Duke Energy Carolinas LLC plans to install aerating turbines in a new 28-MW powerhouse for the Bridgewater project. These turbines, being custom-designed for specific site conditions, are expected to allow Duke Energy to improve dissolved oxygen levels in water downstream from the project while minimizing efficiency losses.
The current 20-MW Bridgewater project is on the Catawba and Linville rivers and Paddy Creek in North Carolina. Three dams – Catawba, Linville, and Paddy Creek – form Lake James. The powerhouse, at the base of Linville Dam, began operating in 1919.
During efforts to relicense the project, owner Duke Energy Carolinas LLC determined that Linville Dam, as originally designed in 1915, needed upgrading to meet current Federal Energy Regulatory Commission (FERC) guidelines for dam stability. Duke Energy hired HDR Engineering Inc. of the Carolinas (HDR|DTA) to conduct feasibility studies and determine various options that could meet current FERC seismic stability requirements. HDR|DTA concluded that the most effective overall strategy was to remove the existing Bridgewater powerhouse and replace it with a new powerhouse further downstream, to make room for the addition of a large reinforcing earth berm on the downstream side of the existing earth dam. This option was preferred over roller-compacted concrete or an engineered retaining wall around the existing powerhouse due to various cost and constructability factors.
The new Bridgewater powerhouse also provided Duke Energy with the opportunity to address continuous minimum flow requirements, as well as turbine discharge aeration at the site to meet environmental requirements agreed to during the relicensing process. If the existing powerhouse was retained, a considerable investment would have been required to upgrade and rehabilitate the existing turbines to address those new environmental requirements.
The Bridgewater project provided a unique collaborative opportunity for a hydro project owner, engineering consultant, and equipment supplier to work together to design a powerhouse that utilizes the latest advances in turbine technology. Duke Energy selected Voith Hydro to supply the new turbines.
Dissolved oxygen problems
Low dissolved oxygen (DO) is a common problem in reservoirs in the southern U.S. During the warmer months of late summer, these lakes can experience thermal stratification, causing the reservoir to separate into distinct layers. The uppermost layer is characterized by high levels of DO. As the depth increases below the thermocline, DO concentrations quickly decline, potentially reaching levels as low as 1 milligram per liter near the reservoir bottom at some sites. At many existing hydropower projects, the turbine intakes are far below the reservoir surface, where these low DO levels are present. This water then passes through the turbines before being discharged into the tailrace downstream of the facility, where low DO levels can have an adverse effect on water quality and aquatic life.1,2
During relatively short periods of time over the course of a calendar year, DO levels in the Bridgewater tailrace possibly could fall below current state water quality requirements of 4 milligrams per liter instantaneous and 5 milligrams per liter daily average. As previously discussed, the turbine upgrades at Bridgewater provided an excellent opportunity to implement some of the most recent aeration technologies.
Project evaluation and design
The need to address DO enhancement with new hydroelectric equipment set the stage for the work to be performed at Bridgewater. In addition to the aeration capabilities, the new plant also will be required to handle the normal role of reservoir level regulation while maintaining minimum flow levels at all times. To maximize energy production and still address these requirements, HDR|DTA and Duke Energy evaluated several different plant configurations with various unit numbers and sizes. The final selection consisted of two larger vertical Francis machines with runner entrance diameters of 2.5 meters and one small Francis unit with runner entrance diameter of 0.9 meter.
HDR|DTA considered several technically acceptable methods to address the DO issues at the project site, with initial studies focusing on upstream options such as surface water pumps and porous line diffusers.
Ultimately, it was determined that the 27-meter distance between the reservoir surface and the Bridgewater intakes was too large for the surface water pumps to be as effective as desired. Oxygen line diffusers are a proven technology and are quite effective but were ruled out due to a combination of initial installation cost and long-term annual operation and maintenance cost when compared against the incremental additional costs of other technically acceptable options.
A number of other aeration options occur within the turbines themselves, where low-pressure regions that develop during operation are vented to the atmosphere, creating a natural flow of air into the water passage. This aeration category is referred to as auto-venting turbines and is a particularly cost-effective method for providing large air flows to the turbines. Three main types of auto-venting turbines exist, including central, peripheral, and distributed aeration. For each aeration method, a unique system is employed to transport the atmospheric air from outside the turbine to different locations within the water passage. Figure 1 gives an overview of each of the three aeration techniques.
