LIBRARY OF THE UNIVERSITY OF ILLINOIS AT URBANA-CHAMPAIGN 629.13 I-£6a no. 11-21; The person charging this material is re- sponsible for its return to the library from which it was withdrawn on or before the Latest Date stamped below. Theft, mutilation, and underlining of books are reasons for disciplinary action and may result in dismissal from the University. UNIVERSITY OF ILLINOIS LIBRARY AT URBANA-CHAMPAIGN *OVfS»73 AUG 3 JUL 12 197! FEB 16 TO JAN27U79 0EC30K79 infeo BOOING UV£ Ofti-l MAR 2 1338 DEC 1 1SP2 FEB 2 7 m JUN MAY ric FEB AUG 161994 6 199^ 3*96 7 1996 y 001 5 9 2008 L161 — O-1096 ^ : : BULLETIN PARACHUTE FLARES as an Aid in Night Forced Landings .u.it»° AERONAUTICS BULLETIN NO. 17 UNIVERSITY OF ILLINOIS INSTITUTE OF AVIATION Leslie A. Bryan, Ph.D., LL.B., Director James M. Hancock, A.B., Editor UNIVERSITY OF ILLINOIS BULLETIN Volume 53, Number 22; November, 1955. Published seven times each month by the University of Illinois. Entered as second-class matter December 11, 1912, at the post office at Urbana, Illinois, under the Act of August 24, 1912. Office of Publication, 207 Administra- tion Building, Urbana, Illinois. by JESSE W. STONECIPHER Chief Flight Instructor MILLER R. GREEN Flight Instructor RICHARD E. MANKUS Flight Instructor OF THE UNIVERSITY OF ILLINOIS PARACHUTE FLARES as an Aid in Night Forced Landings THE LI3RARY OF TK? DEC 16 1955 UNIVERSITY OF ILLINOIS PUBLISHED BY THE UNIVERSITY OF ILLINOIS URBANA 195 5 Foreword This monograph is the seventeenth in a series of bulletins on aviation subjects issued by the Institute of Aviation. It reports the significant re- sults of a project to determine the feasibility of using parachute flares for executing night forced landings in light civil airplanes. It recommends also standard operating procedures in connection with such landings. The entire project was made possible by officials of the International Flare-Signal Division of Kilgore, Inc., of Westerville, Ohio. They pro- vided the parachute flares used and the funds for publication of this bulletin. These funds and flares were provided without any restrictions. The investigators who conducted the project, Mr. Jesse W. Stone- cipher, Chief Flight Instructor, and Messrs. Miller R. Green and Richard E. Mankus, Flight Instructors, are on the staff of the Institute of Avia- tion. All are seasoned pilots, averaging more than 4,000 hours each of flying, and have had extensive experience as both flight instructors and charter pilots. The authors gratefully acknowledge the assistance and advice of the following people in the preparation, conduct, and completion of this study: Mr. Herbert Pinzke, who provided the illustrations; Mr. Ronald S. Geiler, photographer; Messrs. G. J. Bishop, R. S. Long, L. E. Day, and Dr. H. C. Clauser, Kilgore, Inc.; Mr. Karl Aron, The Glenn L. Martin Co.; Messrs. Harry Demeris and Robert Fryer, Linemen, Insti- tute of Aviation; Messrs. F. L. Lancaster and F. B. Schaber, Aircraft Maintenance Shop, Institute of Aviation; and the entire staff" of pilots and flight instructors of the Institute of Aviation who generously donated their time to participate as subject pilots for the simulated night forced landings. In this bulletin, as in all publications of the Institute, the authors have had complete freedom to express their opinions, with the under- standing that they assume sole responsibility therefor. LESLIE A. BRYAN, Director July 1955 TABLE OF CONTENTS FOREWORD INTRODUCTION 4 THE PROBLEM 4 CASE STUDY PROCEDURES 4 FLARES 5 PILOTS 5 AIRPLANES 5 LANDING SITE 8 SIMULATING THE NIGHT FORCED LANDING 9 BRIEFING THE PILOT 11 FLARE CONSIDERATIONS 11 BURNING TIME 15 RATE OF DESCENT 15 DRIFT DISTANCE 15 ILLUMINATION AREA VS. INTENSITY 16 SPECIAL PROBLEMS 17 ELECTRICAL FLARES 19 PISTOL FLARES 24 FLARE VS. AIRPLANE PERFORMANCE 28 A SIMULATED NIGHT FORCED LANDING 31 PHASE I— GENERAL AREA SELECTION AND PREPARATION 31 PHASE II— SPECIFIC FIELD SELECTION AND POSITIONING 32 PHASE III— APPROACH AND LANDING 34 SPECIAL PROCEDURES FOR THE PISTOL FLARE 38 PRACTICE IN A SPECIFIC AIRPLANE 38 TRIAL RESULTS 39 CONCLUSIONS 40 BIBLIOGRAPHY 41 Introduction Since 1946 there has been a