Triggered Lightning Risk Assessment for Reusable Launch Vehicles

Triggered Lightning Risk Assessment for Reusable Launch Vehicles at Four Regional Spaceports Karen Shelton-Mur, FAA/AST Dr. Richard L. Walterscheid, The Aerospace Corporation Presented to COMSTAC RLV Working Group Meeting 18 May 2010 © The Aerospace Corporation 2009

Contributors • • • FAA/AST-300 Lead: Karen Shelton-Mur Volpe Center COTR: Ruth A. Hunter Aerospace Principal Investigator: Dr. Richard L. Walterscheid, 310 -336 -7352, Richard. Walterscheid@aero. org Aerospace Program Manager: Bob Seibold, 310 -336 -1326, Robert. W. Seibold@aero. org Electric Field Estimates, Effects of Plumes, LCC Rationales and Changes: Dr. John C. Willett, AFRL, Ret. LCC Rationales and Changes, Electric Field Measurement Techniques, Preliminary Evaluation of CIP Index: Prof. E. Philip Krider, Department of Atmospheric Sciences, U. Arizona Weather Compilations and Analysis: Dr. Grace S. Peng, Dr. Lynette J. Gelinas, Aerospace RLV and Spaceport Descriptions and Specifications: Glenn W. Law, Aerospace Plume Conductivity and Electric Field Enhancement: Dr. Frederick S. Simmons, Dr. Paul F. Zittel, Aerospace 2

Objectives • To provide an assessment of weather impacts on suborbital RLVs. Specifically, to identify natural and triggered lighting hazards to safetycritical systems on four conceptual RLV designs during launches and reentries at four inland sites and to determine mitigation methods for each hazard. • The sites are: – – • Spaceport America (SA) Oklahoma Spaceport (OS) Mojave Air and Space Port (MJSP) West Texas Launch Site (WTLS) To assess use of icing climatology as a proxy for lightning conditions. 3

Representative suborbital vehicle concepts 4

Four regional spaceports Oklahoma Spaceport (OS) Spaceport America (SA) • Spaceport America near Upham, NM, ~45 mi. N. of Las Cruces, 30 mi. E. of Truth or Consequences. • Western boundary of White Sands Missile Range. • • 5 Oklahoma Spaceport developed at Clinton. Sherman Industrial Airpark at Burns Flat, OK, ~100 miles W. of Oklahoma City. OS encompasses ~3, 000 acres with 2 runways, 13, 500 & 5, 200 ft

Four regional spaceports (cont. ) Mojave Air and Space Port (MJSP) • • Mojave Air and Space Port located on east side of Mojave, CA. 3, 000 acres, 3 runways: 9, 502, 7, 050, & 3, 943 ft. 6 West Texas Launch Site (WTLS) • Culbertson County ~ 25 mi. north of Van Horn, TX. Within larger privately-owned Corn Ranch property

Hazards Associated with Natural and Triggered Lightning 7

Lightning Strikes on Aircraft (RLVs) • • • Direct effects include physical damage from currents Indirect effects include electronic damage from fields Non-metallic vehicles more susceptible to both especially if “non hardened” Source: Uman, M. A. , Art and Science of Lightning Protection, Cambridge, 2008 Source: Lightning Technologies, Inc 8

Indirect Lightning Effects on Aircraft – Based on Fisher et al. , 1999 9

Video of Triggered Lightning Strike to Aircraft • Usually unexpected • Often occurs when natural lightning is absent • Remote sensing unable to detect hazard • At least 80 – 90% of aircraft strikes are triggered Video of B 747 on takeoff from Kamatzu Air Force Base on the coast of the Sea of Japan during winter. (Courtesy of Z. I. Kawasaki) 10

Schematic of Rocket-Triggered Lightning • Triggered lightningits energy from the Lightning draws requires – High electric field ambient electrostatic electric field – Sufficiently large amount of energy stored in the field • – Conductivity of vehiclein many cloud Sufficient fields found causes equipotential to drape over vehicle types – The potential difference that extends over the combined extent of the vehicle and conducting plume is concentrated near tip – Field further enhanced by vehicle curvature – The combined effects can cause the field to exceed its breakdown value Rivera et al. , Protecting Space Systems from Lightning, Crosslink, Aerospace Corporation, 2001. 11

