Vous êtes sur la page 1sur 11

European International Journal of Science and Technology ISSN: 2304-9693 www.eijst.org.

uk


154
Anti-clogging Drip Irrigation Emitter Design Innovation





Ali Abdulqader Mohammed Ali
MSc student
School of Engineering
University of Glasgow
Glasgow
UK

Corresponding Authors Email: ali.al_zaidy@yahoo.com







Abstract
In this study, an innovative drip irrigation emitter was designed using a 3Dcomputer-aided design(CAD).
Therelation between the flow rate and the pressure applied wassimulated using the Computational Fluid
Dynamics (CFD) technique.The model of the emitter was then fabricatedusing rapid prototyping or additive
manufacturing techniques to validatethe flow behaviour experimentally. The results show a 23-25%
difference between the CFD technique and the experiment test data. Thisdifference is judged to be due to
several reasons, including assumptions regarding some flow parameters anddefects in prototype
manufacturing. This study describes the development of a new design of the drip irrigation emitter having
enhanced resistance to clogging when compared with previous drip emitter designs. This study also shows
how selecting the type of rapid prototyping technique is a significant factor in the case of emitterdesign.

Keywords: drip irrigation emitter; CAD; CFD; rapid prototyping

Article Classification: Case study










European International Journal of Science and Technology Vol. 2 No. 8 October 2013


155

1.0 Introduction

The issue of water shortage in agricultural applications has increased dramatically in many parts of the
world. In order to alleviate this problem, it is necessary to improve water-saving irrigation systems (Bralts
& Vincent, 1985). Various kinds of water-saving irrigation system have been developed, including sprinkler
irrigation, micro irrigation, under-membrane irrigation, low pressure irrigation and drip irrigation. Drip
irrigation is generally considered the most effective method with an utilisation rate reaching 90% (Wei,
2011). Drip irrigation can be defined as a slow and uniform application of low-pressure water to root plant
zone through a network of pipes, filters, pumps and emitters. The emitter is considered the most significant
device in this irrigation system(Wei, 2011; Zhang, et al, 2010; Hezarjaribi, et al, 2008; Zhengying, et al,
2012; Bassett et al, 1983). The drip irrigation emitter is a small mechanical device which is designed to
dissipate pressure and constantly discharge a small, uniform flow of water. The reliability of drip irrigation
is defined as the capability of achieving the required task under a specified operating pressure for a given
period of time (Jeznach, 1998).The performance of the emitter is determined by several factors including the
type of the drip emitter and its condition. The drip emitter can also suffer from clogging due to the escape of
some particles (e.g. sand, impurities) from filtering (Fan, et al, 2011; Wei, et al, 2009; Zhengying, et al,
2012). This problem affects the emitter performance significantly. Moreover, drip irrigation emitter
clogging will increase the maintenance cost (Oron, et al, 1991); therefore it is necessary to improve the
design and manufacture of anti-clogging drip irrigation emitters (Fan, et al, 2011; Qingsong, et al, 2006).
The design of an emitter is not trivial. The traditional method to design and improve an emitter takes 4 to 5
months and it can be quite expensive. In this work, a new approach is adopted (detailed in Figure1) to
design a new anticlogging emitter. In particular, three-dimensional software programmes have been used to
design and modify drip emitter designs. Additionally, numerical simulations are adopted by designers to
simulate and define optimum parameters of their work using CAD applications. Moreover, rapid prototyping
and manufacturing (RP&M) or Additive Manufacturing (AM) techniques have been utilised in order to
evaluate the designs rapidly (Celik et al, 2011), and also to subsequently test the developed product.
In this study a new anti-clogging drip emitter design iscreated. This study will determine the relationship
between the emitter discharge and the pressure applied. In addition,it will investigate the water behaviour,
adopting computational fluid dynamics (CFD) simulation and verification techniques. The final product
isfabricated by rapid prototyping additive manufacturing techniques.

