Report

Teaser, summary, work performed and final results

Periodic Reporting for period 2 - UHPE (Integrated Intelligent Bearing Systems for UHPE Ground Test Demo (I²BS))

Teaser

The overall objective of the I2BS project was to develop innovative smart bearings for an Ultra High Propulsion Efficiency (UHPE) Ground Test Demonstrator that not only meet the demo specification but also provide significant safety improvement compared to existing standards...

Summary

The overall objective of the I2BS project was to develop innovative smart bearings for an Ultra High Propulsion Efficiency (UHPE) Ground Test Demonstrator that not only meet the demo specification but also provide significant safety improvement compared to existing standards. I²BS pursues an integrated approach comprising the development of sensor technologies, energy harvesting, wireless communication, data management and algorithms to monitor bearing behaviour in challenging operating conditions (e.g. high temperature, high speed and high thrust). As part of the UHPE Demonstration Project, I²BS will design, develop, evaluate and test interchangeable conventional and smart bearings for the UHPE demonstrator. The bearing design will fulfil all requirements and safety standards for aerospace applications. The ‘smart’ bearings will be able to deliver, in real time, information on the bearing’s main functional characteristics and health including temperature, axial & radial load, ball or roller or cage speed, radial clearance and premise of failure on each part of the bearing.

Work performed

At the beginning of the Project the administrative work including all communication, Dissemination and exploitation plans was performed and completed. The risk Analysis all over this Project was defined by the parties.
All requirements for the sensors, communication Systems, bearings, test rig and testing conditions were defined. First sensors and communication Systems were selected and ordered. University of Southampton performed pre-testing to evaluate sensor behavior at required conditions.
The following sensors were evaluated:
Vibration: Piezoelectric charge mode accelerometer
Load: Resistive strain gauge
Temperature: Thermocouple
Cage Speed:Eddy current probe
Shaft speed:Eddy current probe

For autonomous energy supply, a thermoelectric generator (TEG), ultracapacitors, power management board and microcontroller was tested in laboratory with relevant oil-in and oil-out temperatures. The voltage output of the TEG was connected to the power management board and energy was stored in ultracapacitors. It could be demonstrated that the TEG provides sufficient energy for one sensor to measure for 1s every 3 to 4 minutes. As a conclusion, multiple TEGs or performance optimized TEGs can be used for a main shaft engine bearing to provide the required energy for multiple sensors.
Alternatively, a Switched Reluctance Generator (SRG) can be used to generate power from rotation of the shaft. The rotation of the rotor is achieved by energising and de-energising the phase winding on the stator poles. Several simulations were performed with MagNet infolytica. The simulation is done based on the geometry and restrictions that was discussed between all partners. More simulations will be devloped to find the optimum design for this type of harvester based on the requirements. Some pre-test were already successfully performed. In a next step it is planned to perform some tests under higher Speeds which are similar to a main shaft engine bearing.

A subscale bearing test rig was designed, developed and manufactured. The test rig was ready to test in April 2019. The aim of subscale bearing tests is to evaluate the integrity and performance of the smart bearing monitoring system. The smart bearing system consists of three major parts, i.e. sensing system, energy harvesting and wireless communication. The sensing system, energy harvester and wireless communication system was integrated on the subscale test rig. The subscale testing will help to evaluate the ability of the sensing system to detect different bearing faults. The data collected from testing will help to optimize and develop the algorithms for bearing fault detection. It will help to make informed decision about the optimization of the sensing system and the fault detection techniques. The aim and objectives for these tests are:
• To test a fully integrated sensing system on bearings.
• To test the survivability and performance of accelerometer, thermocouple, strain gauge and speed probe at temperature of between 130 ⁰C – 160 ⁰C.
• The sensors survivability and performance will be tested in the presence of oil circulating through the test bearings.
• To identify the capability of sensing system such as; the identification of the smallest defect size that can be detected by the sensing system, fault detection in the presence of over-rolling frequency/white noise and the detection of slippage.
• To optimize the sensing system and identify the most important parameters to be measured for bearing fault detection.
• To test algorithms and signal processing techniques for bearing fault diagnosis.
The first tests were already completed and the developed algorithm was capable to detect inner ring defects up to a size of 0,1 mm, which is even smaller as a natural initial spall size caused by subsurface or surface intiated fatigue. In a next step intial damages of outer ring and balls will be detected to identify the minimum spall size that can be detected on this compon

Final results

The identified sensors and TEG enables to realize a smart bearing with almost no addtional space required (interchangeable to conventional bearings). The sensors were chosen in a way that the availble space in the bearing chamber can be used to install all required sensors and TEG. So far no technical solutions are known which are installed directly at the bearing or in the bearing chamber with an autonomous energy supply. By means of that we are expecting that initial component failures could be detected at a very early stage (several hours before conventional sensors which are installed in an engine). This will be further examined by the on-going subscale tests. Besides this, we will evaluate the possibility to get a correlation of spall size to sensor signals. With this approach, a remaining life prediction is possible, which enables to extend service intervalls and increase the overall reliability of the whole engine system.