tremendous increase in the use of the light airplane as a means of transportation. Individuals, businessmen, and corporation executives, in varying degrees, have accepted the light air- plane as a supplement to their transportation needs. One factor which has limited the utility of the light single-engine airplane is the concern which many pilots feel over the possibility of a total engine failure at night, thus necessitating a forced landing in dark- ness. There is a wide variation in the opinions expressed by lightplane pilots regarding this possibility. There is little doubt that more use would be made of the single-engine airplane for night transportation if it were possible to provide a reasonable solution to potential night forced landings. As an approach to the problem of night forced landings the staff of the Institute of Aviation in 1950 undertook a survey of the aviation industry for information on the use of the parachute flare. Unfortunately, there was very little information available. The Institute of Aviation, however, did establish a standard procedure for the guidance of Univer- sity of Illinois pilots. In January 1955, officials of the International Flare-Signal Division of Kilgore, Inc., were approached by Institute staff members with a pro- posal to investigate the use of parachute flares as an aid to the execu- tion of night forced landings in light civil airplanes. Kilgore officials indicated their willingness to cooperate, and the study was undertaken by the Institute of Aviation in April 1955. The Problem The project was designed ( 1 ) to determine the feasibility of using parachute flares, both pistol-fired and electrically-fired, for night forced landings in civil aircraft of less than 3,500 pounds gross weight, and (2) if parachute flares were found suitable, to establish standard operat- ing procedures for the safe and efficient use of such flares. Case Study Procedures Numerous conferences were held during which all available informa- tion on flare characteristics was discussed in terms of the specific problems involved in an actual night forced landing. These discussions established certain unknown factors which needed investigation. These factors were: 1. I low much area and how much detail can be seen from altitude using a flare for illumination? 2. What arc the best altitudes for using such flares? 3. What type of flight pattern would be best to obtain optimum assist- ance from the flare? The discussions culminated in a decision to base the entire project on a scries of simulated night forced landings using a variety of pilots as subjects. Standard check lists were prepared upon which the investiga- tors, riding as observers during the execution of the simulated forced landings, could tabulate data on the unknown factors. The check list also included space for checking upon certain information on the flare per- formance furnished by the manufacturer. In addition, it contained space for remarks on unusual incidents as well as space for diagramming the entire process. It was assumed that factors learned during each simulated night forced landing would be taken into consideration in preparation for each subsequent attempt. FLARES A total of 40 parachute flares were fired during the study. This total was composed of 16, Class 3, 1 -minute white light pistol-fired flares; 12, Class 2, l ! /2-minute white-light electrically-fired flares; and 12, 30-second red-signal lights (37-mm. parachute distress flares). All flares used were manufactured by the International Flare-Signal Division of Kilgore, Inc., Westerville, Ohio. In an effort to establish flare reliability and to obtain data on renova- tion recommendations several flares which had been manufactured by the same company two, three, four, and five years earlier were substi- tuted for some of the new flares. The total number of flares fired con- sisted of 12 new red-signal lights; 10 new 1 -minute pistol flares; 6, 1 -minute pistol flares over 3 years old; 6, lV^-minute electrical flares over 2 years old; and 6, l!