.0035 0.0027
M· evap in zone kg/s – 0.006 0.002 0.001 0.0006
Γ kg/s m 0.109 0.0819 0.0454 0.0318 0.0246
Ref – 1294 1014 510 321 225
ESDU 98010
Prf – 2.00 1.87 1.97 2.10 2.22
Nuf – – 0.180 0.209 0.231 0.250
α W/m2 K 5053 5581 6152 6449 6653
Q· W 752 13.54 × 10 3
4514 2257 1354
U W/m2 K 2233 2330 2424 2469 2498
θeff K 15.9 9.0 7.9 6.7 5.5
length of zone m 0.19 5.87 2.14 1.24 0.90
temp. difference across liquid K – 3.8 3.1 2.6 2.1
temp. difference for onb K – 3.8 3.9 3.9 3.8
ESDU 98010
11. REFERENCES AND DERIVATION
References
The references given are recommended sources of information supplementary to that in this Item.
1. WILKE, W. Wärmeübergang an Rieselfilme. VDI–Forsch., Vol. 490, B28, 1962.

2. HARTLEY, D.E. Criteria for the break-up of thin liquid layers flowing isothermally over
MURGATROYD, W solid surfaces. Int. J. Heat Mass Transfer, Vol. 7, pp. 1003-1015, 1964.
3. FROST, W. An extension of the method of predicting incipient boiling on commercially
DZAKOWIC, G.S finished surfaces. Publication 67-HT-61, ASME/AIChemE Heat Transfer
Conf. Seattle, Washington, Aug. 1967.
4. CHUN, K.R. Heat transfer to evaporating liquid films. J. Heat Transfer, Vol. 93,
SEBAN, R.A. pp. 391-396, Nov. 1971.
5. SEBAN, R.A. Transport to falling films. Heat Transfer, Vol. 6, pp. 417-428, 1978.
6. ESDU Baffled shell-and-tube heat exchangers: flow distribution, pressure drop
and heat transfer coefficient on the shellside. ESDU 83038, ESDU
International plc, London, 1983.
7. GANIâ, E.N. Thin film heat transfer. In: Handbook of heat transfer fundamentals (W.M.
Rohsenow, J.P. Hartnell and E.N. Ganic, Eds). 2nd Edn, Section P,
pp. 12-74 to 12-76, McGraw-Hill, 1985.
8. MINTON, P.E. Handbook of evaporation technology. Noyes Publications, Park Ridge,
N.J., 1986.
9. SMITH, R.A. Vaporisers: selection, design and operation. Longmans Scientific and
Technical, Harlow, Essex, 1986.
10. CHEN, S.L. General film condensation correlations. Experimental Heat Transfer,
GERNER, F.M. Vol. 1, No. 2, pp. 93-107, 1987.
TIEN, C.L.
11. ESDU Reflux condensation inside vertical tubes. ESDU 89038, ESDU
International plc, London, 1989.
12. ESDU Condensation inside tubes: condensate film coefficient for vertical
downflow. ESDU 91024, ESDU International plc, London, 1991.
13. PALEN, J.W. Falling film evaporation of binary mixtures. AIChEJ, Vol. 40, No. 2,
WANG, Q.I. pp. 207-214, Feb. 1994.
CHEN, J.C.
Derivation
The derivation lists selected sources that have assisted in the preparation of this Item.
14. STRUVE, H. Heat transfer to an evaporating falling refrigerant film. 12th Congress of
Int. Inst. Refrigeration, Madrid, 1967.
15. CHUN, K.R. Evaporation from thin liquid films. PhD Thesis, University of California,
Berkeley, 1969.
25
ESDU 98010
16. BERNTSSON, K.M. Nonazeotropic mixtures as working fluids in large heat pumps – heat
transfer in a falling film evaporator and system simulations. Dissertation,
Chalmers University of Technology, Götenborg, Sweden, 1986.
17. ÅSBLAD, A. Surface evaporation of turbulent falling films. Int. J. Heat Mass Transfer,
BERNTSSON, T. Vol. 34, No. 3, pp. 835-841, 1991.
18. NUMRICH, R. Heat transfer in turbulent films. Chem. Eng. Technol., Vol. 18, pp. 171-177,
1995.
26
Relative error in Nu between experimental data and prediction
Nu = (0.31/Re 1.32 + Re 1.73Pr 3.78/ 1.2*1012)1/6 [Equation (5.7)]
1
Chun (Derivation15)
Asblad and Berntsson (Derivation17)
0.8
Struve (Derivation14)
Berntsson (Derivation16)
0.6
0.4
0.2
Relative error
27
-0.2
-0.4
-0.6
ESDU 98010
-0.8
-1
0 5000 10000 15000 20000 25000 30000
Re
FIGURE 1 RELATIVE ERROR VERSUS FILM REYNOLDS NUMBER

