SKF4153 PLANT DESIGN EQUIPMENT SIZING
Transcript of SKF4153 PLANT DESIGN EQUIPMENT SIZING
![Page 1: SKF4153 PLANT DESIGN EQUIPMENT SIZING](https://reader030.fdocument.org/reader030/viewer/2022012022/6169c1e411a7b741a34b0682/html5/thumbnails/1.jpg)
EQUIPMENT SIZING
Dr. M. Fadhil A. Wahab
Dr. Muhammad Abbas Ahmad Zaini
SKF4153‐ PLANT DESIGN
![Page 2: SKF4153 PLANT DESIGN EQUIPMENT SIZING](https://reader030.fdocument.org/reader030/viewer/2022012022/6169c1e411a7b741a34b0682/html5/thumbnails/2.jpg)
What to design ??
Consider the following block diagram.
LMcicocpchohihph TAUTTcmTTcmQ Δ⋅⋅=−⋅⋅=−⋅⋅= )()( )()(
⎟⎟⎠
⎞⎜⎜⎝
⎛ΔΔ
Δ−Δ=Δ
2
1
21
lnTT
TTTLM
![Page 3: SKF4153 PLANT DESIGN EQUIPMENT SIZING](https://reader030.fdocument.org/reader030/viewer/2022012022/6169c1e411a7b741a34b0682/html5/thumbnails/3.jpg)
For shell‐and‐tube heat exchangers, tubes are typically ¾‐in. O.D.,
16 ft long, and on 1‐in. triangular spacing. A 1‐ft I.D. shell can
accommodate 300 ft2 tubes outside area, 2‐ft: 1330 ft2, 3‐ft: 3200 ft2.
The tube side is for corrosive, fouling, scaling, hazardous, high T &
P, and expensive fluids. The shell side is for more viscous, cleaner,
lower flow rate, evaporating and condensing fluids.
Shell‐and‐Tube Heat Exchangers
Reference: W.D. Seider, J.D. Seider, D.R. Lewin, Product and Process Design Principles: Synthesis, Analysis and Evaluation, John Wiley and Sons, Inc., 2010.
![Page 4: SKF4153 PLANT DESIGN EQUIPMENT SIZING](https://reader030.fdocument.org/reader030/viewer/2022012022/6169c1e411a7b741a34b0682/html5/thumbnails/4.jpg)
Pumping action‐ cause increase the elevation, velocity and pressure of a fluid.
Main purpose to provide energy to move liquids from one place to another.
Common application is to increase the pressure of liquid.
![Page 5: SKF4153 PLANT DESIGN EQUIPMENT SIZING](https://reader030.fdocument.org/reader030/viewer/2022012022/6169c1e411a7b741a34b0682/html5/thumbnails/5.jpg)
F is molar flow rate; v is molar volume; P is pressure
Due to smaller liquid molar volume, pump requires less power than compressor for the same molar flow rate and increase in P.
Normally the outlet T of liquid increase only slightly.
To increase a pressure of a stream, pump a liquid rather than compress a gas, unless refrigeration is needed.
(Note: to condense a gas through refrigeration and then pump the condensate are expensive)
Power Requirement
![Page 6: SKF4153 PLANT DESIGN EQUIPMENT SIZING](https://reader030.fdocument.org/reader030/viewer/2022012022/6169c1e411a7b741a34b0682/html5/thumbnails/6.jpg)
Capacity (Q) in gal/min (gpm) [conversion, 1ft3= 7.48 gal]
Pump head (H) in ft or m.
H =V
d
2
2g+ z
d+
Pd
ρdg
⎛ ⎝ ⎜
⎞ ⎠ ⎟ −
Vs
2
2g+ z
s+
Ps
ρsg
⎛ ⎝ ⎜
⎞ ⎠ ⎟
for negligible ΔV and Δz and constant ρ ,
H =ΔPρ g
Subscripts d and s refer to discharge and suction, respectively.
Pump Characteristics
![Page 7: SKF4153 PLANT DESIGN EQUIPMENT SIZING](https://reader030.fdocument.org/reader030/viewer/2022012022/6169c1e411a7b741a34b0682/html5/thumbnails/7.jpg)
For heads up to 3200 ft (multiple stages) and flow rates in the range 10 to 5000 gpm, use centrifugal pump.
For high heads up to 20000 ft and flow rate up to 500 gpm, use reciprocating pump.
Less common are axial pumps for heads up to 40 ft for flow ratesin the range of 20 to 100000 gpm and rotary pumps for heads up to 3000 ft for flow rate in the range 1 to 1500 gpm.