2.0 Material and Method
2.1: Design Configuration
The new anti-clogging drip emitter design isdeveloped using SolidWorks software. The design concept is
inspired from the orifice eddy drip-arrows, which consist of six cones connected together by linear channels
and the long-path pre-depositing drippers (refer to Figure 2). The new emitter consists of nine cones; one in
the centre and eight surrounding it in a circle. These cones are connected with each other by pipes to create
the upper part of the emitter. This part is also connected by a pipe to a small cross-section labyrinth
zigzagged channel as shown in Figure 3. The cones and the zigzagged channel represent the heart of the drip
irrigation emitter, which is set inside covers. The upper cover connects with the laterals of the drip irrigation
system. The lower cover has eight small holes to aid in reducing the water pressure as well as ensuring the
discharge of water that occurs in a shower-like formation. When the water passes through the upper cover
coming from the laterals, some particles may escape from filtering. The water will flow in the central cone
then move from cone to cone until it reaches the zigzagged channel. In this part, the particles (sand) will
European International Journal of Science and Technology ISSN: 2304-9693 www.eijst.org.uk


156
settle in the cones due to gravity as well as centrifuge theory; hence the possibility of reaching these
particles in the zigzagged channel and causing clogging of the drip emitter would be significantly reduced.
When the water exits the zigzagged channel, it drops into the lower cover before exiting the emitter as a
shower through the eight holes.
In order to study the water flow behaviour and to determine the relationship between the pressures applied in
the laterals H (kpa) and the emitter discharge q (l/h), the following equation was adopted.
q= k .H
x
.. (1)
Where kdenotes the discharge coefficient and xis theflow exponent. With sufficientdata for H & q;

x =
mIogq
i
.IogH
i
-Iogq
i
.IogH
i
m(IogH
i
)
2
-(IogH
1
)2
(2)
log k =
logq

. (logE

)
2
(logq

. logE

). logE

m(logE

)
2
. (logE

)
2
(S)
Where i represent the iteration number of thedata, and m is the total number of the data
(Zhao, et al, 2009).

2.2: CFD Analyses
Computational fluid dynamics (CFD) can be defined as a numerical analyses to the solution of mathematical
models of fluid flow and heat transfer. The fundamental aim in the use of CFD is to investigate the physical
proceedings that happen in the fluids flow within and around the surrounding parts of the design (Lomax, et
al, 1999). This technique can significantlyreducethe effort and time during experimental tests. CFD has
become a fundamental approach in drip irrigation emitter design recently (Wang, et al, 2010). The irrigation
water usually contains some suspended solids (mostly sand particles); therefore a two-phase flow simulation
isused in this study, by following the steps shown in Figure4.This is to discover the relationship between the
emitter discharge and the pressure applied. This simulation also shows the pressure and velocity distribution
in the device. Moreover, it can reveal the location of the suspended solid particles, which can cause clogging
in the emitter channel.
2.2.1: Theoretical Analyses
The irrigation water in the emitter is assumed to be incompressible and viscous. The water head is
maintained at a steady value over the work period at 20
0
C in the drip irrigation systems. The path wall
surface roughness and the water gravity areconsidered,whilethe surface tension is neglected. In order to
conclude whether the water flow is turbulent or laminar, the Reynolds number can be used (Wei, et al,
2006).
R
c
=
pvR

(4)
Where p represent the density of fluid (Kg/m
3
), Vrepresent the average of the flow velocity, R denotes the
hydraulic radius of flow (m/s) and is the fluid viscosity coefficient (for water 10
-3
kg/ms).The relation
between the velocity (V) and the flow rate (q) is based on the channel cross section(s)
I =
q
s
(5)
R, the hydraulic diameter, can be determined as follows:
R =
wd
2(w+d)
(6)
Where w is the emitter channel width, d is channel depth. Generally, the discharge of most emitter range
from 2 to 8(l/h) under applied pressure of 0.4- 2 bar,so the R
e
rangebetween 100 to 1500 (Zhang, et al, 2007).
In this study both turbulence models and laminar flow are adopted to simulate the flow behaviour at
different operating pressures in the case of comprehensive analysis.
European International Journal of Science and Technology Vol. 2 No. 8 October 2013