/2-minute electrical flares over 3 years old. Table # 1 shows the date of manufacture, date fired, type, serial number, and any irregularities in flare operation. PILOTS Ten staff flight instructors and three pilots not employed by the Uni- versity were used during the course of the study. Only one of the 13 pilots had previously fired parachute flares, and only one had previously experienced a partial power failure necessitating a forced landing at night. Table # 2 shows, by pilot number, the C.A.A. ratings held, the flight experience, and the year each pilot started flying. AIRPLANES All Class 2, lVfc-minute electrical flares were fired from Beechcraft Bonanzas. Four different Bonanzas owned by the University of Illinois were used. 5 TABLE NO. 1. FLARES FIRED Date of Manufac- ture Date Fired Type Serial Number Remarks on Operation 11-13-50 5- 2-55 IV2 rnin. 17420 OK 11-13-50 5- 2-55 1 Vz min. 17421 Flare failed to expel from projectile case, did not gn 11-13-50 5- 2-55 1 V2 min. 17422 OK 7-16-51 5- 9-55 1 min. 6804 OK 7-16-51 5- 9-55 1 min. 6805 Candle burned out in 57 seconds 7-16-51 5- 9-55 1 min. 6802 OK 7-16-51 5- 9-55 1 min. 6803 OK 7-16-51 5-10-55 1 min. 6806 OK 7-16-51 5-10-55 1 min. 6807 OK 8-14-51 4- 8-55 l!/2 min. 17874 Master circuit breaker switch defective 8-14-51 4- 8-55 1 Vz min. 17875 Master circuit breaker switch defective 8-14-51 5- 2-55 V/z min. 17876 OK 12-12-52 6- 6-55 1 Vz min. 18923 OK 12-12-52 6- 6-55 IV2 min. 18924 OK 12-12-52 7- 5-55 T/2 min. 18925 Ejected but failed to ignite 2-13-53 5-20-55 V/z min. 19002 Projectile case failed to eject from projector tube 2-13-53 5-20-55 V/z min. 19003 OK 2-13-53 7- 5-55 IV2 min. 19001 OK 2-55 5-20-55 1 min. 8334 OK 2-55 6-20-55 1 min. 8335 OK 2-55 5-10-55 1 min. 8336 OK 2-55 5-20-55 1 min. 8337 OK 2-55 6-24-55 1 min. 8338 OK 2-55 6-24-55 1 min. 8340 OK 2-55 7- 5-55 1 min. 8342 OK 2-55 7- 5-55 1 min. 8344 OK 2-55 5-10-55 1 min. 8345 OK 2-55 6-20-55 1 min. 8341 OK 2-55 5- 9-55 30 sec. Lot 15 OK 2-55 5- 9-55 30 sec. Lot 15 OK 2-55 5- 9-55 30 sec. Lot 15 OK 2-55 5-10-55 30 sec. Lot 15 OK 2-55 5-10-55 30 sec. Lot 15 OK 2-55 5-10-55 30 sec. Lot 15 OK 2-55 5-20-55 30 sec. Lot 15 OK 2-55 5-20-55 30 sec. Lot 15 OK 2-55 5-20-55 30 sec. Lot 15 OK 2-55 6- 6-55 30 sec. Lot 15 OK 2-55 6- 6-55 30 sec. Lot 15 OK 2-55 7- 5-55 30 sec. Lot 15 OK The Class 3, 1 -minute Pistol Flares were tested under simulated night forced landing eonditions in a Cessna 170, a Cessna 140, a Piper Tri- Pacer PA 22, and a Stinson 150. Prior to the simulated forced landing it was necessary to establish the average rate of descenl of each airplane used. This was done by execut- TABLE NO. 2. SUBJECT PILOT EXPERIENCE Pilot No. C.A.A. Ratings Started Flying Flight Experience Ever Fired Flares Before? Ever Had Night Cross- country Total Night Forced Landing? 1 Commercial — ASMEL, SES, Flight Instructor — Airplane, Instrument 1944 700 2000 4200 No No 2 Commercial — ASMEL, Flight Instructor — Airplane, Instrument 1946 293 1002 2473 No No 3 Commercial — ASMEL, MES, Flight Instructor — Airplane, Instrument 1942 1000 1700 3000 No No 4 Commercial — ASEL, Flight Instructor — Airplane 1942 44 164 906 No No 5 Commercial — ASMEL, Flight Instructor — Airplane, Instrument 1946 152 512 2100 No No 6 Commercial — ASEL, Flight Instructor — Airplane, Instrument 1952 110 240 1200 No No 7 Commercial — ASMEL, Flight Instructor — Airplane, Instrument 1940 1100 2500 5000 No Yes 8 Commercial — ASMEL, Flight Instructor — Airplane, Instrument 1939 115 315 3285 No No 9 Commercial — ASMEL, Flight Instructor — Airplane, Instrument 1951 70 350 950 No No 10 Private — ASEL 1954 25 50 130 No No 11 Private — ASEL 1952 100 400 600 No No 12 Commercial — ASEL, Flight Instructor — Airplane 1951 65 250 610 No No 13 Commercial — ASMEL, SES, Glider, Helicopter, Flight Instructor — Airplane, Helicopter and Glider, Instrument 1940 616 2300 5050 Yes No ing a series of five power-off approaches to a landing from an altitude of 2000 feet actual. The rate of descent for each airplane included the total time, by stopwatch, from completely throttled power to actual touch- down using a 720-degree overhead approach. The airplane was loaded with a full complement of passengers on each attempt and the average rate of descent