Lightning Avoidance • • • Meteorology is primary determinant of triggering potential – Cloud /rain-based Lightning Flight Commit Criteria (LFCC*) • Safe but many false alarms Based on several data sources • Aviation experience • Small rocket experiments • Research on cloud electrification • Airborne Field Mill (ABFM) campaigns Avoid certain cloud types and conditions by LFCCs* – – – Natural lightning Cumulus clouds (depending on cloud-top temperature) Anvils and debris clouds (depending on transparency and age) Disturbed weather (rainy conditions) Thick cloud layers (depending on temperature) * Found in 14 CFR Part 417 – Appendix G 12

Study Results 13

Lightning Strikes Measured within 100 km of SA by the NLDN for July 1999 Each dot represents one lightning strike observed by the NLDN during July 1999 14

Lightning Probability by Day of Year Based on NLDN data WTLS OS MJSP SA • Variation of per cent of hours with lightning within a 100 km radius versus day of year. – Solid red line is running mean 15

Lightning versus Time of Day Percentage of hours with lightning versus universal time (UT) for four months: • MJSP: Local time (LT) = UT – 8 hrs • WTLS: LT = UT – 7 hrs 16

Lightning at Launch Sites Approximate range of days of year (DOY) with greater than indicated probability of lightning within 100 km Site Approx. DOY Range > 20% CCAFS 150 -275 SA 160 -260 Approx. Date Approx. DOY Range >20% Range > 40% Late May/Early Jun 175 -250 to Late Sep/Early Oct Approx Date Range >40% Mid Jun to Mid Sep 190 -240 Mid July to late Aug No significant activity No significant activity OS 150 -240 MJSP No significant activity Late May/Early Jun to Mid Sep 15 Late May/Early Jun to Late Sep/Early Oct No significant activity VAFB No significant activity WTLS 150 -260 Late Jun to Early Sep VAFB = Vandenberg AFB, CCAFS = Cape Canaveral Air Force Station 17

Lightning at Launch Sites (continued) Times of minimum and maximum probability of naturally occurring lightning in July Site Minimum - Local Time (UT) Occurrence frequency Maximum- Local Time (UT) Occurrence frequency CCAFS 05 L (12 UT) 15 14 L (19 UT) 90 SA 6 -10 L (13 -17 UT) 0 17 L (24 UT) 75 WTLS 4 -6 L (10 -12 UT) 0 16 L (22 UT) 50 MJSP 10 L (18 UT) 0 16 L (24 UT) 30 OS 9 -12 L (15 -18 UT) 0 16 L (22 UT) 15 No significant activity at VAFB 18

Climatology Summary • 15 -year climatology of naturally-occurring cloud-to-ground lightning strikes based on data from the National Lightning Data Network showed: – Of the spaceports, MJSP is the most lightning-free site. The most lighting prone site is SA. – Peak season is summer at all locations • All sites except MJSP show strong seasonal variation – Peak hours are mid- to late afternoon. • All sites show strong diurnal variation during summer 19

Risk • Risk is the product of the frequencies of hazardous conditions and the occurrence frequency of triggering fields given the hazard – Event climatology – Effective conducting length of vehicle plus plume – Electric fields found to be associated with the rules 20

Plume effects: electrical effective length Rocket Conducting Plume D Non-conducting Plume L Possible Range of Effective Lengths: (D+L)/2 = Conventional boost phase assumption ≤ D/2 = Aircraft and glide phase D+L = Boost phase if conducting plume acts as potential equalizer Large Uncertainties in Plume Effects 21

