Coming out with a new model such as this is a major undertaking for any aircraft manufacturer, and represents years of effort. Uncovering a major problem during certification flight testing therefore represents a huge setback. LMS Ground Vibration Testing (GVT) and analysis technology helps leading business jet manufacturer develop a new model quickly in a market where there is no room for error.
Now in its 75th year, the Cessna Aircraft Company has delivered over 183,000 airplanes. More than half the general aviation aircraft flying today are Cessnas, including the largest fleet of business jets in the world. The company maintains
this leadership position by leveraging its expertise in aircraft design with advanced technology, enabling the manufacturer to remain in the forefront of its market with new models such as the Sovereign now under development.
This totally new design is an ambitious undertaking to fill a gap in the business jet market between smaller models costing up to $7 million and the big transoceanic jets that top out at $15 million. Industry analysts note that, in this niche, existing designs introduced more than 15 years ago are rather mature, and Cessna sees this as an opportunity for an all-new design.
The $12 million, 10-passenger Sovereign has a cruising speed of Mach 0.75 and is the largest plane in the company’s family of Citation business jets. With the help of LMS technology, Cessna has beaten the competition and expects to capture a larger market share.
Design-right-first-time
Coming out with a new model such as this is a major undertaking for any aircraft manufacturer, and represents years of effort. Uncovering a major problem during certification flight testing therefore represents a huge setback. Required in the US by the Federal Aviation Administration (FAA) and by similar agencies in Europe and around the world, these rigorous tests generally take a year or more and cost millions for flight-ready prototypes. Halting these certification tests and returning to engineering for a re-design could significantly delay the new-model launch and substantially cost the manufacturer in terms of increased development expenses as well as lost revenues and market share.
Clearly, for aircraft manufacturers such as Cessna, errors are not an option. The design has to be right the first time, and any potential problems must be spotted and fixed early in development.
Dangers of flutter
Designers are particularly concerned with potentially catastrophic conditions such as “flutter”: aerodynamic instability where wings, ailerons and other flight surfaces vibrate excessively above a certain airspeed. Driven by aerodynamic forces, the vibration rapidly increases in amplitude and eventually exceeds the ability of the structure to dampen the oscillations. These severe deflections can rip off an aircraft’s wings or tail in seconds.
To spot problems with flutter early in development, aircraft manufacturers for decades have utilized Ground Vibration Testing (GVT) and modal analysis as standard tools. First, an initial Finite Element (FE) model of the aircraft is built. The model is then correlated through GVT, where sensors gather frequency-response data from a prototype plane excited by a set of random-noise shakers. In the correlation process, GVT data are compared to the modes and responses predicted by the FE model, which is then adjusted to conform. This final, tuned dynamic model then is used for modal analysis to predict the airspeed where critical modes become unstable, generally through specialized modules in FE codes such as MSC.Nastran.
Fixing a troublesome bottleneck
Correlation of the FE model with GVT data has always been a troublesome bottleneck in this flutter-prediction process. Craig Mundt, a Dynamics Specialist at Cessna Aircraft Company in Wichita, Kansas, explains that aircraft manufacturers use separate systems for test and analysis. As a result, GVT and FE data comparisons occur across platforms and can be difficult to make.
“Historically, engineers would lay down plots of data side by side for a visual inspection,” says Mundt. “Comparing these pictures of data was tedious and time-consuming, as was trying to set up the test to align with data from the FE model. Weeks of effort were required, while the costs of keeping a prototype plane on the ground for weeks or months is enormous.”
Flutter prediction requires determination of vibration modes for wing bending and torsion as well as control surface rotation. In addition, the model must also sort out higher-order modes in predicting dynamic loads, due to conditions such as wind gusts and engine vibration on flight surfaces as well as the fuselage and major structural components.
An integrated approach
To turn their complex requirements into a competitive advantage, Cessna began using an integrated LMS system in 1997 for GVT and dynamic model correlation. The system was initially used on the company’s Encore aircraft and subsequently on the Citation CJ2 model in 1999. The LMS system is currently being applied in the development of the Sovereign.
The portion of the system that ties together the test, analysis and correlation is LMS Gateway: a platform that combines test and FE data into a single dynamic model, provides a rapid comparison between test and FE results, and helps to identify root causes of problems when analysis results do not meet functional requirements.
For pre-test preparation and planning, a Pre-Test module in LMS Gateway enables engineers to find the best locations for sensors and shakers on the prototype plane. Next, the test is performed using a 192 channel LMS CADA-X system for gathering vibration data from series of sensors mounted on the aircraft excited by a set of four random-noise shakers. From these data, mode shapes are generated within CADA-X from the test data and fed into Gateway for comparison with expected vibration modes determined through FE analysis.
The comparison is displayed on a Modal Assurance Criterion (MAC) matrix showing where the test modes and analysis modes align and where they diverge, indicating exactly where the dynamic model must be adjusted. The model is modified, another comparison performed, and if these results match up, the final correlated model is ready for processing through MSC.Nastran to accurately compute flutter airspeed.
Major benefits for Cessna
“The LMS solution brings significant advantages for us in integrating the testing and correlation operation, and in being able to develop a highly accurate dynamic model,” says Mundt. “We are now able to define the tests in a day or two. And when we are running the GVT, the flutter analyst is sitting there side by side with the testing personnel. We receive results immediately and do not have to wait while files are formatted and translated from one program to another.”
Mundt particularly likes the convenience and accuracy of bringing test and analysis together in Gateway. “Comparing test and analysis plots was tedious, and there was always the chance that points would be missed,” he says. “With the animated display and color-coded MAC in Gateway, there is less chance for error.” According to Mundt, besides saving time for Cessna, the LMS integrated system allows a more accurate prediction of flutter speed, “LMS lets us produce a better dynamic model. And a better model means you have a better handle on where the risks are when you go to flight-flutter-test an airplane.”
For Cessna, this translates into a clear business value and competitive advantage in a fiercely competitive industry segment where fast time to market and an absolutely reliable product is imperative.
“The integrated LMS solution is a key tool allowing us to get the Sovereign to market faster,” Mundt says. “Without LMS, it would take us longer to complete the same tasks. This system has helped us improve our processes and quality.”