for the five trials was used. The average rate of descent obtained is shown in Table # 3. LANDING SITE Because of possible flare misfire and potential fire hazard to the prop- erty of others, all flares were dropped within the boundaries of the Uni- versity of Illinois Airport. This enabled two linemen, properly briefed on the extinguishing of flares, to stand by with a fire truck and sand to control such possibilities. No fires occurred although one flare did hit the ground while still burning; it was extinguished with two buckets of sand. A sod area of approximately 70 acres of blue grass, located near the center of the Airport, was designated as the simulated forced landing TABLE NO. 3. FLARE DATA AND PERFORMANCE CHARACTERISTICS Showing Source of Information Source Data or Characteristic Class 3 1 -minute Pistol Class 2 1 V2 -minute Electrical C.A.A. Minimum allowable burning time in minutes. C.A.A. TSO-C24, dated 10/10/50. 1 l'/2 C.A.A. Minimum allowable Light Intensity in candle- power. C.A.A. TSO-C24. 70,000 110,000 C.A.A. Maximum allowable rate of descent in feet per minute. C.A.A. TSO-C24. 550 550 C.A.A. Minimum number of flares required by C.A.R. Parts 42 and 43 for irregular or non-scheduled commercial night operation beyond 3-mile ra- dius of airport. 5 3 C.A.A. Maximum number of years recommended prior to factory renovation of flare unit. Aviation Safety Release No. 321, dated 7/1/49. 3 3 Mfg. Approximate retail cost per flare unit. $21.00 $35.00 Mfg. Weight in pounds and ounces per flare unit. 1 lb. 14 oz. 4 lb. 5 oz. Mfg. Momentary recoil in pounds. 100 150 Mfg. Horizontal distance in feet flare is projected upon firing. 67-82 67-82 Mfg. Parachute diameter in inches. 48 60 Mfg. Safety device to prevent accidental firing of flare unit. Safety catch on pistol Double switch arrangement Mfg. Safety device to prevent flare from igniting before leaving airplane. Same as center fire shotgun shell Interrupter fuse field. This area, triangular in shape, was approximately 2700 x 2000 x 1900 feet. It was bounded on all three sides by eoncrete runways. SIMULATING THE NIGHT FORCED LANDING Eaeh subjeet pilot was given a one-hour briefing session immediately prior to the simulated attempt at a night forced landing. The briefing session included details on flare performance characteristics in terms of the performance of the airplane to be used. Specific altitudes, times, rates of descent, and methods of firing were covered and a diagram of the flight pattern was drawn to indicate critical points. When the pistol-fired flares were to be used the pilot was allowed to fire a 30-second red-light distress flare on the ground following the brief- ing session. This procedure was adopted to acquaint each pilot with the TABLE NO. 3. Continued Source Data or Characteristic Class 3 1 -minute Pistol Class 2 1 V2 -minute Electrical Mfg. Minimum release altitude in feet to insure flare extinguishment before striking ground. 800 1000 U of 1 Recommended angle of fire. As nearly horizontal as possible Horizontal U of 1 Recommended maximum release altitude for minimum usable illumination of ground objects. (Actual altitude in feet) 1800 2300 U of 1 Recommended maximum release altitude (in feet) for minimum usable AREA of illumination at time of flare extinguishment. 1000 1200 U of 1 Desirable minimum altitude of flare to obtain minimum AREA of illumination for landing touch- down. 400-500 400-500 U of 1 Radius in miles of usable ground illumination at maximum recommended release altitude. '/2- 3 /4 3 /4-l U of 1 Maximum drift distance of flare in miles from release to extinguishment in 30 mph wind. Vi 3 /4 U of 1 Functional reliability of flares over 3 years but less than 5 years of age. 92% 92% U of 1 Functional reliability of flares over 2 years of age but less than 3 years. 