For liquid water,
Head of 3000 ft correspond to ΔP of 1300 psi,
Head of 20000 ft correspond to ΔP of 8680 psi.
Heuristics
![Page 8: SKF4153 PLANT DESIGN EQUIPMENT SIZING](https://reader030.fdocument.org/reader030/viewer/2022012022/6169c1e411a7b741a34b0682/html5/thumbnails/8.jpg)
For liquid flow, we need to include the following when determining the required pumping head.
A pipeline pressure drop of 2 psi/100ft of pipe
A control valve pressure drop of at least 10 psi
A pressure drop of 4 psi per 10‐ft rise in elevation
Estimate the theoretical horsepower (THp) for pumping liquid using,
Heuristics (cont’d)
THp = (gpm) pressure increase, psi( ) /1714
Reference: W.D. Seider, J.D. Seider, D.R. Lewin, Product and Process Design Principles: Synthesis, Analysis and Evaluation, John Wiley and Sons, Inc., 2010.
![Page 9: SKF4153 PLANT DESIGN EQUIPMENT SIZING](https://reader030.fdocument.org/reader030/viewer/2022012022/6169c1e411a7b741a34b0682/html5/thumbnails/9.jpg)
Compressor: to increase the velocity and/or pressure of gases.
Presence of liquid can damage the compressor blades.
Centrifugal, positive displacement and momentum transfer.
If exit T exceeds 375 oF, a multistage compressor with intercoolers must be employed.
Expander or expansion turbine: used in place of valve to recoverpower as pressure is decreased.
The gas T is reduced, check for possible condensation avoid impeller erosion.
Often chilling of gas is more important than the power recovery.
For high P liquid, the power recovery through turbine is not economical.
![Page 10: SKF4153 PLANT DESIGN EQUIPMENT SIZING](https://reader030.fdocument.org/reader030/viewer/2022012022/6169c1e411a7b741a34b0682/html5/thumbnails/10.jpg)
THp (adiabatic) for compressing a gas.
SCFM= std cubic ft per min at 60oF and 1 atm
T1 inlet gas T in oR
P1 and P2 are inlet and outlet P (absolute)
a=(k‐1)/k where k=Cp/Cv is specific heat ratio
T2 is estimated as,
Thp = SCFM T1
8130a⎛ ⎝
⎞ ⎠
P2
P1
⎛ ⎝ ⎜
⎞ ⎠ ⎟
a
−1⎡
⎣ ⎢
⎤
⎦ ⎥
T2
= T1
P2
P1
⎛ ⎝ ⎜
⎞ ⎠ ⎟
a
![Page 11: SKF4153 PLANT DESIGN EQUIPMENT SIZING](https://reader030.fdocument.org/reader030/viewer/2022012022/6169c1e411a7b741a34b0682/html5/thumbnails/11.jpg)
Determine the tower operating conditions (T,P), and the type of condenser.
Determine the equilibrium number of stages and reflux required.
Select an appropriate contacting method (plates or packing).
Determine the number of actual plates or packing height required, as well as the locations of feed and product.
Determine the tower diameter.
Determine other factors that may influence tower operation.
![Page 12: SKF4153 PLANT DESIGN EQUIPMENT SIZING](https://reader030.fdocument.org/reader030/viewer/2022012022/6169c1e411a7b741a34b0682/html5/thumbnails/12.jpg)
Proximity of critical conditions should be avoided.
Typical operating P is 1 to 415 psia (29 bar).
For vacuum operation P > 5 mmHg.
Normally total condenser is used (except for low boiling components and where vapor distillate is desired).
Preliminary material balance to estimate the distillate and bottom product compositions.
Distillation Tower Operating Conditions
![Page 13: SKF4153 PLANT DESIGN EQUIPMENT SIZING](https://reader030.fdocument.org/reader030/viewer/2022012022/6169c1e411a7b741a34b0682/html5/thumbnails/13.jpg)
Assume cooling water available at 90oFAlgorithm: Tower Pressure & Condenser Type
PD : Dist PPB: Bottom PPB=PD+10 psia
Reference: W.D. Seider, J.D. Seider, D.R. Lewin, Product and Process Design Principles: Synthesis, Analysis and Evaluation, John Wiley and Sons, Inc., 2010.
![Page 14: SKF4153 PLANT DESIGN EQUIPMENT SIZING](https://reader030.fdocument.org/reader030/viewer/2022012022/6169c1e411a7b741a34b0682/html5/thumbnails/14.jpg)
Other Things to Consider
If top stream contains both condensable and non‐condensable components, the condenser is designed to produce both vapor distillate and liquid distillate.