ESDU 98010
APPENDIX A VAPOUR SHEAR ENHANCEMENT
A1. Discussion
If a liquid film is subject to both shear forces and gravity forces, the relevant correlations for gravity and
shear controlled flows may be combined in one expression in a form proposed by Chen et al.10 The
correlation is
1⁄ 2
1⁄ 3
α c, x
2
ηl
1⁄ 3  Re f, x Pr f 
1.73 3.78 1.5 *
Pr f τ i , x
 ---------------
0.31
+ --------------------------------- + ---------------------
Nu x = ----------- -------------------------------- =  1.32 12  , (A1.1)
λl ρ l ( ρ l – ρ g )g  Re f, x 1.2 ×10 
771.6
Wavy-laminar Turbulent Shear

gravity gravity flow
component component component
·
where Re f, x = 4Γ/η f = 4M f /πDη f (A1.2)
1.4
*  xg  1.8
τi,x = Λ --------------- Re f, x , (A1.3)
 1 – x g
1.177 0.156
0.252η f ηg
and Λ = ---------------------------------------------- , (A1.4)
2 2 ⁄ 3 0.553 0.78
D g ρf ρg
which is the dimensionless group presented by Chen et al.10

–6 –4
Typically, Λ lies in the range 5 ×10 to 2 ×10 .
The dimensionless interfacial shear stress is defined by
τ i,x g c
τ*i = ------------------------------------
2/3
-. (A1.5)
ρ f ( gη f /ρ f )
Equation (A1.1) applies to gravity controlled and varpour shear controlled flows and to the transition
between them.
Further information on vapour shear enhancement in vertical downflow is given in ESDU 9102412.
28
ESDU 98010
APPENDIX B EVAPORATION FROM GRAVITY-CONTROLLED TURBULENT FILMS AT HIGH
PRANDTL NUMBERS
B1. Discussion
The available data underlying the development of Equations (5.6) and (5.7) were confined to 1.8 < Pr < 5.7 .
There are, however, instances where falling film evaporators are required to operate with Prandtl number
fluids well in excess of the recommended extrapolation limit of about 7 (see Section 5.2.2.4) but with the
liquid film in turbulent flow. In the absence of available data specific to evaporation, a correlation presented
by Numrich18, based on condensation data for amyl alcohol (at Pr = 37.6 ) has been proposed, as follows.
0.44 0.4
Nu x = 0.003Re f, x Pr . (B1.1)
Combination of Equation (B1.1) with Equation (5.5), the equation for a gravity-controlled wavy-laminar
film, gives
1/6
 Re f, x Pr f 
2.64 2.4
Nu x = --------------- + ----------------------------- 
 0.31
. (B1.2)
 1.32 15 
 Re f, x 1.37 × 10 
In laminar flows local values of the film coefficient are the same for evaporation and condensation at a
particular film Reynolds number since they both depend on heat conduction through the film. Similarly, in
turbulent film flow there is conduction through the laminar sub-layer but outside the layer the mechanism
is one of convection. If the convective process may be assumed to be independent of the direction of heat
transfer, then the heat transfer coefficients will be the same in both cases.
Additional data in the range 7 < Prf < 50 are sparse. The limitations of Equation (B1.1) and (B1.2) are
apparent from Sketch B1.1 for Prf = 5.0 and Sketch B1.2 for Pr f = 40 .
Sketch B1.1 is a plot of Equations (5.6) and (B1.1) at a Prandtl number of 5. The Numrich equation
underpredicts the data by between 20% and 45%.
Sketch B1.2 is a similar plot at a Prandtl number of 40. Equation (5.6) overpredicts the Numrich data by
between 120% and 220%.
In the absence of data in the range 7 < Prf < 30 , no definitive method valid for this range can be stated.
29
ESDU 98010
0.5
0.4
Eqn (5.6)
Nux
0.3
Eqn (B.1)
0.2
0.1
0 2000 4000 6000 8000 10000
Re
Sketch B1.1 Comparison of Equation (5.6) predictions with those of Equation (B1.1): Prf = 5.0
1.6
Eqn (5.6)
1.4
1.2
1.0
Nux
0.8
Eqn (B.1)
0.6
0.4
0.2
0 2000 4000 6000 8000 10000
Re
Sketch B1.2 Comparison of Equation (5.6) predictions with those of Equation (B1.1): Prf = 40
30
ESDU 98010
APPENDIX C LIQUID FILM BREAKDOWN
C1. Discussion
Determination of the minimum tube wetting rate implies the determination of the heat flux above which a
liquid film breaks down at a particular liquid flow rate. Relevant factors include:
(a) liquid flow rate per wetted perimeter, Γ ,
(b) liquid film Reynolds number, Re f ,
(c) surface tension of the liquid, together with the liquid/surface contact angle,
