Though NASA has built small, high-speed aircraft called scramjets that can travel at speeds up to Mach 10, this engine technology is not ready for commercial aviation. More work must be done to make the engines more reliable, safe and efficient, and hypersonic engine airflow models, such as the ones developed by Martin, may be the key.
At such extreme speeds, it is nearly impossible to model airflow in a traditional wind tunnel, so computer models become essential for preliminary work, explained MAE professor and department chair Alexander Smits, who is a member of Martin’s team.
“These simulations will lead to a better understanding of the physics involved, which will allow us to develop better laws and models,” Martin said in a statement from the engineering school. “This new understanding may lead to novel design concepts that enable hypersonic travel as well as safe and affordable access to space.”
Martin’s team has succeeded in creating a model that is able to recognize shockwaves. Typically, this is a major challenge in modeling airflow, but it is critical for reaching a high level of computational precision. These shockwaves are extremely small, powerful and abrupt disturbances in the airflow, Smits explained.
“If these [shockwave] interactions are very strong, the airflow can separate, and the engine can un-start, which is a sudden loss of thrust. This happens even in hypersonic aircraft such as the SR-71 Blackbird,” Smits said. In fact, un-start can cause engine failure, rapid decrease in speed and loss of control in an aircraft.
“We need to understand what happens in the engines, and we need to make sure that these interactions are controlled and don’t lead to the most serious levels of un-start,” Smits said.
Martin’s computational models may aid in the development of improvements in hypersonic engine fuel efficiency, which has been a key deficiency to date, Smits noted.
He added that the use of computers has made it practical to process a large amount of numerical data and solve many of the large differential equations that describe the airflows, but that these processes are not without their own issues.
“You end up with a huge amount of data, and it’s a tremendous task to handle such a large database of numbers,” Smits explained, adding that the amount of data has made the analysis process very challenging.
“Typically, the kind of tools Professor Martin develops are at the forefront of science,” Smits said. “This is not something you can give to someone to build an aircraft engine since it involves highly specialized tools that have lots of inputs and outputs, and you have to be totally familiar with the whole process.”
Engineers who would work on designing the specifications for aircraft engines could not easily use Smits’ theoretical models in their current form, he said, explaining that it would be like putting someone who just got his license into a Formula 1 car and expecting him to win a Grand Prix race.
“[Martin’s modeling codes] can be used to develop more engineering-type codes that come close to giving the same results as the Formula 1 car but are much more oriented toward design requirements,” Smits said.

Though these engineering codes won’t be as accurate as Martin’s theoretical models, they are much more user-friendly, and they are necessary for practical applications, he added.