Distributed aeration draws air from several pipes positioned above the headcover (typically within the wheel pit), where it is collected in a continuous chamber above the runner crown. The air then passes into the hollow runner blades before entering the water passage through a series of slots positioned along the discharge edge of the blades.3 During central aeration, air is transported from the region above the headcover to various locations around the side of the deflector, or through the turbine shaft and out the deflector bottom. Peripheral aeration occurs further downstream, near the entrance to the draft tube. Typically, the draft tube access corridor provides a convenient location for the placement of the inlet piping when retrofit into an existing plant. This air is transported through a manifold system that distributes the air around the outside of the draft tube. The air exits the manifold through a continuous slot or multiple orifices positioned along the inside surface of the draft tube cone.4
Because each auto-venting turbine aeration method utilizes different injection locations within the water passage, each technique has distinct characteristics induced by the local flow patterns under the runner. These factors influence bubble size and distribution within the draft tube, which have a significant effect on aeration effectiveness, as well as how the aeration influences the operating efficiencies.
As central air is drawn through the deflector, the incoming air forms a column of bubbles that continues into the draft tube. Figure 2 shows a representation of the bubble distributions that develop during central aeration. Note that these illustrations correspond to full load operation of the turbine.
Although central aeration is relatively inexpensive to retrofit an existing turbine, the bubble column that results remains confined to the central portion of the draft tube cone, preventing the air/water mixing that facilitates the transfer of DO. This aeration method is most effective during part load, where the swirl present within the flow exiting the runner lowers the static pressures near the air injection locations on the deflector. Central aeration also tends to disturb the flow within the draft tube, which can result in large effects on turbine efficiency. These DO transfer efficiencies, as well as the effect that each aeration method has on operating efficiency, will be discussed below.
Injecting atmospheric air from the inner wall of the draft tube cone results in a curtain of bubbles that moves along the water passage periphery. Figure 3 shows the bubble distributions corresponding to peripheral aeration at full load.
As these bubbles continue through the draft tube and into the tailrace, the mixing that occurs creates a bubble distribution that is more uniform than that of central aeration. Despite the increased mixing, the bubble distribution is not quite uniform throughout both channels of the draft tube. Peripheral aeration is effective at medium to high loads and has intermediate effects on turbine operating efficiencies.
Aeration through the runner bucket utilizes the low static pressures that develop at the blade trailing edges to draw air into the turbine. Another key feature of the aerating runner is the velocity gradients that develop adjacent to the air injection locations. These regions of shear help to break up the incoming air into a fine cloud within the draft tube cone. Figure 4 shows a representation of the bubble distributions associated with an aerating runner operating at full load.
The air injection slots located at the discharge edge of the runner buckets provide a uniform bubble distribution that is completely disbursed throughout the draft tube. These smaller, well-disbursed bubbles maximize the contact area between the air and the surrounding discharge. Of the three auto-venting alternatives, distributed aeration has been proved to provide the most effective DO enhancement while having the lowest effect on operating efficiency.
Although distributed aeration would be the method of choice for the new Bridgewater powerhouse project, manufacturing limitations associated with the smaller runners at Bridgewater prohibit the use of the aerating runner.
The previously mentioned bubble characteristics play a significant role in determining the relative aeration effectiveness for each method. This effectiveness can be evaluated by comparing DO uptake efficiencies for each method.
DO Uptake Efficiency = DOuptake/(DOsaturation -DOinitial)
– DOuptake is the amount of DO that is achieved during aeration;
– DOsaturation is the largest amount of DO that can occur under the given site conditions (i.e., water temperature and atmospheric pressure); and
– DOinitial is upstream DO content in the water present within the intakes, penstock, and spiral case.
Figure 5 provides DO uptake efficiencies corresponding to auto-venting turbine field data for Francis turbines, as a function of Q/Qopt (void fraction φ = 3). Qopt represents the discharge rate that produces the peak operating efficiency of the turbine.
– Qair is the volume flow rate; and
– Q is the discharge for each operating point.
Within the normal range of operation for a typical Francis turbine (Q/Qopt of 0.8 to 1.2), distributed aeration provides the largest DO uptake efficiencies, followed by peripheral and central aeration, respectively. However, in the region below Q/Qopt of 0.8, central aeration provides DO uptake efficiencies that are significantly larger than distributed and peripheral aeration. In this part of the operating range, the increased swirl in the flow exiting the runner produces a bubble distribution that provides favorable mixing between the air injected at the deflector tip and the surrounding water.
Efficiency losses when aerating are dependent on several parameters, including turbine draft tube shape, quantity of air (or the void fraction), location of air admission, bubble size, and bubble distribution. Figure 6 presents performance trends derived from field tests at numerous aeration projects for Q/Qopt of 1.0. Note that changes in efficiency (η) are defined as ηnon-aerating – ηaerating and are given as a function of void fraction (φ).
For Q/Qopt of 1.0, efficiency drops are shown to have a large dependence on void fraction. Below a void fraction of 1, these efficiency effects remain small for all aeration methods. However, as the void fraction becomes larger, the efficiency drops associated with central aeration can reach significant levels. The aforementioned DO uptake efficiency relationships, as well as how the aeration methods affect turbine performance, both were utilized to optimize the final aeration solution for Bridgewater.
