Detailed Uncertainty Analysis for Ares I Ascent Aerodynamics Wind Tunnel Database

Abstract

A detailed uncertainty analysis for the Ares I ascent aero 6-DOF wind tunnel database is described. While the database itself is determined using only the test results for the latest configuration, the data used for the uncertainty analysis comes from four tests on two different configurations at the Boeing Polysonic Wind Tunnel in St. Louis and the Unitary Plan Wind Tunnel at NASA Langley Research Center. Four major error sources are considered: (1) systematic errors from the balance calibration curve fits and model + balance installation, (2) run-to-run repeatability, (3) boundary-layer transition fixing, and (4) tunnel-to-tunnel reproducibility. Nomenclature A101Ares This material is declared a work of the U.S. Government and is not subject to copyright protection in the United States.https://ntrs.nasa.gov/search.jsp?R=20080024105 2018-05-11T10:44:17+00:00Z Their primary differences are in the launch abort system, the blast shield covering the Orion Command Module, and the various protuberances. The identifiers for those configurations that are used in this paper are A101 for an earlier configuration and A103 for the current configuration. The designs are sufficiently different that the aerodynamics changes must be accounted for in the database. An artist's sketch of the current configuration at launch is shown in Figure 1. Note that there are three first-order loading regions for longitudinal aerodynamics: (1) crew capsule/service module, (2) interstage, and (3) aft skirt. Each protuberance creates a loading region which is secondorder for longitudinal aerodynamics, but which is first-order for lateral-directional aerodynamics.Each configuration was tested at the Boeing Polysonic blow-down wind tunnel (PSWT) for the Mach range, . The PSWT testing was conducted in the transonic test section which is four-foot square and has porous walls. Each configuration was also tested at the NASA Langley Research Center Unitary Plan Wind Tunnel (UPWT) for the Mach range, 1.6. The UPWT tests were conducted in two test sections: test section 1 for and test section 2 for 2.5 0.5The UPWT is continuous flow and both test sections have solid walls four-foot square. The PSWT tests used the NTF-107 force balance while the UPWT tests used the UT-39B balance. The full-scale calibration ranges (same as the maximum loading ranges) for the balances are given in Table 1. Also, given in Table 1 are the standard errors derived from the calibration curve-fit residuals. For all four tests, the force balances were attached to a straight sting which was attached to a roll motor and then to the tunnel mounting system. Base and cavity pressures were measured to correct them to free stream . Both pitch runs at constant zero roll angle and roll runs at constant pitch angles were obtained. The PSWT tests used continuous pitch and roll with data acquisition (digitization) at 100 frames a second that was post-processed with a digital filter at 20 Hz. The UPWT tests used pitch-pause and roll-pause data acquisition, acquiring data at 30 frames per se...

Assessment of CFD-based Response Surface Model for Ares I Supersonic Ascent Aerodynamics

“…Figure 7 shows the A103 validation residuals (CFD minus WT DB data), the estimated bias uncertainty term (the solid purple line with square markers), and the random uncertainty term (the dotted purple bounds about the bias). The uncertainty intervals for the A103 AFMA DB 24 are shown as dashed red lines in the figure for reference and are symmetric about zero since a validation residual equal to zero indicates no difference from the DB value. Table 4 lists the computed estimates for the code-to-code uncertainty and the bias and random terms of the CFD validation uncertainty at 8 degrees total angle of attack, reported as percentages of a reference value for each coefficient.…”

“…In the figure, the CFD data are shown as red circles with the estimated code-to-code uncertainty shown as error bars, the DB data are shown as the blue solid line with diamond markers, and the shaded blue area bounded by the blue dashed lines represents the DB uncertainty intervals. 24 For each available A103 CFD solution (74 total for M ≥ 1.6), the UQ team computed a validation residual by subtracting the A103 AFMA DB data from the corresponding CFD data. The UQ team divided the CFD validation uncertainty into both a random and a bias term because the Ares aero team observed a consistent bias between the USM3D CFD solutions and WT data for the longitudinal coefficients over the course of the project (e.g., USM3D consistently predicted CAPF to be lower than the measured wind tunnel data).…”

Asymmetric Uncertainty Expression for High Gradient Aerodynamics

“…The article by Hemsch et al 13 describes in detail the process involved in the estimation of aerodynamic uncertainties for the Ares I launch vehicle during the ascent phase of flight. In most instances, when the data trends only exhibit small gradients, a pure magnitude uncertainty is largely sufficient to cover for potential phase errors, even though unaccounted for in the analysis.…”

Quantification of the Uncertainties for the Ares I A106 Ascent Aerodynamic Database

“…This approach is consistent with the way the A wind tunnel database was constructed in the previous design phase (A103). 1 The A106 database is a function of Mach number, pitch angle, and roll angle. The buildup of the final A106 ascent database is defined in Eq.…”