MODELING FLOW IN LOCK MANIFOLDS
MODELING FLOW IN LOCK MANIFOLDS Richard L. Stockstill, Jane M. Vaughan, and E. Allen Hammack U.S. Army Engineer Research and Development Center Coastal and Hydraulics Laboratory
Evaluation of Lock Manifolds • Hydraulic design of navigation locks depends on knowledge of the performance of particular components. • Performance measures are often times quantified with coefficients such as discharge coefficient or energy loss coefficient. • Lock components are such things as manifolds, gates, and valves. • Manifolds vary in function from intakes to filling and emptying manifolds to outlets.
Filling and Emptying Manifolds In-Chamber Longitudinal Culvert System
Side-Port System
Common Manifolds
Culvert Locations for the Side-Port and ILCS Filling and Emptying Systems
Computational Model of Webbers Fall Lock– Arkansas River
Computational Model of Cannelton Lock – Ohio River
Physical and Numerical Models
Calculations The solution method uses a series of energy equations expressed for the fluid path through each port. 2 2 n1 Qp ( n )2 Q Q (i) (i) c c H K f (i) Kc ( i ) Kp(i) 2 2 2g Ac ( i ) 2g Ac ( i ) 2 g Ap ( n ) 2 1 1 n
where H = head loss across the manifold, K f (i) = friction loss coefficient for culvert upstream of port i, K c (i) = loss coefficient for flow across port i , K p (i) = port i loss coefficient, Qc (i) = discharge in culvert upstream from port i, and Q p (i)= discharge through port i
K i = fi
Li Di
Loss Coefficients Friction loss coefficient
Port loss coefficient
K f ( i ) = f( i )
L( i ) D( i )
H L( i ) = K1 ( i )
Coefficient for loss across the port
Vc( i ) 2 2g
= K 3( i )
V p( i ) 2
H L( i ) = K 2( i )
2g
Vc( i ) 2 2g
ILCS Single Port Laboratory Tests Port A 250
Port B
K=H'/(V1^2/2g)
200
150
100
Ports w/o Extensions 50
Ports w/ Extensions Power (Ports w/ Extensions) Power (Ports w/o Extensions)
0 0.0
0.1
0.2
0.3
0.4
0.5 Q3/Q1
0.6
0.7
0.8
0.9
1.0
ILCS Single Port Computational Model
Computational Model Simulated what was previously Tested in a Physical Model
Single Port Computational Model
Computational Mesh
Velocity Contours
ILCS Port
Single Port Loss Coefficients: Numerical and Laboratory Model Results Turbulence Models: • k-e and k-w results compared well with lab data • k-e realizable gave loss coefficients that were an order of magnitude higher than lab data
Physical Model Manifold Data Lock 25 Wall Manifold (Left Wall)
1/8 in. diameter pitot tube (Dwyer Instruments model 166)
Lock 25 Single Port
Loss Across the Port Culvert Velocity Head
Loss Through Port Culvert Velocity Head
Loss Through Port Port Velocity Head
Results
Conclusions • Loss coefficients can be determined using detailed 3D computational models. • Calculation of flow in lock manifolds using energy equations relies on accurate loss coefficient data. • Problem is very nonlinear because the coefficients are dependent on an unknown variable, velocity (which is nonlinear). • Understanding of effective use of 3D computational models is on-going
Questions?
Evaluation of Lock Manifolds • Hydraulic design of navigation locks depends on knowledge of the performance of particular components. • Performance measures are often times quantified with coefficients such as discharge coefficient or energy loss coefficient. • Lock components are such things as manifolds, gates, and valves. • Manifolds vary in function from intakes to filling and emptying manifolds to outlets.
Filling and Emptying Manifolds In-Chamber Longitudinal Culvert System
Side-Port System
Common Manifolds
Culvert Locations for the Side-Port and ILCS Filling and Emptying Systems
Computational Model of Webbers Fall Lock– Arkansas River
Computational Model of Cannelton Lock – Ohio River
Physical and Numerical Models
Calculations The solution method uses a series of energy equations expressed for the fluid path through each port. 2 2 n1 Qp ( n )2 Q Q (i) (i) c c H K f (i) Kc ( i ) Kp(i) 2 2 2g Ac ( i ) 2g Ac ( i ) 2 g Ap ( n ) 2 1 1 n
where H = head loss across the manifold, K f (i) = friction loss coefficient for culvert upstream of port i, K c (i) = loss coefficient for flow across port i , K p (i) = port i loss coefficient, Qc (i) = discharge in culvert upstream from port i, and Q p (i)= discharge through port i
K i = fi
Li Di
Loss Coefficients Friction loss coefficient
Port loss coefficient
K f ( i ) = f( i )
L( i ) D( i )
H L( i ) = K1 ( i )
Coefficient for loss across the port
Vc( i ) 2 2g
= K 3( i )
V p( i ) 2
H L( i ) = K 2( i )
2g
Vc( i ) 2 2g
ILCS Single Port Laboratory Tests Port A 250
Port B
K=H'/(V1^2/2g)
200
150
100
Ports w/o Extensions 50
Ports w/ Extensions Power (Ports w/ Extensions) Power (Ports w/o Extensions)
0 0.0
0.1
0.2
0.3
0.4
0.5 Q3/Q1
0.6
0.7
0.8
0.9
1.0
ILCS Single Port Computational Model
Computational Model Simulated what was previously Tested in a Physical Model
Single Port Computational Model
Computational Mesh
Velocity Contours
ILCS Port
Single Port Loss Coefficients: Numerical and Laboratory Model Results Turbulence Models: • k-e and k-w results compared well with lab data • k-e realizable gave loss coefficients that were an order of magnitude higher than lab data
Physical Model Manifold Data Lock 25 Wall Manifold (Left Wall)
1/8 in. diameter pitot tube (Dwyer Instruments model 166)
Lock 25 Single Port
Loss Across the Port Culvert Velocity Head
Loss Through Port Culvert Velocity Head
Loss Through Port Port Velocity Head
Results
Conclusions • Loss coefficients can be determined using detailed 3D computational models. • Calculation of flow in lock manifolds using energy equations relies on accurate loss coefficient data. • Problem is very nonlinear because the coefficients are dependent on an unknown variable, velocity (which is nonlinear). • Understanding of effective use of 3D computational models is on-going
Questions?
