Design of Pump as Turbine Experimental Test Facility

This paper presents the design process of experimental test facility for pump as turbine hydropower system. Three design possibilities that related to the PAT condition of operation was developed and analyzed by using CFD Software. It is found that the First Variant with a straight flow to the PAT will produce higher velocity, which is needed to generate more rotation of the shaft generator, in order to generate more electric power. The strength of PAT construction was analyzed by using FEM software. It was found that the maximum stress is 6 MPa and can be concluded that the construction is appropriate to the design requirement.


Introduction
Hydropower is the cleanest energy provided by nature. Utilization of hydropower system becomes worldwide, especially in the area that has the supporting resources. There are many types of turbine used in hydropower system, such as Pelton, Kaplan turbine, Francis reaction turbine, Pump as Turbine (PAT), etc. PAT technology that first introduced in 1931 [1] is one of a micro hydropower system that used an available commercial pump as a turbine in order to generate rotational movement. PAT has several advantages such as 53% cost equipment reduction compare to another hydropower system [2], the power source availability, simple design and easy to install [3], components are available in the market [3], and long life time up to 25 years.
In order to perform researches on PAT characteristic and technology, a laboratory scale PAT installation was required. Figure 1 shows an open PAT experimental test facility used other researchers. There are two types of test facility that used by other researchers. The first type is a test facility that has a direct flow from the pump to PAT, used by Jain et al. [4], Yang et al [5], etc. The second type is a test facility that has a tank which provide a water drop from a certain height, used by Giosio et al. [6], Singh and Nestmann. [7], Suarda et al. [8].
The second type of test facility which has a reservoir tank, is considered closer to the actual condition of PAT operation. However, there might be three actual conditions that happen. First is, the condition where the area has a waterfall and the PAT could be placed right below the waterfall. The second condition might happen when the PAT could not be placed below the waterfall. The third condition might happen when there is only a water stream with certain inclination. This paper describes the design and analysis of the three conditions (Variant) mentioned by using Computational Fluid Dynamic (CFD) Software. One of the variant, which has an optimum output flow parameter was then selected and developed.

The design
Verien Deutscher Ingenieure 2221 (VDI 2221) design method was used in designing the PAT test facility. Three variants were developed in order to find the optimum installation design. The fluid flow of the design was analyzed by using CFD Software. Meanwhile the strength of PAT construction was analyzed by using a Finite Element Analysis Software.

First variant
Three variants were developed in order to get the best design for the experiment test facility. The first Variant has a straight flow from the tank to the pump, which used as a turbine, as shown in Figure 2. The first design condition might occur in a remote area which has a waterfall resource and the pump could place below the water reservoir tank. Clean water (assumed as H 2 O Constant) flows from the PVC tank into the PVC pipe. Due to the space limitation in the laboratory, the tank was placed 3.7 meters' height from the ground level. The velocity of water drop from the tank was calculated by using Equation (1). The Top Tank was designed with a volume of 640 liters, and the water density is 1000 kg/m 3 . The inlet of the pump has a diameter size of 3 inches. A 4 inches' diameter pipe with height of 1.2 meters was used to connect the top tank and a 3 inches diameter pipe before the water flow into the pump. V is the flow velocity; g is gravity of 9.81 m/s 2 . The cross sectional area A of the pipes was calculated by using Equation (2) where d is the diameter of pipe. Q is flow-rate (m 3 /s). The pressure and Reynold number were calculated by using Bernoulli Equation (4) and Equation (5). .
The calculation result found that the inlet parameters before entering the 4 inches pipe, which is velocity, pressure, and flow-rate was 8.52 m/s, 137,622 Pa, and 0.08094 m 3 /s, respectively. These values were entered as the input parameter for flow model simulation in CFD.

Second variant
The Second Variant has an L shape pipe joint (elbow) between the tank and a pipe that flow to the pump, as shown in Figure 4. This design condition might be happening on a remote area which has a waterfall resource, but unable to place the pump below the water reservoir tank. The input parameter was calculated by using Equation (1) -Equation (5), the same as the input parameter for the first Variant, which is 8.52 m/s, 137,622 Pa, and 0.08094 m 3 /s for velocity, pressure, and flow-rate was, respectively. The CFD analysis is shown in Figure 5.

Third variant
The Third Variant has a 45° inclination as shown in Figure 6. This design condition might be happening in a hilly area with a certain inclination on the landscape and has no waterfall resource.  3 Design implementation Figure 8 shows the implementation of experimental test facility that designed and analysed by CFD software. Various condition was performed to acknowledge the power that generated by the PAT generator. The construction of the experimental test facility was made of St-37, L shape with dimension of 50x50 mm and thickness of 4.62 mm. In order to estimate the strength of the installation, FEM was performed as shown in Figure  8(b). It is found that the maximum stress on the construction is 6.62 MPa and considered safe for the experimental facility.

Conclusion
In developing the experimental test facility, three variants that represent the possible conditions of the PAT working area were designed and analysed by using a CFD software and calculation. It is found that the first Variant, which has a straight flow is the best design among two others. The construction was made by using L shape scaffold from St 37 with 4.62 mm thickness and 50 mm width. The strength of the construction was analysed by using FEM software. It was found that the maximum stress is 6 MPa. It can be concluded that the construction is appropriate to the design requirement.