New Tool Aids in Turbine Selection Process

By James L. Gordon and John P. Christensen

An online tool, called HydroHelp 1, aids hydro project developers in the work of choosing the best combination of turbine and powerhouse design for a particular site. More than 100 developers have used this tool during preliminary project design.

Optimum conceptual design of a hydroelectric power plant for a specific site requires a careful evaluation of turbine technology, generating equipment costs, and experience cost data for various plant configurations. The HydroHelp 1 computer program is an evaluation tool that aids hydro project developers in the work of choosing the best combination of turbine and powerhouse design. The program brings together the necessary analytical tools, up-to-date technical information, and experience cost data to enable the user to develop a preliminary design for a hydro powerhouse. This is only a preliminary design and should be verified and expanded to produce a final design.

The program was developed with assistance from CANMET Energy Technology Centre, an energy, science, and technology research organization in Canada. HydroHelp 1 is available free of charge on the Internet at www.

How the tool aids with turbine selection

There are many different hydro turbines available, ranging from horizontal axis very-low-head mini-bulb units to large high-head impulse units. For example, for potential sites with low to medium combinations of flow (5 to 320 cubic meters per second) and head (3 to 444 meters), 28 turbine choices are available. Narrowing the specifications still results in several options. For instance, at 220 meters of head and 5 cubic meters per second of flow, nine turbine types are available. Because of the large number of available turbine types, selecting the optimum turbine for a particular site can be a difficult process.

The HydroHelp 1 program can aid in the process when selecting turbines with a capacity larger than about 1 MW. All algorithms used in the program are based on data contained in articles written by Jim Gordon.1,2,3,4,5,6,7

Initial turbine choices

The HydroHelp 1 program runs on Microsoft Excel 2003 and occupies 2 Megabytes of memory. The print-out that is produced when users complete the program is four pages long, plus an optional cover page.

The program requires the user to enter 13 parameters describing the turbine requirements. These parameters are either known from the site conditions or chosen by the program user (such as desired number of units, generator power factor, and inflation ratio). These 13 inputs provide adequate information for initial turbine selection based solely on equipment cost. The parameters, and data entered for illustrative purposes, are:

  1. Full head pond elevation (in meters): 405.0
  2. Low head pond elevation (in meters): 404.0
  3. Conduit percent head loss: 2.5
  4. Normal tailwater elevation (in meters): 380.0
  5. Flood tailwater elevation (in meters): 398.0
  6. Design plant flow (in cubic meters per second): 80.0
  7. Desired number of units: 2
  8. Summer water temperature (in degrees Celsius): 15
  9. Transmission system frequency (in Hertz): 50
  10. Generator power factor: 0.90
  11. Maximum allowable gearbox power (in megawatts): 2
  12. Generator quality (utility or industrial): industrial
  13. Inflation ratio since 2008: 1.01

    Note the large rise in flood tailwater in the above example. This is provided to illustrate the effect of such a flood level on equipment selection.

    To assist the user, several comment cells in the program provide assistance with regard to data input. For example, the comment on gearboxes is: “Speed increasers are available up to about 13 MW. However, their use is debatable above about 4 MW, and they should not be considered above about 10 MW. Lower gearbox capacity to below turbine output to see effect on cost.” The comment associated with generator quality instructs users to select an industrial generator for a site with a capacity with less than 10 MW. With the comment cells, a user’s manual is not required.

    HydroHelp 1 uses the above data to calculate the operating envelope for each type of turbine detailed in the program. The program then checks to determine whether each type of turbine fits within the operating envelope – based on number of units, turbine capacity, speed, runner size, head, flow, etc. The program discards as choices any turbines deemed unsuitable for the particular situation. For example, the number of units is required because the program will not select a fixed-blade propeller unit if there are fewer than three turbines for the site, due to the poor propeller efficiency at low flows. The program also discards fixed-blade propeller units as a choice if there is a significant variation in the headwater level.

    Upon conclusion of its calculation, the program provides a list of suitable turbine types, along with an estimate of the water-to-wire generating unit cost.

