Example Shell-and-Tube Exchanger Using The Kern, Bell-Delaware, Wills-Johnston PDF
Example Shell-and-Tube Exchanger Using The Kern, Bell-Delaware, Wills-Johnston PDF
Example Shell-and-Tube Exchanger Using The Kern, Bell-Delaware, Wills-Johnston PDF
■ Solution
The procedures for all three methods have been outlined briefly in the shell-and-tube section.
Appendix D provides the following data for water:
Kern Method
Determine the flow area at the shell centerline. The gap between tubes PD is given as 0.00635 m. The
cross-flow area along the centerline of flow in the shell is given by Eq. (14-32).
Ds PD L B 0.584(0.00635)(0.1524)
Ss = = = 0.02225 m2
PT 0.0254
Determine De from Eq. (14-33).
4 PT2 − π Do2 /4 4[(0.0254)2 − (π/4)(0.019)2 ]
De = = = 0.02423 m
π Do π(0.019)
To obtain the heat-transfer coefficient at an average water-film temperature requires evaluation of the
Reynolds and Prandtl numbers.
De G s 0.02423(494.4)
Re = = = 16,550
µf 0.000724
Cp µ 4.179(0.000724)
Pr = = = 4.86
k f 0.000623
718 C H A P T E R 14 Heat-Transfer Equipment—Design and Costs
Calculate the pressure drop on the shell side, assuming no effect for any type of fluid leakage. The
number of baffles on the shell side is obtained from Eq. (14-36).
Ls 3.66
NB = −1= − 1 = 22.2 or 22
L B + tb 0.1524 + 0.005
For a shell-side Reynolds number of 16,550, Fig. 14-44 provides a value of 0.062 for the friction fac-
tor. The pressure drop is obtained from Eq. (14-35) as
4G 2s Ds (N B + 1)
ps =
2ρ De (µ/µw )0.14
s
4(0.062)(494.4)2 (0.584)(22 + 1)
= = 16,420 Pa
2(995)(0.02423)(0.000803/0.000657)0.14
Bell-Delaware Method
The first step in this method is to calculate the ideal cross-flow heat-transfer coefficient. Calculate
Vmax from Eq. (14-39) and obtain Sm from Eq. (14-40) to substitute into Eq. (14-22).
(D OTL − Do )(PT − Do )
Sm = L B Ds − D OTL + where D OTL = Ds − b = 0.549
PT
(0.549 − 0.019)(0.0254 − 0.019)
= 0.1524 0.035 + = 0.0255 m2
0.0254
ṁ T 11
Vmax = = = 0.4335 m/s
ρ Sm 995(0.0255)
ρVmax Do 995(0.4335)(0.019)
Re = = = 10,205
µ 0.000803
Cp µ 4.179(0.000803)
Pr = = = 5.449
k 0.000616
The ideal heat-transfer coefficient is given by
k
hi = a Rem Pr0.34 F1 F2
Do
where constants a and m are obtained from Table 14-1 for a staggered tube array, F1 from
Eq. (14-22b), and F2 from Table 14-2.
0.616 5.449 0.26
hi = (0.273)(10,205)0.635
(5.449)0.34
(0.99)
0.019 4.345
= 5807 W/m2 ·K
Design of Key Heat Exchanger Types 719
The actual shell-side heat-transfer coefficient is obtained from Eq. (14-41). This requires obtaining
values for JC , JL , and J B using the appropriate correction factors to account for baffle configuration,
leakage, and bypass. Equation (14-42) permits calculation of Fc
1 2(Ds − 2L c ) −1 Ds − 2L c −1 Ds − 2L c
Fc = π+ sin cos − 2 cos
π D OTL D OTL D OTL
Ds − 2L c 0.584 − 2(0.146)
= = 0.5318
D OTL 0.549
1
Fc = [π + 2(0.5318) sin(cos−1 0.5318) − 2 cos−1 0.5318] = 0.6437
π
From Fig. 14-45
JC = 0.55 + 0.72Fc = 0.55 + 0.72(0.6437) = 1.013
To obtain JL , calculate the leakage areas Ssb and St b from Eqs. (14-43a) and (14-43b), respectively.
sb −1 2L c
Ssb = Ds π − cos 1−
2 Ds
0.005 2(0.146)
= (0.584) π − cos−1 1 − = 0.003058 m2
2 0.584
t b 1 + Fc
St b = πDo NT
2 2
0.0008 1 + 0.6437
= π(0.019) (384) = 0.007535 m2
2 2
The correction factor JL is obtained from Fig. 14-46, utilizing Ssb and St b .
LB 0.1524
Fbp = (Ds − D OTL ) = (0.035) = 0.2092
Sm 0.0255
Note that Fbp = Sb /Sm , and Fig. 14-47 can be used to obtain a J B value of 0.935 when Nss /Nc = 0.2.
The corrected heat-transfer coefficient from Eq. (14-41) is then
h = h i Jc JL J B
= 5807(1.013)(0.56)(0.935) = 3080 W/m2 ·K
720 C H A P T E R 14 Heat-Transfer Equipment—Design and Costs
Evaluation of the pressure drop using the Bell-Delaware method is similar to the process for obtain-
ing the heat-transfer coefficient. The ideal cross-flow pressure drop through one baffle space is
obtained with the use of Eq. (14-46).