(d) nature and condition of the tube surface,
(e) liquid composition in the case of liquid mixtures,
(f) interfacial shear.
Experimental data are far from complete and further research is necessary before definitive procedures for
the prediction of the onset of film breakdown can be established.
The procedure of Minton8 and the recommendation of Smith9 have received some currency.
Minton recommends the following expressions.
To induce a film at tube entry
3 1/5
Γ min = K min ( η f ρ f σ ) (C1.1)
where K min = 0.51 with SI units or 2250 with British units.
Once a film has been formed a lower flow rate will maintain a continuous film. At the bottom of the tube
the minimum terminal flowrate is given by
3 1/5
ΓT = KT ( ηf ρf σ ) (C1.2)
where K T = 0.063 with SI units or 277 with British units.
Equations (C1.1) and (C1.2) are adaptations of an expression developed by Hartley and Murgatroyd2.
Limitations implicit in the development of Equations (C1.1) and (C1.2) are that the data of Hartley and
Murgatroyd were for sub-cooled water at temperatures well below surface evaporation temperature and in
the absence of heat transfer.
Smith9 simply recommends a minimum Reynolds number, Re f, min , at the bottom of the tubes of 100.
31
ESDU 98010
KEEPING UP TO DATE
Whenever Items are revised, subscribers to the service automatically receive the material required to update
the appropriate Volumes. If you are in any doubt as to whether or not your ESDU holding is up to date, please
contact us.
Please address all technical engineering enquiries and suggestions to:
ESDU International plc Tel: 020 7490 5151 (from the UK)
+44 20 7490 5151 (from outside the UK)
Fax: 020 7490 2701 (from the UK)
+44 20 7490 2701 (from outside the UK)
E-Mail: esdu@ihs.com
Website: www.ihsesdu.com
For users in the USA, please address all Customer Service and Support enquiries and suggestions to:
IHS Engineering Products Tel: 1 800 447 3352 (toll free number)
and Fax: 1 303 397 2599
Global Engineering Documents Website: www.ihs.com
global.ihs.com
ESDU 98010
Falling film evaporation in vertical tubes
ESDU 98010
ISBN 978 1 86246 047 8, ISSN 0141-402X
Available as part of the ESDU Series on Heat Transfer. For information

on all ESDU validated engineering data contact ESDU International plc,
27 Corsham Street, London N1 6UA.
ESDU 98010 provides a guide to the design and operation of falling film
evaporators under steady state conditions with pure component fluids.
Correlations derived from an analysis of experimental data extracted from
the literature are provided for predicting the local film heat transfer
coefficient for flow inside plain vertical tubes of circular cross section. The
background to the correlations is described fully, and a calculation
procedure that predicts, using the new correlations, the tube length
required for a particular duty is laid out and illustrated fully by a worked
example. The effect of vapour shear is considered and limits due to
minimum wetting rate and the onset of nuclear boiling are defined. The
leading design details and practical operating characteristics of falling film
evaporators are discussed. The mechanism of heat transfer is discussed,
the various flow regimes that can occur in the annular film over the
practical range of operating conditions and at specific locations within the
evaporator are outlined and their effects on the film coefficients, for which
equations are provided, are explained. The scope of the data bank that
was used to derive the new correlations is stated and the performance of
the new equation for Nusselt number against the data is illustrated
graphically. A full bibliography is provided.
© ESDU International plc, 2008
All rights are reserved. No part of any Data Item may be reprinted, reproduced, or
transmitted in any form or by any means, optical, electronic or mechanical including
photocopying, recording or by any information storage and retrieval system without
permission from ESDU International plc in writing. Save for such permission all
copyright and other intellectual property rights belong to ESDU International plc.