157
2.2.1.1: Laminar flow model
Continuity equation;
u
i
x
]
= u (7)

Navier-Stokes equation;

p

x
i
(u

u
]
)=
p
x
i
+

x
]
[p (
u
i
x
]
+
u
i
x
i
) ] (8)

Where prepresent the density of fluid (Kg/m
3
) andu denote dynamic viscosity in the
i,j
th
direction.

2.2.1.2: Turbulence flow model

The RNG Eq k- turbulence model isadopted in this study as it is generally used for most engineering
calculations.
The governing equations include Eq. (7), Eq (8), k -equation, equation
The turbulence kinetic energy k and its rate of dissipation can obtained from following
equations

( pk)
kt
+
(pku
i
)
x
i
=

x
]
[keff
k
x
]
] + 0
K
+ p (9)
(ps)
kt
+
(psu
i
)
x
i
=

x
]
[
c]]
s
x
]
] +
C
1s
-
k
G
k
C
2s p
s
2
k
(10)
Whereeff = +t

pt = pCp
k
2
e
, Cp = u.u84S onJ o k = o e = 1.S9
C
1s
-
= C
1s
( 1-

0
)
1+[
3
, C
1s
= 4.377
C
2s
= 1.68, =

2E
]
E
]
k
s
,
0
= 4.377 , = 0.012
Where G
k
represent the generation Kinetic energy due to the mean velocity gradients;
t
is the turbulent
viscosity; and
eff
denote the effective viscosity ; C, C
1s, C
2s are model constants; ,
0
are strain rate; the
quantities
k
and s are the inverse effective Prandtl number for k &s respectively.

2.2.2: Two phase analyses

The twophase analyses areachieved. The first phase is water with 988.2 kg /m3 in density and the viscosity
is 0.001003 kg/m.s. The second phase is suspended particles (usually sand with 2500 kg/m3 density).
Aluminium isselected to represent particles suspended in the simulation. The particle size in irrigation water
typically ranges from 0.01 to 0.1 mm diameter and the concentration between zero to one gram per litre.
The sand grain can be fitted using Rosin-Rammler distribution (Zhengying, et al, 2012). Based on sediment
dynamic principles, the movement of sand isrelated to the fluid vortices in the liquidsolid two-phase flow.
The capability of solid diffusion motion in the water can be measured by using Stokes number (St).
St =
:
p
:
k
(11)
European International Journal of Science and Technology ISSN: 2304-9693 www.eijst.org.uk


158
Where x
p
is defined as time of particle relaxation and x
p
=p
p
J
p
2
/(18),x
k
defined as Kolmogorov
characteristic time x
k
= pe. According to Wei et al, cited in (Zhengying, 2012), the sediment sands with
small (0.15-0.59) St No. or middle (1.33-2.36) can be easy entrapped into the water vortices.

