Calculation of Pigging Effectiveness for Petroleum (product) Pipelines

Transcription

1 International Journal of Scientific and Research ublications, Volume 3, Issue 9, September 03 ISSN Calculation of igging Effectieness for etroleum (product) ipelines Mohd Sadaf * * Operations Deptt., IOCL (pipelines), Mughalsarai Abstract- igging is a process of pushing a deice (IG) equipped with metal wire brushes to clean the deposits on the inner walls of the pipeline. This study depicts a concise expression for calculating igging effectieness in petroleum product(diesel, etrol, Kerosene) pipelines especially where smaller batches of products are being pumped i.e. where the pipe length between two successie pump stations is not filled with a single product. As the pipeline has more than one products(or same products but from different refineries)filled in between successie pump station, the Bernoulli s equation can t be applied across the whole length. The study tries to propose a method for getting an approximate alue of pressure across indiidual batches. Comparing friction factor for getting igging effectieness sometimes gies a negatie alue. This study tries to find out the probable reasons& suggests a new expression for calculation of igging effectieness. Index Terms- Friction factor, etroleum product pipelines, igging, Turbulent flow, interface. I. INTRODUCTION For a fully deeloped flow in a pipeline, the pressure head drop ( h) across the pipe length (l) of pipe diameter (D) with a steady flow elocity () is gien by the equation: h= fl /gd.(i) (Darcy Weisbach eqn.) where f is the friction factor that is characteristic of the surface roughness of the pipe & the Reynolds no. Thus, f α h/, since olume flow rate Q = A. where A is cross sectional area of the pipe, this implies f α h/..(i) Thus we see that the relation of f with the flow rate is not linear rather it is inersely proportional to square of flow rate ( In oil industry product pipelines, different products (Diesel, etrol, Kerosene) are pushed through pumping sets one after the other in the same pipe. Thus an interface (mixture) is formed at the junction of two different products. The pressure fall due to frictional losses, is further boosted at intermediate pump stations to push the products further to the next pump station. Mughalsarai is one such station coming after atna pump station of the Barauni-Kanpur ipeline (BKL) of Indian Oil Corporation Ltd. The pipeline is designed to hae turbulent flow to reduce interface generation (mixing). igging is a process of pushing a deice (IG) equipped with metal wire brushes to clean the deposits on the inner walls of the pipeline. Thus itis supposed to cleanse the deposits (carbonates, sulphates, iron oxides, greasy sludge) on the inner walls of the pipes that should decrease surface roughness & hence friction factor f. Now, after efficient igging the alue of f should decrease. As the flowrate is constant so h = {(ressure drop across each batch length)/ρ.g}; where ρ is the density of fluid & g is acceleration due to graity. Currently the pigging efficiency (effectieness) has been calculated as: = X 00% (As per Operations manual of BKL, IOCL) Abr.: bp- before igging ; ap- after igging; Before igging After igging Date a) Suction ressure at upstream station.(kg/sq.cm) 6.5 3

2 International Journal of Scientific and Research ublications, Volume 3, Issue 9, September 03 ISSN b) Flow rate (KLS/hr) ( ) 5 58 c) Back. r. At downstream station.(kg/sq.cm) (c) minus (a) = Flow rate (KL/KCS) X = 5/43.5= 5.77 Y=58/5=5.86 KCS = Kg/Cm Here the efficiency(effectieness) much depends on CDR injection So pigging efficiency is not realistic. % igging efficiency (effectieness) = (X-Y/X)* 00 =.56% The initial correction seems to be: calculation of igging effectieness by comparing the friction factor before & after igging as being followed at some of the companies. Research Elaborations: Using notations as aboe, initially, pigging effectieness may be calculated as: = X 00% Taking the data from Table.below, we can now compare the friction factor before & after igging : assuming same density of fluid before & after igging, from the first column of table. i) = ( =43.5 x ; Q= 5 kl/hr thus f = = x = 44 x ; Q = 58 kl/hr thus f = = x (, are constants) Thus IG effectieness may be calculated as = x 00% = 4.3% Data for Allahabad-Kanpur section of BKL, as per igging reports for different quarters: Before IGging Date Back pressure Flow rate Suction pressure After IGging Back pressure Flow rate Suction pressure

3 International Journal of Scientific and Research ublications, Volume 3, Issue 9, September 03 3 ISSN Similarly: ii) iii) i) ) i) Table. Note the negatie alues of effectieness. Fig.: Moody s chart for flow through pipes Fig. on the ertical axis we hae the friction factor f & relatie roughness є/d, where є is the surface roughness of the inner surface of the pipe. On the horiontal axis we hae the Reynolds number (Re). The following characteristics are obsered from the Figure. For laminar flow, Re < 300, f = 64/Re, which is independent of relatie roughness. For ery large Reynolds numbers, f =function of (e/d) which is independent of the Reynolds number. For such flows, commonly termed wholly turbulent flow, along the wall pipe, exists the laminar sub layer so thin that the surface roughness completely dominates the character of the flow near the wall. If we calculate the Reynolds no. for BKL : kinematic iscosity of Diesel (-5 centi-stokes), Diameter of ipe =, aerage elocity V = flow rate/x-section area= (85 kilolitres/hr)/(3.4x( /4)=. m/sec (say) Re= (V X Dia.)/ kinematic Viscosity = to From Fig., clearly it lies in the turbulent region but to the left of the wholly turbulent flow regime. Most of the pipelines would lie in this region only. Limitations: Effectieness sometimes comes negatie. Actually, the region of the Moody s chart we are interested in has friction factor as a function of both relatie roughness & Reynolds number. If the Reynolds number reduces after igging (due to ariation of flow & kinematic iscosity of the particular batch of diesel), then een if the IG has reduced the surface roughness alue(є), there can still be a rise in friction factor after igging& thus negatie effectieness. In that case we should try to separate the effects of iscosity (& thus Reynolds number) from the effect of