ρVmax
2
pc = (K a + Nc K f ) assume K a = 1.5
2
Ds 2L c
Nc = 1− PT P = 0.866PT , for triangular array
PT P Ds
0.584 2(0.146)
= 1− = 13.27
0.866(0.0254) 0.584
A value of 0.495 for K f is obtained by using the following relation given in the footnote of
Table 14-10:
Next, calculate the number of effective cross-flow rows in the window zone from Eq. (14-48).
0.8L c 0.8(0.146)
Ncw = = = 5.31
PT P 0.866(0.0254)
Now calculate the window zone pressure drop for Re > 100.
(2 + 0.6Ncw )ṁ 2T
pw =
2Sm Sw ρ
(11)2
= [2 + 0.6(5.31)] = 375 Pa
2(0.0255)(0.03298)(995)
Finally, estimate the leakage and bypass correction factors R B and R L . To obtain R B , use the calcu-
lated values of Fbp and Nss /Nc = 0.2 with Fig. 14-48. This gives a value of 0.82 for R B . For R L use
the area ratio values of (Ssb + St b )/Sm and Ssb /(Ssb + St b ) with Fig. 14-49 to obtain a value of 0.365
for R L .
Design of Key Heat Exchanger Types 721
0.036(2)(0.005)/0.005 + 2.3[2(0.005)/0.005]−0.177
=
2(995)(0.004567)2
= 50.75
Calculate the tube-to-baffle clearance resistance coefficient n t from Eqs. (14-58) and (14-59).
t b t b
St = NT π Do +
2 2
= 384π(0.019 + 0.0004)(0.0004) = 0.00936 m2
0.036(2tb /t b ) + 2.3(2tb /t b )−0.177
nt =
2ρ St2
0.036(2)(0.005/0.0008) + 2.3[2(0.005/0.0008)]−0.177
=
2(995)(0.00936)2
= 11.02
1.9e0.6856 Sw /Sm
nw =
2ρ Sw2
1.9 exp[0.6856(0.03298/0.0255)]
nw = = 2.13
2(995)(0.03298)2
722 C H A P T E R 14 Heat-Transfer Equipment—Design and Costs
The bypass flow resistance coefficient n b is calculated from Eqs. (14-61) and (14-62).
Sb = (b + pp )L B assume pp ∼
=0
= (0.035 + 0)(0.1524) = 0.00533 m2
Nss
Nss = Nc = 13.27(0.2) = 2.65 ∼
=3
Nc
a(Ds − 2L c )/PT P + Nss
nb =
2ρ Sb2
Since Nc = (Ds /PT P )(1 − 2L c /Ds ), this can be rearranged and simplified to
a Nc + Nss
nb = where a = 0.133 for triangular arrays
2ρ Sb2
0.133(13.27) + 3
= = 84.2
2(995)(0.00533)2
For a first approximation assume that the fraction Fcr of the flow that is in cross-flow over the bundle
is 0.5 to initiate a calculation for the flow resistance coefficient n c . For an Fcr of 0.5
Do ṁ T Fcr
Re =
Sm µ
0.019(11)(0.5)
= = 5103
0.0255(0.000803)
The flow resistance coefficient n c is evaluated by using Eq. (14-64) where K f is obtained from the
relation given in Table 14-10 for a triangular tube array with 103 < Re < 106 .
K a + Nc K f
nc = assume K a = 1.5
2ρ Sm2
1.5 + 13.27(0.6227)
= = 7.55
2(995)(0.0255)2
−2
−1/2
n cb = n −1/2
c + nb
= (7.55−1/2 + 84.2−1/2 )−2 = 4.47
n a = n w + n cb = 2.13 + 4.47 = 6.60
−2
−1/2
n p = n a−1/2 + n −1/2
s + nt
= [(6.60)−1/2 + (50.75)−1/2 + (11.02)−1/2 ]−2 = 1.47
Design of Key Heat Exchanger Types 723
(n p /n a )1/2
Fcr =
1 + (n c /n b )1/2
(1.47/6.60)1/2
= = 0.363
1 + (7.55/84.2)1/2
Repeat the above calculations beginning with the Reynolds number evaluation to determine a new
value for Fcr until a convergence value for Fcr is obtained. The iteration results are shown below.
Iteration attempts
1 2 3 4
Fcr (initial) 0.50 0.363 0.355 0.354
Re 5103 3705 3618 3614
Kf 0.6227 0.6729 0.676 0.676
nc 7.55 8.06 8.09 8.09
n cb 4.47 4.70 4.72 4.72
na 6.60 6.83 6.85 6.85
np 1.47 1.47 1.474 1.474
Fcr (calc.) 0.363 0.355 0.354 0.354
The iteration establishes Fcr at a value of 0.354 and fixes the Reynolds number for this calculation of
the heat-transfer coefficient from Eq. (14-22) with constants a and m listed in Table 14-1, and F1 and
F2 obtained from Eq. (14-22a) and Table 14-2, respectively.
k
h= aRem Pr0.34 F1 F2
Do
0.26
0.616 5.449
= (0.273)(3614)0.635 (5.449)0.34 (0.99)
0.019 4.345
= 3004 W/m2 ·K
For the pressure drop calculation determine the various flow fractions.
Equation (14-66) for shell-to-baffle leakage flow:
1/2
np 1.474 1/2
Fs = = = 0.1704
nt 50.75
Note that the Kern method provides higher values for the heat-transfer coefficient and pressure drop
on the shell side. The Bell-Delaware and Wills-Johnston methods provide similar results for the heat-
transfer coefficient.