Overview

Welcome to the Control Systems Test Rig (CSTR), proudly developed and operated by the Control Systems Test Rig (CSTR) group, formally titled the Controls, Electronics and Auxiliaries (CEA) group.

The CSTR is a student-built gas turbine engine that uses kerosene as a fuel. The purpose of the CSTR is to generate accurate, repeatable data that can be used to validate a dynamic model, which was developed from theory and software. A dynamic model for the CSTR has been previously developed in Simulink, Matlab. The data collected from the CSTR is used to validate the CSTR dynamic model. Validation of this model will confirm the theory and software that will be used to develop a model for the 100 MW full scale power plant being developed by the project team.

The CSTR group works with professors, industry professionals, graduate students and volunteers from all disciplines of engineering and receives funding from the department of Mechanical and Aerospace Engineering and the Carleton Student Engineering Society.

Physical Description and Auxiliaries

The CSTR is made from a combination of OEM and specifically designed components. It is entirely mounted on a movable cart for ease of transportation to the test site. The fuel and oil systems, compressor, turbine etc. are all automotive industry components. The starter motor is a router, with a shaft that ends in a hex bit to engage the compressor. The combustor is from a Garrett auxiliary power unit. The exhaust pipe, exhaust nozzle and bellmouth inlet were all designed specifically for the CSTR. It is hoped that the addition of a bellmouth and exhaust nozzle for load/no load condition will generate data that allows for more accurate evaluation of the dynamic model.

Data Collection and Electronics

In the past, equipment such as PLC controllers, instrument boards and a desktop computer running LabVIEW software were used in data collection and fuel scheduling as they were available through donation. As equipment became available, it was added to the existing system (this includes pumps and other hardware as well as data collection hardware).

The result was a setup that was not ideal for many reasons. It was difficult to understand the wiring as it was done over multiple years, making modification and set-up difficult. It was also fragile, having many loose connections and open circuits. The signals produced a lot of noise and therefore the data collected was not as reliable as we would have liked.

Previously the CSTR group completely re-wired the CSTR for proper instrumentation and to acquire data acquisition hardware capable of accurate and reliable measurements. The modifications are also intended to make the CSTR more robust, faster to set up and safer.

Previous modifications have included: full electrical and signal re-wiring using industry standards, new electrical panels, proper electrical connectors, a USB data acquisition device from National Instruments, thermocouple amplifiers and a laptop that is now used to operate the test rig.

This year the CSTR group has continued to increase the safety and reliability of the rig by including safety features such as properly grounding all electrical components. As well, the issue of a noisy signal from the RPM signal was resolved this year to allow accurate data to be collect. Finally, a new load cell was procured this year, which required calibration and was successfully implemented on the rig.

Each of these modifications have resulted in an improvement in the ease of use of the control system, the safety of the CSTR, the robustness of the system as a whole and the reduction of signal noise without the use of software filters.

Programming and Control

A fuel schedule for the CSTR was previously developed to obtain data that resembles a modeled fuel schedule in the dynamic model. This will allow data collected for the test rig to easily be compared to the dynamic model. A nozzle schedule is currently being incorporated to work with the fuel schedule to facilitate data collection during a test run.

Previously, fuel control for the CSTR was through the programming of a PID and manual use of a knob on a LabVIEW interface. Currently control is still done through LabVIEW software on the laptop, which is connected by USB to the CSTR, but the fuel flow is adjusted by pre-determined increments with either a graphic knob or by typing in the desired fuel flow.

To date the data collected from the fuel flow schedule on the CSTR has agreed with the same schedule for the dynamic model. In order to further validate the model the test rig will be run through a series of load shed tests.