CHE 611A Transport Phenomena
Dr Anurag Tripathi
Chemical Engineering
I IT
Kanpur
Lecture
Using Energy Balance Gen
to Solve problems with
two independent variables
Forced Temp profile in fully developed
Convection
laminar flow in a tube
Viscous fluid with const Mike go
F IT
2 0
PIE 1 4672 I
ty 0 0 Electrical ly
coil of
GHC heating
gIzt rd
F fo egz at wall
Postulate F TOR No 0 dependence
f
aee
o.edu zrEErItIE eaf
oses tle
DTh a
fCf Eon.of
9skdesdessipatio
motion can
be solved to give E fPofaPDyR2
tr
Omar
To solve energy ego assume conduction in z dir to be
small enough i e
I
PDE Rtorder in 2
iH
2ndorder in r
zh
BCI at zeo F T BCI at F Rg fyIg
why vesigns
BC3 at r o F finite
Nondimensionalisethe energy balanceeger using
i 0 suggest
profaned BefagBffm
these in
using the energy bal eau
leads to
eepk.mx I
II Frfft
eepe.m.li
579 Ez EE sf ha
C5 ecpk.gg afz ffzKFD
Obvious choice of E 2 as dimensionless
Lego
binary
axial position to get the energy bed eger in dimensionless
form
533oz
Is RED
Bct at 5 0 0 0
BCI at 5 1 2 I
B at 5 0 O finite
OR use leap symmetry at the center as BC3
at fog 20125 0
Can with 3Bcs need to be solved to get the complete
solution Example 12.2 1 shows how the problem is solved
using Eigenfunctions the Sturm Liouvilleproblem
fr
Another approachto solvethis problem is to resort to the
asymptotic solutions that are valid for very small
or very large values of e i e dimensionless a
coordinate
for very small valuesof
Solution G is given in the
example 12.2 2 of the fat book
we solvethe other asymptoticsolution corresponding to
large values off i e solution valid for 2770
In this limit as the fluid is suffcienty far off
from the beginning of the heated section the flow
will be THERMALLY FULLY DEVELOPED This means
that while the temperature will keep changing with
a r fog the shape of the temperature
profile will not change
t t t t t t t
T T T T T T R T T T
j
t z
t
idea Tbz
energy ives
EoGTR4 miCp z 1T whereThe bulk
temp infalse
Tba T t 9 RIR 2i
me
forthermallyfullydevelopedflow if the Tempdependenceon
r CTeupprofileshape doesnot charge linear variation in
bulk leap indicates that Tarp variation can be written as
a summation of an r dependent part a 2 dependent
part that is linear in 2 Thus we propose a solution
Galid onlyfor large values of E
BC 5 0 0 0
OCS E Get 4 B12 5 1 yes
Valid onlyfor 5 20 Be 5 0 G finite
BCI 9 0 0 0 Cannot be used
IT
2 0
samescope
small a
Moses
wi
ur
Matale
Thfame shape
Temp profile
Largez ef shifted
upwards
A modified BC is needed to satisfy the constraint that
0 0 at O is Integral
form of energy balance
is used to equate the heat transferfrom the coil
to the flowing fluid
R2Eo e GET G rdr
Ee aE eepo omaacts ds
BI
T flocs C 5 goes
F IT
Ze O
ME
e
is 2171229
4 IT
2
2
d
Energy loving is
an freep verdr
Using off E if Gf in the PDE given
by our becomes
gtda f.de coCSD since
this ODE can be solved to get
4 65 tea
5,5ft
Glens
o e Get co S tGla StG
BCIis at 5 0 D nite 4 0
BCI at E I 6GHz 1,1 1 KEI
B not3 I a Ci Sdf using 9 028 9
gives Cz 2I7
zI Exact as Gil
For T Zo I local value of 0 with in a 2
Bulk meantemp for internalflow is defined as
Tb
Cfz Ik T
Mrd fR arc aardr
IT
Bulk temp or
J
Tt
Cup mixingleap
A C 4e 12
1,2192
z
Let thetemp of the tube wall Tarpoftreflaid b w
be Tw O Then zoode chopped
e mixed in a
Tb ee9qLI cee Tw
O rq
eofCoc.etae _eqIcfG.e
ie
a et 49
GfI D
In many heat transferproblems involving convection it is
customary to
reportresults interms of Nusselt number Nu
for internalflows Newtons law of cooling is used
with balkteup instead of the ambient leap
Hence go how To hqgRzL
Nu
hf
o
Look out
0 dependsonly
upon the two
The dimensionless temp
dimensionless coordinates f E
IOG.at I
eEpTa
ETq e E Emr
KE
Fa Rte
since Pr Prandtl number we have
9 2
12 HRs
RePr Pe
Pedet number
Pe Pr Re
Trees Pectetnumberdetermines the tempvariation in
forced convection problems In general the Patel mender
is expected to play dominant role in many forced
convection problems
Higher value of f 41 is obtained at
Same zlocelion in etube of given size for low Pe
From the expression
Tb IT t 47 Bulk temp or
Cup mixingleap
it is clear that higher bulktemp will be obtained
for low Pe i e for low Re This is expected as reduction
in Re is similar to reducing value of leading to
larger residencetime for a fluid parcel in a given
length segmentleading to increased bulkleap