100% 92% U of 1 Functional reliability of flares less than 2 years of age. 100% 100% U of 1 Recommended airplane position in relation to flare for optimum vision of ground objects. Below Below problem of recoil. Since exact recoil data were not available, it was impossible to answer this question to the pilot's satisfaction. It was felt that a preliminary ground firing would prepare him more adequately for pistol recoil. Twelve distress flares were fired. Each pilot agreed that the recoil of the distress flare was greater than that of the 1 -minute pistol-operated parachute flare. This response was not surprising since the distress flare is designed to propel the candle and parachute to an altitude of 200 feet vertically, whereas the 1 -minute parachute flare is only propelled a distance of 67-82 feet horizontally. One pilot sustained a sprained wrist as a result of firing the distress flare; no injuries were sustained in firing the 1 -minute parachute flares. Immediately following the briefing period the pilot was given a "dry run" simulated night forced landing. This consisted of flying through the CHECK LIST Flare Project Date. Moon Phase_ Takeoff Time. Flare # Type. Airplane Type N Official WX_ Flare #1 Flare #2 Flare #3 Time of Emergency Firing Altitude Time Flare Fired Time Flare Ignited Time Flare Extinguished Line Fire Relative Airplane Position Flare Relative Landing Area. Position Flare Relative Airplane Objects Visible by Flare Illumination . Time Airplane Landed Flares Ext. Before Striking Ground . . , Remarks or special observations:.^ Draw track of flight pattern used on reverse side of this sheet, and iden- tify landing area and positions where flares were fired. 10 complete procedure outlined during the briefing session except that no flares were fired. The pilot shouted "bang" at the flare-firing point and an observer riding in the airplane recorded the time with a stopwatch. Altitudes, airplane rate of descent, and patterns were also noted by the observer. Upon completion of the dry run any errors noted were called to the pilot's attention and the proper correction of procedure was sug- gested to him. In only one case was a second dry run necessary. After completion of the dry run the pilot was instructed to climb to a specified altitude and take up a given heading in straight and level flight. Airport attendants on the ground were instructed by radio to turn off airport lights. At a predetermined position the observer in the front seat closed the throttle and said "forced landing." The subject pilot then executed the previously practiced procedures in an attempt to land within the specified area using only flares and landing lights. This pro- cedure was based upon the use of two flares. The first flare was to enable the pilot to select the specified landing area and maneuver into a posi- tion and an altitude for an approach. The second flare was used to illuminate the selected area for the approach and landing. The observer recorded altitudes, times, patterns, and other pertinent data on the check list during the execution of each attempt. The briefing section of this bulletin covers in detail the discussion, photographs, illus- trations, and tables used in briefing each pilot for the simulated night forced landing. Briefing the Pilot FLARE CONSIDERATIONS There are three general types of parachute flares available as emer- gency illumination for night forced landings. The Class 1, 3-minute flare, is expensive and perhaps too bulky and heavy for installation in the av- erage lightplane. The other two varieties, however, lend themselves well to the problems of the lightplane owner. One of these, the Class 2, \ l A- minute flare, is permanently installed in the airplane and is fired electri- cally. The Class 3, 1 -minute flare is available either as a rack-mounted electrically-fired unit or in a portable case with a 37-mm. pistol. In this study only the Class 2, lVfc-minute electrical flare and the Class 3, 1 -minute pistol flare will be considered. In selecting the flare best suited for the individual need, the pilot should consider the following factors: 1. Weight 2. Burning time 3. Intensity and area of illumination 4. Cost 5. Loading, aiming, firing, unloading, and reloading tasks Photo No. 1. Components of the Class 2, 1 V? -minute electrical flare. Reading from upper right to lower left, (1) ignitor housing, (2) electrical ignitor, (3) wadding, (4) projector tube, (5) wadding, (6) projector tube cap, (7) components of the delay fuse and safety interrupter, (8) projectile tube and cap, and (9) flare candle with parachute attached and three wadding discs. Photograph #1 shows the components of the l!