    HydroHelp 1 includes a sub-routine in the process to estimate the likely optimum head loss based on conduit length, gross head, plant operating capacity factor, and the design standard – either utility (lower loss) or industrial (higher loss). The user has the option of using the head loss recommended by the program or any other loss as input for the conduit percent head loss.

    The HydroHelp 1 program recommends a specific turbine type and total power plant capacity for a specific site. The program calculates basic parameters for the selected turbine, including runner speed and centerline elevation. The green in the figure corresponds to cells that have associated comments.
    Click here to enlarge image

    The program also calculates some basic parameters for the selected turbine – such as speed, required powerhouse crane capacity, and generating unit capacity (see Figure 1 on page 40). The crane capacity is based on lifting the rotor and stator for horizontal units and the rotor only for vertical units.

    The example shown in Figure 1 illustrates the range of turbine options available at a particular flow and head. HydroHelp 1 displays a list of the suitable turbines, which in this case includes horizontal axis mini-bulb Kaplan, horizontal axis “S” type Kaplan, and vertical axis “Saxo” axial flow Kaplan. (An S unit has a water passage shaped like an “S,” and a Saxo turbine has a water passage that resembles a saxophone.) If the user prefers not to choose a particular turbine the program selects, the user can de-select that turbine by inserting a zero adjacent to the turbine type selected by the program. The program then reverts to the next most suitable turbine, based on cost.

    Turbine costs generated by HydroHelp 1 can only be regarded as an estimate and can easily be affected by conditions not analyzed by the program. For example, a manufacturer may have a suitable turbine design that was developed for another site. This reduces the cost of providing a turbine because all the engineering work has been completed. For this reason, the program provides the relative cost of all suitable turbines, allowing the user to consider these as alternatives. Equipment costs generated by the program are based on water-to-wire generating units from European manufacturers. Lower costs can be expected from manufacturers in Asia.

    The cost estimating algorithms for hydropower equipment require annual updating because equipment costs currently are escalating at a higher rate than general cost inflation. Equipment cost inflation from mid-2006 to November 2008 was estimated at 40 percent to 60 percent. This was caused by the large cost escalation of materials, increasing transportation costs, a high demand for hydro turbines worldwide, and a longer-term decline in the number of hydro equipment manufacturers.

    Refining choices using powerhouse cost data

    The shape and cost of the powerhouse for a particular site varies depending on the type of turbine. This is particularly important at heads lower than about 30 meters, where there is a large variety of turbines available. For example, at a site that would experience a large rise in tailwater during floods, either a vertical or horizontal axis turbine could be used. However, with a horizontal axis turbine, a significant amount of extra concrete would be required to anchor the powerhouse to withstand the high water levels. A better, more cost-effective option would be to choose a vertical axis turbine, which would require a taller and narrower powerhouse configuration. The smaller footprint of this configuration would reduce the amount of concrete needed, and thus cost.

    Click here to enlarge image

    HydroHelp 1 features an option of customizing turbine selection based on the powerhouse cost. This requires inputting some additional data, which is easily obtained from a casual site inspection and discussions with a reliable contractor:

    • Cost of overburden excavation (in US dollars per cubic meter);
    • Cost of rock excavation (in US dollars per cubic meter);
    • Cost of concrete (in US dollars per cubic meter);
    • Cost of walls and roof (in US dollars per square meter);
    • Cost of steel superstructure (in US dollars per ton);
    • Average rock level at powerhouse (in meters); and
    • Average depth of overburden at powerhouse (in meters).

    By adding data about powerhouse cost to HydroHelp 1, a user can refine the choice of turbine type for a specific site (bottom line) and estimate costs. The green in the figure corresponds to cells that have associated comments.
    Click here to enlarge image

    Using this data, the program calculates basic dimensions and cost for the powerhouse (see Figure 2 on page 41). The program assumes that the powerhouse is founded on sound rock at about turbine level. If the rock is below this level, a negative quantity is generated and the user sees a comment warning to avoid setting the powerhouse too far above rock level.