2.3: Simulation results

Based on the above mathematical and geometric models, the SolidWorks package isutilised to solve the two-
phase solid-liquid model defined above. The results of the simulation show that the suspended particles will
settle in the cones (refer to Figure 5) due to the gravity and centrifugal forces and the probability of moving
down into the zigzagged channel causing emitter clogging issignificantly reduced. The simulation also
shows the pressure distribution in the emitter. Figure 6 shows the water pressure declining significantly in
the zigzagged channel, which indicates that this part of the system indeed plays a significant role in the
operation of the device. Figure 7also shows that other parts of the system (e.g. cones) have a much smaller
effect on the overall reduction of water pressure. The solutions were run several times at different pressures
(50,75, 100, 125 and 150) kpa to obtain the different velocities as shown in Table 1.
3.0 Rapid Prototyping
The traditional technique used to fabricate the emitter depended on mould manufacture and machining the
emitter sample to do experiments. This was expensive and took a long time.Therefore,rapid prototyping and
manufacturing (RP&M) technology was adopted. This technology has been available since the1980s. The
essential idea of RP equipment is based on the notion of layered or additive manufacturing. A complex 3D
CAD model is firstly sliced, and then converted into 2D planes or layers. RP equipment builds the part layer
by layer and finally fabricates the complete item (Ryder, et al, 2002).Emitters are commonly manufactured
from high or medium density polyethylenematerial. In order to test the emitter experimentally, a model
isfabricated using Fused Deposition Modelling (FDM) technology (Figure 8).The CAD model
isthentransferred into an STL file. Subsequently,the data isapplied in the HP designjet 3D machine. In this
equipment, a filament of ABS plastic, which has strength up to 80% of the origin material,is inserted into an
extrusion head (liquefier head) and then heated to just above its melting point. This solid is then ejected
through the tip of the liquefier head and put onto the parts as a run, a solo strand of material; these runs are
extruded side by side to generate the layer of information. As the most recent layer is completed, the build
table indexes downward a single layer and the process repeat until the layer of information is finished.
Items with downward facing surfaces need supports, for most processes these supports are created
automatically (Ryder, et al, 2002). These support materials seem to cause clogging in the cones as well as in
the zigzagged channel (refer to Figure 9), therefore.The design ismodified. Moreover, each of the cones
part and zigzagged channel iscut into two sections. The two covers are modified as well. The emitter isre-
fabricated, with six parts being manufactured for test purposes. The supported material in this case is
removed. After attaching the two sections of cone parts and the zigzagged channel using adhesive glue, the
device isassembled and ready for the experiment.

4.0 Experimental test
An experimental test was undertaken in a laboratory environment toverify and compare the CFD analyses
results with the test data. Sand (0.5 gram per litter) was put in a tank of tap water, which was then connected
with a pump, regulator, gauge pressure and the emitter respectively. The discharge water was measured in a
beaker. The first 5 minutes were ignored to ensure that the flow reached the stability state. The test was run
European International Journal of Science and Technology Vol. 2 No. 8 October 2013


159
under five different pressures. It was observed that the lowest Standard deviation (0.06) occurred at a
pressure of150 kp(see Table 2). The results show differences between the CFD analyses and experimental
test (see Figure 10).