The PD is calculated at 120oF for the condensable components
in liquid distillate.
For vacuum operation, the vapor distillate is sent to vacuum pump.
If refrigerant is used, always consider placing water‐cooled partial condenser ahead of it (to reduce coolant requirement).
![Page 15: SKF4153 PLANT DESIGN EQUIPMENT SIZING](https://reader030.fdocument.org/reader030/viewer/2022012022/6169c1e411a7b741a34b0682/html5/thumbnails/15.jpg)
Design of Distillation Tower
What to design ??
Consider the following scenarios:
At total reflux ratio
Minimum no. of stages
High utility cost
At minimum reflux ratio
Infinite no. of stages
Low utility cost
PD
V
L
D
R=L/D
![Page 16: SKF4153 PLANT DESIGN EQUIPMENT SIZING](https://reader030.fdocument.org/reader030/viewer/2022012022/6169c1e411a7b741a34b0682/html5/thumbnails/16.jpg)
Valid for single feed, distillate and bottoms, i.e. Ordinary Distillation.
To estimate:
Reflux ratio
No. of equilibrium stages & feed location.
Quite accurate for ideal mixtures of narrow boiling range.
Not for non‐ideal mixtures, azeotropes and mixtures of wide‐boiling range (need to use rigorous model).
Fenske‐Underwood‐Gilliland (FUG) Method
![Page 17: SKF4153 PLANT DESIGN EQUIPMENT SIZING](https://reader030.fdocument.org/reader030/viewer/2022012022/6169c1e411a7b741a34b0682/html5/thumbnails/17.jpg)
Step 1: Use Fenske Equation to determine minimum number of equilibrium stages (i.e. at total reflux, D=0, R=∞)
d and b are component molar flowrates at distillate and bottom respectively. HK (heavy key), LK (light key), α is relative volatility
Nmin
=Log d
LK
bLK
⎛ ⎝ ⎜
⎞ ⎠ ⎟
bHK
dHK
⎛ ⎝ ⎜
⎞ ⎠ ⎟
⎡
⎣ ⎢
⎤
⎦ ⎥
logαLK , HK
Step 2: Also use Fenske Equation to determine the distribution of non‐key component between distillate and bottom (d/b) streams at total reflux. Good estimate for the distribution (d/b) at finite reflux conditions.
HKNK
HK
HK
NK
NK
db
bd
LogN
,min logα
⎥⎦
⎤⎢⎣
⎡⎟⎟⎠
⎞⎜⎜⎝
⎛⎟⎟⎠
⎞⎜⎜⎝
⎛
=
![Page 18: SKF4153 PLANT DESIGN EQUIPMENT SIZING](https://reader030.fdocument.org/reader030/viewer/2022012022/6169c1e411a7b741a34b0682/html5/thumbnails/18.jpg)
Step 3: Using Underwood Equation to determine minimum reflux ratio (Rmin) that correspond to infinite number of equilibrium stages (N=∞).
∑ −=−
θαα
i
iFi xq1
∑ −=+
θαα
i
iDi xR 1min
Solve Ө by trial and error. Note: αHK (equals 1) < Ө < αLK.
Refer C.J. Geankoplis textbook for detail calculations.
Step 4: Using Gilliland correlation to estimate the actual number of equilibrium stages (N) at a specified ratio of R/Rmin. Note: R = 1.1~1.5 Rmin.
![Page 19: SKF4153 PLANT DESIGN EQUIPMENT SIZING](https://reader030.fdocument.org/reader030/viewer/2022012022/6169c1e411a7b741a34b0682/html5/thumbnails/19.jpg)
Step 5: Estimate the feed location by using Fenske Equation.
Calculate NR,min for rectifying section (between feed and distillate) and NS,min for stripping section (between feed and bottom).
Assume that, NR,min/NS,min =NR/NS; also N=NR+NS.
NR, min
=Log d
LK
fLK
⎛ ⎝ ⎜
⎞ ⎠ ⎟
fHK
dHK
⎛ ⎝ ⎜
⎞ ⎠ ⎟
⎡
⎣ ⎢
⎤
⎦ ⎥
logαLK , HK
NS , min
=Log f
LK
bLK
⎛ ⎝ ⎜
⎞ ⎠ ⎟
bHK
fHK
⎛ ⎝ ⎜
⎞ ⎠ ⎟
⎡
⎣ ⎢
⎤
⎦ ⎥
logαLK , HK
Alternatively, use Erbar‐Maddox correlation and Kirkbride equation (refer to Geankoplis text book).
