Validating the Design of an Intake Structure

By Narasimhan Raghavan and Mundath Valiyathodiyil Ramachandran

Engineers designed an intake structure for a new 100-mw powerhouse in India, then validated the design by performing a physical hydraulic model study. The model study proved to be an effective way to confirm the adequacy of the initial design.

Intake structures are used to withdraw water from rivers or reservoirs. These intakes can create swirling flows, which lead to formation of air-entraining vortices and increase head loss. This situation can be avoided through proper design of the intake, including a sufficient entry size and submergence depth.

Generally, use of empirical calculations or numerical model studies alone cannot provide an adequate submergence depth to avoid vortices. Physical hydraulic model studies are needed to provide additional information.

Kerala State Electricity Board is building a 100-mw extension project at the existing Kuttiyadi power station in India. Larsen & Toubro Limited and Bharat Heavy Electricals Ltd. (BHEL) won the engineering, procurement, and construction (EPC) contract for the project. To design the power intake, Larsen & Toubro performed empirical calculations and then a hydraulic model study. The result of this work was an intake design that provided adequate submergence to operate without inducing air-entraining vortices.

Background on Kuttiyadi

Kakkayam Dam, on Kuttiyadi River in the state of Kerala, is the site of two existing hydroelectric facilities, 50-mw Stage I and 75-mw Stage II. For this Stage III project, Kerala State Electricity Board determined to build a third powerhouse with a capacity of 100 mw (two 50-mw vertical shaft Pelton turbine-generating units supplied by BHEL), using the existing Kakkayam Reservoir. This reservoir has live storage of 34 million cubic meters.

To feed this new powerhouse, water will be conveyed through a horseshoe-shaped headrace tunnel, a low-pressure penstock, and a surface penstock. The project will have a gross head of 666 meters. It is expected to produce about 240,500 megawatt-hours each year.

The proposed intake design

The proposed intake structure for the new powerhouse, based on a traditional design concept, consists of a trapezoidal approach channel, a trashrack, an elliptical bellmouth entry converging into a rectangular approach tunnel, a circular intake shaft with a service gate, and an emergency gate for regulating entry of water into the conductor system. (See Figure 1 on page 32.)

The plan view (top) and elevation (bottom) of the proposed intake structure for the new 100-mw powerhouse at Kakkayam Dam show the trapezoidal approach channel, trashrack, elliptical bellmouth entry converging into a rectangular approach tunnel, and circular intake shaft.
Click here to enlarge image

Water is withdrawn from Kakkayam Reservoir between the full reservoir level of Elevation 757 meters and the minimum drawdown level (the lowest possible water level under standard operating conditions) of Elevation 743 meters. The new intake is located 20 meters upstream of the dam that impounds the reservoir. The proposed intake also features a trashrack and stoplog gates. The proposed piers for the trashrack are semi-circular in shape, placed 0.8 he (height of opening) from the centerline of the intake, as recommended in the Indian Standard Code.

The intake is of the bottom-suppressed type, with the invert of the intake below the general approach bed level of the reservoir for better submergence.1 The sill level is to be placed at Elevation 736.5 meters. The top curve of the entrance is bellmouth shaped, with an elliptical profile, allowing streamlined entry of water into the approach tunnel. The bellmouth profile is designed as per the guidelines of the United States Department of the Interior’s Bureau of Reclamation2 because the Indian Standard Code does not provide specifications for a bottom-suppressed intake. The bellmouth entrance follows an elliptical profile in plan. An intake service gate will enable inspection and maintenance of the system.

Studies to validate the design

The designers performed standard empirical calculations to validate the initial intake design, then used results of hydraulic model studies to choose the appropriate submergence depth.

Empirical calculations of submergence and vortex phenomena

In intake structures for hydro facilities, phenomena of free vortex formation are common. Figure 2 on page 34 shows the various types of vortices to which an intake can be subjected.3 The deleterious effects of such vortex phenomena are well known. These vortices introduce air bubbles into the stream and result in loss of head at the intake. They affect the function of the water pumps operating in the intake and/or adversely affect the streamlined full-design flow into the water conductor system. Any entrapped air moving along the boundary of the tunnel may cause negative pressure and damage the concrete lining. Also, for proper operation of the turbines or pumps, it is imperative that the water be free of entrapped air.