5.0 Discussion and Conclusion
Comparing the experimental discharge values with the CFD results, there are some differences but are
within a generally acceptable range (10 % as maximum). The experimental data were greater (23-25) % than
the CFD data. These differences are likelyto be due to various reasons. These include the quality of the mesh
model, the assumption made on some of the hydraulic parameters values in the CFD analyses, which
requires more investigation. In addition,there could be some deficiencies in the test equipment, which would
affect the accuracy of the result, and some manufacturing defects, such as the deformation in size and
surface roughness, which was caused by theremovingofthe RP support material inside the cones, pipes and
zigzagged channel.Although some differences between the CFD simulations data and experimental results
are observed, the new design shows good performance in terms of the efficiency of anti-clogging and it
provides a reasonable discharge value under different pressure settings. The new design also provides a
better anti-clogging achievement because the cones shape combines to create fast vortices, which leads to
the deposition of the sand particles in the bottom and corners, due to thegravity and centrifugal force. This
suggests that a longer service life service can be anticipated, due to a decreasein the probability of clogging.
As discussed earlier, the emitter is the most significant part of the drip irrigation system. Moderntechnieqes
include CAD, Rapid prototyping RP and CFD simulation, has important advantages intime saving and
reduction of cost consumption significantly. CAD isused to create a new design of emitter within a short
period of 4 to 5 months. The modification of the emitter design isachieved quickly, and contributes to the
many benefits of CAD. The two-phase CFD analysis was done to investigate the flow behaviour of the
irrigation water, which is assumed to contain a small amount of suspended particles. The results show that
the particles (sand) will deposit in the cones to avoid clogging the small section of the zigzagged channel. In
addition to that, CFD simulations clarify the pressure distribution in the emitter, which shows that the
majority of pressure drop occurs in the zigzagged channel. The relation between the flow rate and pressure
applied has been observed by this simulation. In order to verify this relation and confirm that the clogging
chance is minimised, the device is manufactured by a RP machine. Inthe first attempt, the cones and
channel, in addition to the pipes, were clogged by supported materials during the fabrication
process.Therefore,the emitter design ismodified and the RP manufacture was repeated. Finally, the second
version of emitter was ready to be tested after the removal of the support material. An experimental test was
achieved in the laboratory. The results show largedifferences due to the changes in size and shape of the
emitters parts. Therefore, re-manufacturing of the device by other RP techniques such as 3DSystems
stereolithography (SL) is recommended for future work. In order to obtain a better clarification of the flow
behaviour and the relation between the flow rate and the pressure the quality of the mesh, a significant effect
on the results is observed and the experiment can be improved. Further investigationis needed to find out the
actual values of the hydraulic parameters, where assumptions are made in this study due to time limitations.
Moreover, a specific emitter test machine can be utilised in the experiment.
Acknowledgment
The author wishes to thanksDr.Graham Green for his supervision and help to improve this research.


European International Journal of Science and Technology ISSN: 2304-9693 www.eijst.org.uk


160
References
Bassett, D.L.,Fangmerier, D.D. &Stelkoff, T. (1983).,Hydraulics of surface irrigation,In JensenM.E.
(Ed.),Design and operation of Farm Irrigation System,Am.Soc.Agric.Eng.,St Joseph,MI,pp. 47-98.
Bralts,V.& Vincent, F. (1985),Micro computer Aided Design of Drip /Trickle Irrigation Submain
Units,in:Drip /Trickle Irrigation in Action,Proceedings of theThirdInternatiional
Drip/TrickleIrrigationCongress,pp.452-458,ISBN091-6150-73-9 Silver Spring,MD,USA.November,1985.
Celik, H., Karayel, D., Caglayan,N., Rennie, A., & Akinci, L.( 2011),Rapid prototyping and flow
simulation applications in design of agricultural irrigation equipment: Case study for a sample in-line drip
emitter, Virtual and Physical Prototyping, Vol.6,No.10, pp.47-56.
Fan, J., Chunman, H., Jianghua, X., & Yijun, W. (2011),Design and optimization of drip Irrigation emitters
using directly of sewage, The 6th International Conference on Computer Science & Education
(ICCSE2011) ,Super Stare Virgo, Singapore.pp.483-488.
Hezarjaribi, A.,Dehdhani, A., Helghi, M. &Kiani, A.(2008),Hydraulic performances of Various Trickle
Emitters, Journal of Agronomy , Vol.7,No.3,pp.265-271.
Jeznach, J., (1998) Reliability of drip irrigation systems under different operation conditions in Poland,
Agricultural Water Management, Vol.35No.3,pp. 261226.
Lomax, H., Pulliam, T. H. &Zingg, D. W. (1999), Fundamentals of computational fluid dynamics. NASA
Ames Research Centre and University of Toronto Institute for Aerospace Studies.
Oron, G., Demalach, J., Hoffman, Z. &Cibotaru, R.( 1991).,Subsurface Microirrigation with Effluent. J.
Irrig . Drain. Eng. ASCE ,Vol.117, No.1,pp.2536.
Qingsong, W., Gang, L., Jie , L., Yusheng, S., Wenchu ,D. &Shuhuai, H. (2006).,Study on hydraulic
performance of drip emitter by computational fluid dynamics, Agricultural water
management,Vol.84,No.2,pp.130136.
Ryder, G., Ion, B., Green, G., Harrison,D. & Wood, B.,(2002) Rapid design and manufacture tools in
architecture, Automation in Construction, Vol.11,No.3, pp.279-290.
Wang, D., Zhang, X., Li, G., &Ma,Y.(2010) Influencing Factors of CFD Simulation Precision for the
Emitter Channel,International Conference on Mechanical and Electrical Technology (ICMET).10-12
September,2010, Singapore,pp.641-645.
Wei, Q.,Shi, Y., Dong, W., Lu, G.,& Huang, S.(2006),Study on hydraulic of drip emitters by computational
fluid dynamics, Agricultural water management,Vol.84,No.1,pp.130-136.
Wei, Z., Cao, M., Yang, Y., &Lu, B.( 2009).,Two-phase flow analyses and experimental investigation of
micro-PIV for emitter micro-channels,Seventh International Conference on CFD in the Minerals and
Process Industries, Melbourne, Australia,9-11 December 2009.pp.1-5.
Zhengying Wei (2011). Application of RP and Manufacturing to Water-Saving Emitters, Advanced
Applications of Rapid Prototyping Technology in Modern Engineering, Dr. M. Hoque (Ed.), ISBN: 978-
953-307-698-0, InTech, DOI: 10.5772/22401. Available from: http://www.intechopen.com/books/advanced-
applications-of-rapid-prototyping-technology-in-modern-engineering/application-of-rp-and-manufacturing-
to-water-saving-emitters

