Development of a variable-fill hydraulic dynamometer test bench
Abstract
The WJ Engineering company required a torque absorption device to be used in their newly developed Hydro Testing Facility. The torque absorption device is intended to load hydraulic turbine models that will be tested in the facility, with the purpose of validating and verifying the turbine’s performance and control system. To do this, the torque absorption device must be able to measure the torque and power delivered by the models and be able to change the load during continuous operation. The variable-fill hydraulic dynamometer was subsequently identified as possible candidate for application.
The hydraulic dynamometer is a torque measuring device that converts mechanical energy into heat by means of friction, turbulence and incidence losses of a fluid mass circulating inside working compartments within a rotor (connected to the machine under test) and two stators (fixed to the casing). The conversion of energy produces torque that acts on the dynamometer casing which is resisted by a torque arm with load cell, thus enabling torque measurements.
The fluid pressure generated inside the working compartments forces a throughflow of fluid, which dissipates heat to allow for cooling of the dynamometer. The torque absorption and throughflow of the dynamometer can be controlled by an outlet control valve. The entire setup, with the dynamometer, the control valve and an external cooling cycle is known as a variable-fill hydraulic dynamometer test bench.
These test benches are not readily available off-the-shelf and therefore the aim of this study was to develop one suitable for the testing facility. There exists little open literature on the simulation and design of variable-fill hydraulic dynamometers, and therefore almost no established theory to explain and predict their behaviour. However, a one-dimensional theoretical model presented by Hodgson (1991) and J.K. Raine & P.G. Hodgson (1991,1992) claimed to accurately simulate the behaviour of variable-fill hydraulic dynamometers. Their method was implemented during this study to analyse and develop the test bench, and the method was evaluated for its suitability as design tool and applicability in system simulation.
WJ Engineering specified device requirements of a torque loading up to 300 [Nm] at an oper-ating speed of 1200 [rpm] and power absorption capability of 38 [kW]. Due to time constraints WJ Engineering already designed and manufactured the rotor and stators prior to the current study. These rotor and stators were subsequently analysed and used as a basis from which the additional test bench components were developed. Simulation of the WJ Dynamometer by means of the theoretical model, predicted that it would achieve a maximum torque absorption of 1154 [Nm] at 1200 [rpm], hence a 145 [kW] power capability. The 300 [Nm] and 38 [kW] torque and power requirement from WJ Engineering was met at dynamometer percentage fill conditions of as low as 30% according to the simulation.
Additional requirements included a throughflow of 1 [l/s] at low torque and 4 [l/s] at high torque for proper cooling. It was determined that a control valve with a pressure drop of 39 – 45 [kPa] at 1 [l/s] flow rate, and 5 – 15 [kPa] at 4 [l/s], should allow the required throughflow. A control valve and flow meter were manufactured and tested to ensure specifications were satisfied.
After commissioning the test bench, it was found that torque absorption far exceeded the original specification, with an ability to absorb torque between 313.9 to 372.8 [Nm] at a rotational speed
as low as 800 [rpm]. Even though the torque was above specification, the test bench did not achieve the full power requirement, only reaching a power absorption of 32 [kW] within load cell limits. The throughflow achieved during testing ranged from 0.8 to 1.8 [l/s] and consequently the dynamometer overheated after a limited period of operation. The outlet control valve was also found to exhibit unsatisfactory control behaviour with relation to the torque absorption of the test bench.
Based upon these results it was concluded that the test bench, without further modifications, was not yet suited for use in the WJ Hydro Testing Facility. The theoretical model was deemed to possibly be a sufficient analysis and design tool, since the results fell within the range of predicted values of the simulations, but this cannot be confirmed due to the improper flow control characteristics of the outlet control valve which influences the torque characteristic of the dynamometer. It is suspected that the lack of control on the inlet flow of the dynamometer could be the reason that adequate throughflow for cooling and torque control was not reached. The lack of inlet flow control and other shortcomings thus warrant future research.
Nevertheless, a functioning variable-fill hydraulic dynamometer test bench was successfully developed with the capability to absorb load from devices such as a hydraulic turbine, albeit for a slightly smaller power envelope than originally intended.
Collections
- Engineering [1403]