/2-minute electrical flare. From upper right to lower left the first three items are the ignitor housing, electric ignitor, and the wadding. The ignitor ignites the black powder contained in the crimped end of the long projector tube. When the black powder is ignited the projectile tube (lower metal tube) is expelled from the projector tube to a distance of 67-82 feet horizontally. At the same time the projectile is expelled a delay fuse in the fuse hous- ing (right end of projectile tube) is ignited. The burning of this fuse is prevented from entering the black powder charge of the projectile tube by an interruption device shown just below the fuse housing (consisting of the spring, the interrupter, and the interrupter button). If the projec- tile fails to clear the airplane the delay fuse burns out in Vh seconds and all further possibility of ignition of the flare candle ceases. When the projectile leaves the airplane the interrupter allows the delay fuse to burn into the black powder charge contained in the projec- tile tube. The black powder ignites the flare candle and expels both the candle and the parachute from the projectile tube. Gaps and wadding for both the projector tube and the projectile tube are shown at the 12 Photo No. 2. Components of Class 3, 1 -minute pistol flare. Reading from upper right to lower left, (1) 37-mm. pistol, (2) projectile shell, (3) wadding, (4) projectile tube, (5) flare candle with parachute attached, (6) five wadding discs, (7) push stick, (8) projectile tube cap. extreme left of the photograph. (Note that photograph #3 shows the de- tailed operation of the safety device to prevent flare ignition until the projectile case has been expelled from the airplane.) Photograph #2 shows the major components of the Class 3, 1 -minute pistol-fired flare, and, from upper right to lower left, the order in which the flare is assembled. In the upper right hand corner is the 37-mm. pistol that has, much like the .45 Colt Automatic, a safety device on the back of the grip. To fire the flare the safety device and the trigger must be squeezed simultaneously. To the left of the pistol muzzle is the shell which contains the powder and the small ring against which the projec- tile tube rests in the shell. The long metal cylindrical tube just below and to the left of the pistol is the projectile tube. This tube contains the flare candle and the parachute shown just below it. The entire projectile tube is expelled a distance of 67-82 feet when the trigger is squeezed. A delayed action fuse in the end of the projectile tube then ignites and expels the flare candle with parachute attached. The five circular discs are used for wadding and spacing the parachute and candle. The three-inch wooden stick shoves the projectile tube cap off the projectile tube. 13 ^PROPELLING CHARGE MAGAZINE (4)INTERRUPTER SPRING (6) EXPELLING CHARGE MAGAZIN Photo No. 3. Details of interrupter (safety device) on electrical flares. The interrupter will not allow flare candle to ignite until after the projectile case has been ejected from the projector tube. 14 Both type flares must meet certain minimum specifications laid down In the C.A.A. These specifications are included in the flan- data table on page 8. BURNING TIME The burning time of the flare is important because it sets the amount oi time during which the pilot will be able to see illuminated objeets on the ground. The pilot should not assume that it is absolutely necessary to have illumination during the entire period from engine failure to com- pletion of landing roll. Even though the flare burns out on final approach the pilot will still have had the advantage of the illumination to select the most suitable field, maneuver over it, and plan his approach. Seven of the test landings were completed with no assist from the final flare. If the pilot has selected the most suitable field and has had the use of the flare in planning his approach, the landing lights