Hydro plant intakes can be subjected to various types of vortices, which can increase head loss and cause damage to the equipment as a result of entrained air.
Click here to enlarge image

To avoid vortex formation and prevent air entrainment, it is crucial that the entry location have adequate water seal or submergence, even at the lowest water level possible in the system. Certain equations suggest empirical guidelines for submergence criteria. The Indian Standard Code makes recommendations for the geometry of the intake using a dimensionless factor, S/d:4

Equation 1:
S/d = 2.0 F + 0.5
  • S is the depth of the intake centerline from the minimum drawdown level;
  • d is diameter of the conduit; and
  • F is the Froude number (the ratio of inertial forces to gravitational forces) of flow in the conduit.
  • Other empirical relations are available for defining the limiting value of S/d:3
  • Gordon (1970), S/d = 2.3 F + 0.5
  • Amphlett (1976), S/d = 3.3 F — 0.5
  • Knauss (1987), S/d = 2.3 F + 1.5

    Using the various empirical relations, the designers calculated the required submergence depth for this project, using an estimated Froude number for flow in the conduit of 0.466. Table 1 shows the results.

    Click here to enlarge image

    Considering space limitations and past experience, the actual S/d designers adopted for the proposed structure was 1.61. Figure 3 on page 34 shows a plot of the various empirical relations. From these results, it is apparent that the empirical relation from Knauss gives a conservative value of S/d as compared to the stipulations in the Indian Standard Code. The submergence provided in the code compares well with the relations of Gordon and Amphlett.

    A plot of the relationship of dimensionless submergence number S/d to the Froude number shows how the Gordon and Amphlett calculations correlate well with the value provided by the Indian Standard Code.
    Click here to enlarge image

    However, empirical equations alone cannot ensure freedom from vortices because of the complex factors involved in vortex formation. These factors include approach geometry, flow conditions, velocity at the intake, geometric features of the trashrack structure, relative submergence depth, and withdrawal Froude number. A physical hydraulic model study can take into account all these factors. Hence, the designers decided a hydraulic model study should be developed.

    Hydraulic model studies

    Scale modeling to study vortex phenomenon is difficult because there is no single similitude law that accurately correlates model vortices to those observed in the prototype. However, vortex phenomena can be qualitatively assessed with some reliability.5 A Froude-based model does not accurately reproduce hydraulic phenomena such as vortices, which are also affected by Reynolds and Weber numbers. Consequently, vortex evaluation from a model is only qualitative. However, based on 16 cases where model-prototype data was available, it was seen that, in general, final intake designs developed from Froude model tests to be vortex free were vortex free in the prototype as well. Those that had weak vortices in the model had weak vortices in the prototype.6 There was no reported case in which a negligible model vortex corresponded to a strong prototype vortex that produced operational problems.

    The model study of the proposed intake structure was carried out with the following scope:

    • Assess adequacy of the bellmouth opening for passing the discharge;
    • Check for vortex formation; and
    • Assess the adequacy of the air-vent at the gate.

    The hydraulic model test was carried out at Central Water and Power Research Station (CWPRS) in India.7 The model was constructed with a geometric scale of 1:12, based on Froudian similitude. This scale was conducive to a small model size, suiting the available space, but at the same time large enough to avoid viscous effects. Parts of the model were constructed using transparent Perspex material to facilitate flow observation at the power intake, gate structure, and tunnel portion. The sections modeled were those at 2.18 meters on the downstream side of the gate and at 7.5 meters on the upstream side of the gate, which correspond to 26 meters on the downstream and 90 meters on the upstream in the prototype.

    The model was tested in an open loop mode. Two mono-block pumps recirculated the water in the model. A common header was used to connect suction pipes of the two pumps to the Perspex tunnel. Flow rate was measured using rotameters and adjusted to Froudian flow condition. To adjust the water level in the model, inlet and outlet drain connections were used. Water level was measured with a level gage.

    The model tests were conducted at a discharge of 42.86 liters per second, which corresponds to the maximum discharge of 21.38 cubic meters per second (cms) in the prototype.

    At minimum drawdown level, no air-entraining vortex and no submerged vortex from bottom or side walls was observed in the intake structure. There was also no vortex formation in the gate slots. On lowering the water level below minimum drawdown level, no vortex/ swirl formation took place until the water reached Elevation 742 meters. Thereafter, occasional air-entrainment and Type 2 and Type 3 vortex formations were observed. Hence, the model flow was deemed to be satisfactory with respect to vortex and swirl formation at minimum drawdown level.

