Skip to main content

Blade Tip Timing System Validator

Periodic Reporting for period 1 - Batista (Blade Tip Timing System Validator)

Reporting period: 2019-09-02 to 2020-09-01

One of the main objectives in the design of a rotating component is to obtain the vibration response to ensure that the mechanical loads encountered in operation do not result in excessive stress that can cause fatigue failures. Today this is achieved through modelling, laboratory tests and engine testing with the most common method being the use of strain gauges, the application of which are expensive and often unreliable in a gas turbine environment.

There are a number of reasons for requiring a new technology and most hinge on both the lead-time and cost of application. Today’s gas turbine development programs are up to 50% shorter than those of a decade ago and the machines operate at increasingly higher speeds and temperatures. Instrumentation failure is high, as is the mortality of a strain gauge in such an environment.

The importance of adopting new methods is driven mainly by the cost of development programs and time-scales. More information about blade vibration can be gathered leading to new opportunities to develop trending of data and extracting further useful information from the data which is essential for long term product health. The potential reduction in down-time through scheduled maintenance can be replace by planned maintenance, this in turn leads to higher reliability lower operating costs and a reduction in the carbon footprint due to more efficient operation of the plant.

Often equipment is produced that have many features in order to offer a unique selling point to potential customers, in many cases these functions can have a marked effect on the measurement error and uncertainty.
Batista looks at the small number of these functions that are in use and has configured a test set-up to exercise the most common configurations. This uses a range of sensors including optical, inductive capacitive and magnetic sensors to provide an over view of their real capability v's the manufacturers stated capability which are often very different. Varying the timing resolution and channel bandwidth will provide data on the uncertainty associated with these functions

To this end we have produced a rig test comprising of an Aero - compressor rotor stage and equipped it with a range of blade tip sensors. The rotor has strain gauge attached that are monitored and recorded through a telemetry system. A magnetic and air-jet excitation system will be used to induce blade vibration that will be captured by the probes, strain gauges and a Laser Doppler Velicometer allowing correlation and verification of the vibrational responses.

Standard blade calibration techniques that are used today in strain gauge systems are being modified to introduce the BTT requirements. These include the FE modelling of the blade and disc, static modal excitation using air jets and correlation of tip deflection, strain gauge levels and FE predictions at zero speed. By incorporating this data in to BTT simulator that is based upon the FE model it will allow us to predict the best set-up and expected measurement uncertainty based upon real blade geometry.

By incorporating all of the above activities we can define a standard process for others to follow, and improve upon, that controls the way in which BTT is applied and thus support the certification process in a controlled way. This will open the way for smaller industries that cannot afford the cost of technique validation, to use the BTT methods in support of their product development.
The initial work consist of five parts

1. Obtaining the test article and adapting it to the test cell.
A Viper 522 engine has be procured and the rotor 1 stage removed, This is a small rotor consisting of 29 bade with an overall diameter of approximately 0.5m and a blade length of 120mm.
An adapter has been designed and manufactured to attach the assembly to the test rig capable of speeds up to 6000 rpm.
BTT probe holders have been designed and manufactured to hold the array of sensors applied to the experiment.
A telemetry unit has been designed, manufactured and tested to transfer the strain gauge data via a slip-ring.

2. Calibration process work
The rotor disc and blades have been separated and an FE model created for both.
Predictions for best axial positions of the BTT probes and optimum position for the strain gauges to cover the first three modes have been identified.
The axial position data has been imported into the BTT configuration tool allowing optimum positioning and an uncertainty figure to be calculated based upon the sensor aperture.
The blades have been excited in the first three modes using an air-jet.
Identification of the drivers of uncertainty are being documented.
A set up process is being prepared.

3. Instrumentation set-up
A BTT system has been obtained and modified to allow bandwidth and timing resolutions to be changed to assess their effect on the overall results
Telemetry acquisition software has been produced and validated
Post-processing software design has commenced to import all of the relevant test data.
Real-time derived once per rev code has been produced and validated (not in the original proposal, but added as a requirement resulting from a literature search)
Sensors and probes have been offered by a third party vendor.

4. Test facility set-up.
The rotor and adapter have been trial fitted to the rig motor.
Magnetic excitation of a static blade has been undertaken.
Rig running limiting parameters continue to be identified.

5. Program management
This function has operated throughout the project
Adapted the program due to the Covid-19 issues
re datummed deliverables due to changes in available resources.
Maintained the end data with no additional costs.
Agreed a no cost extension due to furloughed workers.
Shared up to date information between partners.

Overall the project is on plan and within budget but with a later than expected delivery of some milestones due to the Covid-19 issues. Overall the re-planning has put more pressure on the compilation of the results and the final report
The use of a non-optical probe provided by a third party with demonstrated claims of 10X bandwidth over current technology will be a game changer for BTT probe technology.
Probes with proven BTT capability that require low or no maintenance and operate at elevated temperatures are essential to future health monitoring application.
If the chosen technology is successfully validated by Batista then this will change the state of the art, if not it will produce a method for evaluating future technologies that can be employed by sensor designers and manufacturers.
The work being undertaken in Batista has led to a proposal for continuation of the research theme, using assets purchased in this project for EPSRC academic funding for two years in the UK and an accepted continuation of research over a four year period in the Czech Republic.
The project will add clarity the effects of the myriad of processing function and BTT instrumentation offered through the supply chain.
Significant interest has be aroused in the field with over 1400 hits (an increase of 10 times over expected hits) on the Batista pages of the web-site demonstrating the interest in this project.
Strain gauges applied to blades
Disc mounted on rig for checks
Rotor disc mounted on rig adaptor
Rotor under test
Blade in the calibration jig