,()/

,. \.....-

~o/ ?J /

Program

determines

two-phase flow
We chose to approximate the boundaries of the Baker graph by eight straight lines, as is shown in Fig. 1. The equations of those hiles are: Cl: lag By = 3.698 C2: lag By = 4.261
C3: C4: C5: C6: lag lag lag lag By By By By = = = = 4.959 4.477 4.019 1.935
-

BASIC computer program determines which of seven possible flow patterns a gas-liquid mixture will have as defined by the Baker graph
C. E. Yamashiro, lo G. Sala Espiell and 1.H. Farina, IPAKO S.A., Ensenada, Argentina

0.163 lag Bx 0.642 lag Bx

- 0.410

WHEN A MIXTUREof a gas and a liquid flows along a

- 0.241 lag Bx + 1.057 lag Bx

C7: lag By = 6.527

1.072 lag Bx

"'-'

C8: lag By = 3.301

0.197 lag Bx
we divided the "

'-'"

1. Dispersed.When nearly all the liquid is entrained as spray by the gas. 2.. Annular.The liquid forms a film around the inside wall of the pipe, and the gas flows at a high velocity as a central coreo 3. Bubble. Bubbles of gas move along at abolir the same velocity as the liquido 4. Stratified. The liquid flows along the bottom ofthe pipe and the gas flows above ayer a smooth gas-liquid interface. 5. Wave. Is similar to stratified except the interface is disturbed by waves moving in the direction of flow. 6. Slug. Waves are picked up periodically in the form of frothy slugs that llave at a much greater velocity than the average liquid velocity. . 7. Plug. Alternate plugs of liquid and gas llave along the plpe. Bakerl has correlated the seven flow patterns by means of two numbers, the "Baker parameters," Bx and By- Thus the Bx, By plane shows seven regions, one for each flow patterh. Although the borders of the regions are drawn as hiles, they are really broad transition zones.

To simplify the computer programming range of Bx into the following five zones:
1. 0.1

<

Bx

>

4.013

2. 3. 4. 5.

4.013 < Bx > 15.00 15.000 < Bx > 40.322 40.322 < Bx> 143.51 143.51 < Bx:> 10,000 have the following expressions:

The Baker parameters Bx = K (WjWc)

(VPL PClpL2/3) (ILLl/3IaL)

By = K2Wc/(7rD2/4) VPL Pc where: W = flowrate P = density IL = viscosity a = liquid surface tension D = internal diameter of pipe L = liquid G = gas K1 and K2 are constants that involve the conversion factors of units. Two seIs are included in our program: K are = 2.1039 data metric and K2 = 25,524.62

'-'"

'-

when the units of the-/

10,000 cQ

1,000

W = Kg/s IL = cp D=m P = Kg/m3 a = dyne/cm K1 = 513 and K2 = 2.16 when the units of the data are English

100 10 B. Fig. 1-Program's approximation of the Baker graph. 100 1,000 10,000

W = IL = D = P = a =

lb/h cp ft Ib/cu3 dyne/cm

Determination of two-phase flow by computar. In a recent paper Soliman3 provided a BASIC program to compute twophase pressure drop based on the excellent work of Kern.2 To determine the pattern of two-phase flow, Soliman reduced the Baker graph to four regions instead of the seven found experimentally. This makes the computer programmirig a great deal easier. However, we think that Ibis is an oversimplification. 46 Hydrocarbon Processing, December 1986

The computar programo Table 1 is a listing of the computer program written in BASIC. It has two options for the input of the data: metric or english units. Table 2 is an example of the program output illustrating one zone in the Baker graph. We think that the errors in using our approximations are of the same magnitude as the inherent errors of the original Baker graph.

~~

..,

---

TABLE1
,.10 11 12 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170 180 190 200 210 220 230 ;,40 250 260 270 280 290 300 310 320 330 340 350 360 370 371 380 381 :\90 400 410 420 430 440 450 460 470 480 490 500 510 520 530 540 550 560 570 580 590 600 610 620 630 640 650 660 670 680 690 700 710 720 730 740 750 760 770 780 790 800 810 820 830 840 850 860 870 880 890 90 O 910 920 930 940 950 960 970 980 990 1000 1010 1020 1030 1031 1040 REM PROGRAMA "BAKER" EN CINTA 01-F REM IPAKO INe, DE PROCESO -1984-' REM REM DEFINES THE PRINTER PRINTER IS 7,I',WIDTH(76) OPTION BASE 1 REM REM DATA INPUT REM INPUT "ENGLISH UNITS? <Y/N)",S$ INPUT "LIQUID FLUW RATE?" ,G1 INPUT "GAS FLOW RATE?" G2 INPUT "LIQUID DENSIrY?:', RI INPUT "GAS DENSITY?",R2 INPUT "LIQUID VISCOSITy?" ,VI INPUT "LIQUID SURFACE TENSION?", T INPUT "P IPE DIAMETER?", D REM REM DATA PRINTING REM U$=" METRIC UNITS" IF S$="Y" THEN U$=" ENGLISH UNITS" PRINT U$ ,LIN(2) IF S$="Y" THEN 290 PRINT "LiquId flo" rate(Kg/s): prnNT "Ga', flo" rote(Kg/s):""""", PRINT "LiquId denslty(Kg/M3):"""", PRINT "Gas densIty(Kg/M3):"""""" PRINT "PIpe dIaMeter(M):"""""", GOTa 340 "G1 "G2 " 'R 1 ";R2 " D