    Accuracy of the tool

    To verify the accuracy of HydroHelp 1, the authors tested the results against actual data from about 15 hydro projects. In all cases, this testing confirmed the turbine selection recommended by the program.

    Although the program is intended for use by small hydro developers, it can be used on large projects. The 22,400-MW Three Gorges project in China is an example of a test of the HydroHelp 1 results against published information (see Table 1).8,9,10,11 Because the rated head for this site is known (80.6 meters) and not required to be computed by the program, input data for the program was narrowed to:

    • Normal tailwater elevation: 62.0 meters;
    • Design plant flow: 1,880 cubic meters per second;
    • Two units desired (for test purposes only);
    • Summer water temperature: 15 degrees C;
    • System frequency: 50 Hz;
    • Generator power factor: 0.9;
    • Generator quality: utility; and
    • Inflation ratio since 2008: 1.01.

    Note that the maximum allowable gearbox power does not apply for the Three Gorges project because of the large turbine size.

    A dynamic tool

    By the end of November 2008, more than 120 users from Argentina to Zambia had downloaded the HydroHelp 1 program. Of these users, about half are consultants, a third are from utilities, and the remainder represent manufacturers, universities, or individuals.

    Program users are encouraged to provide feedback to the developers of HydroHelp 1 regarding the program and the cost data. This feedback will be used to further refine and update the cost algorithms.

    Some enhancements scheduled for the next program update, to be performed in mid-2009, include producing more data on the powerhouse. This will include the roof level, the powerhouse length, and an allowance for rental of a mobile crane for unit installation (for when equipment has to be lowered through a roof hatch, due to very high tailwater flood levels).

    The program also will be updated as the turbine operating envelope changes and data from new turbine types becomes available.


    The authors thank CANMET for support provided during development of HydroHelp 1.


    1. Gordon, James L., “Hydraulic Turbine Sizing,” Hydro Review, Volume 9, No. 1, February 1990, pages 74-78.
    2. Gordon, James L., “Hydraulic Turbine Efficiency,” Canadian Journal of Civil Engineering, Volume 28, No. 2, April 2001, pages 238-253.
    3. Gordon, James L., “Powerhouse Concrete Quantity Estimates,” Canadian Journal of Civil Engineering, Volume 10, No. 2, June 1983, pages 271-86.
    4. Gordon, James L., “Submergence Factors for Hydraulic Turbines,” Journal of Energy Engineering, Volume 115, No. 2, August 1989, pages 90-107.
    5. Gordon, James L., “Powerhouse Superstructure Steel Weight,” International Water Power and Dam Construction, Volume 60, No. 8, August 2008, pages 32-33.
    6. Gordon, James L., “A New Approach to Turbine Speed,” International Water Power and Dam Construction, Volume 42, No. 8, August 1990, pages 39-46.
    7. Gordon, James L., “Estimating Hydro Powerhouse Crane Capacity,” International Water Power and Dam Construction, Volume 30, No. 11, November 1978, pages 25-26.
    8. Huichao, Dai, and Tian Bin, “Equipping the Largest Hydro Plant in the World: Insights from Three Gorges,” HRW, Volume 13, No. 4, September 2005, pages 14-16.
    9. Youmei, Lu, “Innovations and Firsts in Equipment at Three Gorges,” HRW, Volume 9, No. 5, November 2001, pages 26-28.
    10. Guan, L.C., “The Three Gorges Project,” Symposium, Hong Kong, 1993.
    11. Li, S.H. et al, “Challenges to Turbine Development,” International Water Power and Dam Construction, Volume 59 No. 8, August 2007, pages 30-37.

    Jim Gordon is a hydro consulting engineer and John Christensen, P.E., is president of Christensen Associates Inc. They worked together to develop the HydroHelp 1 tool.

Previous articleHR Volume 28 Issue 2
Next articleEquipment contract awarded for Austria’s 480-MW Limberg 2

No posts to display