Moser indicated that there are several needs, including: math models for uncertainty, algorithms and software tools for computing with uncertainty, and characterization of experimental uncertainties. He presented an example of a heat flux gage where he was able to get an idea of the uncertainty in the measurements of that system. Another example he provided was on the NASA Orion vehicle, where uncertainty quantified in results lets NASA make more rational decisions regarding margins in the system design. Moser concluded with several recommendations: (1) ensure that rigorous code validation is applied to computational simulations; (2) develop modeling software with modern a posteriori analysis and adaptivity; (3) develop/adopt formulations and software tools for uncertainty quantification; and (4) tightly integrate physical modeling, uncertainty analysis and experimental programs to ensure reliable uncertainty assessments. During the discussion following the presentation, Moser was asked by a workshop participant how one deals with the absence of some physics. Moser responded that this is a challenge, and that in general you calibrate with the data available to the extent that you can do so. When asked about his comments on the NASA roadmap, Moser indicated that what he thought was missing was defining what is needed to simulate, and how it should be done. He commented that obtaining data to quantify uncertainty and reliability calculations can be difficult. Finally, responding to another question on the role of modeling and simulation as part of the design process, Moser suggested that in some situations modeling and simulation might be used to provide confidence in the system to be fielded.
Session 3: Thermal Protection Systems
The session on Thermal Protection Systems started with a presentation by one of the panel members, Don Curry (Boeing). Curry started with a table showing historical thermal protection system (TPS) mass fractions for several human-rated vehicles. In general, this was on the order of around 10 percent. Curry provided some discussion on different ablative materials, including the Apollo AVCO and Ames’ PICA materials. For carbon phenolic TPS, Curry noted that in many cases this is the only feasible TPS material for specific missions, yet the difficulty obtaining aerospace-grade Rayon is a significant issue for future missions. In terms of TPS testing, Curry mentioned how reusable TPS materials have had mission lifespans quantified via arcjets and other testing. He also provided some data on AVCO used for Apollo; in this case thousands of hours of testing were performed, along with multiple facilities. Curry noted that this was all necessary in order for the material to be available in time. Curry discussed the importance of testing, noting that the final design values for many properties (e.g., thermal conductivity of char) come from arcjet testing; likewise understanding material properties such as compression, shear, etc., is required. Related to TPS design, Curry highlighted many important considerations, including: aerothermal environment, strength/stiffness, thermal gradients, venting characteristics, outgassing, space environment, damage tolerance, repairability, and refurbishment. Finally, Curry concluded by emphasizing that test facilities are important to TPS development, and that eliminating facilities will lead to significantly increased risk. After his presentation, Curry was asked why Orion did not use PICA, as it has a low density but high heat of ablation. Curry’s response was that PICA is a tile system, and can potentially crack due to tension in the structure (there are also typically gaps between tiles to account for this). Curry also noted that in the past, PICA has had problems with shockloads during separation pyros.
Next, Bill Willcockson (Lockheed Martin) gave a presentation on TPS materials. Starting out his discussion talking about past experience in robotic missions, Willcockson noted that Viking had hundreds of tests (potentially up to a 1,000), and that tests might be a good metric relative to human missions. For Jupiter/Venus entry (e.g., 10,000 W/cm2 class), he noted that carbon phenolic can no longer be made, and that these types of materials cannot be tested without arcjet facilities. Relative to affordability, he commented that PICA is three times the cost (process-wise) versus SLA-561V, and that Lockheed Martin has been developing new materials such as MonA to address this. He noted that while SLA-561V was developed in the 1970s, it is important to keep older technologies like this up to date to avoid obsolescence issues. Regarding flexible materials, Willcockson suggested that these are at relatively low TRLs and maturing slowly. During his wrap-up, Willcockson highlighted the importance of investment in TPS; in particular: (1) the need for a resurrection of carbon phenolic, which may require rebuilding facilities as well; (2) the need for a mechanism to take advantage of experienced folks at large companies (e.g., similar to the SBIR program); (3) continued funding to maintain existing TPS materials; and (4) NASA program