Simulation results • Notice that, except very close to the nozzle, the exhaust time-of-flight greatly exceeds the electrical relaxation time at all locations – Relaxation means electrons can respond to the electric field making it act as a conductor. – When the relaxation time is too long the plume behaves as an insulator • Suggests that the effective electrical length of this plume is much greater than 40 m. Plot of time of flight (black) and electrical relaxation time tl (red) in an exhaust parcel vs. downstream distance at a rocket altitude of 9. 4 km. 22

The level of hazard depends on cloud type and temperature Altitude Cloud top and isotherm heights for selected cloud types for July at MJSP and WTLS Cb = Cumulonimbus Day 5 C, -10 C, -20 C 23

Overall triggering probability by launch site and vehicle Annual average of unconditional probability of triggering summed over all hazardous conditions Launch Site Vehicle Number Vehicle Type 1 HTHL w/ Jet Engines 24 2 Air Launch 3 HTHL w/ Rocket Engines 4 VTVL

Risk results • • Seasonally and time of day dependent Depends on vehicle – Least for concept 2 (Air Launch) • Depends on site – Least for MJSP 25

Lightning versus Current Icing Potential (CIP) Index • The meteorological conditions that produce high electric fields in clouds are very similar to the conditions present in an in-flight icing hazard to aircraft – Aircraft icing is most severe in mixed-phase clouds that contain both supercooled water drops and ice particles in close proximity • The Current Icing Potential (CIP) was developed for the FAA by NCAR and quantifies the potential for clouds to produce icing on aircraft – This index combines data from a number of sources (radars, aircraft, satellites, radiosondes, etc. ) and applies the relevant physics to identify supercooled water in clouds 26

Correlation between CIP and lightning CIP at 2 UT on 2 April 2006 (left panel) and CG lightning flashes (right panel) detected by the NLDN in during a 1 -hour interval following the time of the CIP map. 27

Correlation between CIP and lightning (continued) • The variation of lightning occurrence within 100 km against CIP bins for four months over all years and over all four Spaceports • Based on a four-year CIP data set Occurrence Frequencies Mo. Bin 0. 2 -0. 4 -0. 6 -0. 8 -1. 0 Jan 0. 00 0. 01 0. 04 0. 05 0. 04 Apr 0. 04 0. 15 0. 21 0. 45 Jul 0. 16 0. 40 0. 41 0. 44 0. 56 Oct 0. 04 0. 20 0. 19 0. 28 0. 48 Totals • 0. 0 -0. 2 0. 07 0. 15 0. 19 0. 22 0. 35 We conclude that the CIP index and occurrence of lightning are highly correlated – When combined with other data, may be a useful operational indicator of lightning potential 28

Conclusions • • • The nature of the lightning hazard – Triggered lightning is the primary threat Lightning and cloud climatologies show lightning frequency is – Greatest for SA, least for MJSP – Greatest in July, afternoon/evening Estimate of triggering conditions – Depends strongly on plume length – Simulations indicate that conducting plume may be much longer than usually assumed Risk assessment – MJSP site and Concept 2 vehicle (Air Launch) have least overall risk Relevancy of existing LFCC – Relief for small and moderate cumulus clouds seems promising Lightning versus Current Icing Potential (CIP) index – Strong correlation between the CIP index and lightning occurrence 29

Backup Charts 30

Applicability of LFCC to RLVs • Most important rules should apply to all vehicles: – Natural lightning • Direct/indirect threat natural/triggered lightning – Large cumulus and cumulonimbus – Anvils and debris clouds – Surface electric fields (where measured) • Best ground-based indication of hazard aloft – Smoke-plume cumulus • Not sufficiently understood 31

Vehicle Hardening • Direct effects – Design external surfaces and associated hardware (air data probes, antennas, radomes, navigation lights, etc. ) to withstand lightning currents – Design fuel system to withstand lightning currents • Indirect effects – Make topology of aircraft wiring such that induced transients or (or ‘pick-up’) are minimized – Apply transient surge suppressors at key points in avionics systems – Check design by applying suitable tests 32

Lightning probability by day of year • Variation of per cent of hours with lightning within a 100 km radius versus day of year. – Based on NLDN data – Solid red line is running mean WTLS MJSP CCAFS 33