    The observed flow was stable, with the flow passage running full, and was smooth at all locations, thus establishing the adequacy of the intake size. Water level in the air vent pipe was stable under varying flow. No air demand from the tunnel was noticed. Based on these findings, an air vent of 450-millimeter-diameter was determined to be adequate when passing the maximum discharge of 21.38 cms. Hence, the shape and size of the proposed intake structure were deemed to be adequate for efficient and safe entry of water into the water conductor system.

    CFD analysis for hydraulic studies

    Computational fluid dynamics (CFD) analysis is a tool that can be used to provide insight into flow phenomena and hydraulic designs of an intake structure. CFD analysis involves the numerical solution of the governing equations of fluid flow, the Navier Stokes equations. It can give a three-dimensional representation of the fluid flow domain.

    The vortex formation phenomena can be simulated in a model and flow behavior can be predicted. As mentioned earlier, vortex formation is a function of a number of parameters that need to be modeled precisely. Although CFD presents possibilities for accurate hydraulic analyses, it still needs to be explored in depth to establish full credibility with designers to serve as a tool for replacing physical model studies.

    In this case, CFD analysis was performed using three-dimensional tetrahedral elements (in excess of 1.6 million) using FLUENT software. However, a large number of iterations were required to be performed, which was not compatible with the time available. Hence, a solution using empirical formulae and the conventional design method was adopted, with the physical model proving the design.


    Physical hydraulic model studies confirmed that the proposed power intake was capable of sustaining efficient and safe operation with a discharge of 21.38 cms over the full range of water levels. Hence, the provided submergence depth was adequate for vortex-free entry of water.

    As the curve in Figure 3 shows, the adopted value of the dimensionless parameter S/d lies above the curve as per Indian Standard Code recommendations. The empirical relations given by Gordon and Amphlett compare well with the provided submergence, whereas the Knauss relation gives a higher value. The selection of the S/d ratio was verified and found to be acceptable through model studies.

    Installation of the new intake will be complete by July 2007. The project is to be commissioned in mid-2008. s

    . Raghavan, vice president and head of hydro projects with Larsen & Toubro Limited, is responsible for construction of all hydro projects for the company. M.V. Ramachandran is Larsen & Toubro’s project manager for the Kuttiyadi project.

    Messrs. Raghavan and Ramachandran may be reached at Larsen & Toubro Limited, ECC Division, Mount-Poonamallee Road, Manapakkam, Chennai 600 089 India; (91) 44-22528801 (Raghavan) or (91) 496-2698044 (Ramachandran); E-mail: or mvrchandran@


    1”Manual on the Planning and Design of Hydraulic Tunnels,” Publication No. 178, Central Board of Irrigation and Power, Government of India, 1984.

    2”Design Standards No. 6 – Turbines and Pumps,” U.S. Department of the Interior’s Bureau of Reclamation, 1971.

    3Sruvastava, Y.N, A.K. Agrawal, S.L. Patil, and P.B. Deolalikar, “Studies on Performance of Horizontal Power Intakes,” ISH Journal of Hydraulic Engineering, Volume 10, No 1, March 2004, pages 65-76.

    4Hydro Power Intakes – Criteria for Hydraulic Design, IS 9761, Bureau of Indian Standards, 1995.

    5Vermeyen, Tracy B., “Glen Canyon Dam Multi Level Intake Structure Hydraulic Model Study,” Water Resources Research Laboratory, Colorado, United States, July 1999.

    6Hecker, G.E, “Model Prototype Comparison of Free Surface Vortices,” Journal of Hydraulic Engineering, Volume 107, No. HY10, October 1981.

    7“Report on Hydraulic Model Study of Kuttiyadi Hydroelectric Project,” Central Water and Power Research Station, Pune, India, May 2005.

    This article has been evaluated and edited in accordance with reviews conducted by two or more professionals who have relevant expertise. These peer reviewers judge manuscripts for technical accuracy, usefulness, and overall importance within the hydroelectric industry.

    Previous articleLeaders of the pack: Second-generation biofuel pioneers
    Next articleWhat’s Happening in the Biodiesel from Algae Sector?

    No posts to display