Falling Film Evaporation in Vertical Tubes

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Falling Film Evaporation in Vertical Tubes

Uploaded by

Copyright:

Available Formats

ESDU 98010

Issued April 1998

Falling film evaporation

THE PREPARATION OF THIS DATA ITEM

Mr C.R. Brecknell — Wellman Process Engineering Ltd

Mr G.H. Walter — Independent.

1. NOTATION AND UNITS 1

3. PURPOSE AND SCOPE 5

4. LEADING CHARACTERISTICS OF FALLING FILM EVAPORATORS 6

5. ANNULAR FILM COEFFICIENT 6

8. INFORMATION AND DATA REQUIREMENTS 16

11. REFERENCES AND DERIVATION 25

APPENDIX A VAPOUR SHEAR ENHANCEMENT 28

APPENDIX B EVAPORATION FROM GRAVITY-CONTROLLED TURBULENT FILMS AT HIGH

APPENDIX C LIQUID FILM BREAKDOWN 31

1. NOTATION AND UNITS

C 0, C constants in Equations (5.2) and (5.3) respectively – –

cp specific heat capacity J/kg K Btu/lb °F

D internal diameter of tube m ft

g acceleration due to gravity m/s2 ft/h2

h fg specific enthalpy of vaporisation (latent heat) J/kg Btu/lb

K min , K T constants in Equations (C1.1) and (C1.2) – –

m, n exponents in Equations (5.2) and (5.3) respectively – –

U overall heat transfer coefficient referred to inside W/m2 K Btu/ft2h°F

Issued April 1998 – 31 pages

η dynamic viscosity kg/s m lb/h ft

Λ dimensionless group defined in Equation (A1.4) – –

µ relative error in prediction of Nusselt number defined in – –

λ thermal conductivity W/m K Btu/h ft °F

ρ fluid density kg/m3 lb/ft3

σ surface tension N/m lbf/ft

τi interfacial shear stress N/m2 (Pa) lbf/ft2

τ *i dimensionless interfacial shear stress defined in – –

χ concentration, see Section 8 kg/kg lb/lb

b at boiling or saturation temperature

Notation for temperature distribution

Corresponding thermal resistances and heat transfer coefficients are as follows.

Tubes set proud of tubesheet

Diagonal cut to assist

Concentrated fluid out

3. PURPOSE AND SCOPE

4.1 Tube Dimensions

4.2 Thermal Characteristics

(iii) Film heat transfer coefficients are high.

(iv) Pressure drop through the tubes is often negligible.

5. ANNULAR FILM COEFFICIENT

5.1 Flow Regimes in Vertical Tube-side Falling Films

5.2 Film Coefficients

5.2.1 Average coefficient in sub-cooled zone (tube entry)

where δ is the film thickness given by

TABLE 5.1 Constants for Equations (5.2) and (5.3)

C 0.866 0.073* 0.137

5.2.2 Local film coefficient in evaporation zones

5.2.2.1 Evaporation from a laminar film, Ref, x < 30

5.2.2.2 Evaporation from a gravity controlled wavy-laminar film

5.2.2.3 Evaporation from a gravity controlled turbulent film*

5.2.2.4 Combination of wavy-laminar and turbulent film expressions

5.2.3 Evaporation from a film with vapour shear

5.3 Evaporation in the Presence of a Non-condensable Gas

5.4 Evaluation of Film Transport Properties

5.5 Onset of Nucleate Boiling

7.1 Liquid Distribution in Inlet Channel

Sketch 7.1 Castellated tube entry