Measurements on gas turbines

Gigabit Ethernet makes it possible: a dynamic measurement data acquisition system with a large number of channels, distributed decentrally with cascadable computing power.

The Power Generation Group within Siemens AG is one of the world's leading specialists in systems and components for power generation and energy management. As part of the continuous further development and optimization of its products, the company operates a test field for investigations on gas turbines at its gas turbine plant in Berlin. The demands on the measurement technology are enormous: Typically, more than 300 dynamic channels with sampling rates of up to 50 kHz per channel have to be recorded.

In the course of modernizing the gas turbine test stand, the Würzburg-based measurement technology specialist LTT Tasler was commissioned to supply a new dynamic measurement data acquisition system. The new concept is based on a networked system of acquisition, display, monitor and storage stations based on Gigabit Ethernet technology. The topological structure of the network is arbitrarily selectable within the framework of standard norms, so that all components are interconnected. As a central control instance, a single so-called master node (PC) takes care of the configuration and hardware management of the entire measurement system.

A total of 10 front-end systems of the type LTT180-32 SensorCorder are used for the decentralized acquisition and processing of the dynamic signals. These contain 32 galvanically isolated input channels including integrated combination amplifiers for volt, strain gauge and ICP signals. In strain gauge mode, full, half and quarter bridge circuits are supported. An optional constant current supply allows highly accurate and extremely low noise strain gauge measurements for high temperature applications. Optionally, each individual channel can also be used for ICP sensors and operated in voltage AC or DC mode. The circuitry for the respective sensors is configured via software - a modification of the hardware is not necessary.

The central network management ensures that each frontend within the network always has the required bandwidth for the transmission of the active measurement channels. The measurement data stream from the front ends to the display and storage stations is based on a secured broadcast transfer. This makes it possible for several display and monitoring stations to receive the data stream of a particular front-end system simultaneously. Each of these stations can display, store or process selected measurement channels independently of the others. This makes it possible, for example, to distribute large numbers of channels among several online displays. The same applies to the backup of measurement data. The network load is almost independent of the number of stations on the processing side.

The obvious advantages of decentralized, digital networking are, on the one hand, the significantly reduced cabling effort and, on the other hand, a lower susceptibility to interference. The analog cables from the sensor to the respective front end, which are sensitive to interference, can be kept relatively short. The digital measurement data transmission, on the other hand, takes place via a single, relatively insensitive Ethernet cable, which can also be designed as an optical fiber. Even longer distances (typically e.g. 100 meters) between the measuring object and the evaluation stations can be bridged without any problems or losses. In addition, the independence of the individual display clients allows users to integrate their own software packages via the API provided.  Since the visualization and monitoring of dynamic data takes place almost exclusively in the frequency domain, the decentralization of the FFT calculation to the visualization clients achieves a considerable increase in efficiency and performance with the greatest possible flexibility. A special feature is the possibility of feeding the stored online measurement data into the visualization software again section by section in "replay" mode after the end of the test, in order, for example, to be able to perform a calculation with modified FFT parameters.

For Siemens user Dr.-Ing. Uwe Pfeifer, according to whose specifications the components of the new system were developed, the main advantage is a significant increase in personnel efficiency during installation, configuration and operation of the system. The cooperation with LTT Tasler has also fully proven its worth, he says, as a close and trusting process has resulted in the development of a tailor-made system that is unique in terms of its performance and efficiency.