![Page 20: SKF4153 PLANT DESIGN EQUIPMENT SIZING](https://reader030.fdocument.org/reader030/viewer/2022012022/6169c1e411a7b741a34b0682/html5/thumbnails/20.jpg)
• For column with one feed, one absorbent or stripping agent, and two product streams.
• To estimate minimum absorbent (Lmin) or stripping agent (Vmin) flow rate, and the number of equilibrium stages N.
• Instead of relative volatility α, this method uses Ae=L/KV for absorption and Se=KV/L for stripping.
Kremser Shortcut Method for Absorption and Stripping
![Page 21: SKF4153 PLANT DESIGN EQUIPMENT SIZING](https://reader030.fdocument.org/reader030/viewer/2022012022/6169c1e411a7b741a34b0682/html5/thumbnails/21.jpg)
Minimum absorbent molar flow rate
absorbed be togasin component key offraction is )1(P, and T averageat computedcomponent key of value-K is
)1(min
K
K
A
K
AinK
K
VKL
φ
φ
−
−=
Typical actual absorbent rate L,
L = 1.5Lmin
Absorption Tower
To calculate number of equilibrium stages N, use
φAK
=A
eK−1
AeK
N +1 −1A
eK=
LK
KV
Where,
![Page 22: SKF4153 PLANT DESIGN EQUIPMENT SIZING](https://reader030.fdocument.org/reader030/viewer/2022012022/6169c1e411a7b741a34b0682/html5/thumbnails/22.jpg)
Min stripping agent molar flow rate
stripped be toliquid feed in thecomponent key offraction is )1(P and T averageat computedcomponent key of value-K is
)1(min
K
K
S
K
SK
in
KKLV
φ
φ
−
−=
Typical actual stripping agent rate V,
V = 1.5Vmin
Stripping Tower
To calculate number of equilibrium stages N, use
φSK
=S
eK−1
SeK
N +1 −1 Where, SeK
=K
KV
L
![Page 23: SKF4153 PLANT DESIGN EQUIPMENT SIZING](https://reader030.fdocument.org/reader030/viewer/2022012022/6169c1e411a7b741a34b0682/html5/thumbnails/23.jpg)
Plate efficiency (Eo) to convert Nequilibrium (equilibrium stages) to actual trays (Nactual).
Height equivalent to a theoretical plate (HETP) Height equivalent to a theoretical plate (HETP) to convert Nequilibrium to packed height.
Plate Efficiency and HETP
Nactual
=N
equilibrium
Eo
Packed Height = Nequilibrium
(HETP)
![Page 24: SKF4153 PLANT DESIGN EQUIPMENT SIZING](https://reader030.fdocument.org/reader030/viewer/2022012022/6169c1e411a7b741a34b0682/html5/thumbnails/24.jpg)
Column Height (ft) =
(Nactual ‐ 1) x (Tray Spacing)
+ Height of sump below bottom tray
+ Disengagement height above top tray
Height of Tray Tower
Note:
For structural reasons, tower height must not
exceed 200 ft. If calculated height exceeding
200 ft, consider using tower in series.
![Page 25: SKF4153 PLANT DESIGN EQUIPMENT SIZING](https://reader030.fdocument.org/reader030/viewer/2022012022/6169c1e411a7b741a34b0682/html5/thumbnails/25.jpg)
Values of HETP are usually derived from experimental data for a particular type and size of packing.
Packing vendors/manufacturers can provide HETP values.
Typical values of :
For modern random packing: 2 ft
For structured packing: 1 ft
HETP as a function of nominal diameter of random packings, and specific surface area as recommended by Kister (1992).
Height equivalent to a theoretical plate (HETP)
![Page 26: SKF4153 PLANT DESIGN EQUIPMENT SIZING](https://reader030.fdocument.org/reader030/viewer/2022012022/6169c1e411a7b741a34b0682/html5/thumbnails/26.jpg)
Tower diameter is calculated to avoid flooding (i.e. liquid began to
fill the tower and leave with vapor at top).
The diameter depends on,
Flowrates of vapor and liquid.
Properties of vapor and liquid.
Why Tower Diameter ?