As the air mixes with the surrounding water during aeration, oxygen transfers from the air to the surrounding water. This topic of mass transfer within a two-phase mixture has received considerable attention, and the insights gained from these studies have been employed within various aeration models.5 One of the most successful of these models is known as the discrete bubble model (DBM). During the design stage, the DBM was used to evaluate various air flows and contact times (dwell time) for the bubbles in order to achieve the desired DO update at Bridgewater. A historical review of DO levels in the upper reservoir showed that a goal for uptake of 3 to 4 milligrams per liter would conservatively envelope any anticipated worse case conditions. Deficiencies near this level occurred rarely and for short periods of time. The design basis however was established to cover these extremes. These calculations incorporated individual parameters specific to Bridgewater operation, including site conditions, operating points, predicted air flows, and turbine geometry.
Two design concepts for DO enhancement through turbine discharge aeration were compared. One concept was based on utilizing a conventional draft tube, while the second incorporated a deeper and longer draft tube that would increase the bubble dwell time and provide more DO uptake. The DBM runs verified that the deeper and longer draft tube would be required for the amount of air that would be available for mixing with the turbine discharge. Based on the calculations, Voith Hydro was able to accommodate this draft tube into the turbine design.
During the initial inquiry phase, it was expected that a higher than normal turbine setting would be necessary to develop the lower static pressures within the water passages that would be required to deliver the desired higher air flows. However, during the design phase, design optimization analyses, including the use of the DBM program, showed that a lower turbine setting still provided enough air flow to reach DO uptake goals. Voith Hydro used this lower machine setting to provide a smaller, faster rotating turbine, which decreased the civil and equipment costs.
While distributed aeration would have provided the best overall aeration performance, manufacturing limitations related to runner size prevented the implementation of the aerating runners at Bridgewater. Instead, Voith Hydro designed the turbine equipment to employ a combination of peripheral and central aeration technology. During periods of low flow operation in which DO is required, both the central and peripheral air admission systems will be in operation. During periods of high flow operation in which DO is required, the central aeration system will be turned off to minimize efficiency and power losses and the peripheral aeration will operate alone. Implementation of both methods provides a measure of flexibility in the ability to meet the flow and DO uptake needs of the plant.
HDR|DTA defined the air flow rate and level of DO uptake necessary to achieve tailrace DO goals, and Voith Hydro designed the air intake systems to provide the air flow to the machines. For new equipment and powerhouse designs, it is possible to size and route the piping through the powerhouse to minimize the twists and turns (i.e., head losses), which adversely affect the aeration performance of the system. For rehab projects, this flexibility often is not available and the designs have to accommodate long, inefficient piping routes to transport the atmospheric air to the turbines.
Bridgewater is one of the first new hydro plants designed for aeration. It is an example of how plant owners, engineering consultants, and equipment suppliers can work together to come up with the best possible solution. This collaboration resulted in a unique design that utilizes a combination of advanced aeration technologies that will maximize the DO uptake across the full range of flows at the plant while minimizing the effect on turbine efficiency. The project is in the construction phase, and the units are scheduled to be placed in commercial operation near the end of 2011.
- Fisher, R.K., et al, “Innovative Environmental Technologies Brighten Hydro’s Future,” Proceedings of the XIX IAHR Symposium on Hydraulic Machinery and Cavitation, World Scientific Publishing Co. Pte. Ltd., Singapore, 1998.
- March, P.A., and R.K. Fisher, “It’s Not Easy Being Green: Environmental Technologies Enhance Conventional Hydropower’s Role in Sustainable Development,” Annual Review of Energy and the Environment, Volume 24, November 1999, pages 173-188.
- Patent nos. 5,924,842 and 6,155,783.
- Patent no. 5,941,682.
- McGinnis, D.F., and R.J. Ruane, “Development of the Discrete-Bubble Model for Turbine Aeration Systems,” Waterpower XV Technical Papers CD-Rom, HCI Publications, Kansas City, Mo., 2007.
Rohland, Kevin M., and John C. Sigmon, “Aeration Solutions for New Hydro – The Bridgewater Project,” HydroVision 2008 Technical Papers CD-Rom, HCI Publications, Kansas City, Mo., 2008.
Kevin Rohland, P.E., manager of proposal engineering for Voith Hydro, worked with Duke Energy Carolinas LLC and HDR|DTA to develop the solutions described in this article. Jason Foust, PhD, hydraulic engineer with Voith Hydro, assisted in the development of the final piping layouts for the aeration system. Gregory Lewis, P.E., is technical manager for Duke Energy, which owns the Bridgewater project. John Sigmon, P.E., mechanical engineering manager for HDR|DTA’s Southeast Regional office, has been involved with the Bridgewater Powerhouse project from the conceptual stage through detailed engineering and into the construction phase.
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.
Touring the Bridgewater Plant