Home Projects Heat Transfer

Developing a Workflow to Estimate Heat Transfer for High Altitude Flight

Deliverables

Premise

As part of the Thermal Sciences (AAE 33800) curriculum at Purdue University, students are tasked with creating a project that will address a pressing thermal sciences issue.

With Anish Agarwal, Shreesh Nalatwad, Noah Smith, and Will Stupart, we chose to create an interface capable of estimating heat transfer and approximating necessary insulation thickness to protect internals for high altitude flight vehicles, particularly in the case of sounding rockets. This was especially pertinent due to the volume of rocketry teams at Purdue in addition to the growing popularity of collegiate rocketry across the nation.

Development

We determined three primary methods of heat transfer affecting a sounding rocket: forced convection, convection due to compressibility effects, and conduction. We chose to neglect the effects of radiation for simplicity and due to its presumed negligibility. The rocket experiences forced convection due to the high speeds with which the air contacts the airframe, while the conduction occurs in between the outer and inner airframe walls.

This would normally be a simple calculation if the rocket's altitude was fixed, but it is clear that the thermal properties of air will change depending on altitude while the rocket's altitude also behaves variably. For that purpose, the interface utilizes an iterative process that calculates heat transfer and airframe temperature at each time step based on inputted airframe properties and flight trajectory.

The governing equations for such an interface include simplified versions of Fourier's law of heat conduction and Newton's law of cooling in addition to compressible flow relations. While forced convection due solely to airflow does contribute some change in thermal characteristics of the airframe, the compressibility effects are even more significant, making the outer airframe temperature calculations a combination of the two heating methods. The air properties were linearly interpolated from a table provided by Haifeng Wang in his Thermal Sciences textbook, which allowed the ambient air temperature, density, pressure, kinematic viscosity, and specific heat capacity to be precise at any altitude.

Outcomes

To test the accuracy of the interface, we decided to find publications of heat transfer measurements taken during the flights of sounding rockets. When inputting the parameters of the sounding rockets into the interface, we were able to calculate the peak airframe temperature in MATLAB with ~7% error when compared to the actual peak heat flux in the publication. We also tested the interface's capability to estimate internal temperatures based on controlled insulation parameters; the MATLAB insulation simulations yielded an error of 9.91% for the inner airframe temperature compared to an Ansys Fluent/Transient Thermal Module simulation with the same insulation and airframe parameters. Of course, these simulations could be completed in 5 seconds or less, essentially leading to a potential simulation time reduction of 90% or more when compared to the Ansys Transient Thermal Module.