4 International Journal of Scientific and Research ublications, Volume 3, Issue 9, September 03 4 ISSN surface roughness on friction factor to hae a better idea of IG effectieness because by igging we intend to reduce the surface roughness only. Corrections: We shall calculate i) Reynolds number ( Re=.D/ : for this we shall need the alue of flow rate & kinematic iscosity( of that particular batch of HSD (before & after igging) ii) Friction factor f = hgd/l.(darcy Weisbach Eqn.) Calculation of h: we are considering two batches of Diesel from different refineries (haing different densities & iscosities) As h is directly proportional to the length of pipe(darcy Weisbach Eqn.), pressure must drop linearly. As we always know the line fill(, we can interpolate the approximate alue of, from the known alues of &. Applying Bernoulli s equation across length L of the diesel batch-: g g where k g Disch. ress.at atna; similarly is minor loss that may be neglected. & at interface. h k...(iii) g eleation at atna; Rearrangin g : h...(i) g g g

5 International Journal of Scientific and Research ublications, Volume 3, Issue 9, September 03 5 ISSN Now, if 3 & 3 are pressure& eleation respectiely at Mughalsarai. As pressurehead drops linearly with length of the pipe( h l) & assuming the same for eleation, we hae : h g from eqn.(ii) we hae : 3 3 L X...() ; L L 5 gdh gπ f D h...(i) L 8Q L It is shown below by blue lines that how it is possible in our scenario(kinematic Viscosities arying from -5 centi stokes & Re arying from to 60000) that if we take into consideration, the ariation of Re due to difference in kinematic iscosity due to different batches or temperature ariation we may get an increased friction factor een after effectie igging. Thus the negatie alue of effectieness calculated from friction factor is explained. We can clearly see that: Een if relatie roughness is reduced by 0% (0.00 to ), f may still increase by 4%. Thus, a better idea would be to compare relatie roughness before & after igging. For that we require f & Re before & after igging.

6 International Journal of Scientific and Research ublications, Volume 3, Issue 9, September 03 6 ISSN Once we hae calculated the Reynolds No. (Re) & then friction factor f (eqn. i), we may either make use of Moody s chart, or use Colebrook equation to find relatie roughness before & after igging. Then the effectieness would be calculated as: = X 00% Results: Analysis of IGGING Effectieness calculation for ATNA-MUGHALSARAI section of BKL. March : Date atna Discharge Mughalsarai Suction Flow Rate Linefill at atna mouth(l) ress.(kcs) ress.(kcs) KLs/Hr BR3H-348, 66 KLs BR3H-0, 5079 KLs June : Date atna Discharge Mughalsarai Suction Flow Rate Linefill at atna mouth(l) ress.(kcs) ress.(kcs) KLs/Hr BR3H-7, 4703 KLs BR3H-83, 4473 KLs Mar 3: Date atna Discharge Mughalsarai Suction Flow Rate Linefill at atna mouth(l) ress.(kcs) ress.(kcs) KLs/Hr BR3H-376, 389 KLs BR3H-377/HR3H-55, 507 KLs Note: HR3H & BR3H are Diesel batches(bs-iii) from Barauni & Haldia Refinery respectiely. Obserations: Date of IGging Effectieness calculation Reynolds No. (Before igging /After igging) Friction factor(before igging/after igging) 6.3. / / /.06 00% 6.6. / / / % / / /.06 00% Effectieness based on Relatie Roughness є/d Conclusion: The alue of relatie roughness є/d just after IGGING is coming out approx. equal to ero that may imply that oer last 50 years due to repeated IGGING the pipeline inside surface has become smooth. Due to uniform deposition of sulphate/carbonate, iron particles & bacterial growth& sticky hydrocarbons (as per igging muck sample report# CCF/07976, DATED..0 by Italab t. Ltd., Mumbai) during next three months after igging an apparent roughness deelops, that is reflected by re- increased friction factor. As the Moody s chart was prepared by creating artificial roughness, this apparent roughness can be calculated by the chart & my proposed expression for igging effectieness is alid.

7 International Journal of Scientific and Research ublications, Volume 3, Issue 9, September 03 7 ISSN ACKNOWLEDGMENT I am thankful to my controlling officer who spared me for some time to work for this paper oer & aboe my routine job. The informations & data used here are true to the best of my knowledge. REFERENCES [] F.M. White, Fluid Mechanics, McGraw-Hill, 999. [] Ronald V. Giles, Fluid Mechanics & Hydraulics,3 rd Edition, 994, pp [3] aol Daucik, Temperature Dependance of the Viscosity of Hydrocarbon Fractions, Acta Chimica Sloaca,Vol., No.,008, pp [4] Operations Manual of Barauni Kanpur ipeline AUTHORS MOHD SADAF, B.TECH(MECHANICAL), IOCL, mohdsadaf@yahoo.com, mohammed@indianoil.in