Trend assessment for full-scale fatigue testing

Save time and money with real time trend assessment with the new HBM MD Trend software. Full-scale fatigue testing in the aviation industry often takes years to reach the number of flying hours necessary for certification. Although tests can be carried out on a 24-hour basis, the labor costs are high and there is the disadvantage that the work has to be done in shifts.

Full-scale fatigue tests often run for years to accumulate the equivalent flight hours required for certification. Tests can be run on a 24-hour basis, but this incurs test laboratory costs and the inconvenience of shift work.

Tests are usually interspersed with inspection intervals that can be time consuming and cause delays. If inspections are performed too frequently, testing will take longer than planned; if inspections are carried out too infrequently there is a risk that the test article could sustain serious fatigue damage, requiring extensive repairs and incurring enormous costs for idled program and laboratory personnel. In such situations, the mode of crack nucleation may prove to be indeterminable.

In many cases, if a crack is found by visual inspection, the acquired strain data can be analysed to help determine the nature of the failure. This led HBM, together with BAE SYSTEMS, Lockheed Martin and others, to develop software that can look at acquired strain data in real time and detect changes in the loading and end-point characteristics.

The software has two main functions. First, it locates areas of potential problems to help reduce damage to the test article. In doing so, it provides the design engineer with an early warning that modifications may be necessary. This significantly reduces the time to market. Secondly, the software offers extensive triggering capabilities, including even stopping the test. This allows testing to continue minimally manned or even unmanned, thereby reducing the cost of 24-hour operation.

The software achieves these objectives by storing the data from every strain gauge or other input at each end point or load case, together with the associated load case identifier. A base line is determined from the first load case, or an average of the first few load cases. For subsequent load cases the data is compared with the base line. Triggers are generated for simple changes in magnitude, or from statistical or linear regression calculations.

Potentially, the amounts of data acquired are huge. Tests commonly have 2,000 channels, running flight spectra with thousands of distinct end points repeated during the test. This normally requires a prohibitive amount of data entry to set up trending assessment for a test. This problem has been solved and configuration of a 2,000 channel test takes a few minutes.

Initial tests have produced exciting results. Everything from changes in strain characteristics to the quality of the end-point loading was detected. Increased confidence in test quality and security ensures trend monitoring will become an essential part of fatigue testing in the future. 

Files:

 Trend assessment for full scale fatigue testing (141 kB)