![Page 27: SKF4153 PLANT DESIGN EQUIPMENT SIZING](https://reader030.fdocument.org/reader030/viewer/2022012022/6169c1e411a7b741a34b0682/html5/thumbnails/27.jpg)
Tower inside diameter,
DT
=4G
fUf( )π 1 −
Ad
AT
⎛ ⎝ ⎜
⎞ ⎠ ⎟ ρ G
⎡
⎣
⎢ ⎢ ⎢ ⎢
⎤
⎦
⎥ ⎥ ⎥ ⎥
12
Flooding velocity,
Uf
= C ρL
− ρG
ρG
⎛ ⎝ ⎜
⎞ ⎠ ⎟
12
Diameter of Tray Tower
![Page 28: SKF4153 PLANT DESIGN EQUIPMENT SIZING](https://reader030.fdocument.org/reader030/viewer/2022012022/6169c1e411a7b741a34b0682/html5/thumbnails/28.jpg)
( )( )
( )
0.1 if 2.0=
0.10.1 if 9
1.01.0=
1.0 if 1.0
Also,
/L/G=parameter ratio flow
liquid ofdensity gas ofdensity area sectional-cross insidetower areadowncomer
0.85 to0.75= velocityflooding offraction liquid of rate flow mass=L gas of rate flow mass
5.0
≥
≤≤−
+
≤=
=
====
==
LG
LGLG
LGT
d
LGLG
LG
Td
F
FF
FAA
F
AAfG
ρρ
ρρ
![Page 29: SKF4153 PLANT DESIGN EQUIPMENT SIZING](https://reader030.fdocument.org/reader030/viewer/2022012022/6169c1e411a7b741a34b0682/html5/thumbnails/29.jpg)
( )[ ]
tray on the area hole total=A )2(A= tray theof area active=A
1.0)/(A0.06 tray withsievefor 5.0/A5 = 1.0)/(A tray withsievefor 1=
trayscap bubble and for valve 1= factor area hole
)absorption (typical system foamingfor 0.75-0.5= on)distillati (typical system foaming-nonfor 1=
factor foamingdyne/cmin tension surface is /20)(=factor tension surface
19.4) figure (see plate perforatedfor parameter capacity
parametercapacity
hTa
hh
h
0.2
d
aa
a
HA
F
ST
SB
HAFSTSB
A
AAA
F
FF
C
FFFCC
−
≤≤+≥
=
==
=
==
σσ
![Page 30: SKF4153 PLANT DESIGN EQUIPMENT SIZING](https://reader030.fdocument.org/reader030/viewer/2022012022/6169c1e411a7b741a34b0682/html5/thumbnails/30.jpg)
Tower inside diameter,
DT
=4G
fUf( )πρ
G
⎡
⎣ ⎢
⎤
⎦ ⎥
12
For flooding velocity, use
{ } { }
liquid ofdensity vapor ofdensity 0.70= velocityflooding offraction
32.2ft/s=g 19.1) Table (Seefactor packing 2
)(2
2
====
⎟⎟⎠
⎞⎜⎜⎝
⎛=
LG
P
LLlOH
GPf
fF
ffgFU
Y
ρρ
μρρ
ρ
Diameter of Packed Tower
![Page 31: SKF4153 PLANT DESIGN EQUIPMENT SIZING](https://reader030.fdocument.org/reader030/viewer/2022012022/6169c1e411a7b741a34b0682/html5/thumbnails/31.jpg)
( ) ( ) ( )[ ]
{ }
{ }cP 20 to0.3 from es viscositiliquidfor valid
96.0greater),or in. 1diameter nominal packing random(for function Viscosity
1.4 to0.65 from ratiosdensity for valid
6313.06776.28787.0
function,Density
Tower)Tray in (asparameter ratio flow 10; to0.01=Yfor validln007544.0ln1501.0ln0371.17121.3exp
,
19.0
2)(2)(2
32
LL
L
LOH
L
LOHL
LG
LGLGLG
f
f
FFFFY
Where
μμ
ρρ
ρρ
ρ
=
⎟⎟⎠
⎞⎜⎜⎝
⎛−⎟⎟
⎠
⎞⎜⎜⎝
⎛+−=
=−−−−=
![Page 32: SKF4153 PLANT DESIGN EQUIPMENT SIZING](https://reader030.fdocument.org/reader030/viewer/2022012022/6169c1e411a7b741a34b0682/html5/thumbnails/32.jpg)
References
• J.M. Coulson, J.F. Richardson, R.K. Sinnot, Chemical Engineering Vol. 6, Pergamon Press, 1985.
• L.T. Biegler, I.E. Grossman, A.W. Westerberg, Systematic Methods of Chemical Process Design, Prentice Hall, 1997.
• Monograph, Process Design and Synthesis, Universiti Teknologi Malaysia, 2006/07
• R. Smith, Chemical Process Design, McGraw Hill, 1995.• W.D. Seider, J.D. Seider, D.R. Lewin, Product and Process Design
Principles: Synthesis, Analysis and Evaluation, John Wiley and Sons, Inc., 2010.