I~
I I

TABLE 2
METR lC LJNITS 7,488 1.176 499,888 29,637 ,102 ,11 5,07 LiquId flo" rate(Kg/s):"""" Gas Ho" rate<Kg/s): Liquid denslty(Kg/M3):"""", Gas densIty(Kg/M3):"""""" Pipo dinMeter'(M):"""""", Uq,Jid vlscoslty(cp):"""-",, LiquId surface tensIon(dynes/cM)
RESLJL TS

BLJBBLE X:" '" '" Y: , , , , , , " ENGUSH

" " , , "

'" " " " '" " , , , '" , , , , , , "

, , , , , ,

244,886469269 30180,2515474

UNITS 59428,57 .9333,333 31.204 1.85 4,01967 ,11 5,07

\.......-

'--

'-

'--'

" GI PRINl "LiquId nO" rote(lb/h):"""" PRIN1 "Gas flow rote(lb/h):""""", " G2 " R1 PRINT "LiquId densIty(lb/ft3)"""", " R2 PRINT "Gas denslty(lb/ft3):""""", " D PRINT "PIpe dIaMeter(lnch): , " VI PRINT "LIquId ulscoslty(cp):""""" " T, PRINT "LiquId surface tensIon(dyneslcM) PRINT "RESULTS" PRINT LIN(1) -REM REM DEFINES SYSTEM OF UNITS REM IF S$,""Y" THEN 450 T=T/1000 VI=V1/1000 KI=2,1039 K2=25524,617 GOTa 480 K 1"531 K2=2, 16 D=D*,08333 A=PI*D*D/4 X=K1*(G1/G2)*( (R1*R2)', 5/R1' ,667)* (V1 A, 333/1> Y=K2*GU(A*(R1*R2)',5) IF X)10000 THEN lOtO lF X< ,1 THEN 1010 IF Y>100000 THEN 1010 IF yaOO THEN 1010 L3=LGT< X) L4=LGT(y) CI=3,6<18-, 163*L:\ C2=4, 261-, 642*L3 C3=4,959-,410*L3 C4=4,477 C5=4,OI9-,241*L3 C6=I, 935+1 ,057*L3 C'7=6, 527-1, 072*L3 C8=3,301-,197*L3 IF X)15 THEN 730 IF L4<CI THEN 870 IF L4<C2 THEN 890 IF L4 )C3 THEN 910 IF X)4,OI:! THEN 710 GOTa 950 IF L4<C5 THEN 930 GOTa 950 IF L4< C2 THEN 870 IF X>143,51 THEN 810 IF L4)C4 THEN 910 IF X )40 322 THEN 790 IF L4)C5 THEN 950 GOTa 930 IF L4)C6 THEN 950 GOTa 930 IF L4)C7 THEN 970 IF L4 )C8 THEN 930 GOTa 990 REM REM RESUL TS REM PRINT "ESTRATIFIED" GOTa 1020 PRINT "WAVE" GOTa 1 020 PRINT "DISPERSE" GOTa 1020 PRINT "SLUG" GOTa 1020 PRINT "ANNULAR" GOTa 1 020 PR INT "BUBBLE" GOTa 1020 PRINT "PLUG" GOTa 1020 PRINT "DATA OUTSIDE BAKER GRAPH" PRINT "X: , "X PRINT "Y: " " ".., "Y PRINT LIN(4) END

Liquid flo" rate<lb/h):"""" Gas Ho" rate<lb/h):""""", Uquid density(lb/ft3)"""", Gas denslty<lb/ft3):""-"",,, Pipo dioMeter(lnch):"",-",,:, Liquid vi""oslty(cp):"",.."" UquId surface tension(dynes/cM) RESUL T5

UN (2)

BUBBLE X: , , , , , , , , " Y:"" " "',"',

, , , , , , , , , , , , , .-, , , , , """ " '" " "

244,839562172 30111,0615956

Although the program was originally written for a HP9835 and IBM PC, we rewrote it in a version simple enough to run in most personal computers. It is easily converted to a subroutine by changing the GOTO 1020's to RETURN's.

LITERATURE

CITED
1969, pp. 105-116. 1984, pp. 155-157.

Kern, R., Hydrocarbon Processing, Vol. 48, No. 10, October 3'Soliman, R., Hydrocarbon Processing, Vol. 63, No. 4, April,

: Baker,0., GilandGasJournal, July 26, 1954.

The authors
Carlos E. Yamashiro is a process engineer with IPAKO S.A., Ensenada, Argentina. He is responsible for optimization and troubleshooting of the high pressure PEplant. He holds a as degree in chemical engineering from the University of La Plata.

Luis G. Sala Espiell is a computar analyst with IPAKO S.A., Ensenada, Argentina. He is engaged in developing software for engineering needs, assisting the computar users and studying computar control problems. He holds a as degree in electronic engineering from the University of La Plata.

Italo H. Farina is a process engineering superintendent for IPAKO S.A., Ensenada, Argentina. He is responsible for supervising the process engineering of ethylene and PE plants and developing computar resources for engineering. He is a part-time professor of chemical engineering at the University of La Plata, Argentina. He holds a as degree from the University of La Plata and an MS degree fram the University of Minnesota, both in chemical engineering.

Hydrocarbon Processing, December 1986

47