Zhao, W., Zhang, J., Tang, Y., Wie, Z.& Lu,B.(2009),Research on Transitional flow characteristics of
Labyrinth-channel emitter, Computer and Computing Technology in Agriculture,Vol.2,No.2,pp.81-89.
Zhengying, W., Meng, C., Xia, L., Yiping, T., &Bingheng, L.(2012),Flow Behaviour analysis and
experimental investigation for emitter micro-channels, Chinese Journal of mechanical
engineering,Vol.25,No.8, pp.1-9.

European International Journal of Scien


Appendix
Figures
Figure1: The modern approach to design
Figure 2: pre-depositing emitter and edd
Figure3: The new ant-clogging drip irrig
ence and Technology Vol. 2 No. 8
gn drip irrigation emitter from (Celik, et al, 2011)
ddy drip-arrow (Qingsong, et al, 2006)
rigation emitter
October 2013

1)


European International Journal of Scien


Figure 4: Flow rate prediction using of C





Figure5: Distribution of sand particles in
Input
Inp
ence and Technology ISSN: 2304-9693
f CFD after (Wei, 2011)

in the emitter

Output
Sands
Input
Output
www.eijst.org.uk

European International Journal of Scien


Figure 6: Pressure distribution in the emi
Figure7: Velocity distribution in the emi
Figure8: (FDM) technology

Figure9: The emitter fabricated by the FD
Inpu
ence and Technology Vol. 2 No. 8
mitter

mitter
FDM Rapid Prototyping technique
Input
Output
October 2013


European International Journal of Scien


Figure 10: The curves pressure-discharge


Tables:
Table 1: Simulation results
Pressure(kpa)
50
75
100
125
150



Table2: Experiment test results
Pressure( kpa) 50 75
Flow rate(l/H) 3.15 4.2
3.2 4.0
3.4 4.15
3.3 3.9
3.4 4
Average 3.29 4.05
Standard
Deviation
0.11 0.1

ence and Technology ISSN: 2304-9693
rge relation
Velocity (v) Flow rate (q)
3.85 2.7
4.67 3.3
5.2 3.7
6.12 4.3
6.8 4.8
100 125 150
4.9 5.2 6
4.75 5.3 5.9
4.8 5.35 5.85
4.8 5.35 5.85
4.9 5.4 5.9
4.83 5.32 5.9
0.07 0.08 0.06
www.eijst.org.uk

Vous aimerez peut-être aussi