will give enough illu- mination to complete the landing with a reasonable chance of avoiding major obstacles on the ground. RATE OF DESCENT The maximum rate of descent of both types of flares is set at 550 feet per minute. In the simulated tests it was discovered that the rate of descent was somewhat less than 550 fpm. The slower rate of descent was probably a result of the lightening of the load on the parachute as the candle burned. It would appear that a maximum rate of descent of 550 feet per minute, or less, is desirable. Most airplanes have a higher rate of descent and can therefore descend below the flare, thus providing the pilot better ground visibility. A light from above or behind normally provides a viewer with better illumination. Occasionally it was necessary to dive the airplane slightly, immediately following firing the flare, in order to get below the flare more quickly and thus acquire a longer period of good illumination. DRIFT DISTANCE The distance the flare will drift during its burning time depends en- tirely upon the wind velocity. In a 15-mph wind the 1 -minute flare will drift V4 mile and the l!/2-minute flare will drift % mile during the burning time. In 30-mph wind these distances become Vi mile and % mile respec- tively. These drift distances are important both in selecting the field and in making the approach. In firing the field selection flare the pilot should know that his illumination will move downwind and thus should favor a downwind rather than upwind direction when searching for a suitable field. The drift of the second flare is of even greater importance. Unless the second flare is positioned correctly when fired it may be' too far downwind to be of any use after turning on final approach. It was for 15 this reason that 180-degrce overhead and/or 360-degree overhead ap- proaches were selected as best suited for use with parachute flans. These two patterns may be modified if necessary to allow the pilot to fire the approach flare at the extreme upwind end of the selected field and still make it possible to touchdown before the flare has drifted beyond the downwind end of the field. ILLUMINATION AREA VS. INTENSITY The illumination provided by both types of flares must be considered in terms of intensity and area. As the area of illumination is increased the intensity of illumination at any given point is decreased. In other words, the factors of illumination area and illumination intensity work in inverse proportion. Illustration #1 depicts this concept including the factor of drift and the effect all three factors have on what the pilot can expect to see in terms of drift, area, and intensity of illumination. Information provided by the manufacturer recommended 1200 feet actual altitude for best over-all illumination by the lV^-minute flare, and 800 feet for the 1 -minute flare. The early attempts at simulated forced landings were based on these altitudes but it was soon discovered that they did not give the necessary flexibility during a forced landing. "Usable" illumination on the ground was obtained with the lV^-min- ute flare up to 2300 feet actual altitude, and up to 1800 feet actual altitude with the 1 -minute flare. Under the l^-minute flare (110,000 candlepower), released at 2000 feet and with the airplane slightly below the flare altitude, there will be reasonable illumination of ground objects for a radius of 3 A to 1 mile in every direction from the flare's position. The 1 -minute flare (75,000 can- dlepower) released at 1500 feet above ground under the same conditions will give reasonable ground illumination for a radius of Vi to 3 A of a mile. This illumination will move downwind as the flare drifts and thus will supply an elliptically-shaped area of illumination during the burning period. In applying the factors of drift and descent, the area of illumina- tion becomes egg-shaped with the small end of the egg occurring at the end of the burning time of the flare. Light colors, due to reflective properties, stand out much better than dark colors. Concrete and white buildings, for example, are much easier to see than plowed fields or green foliage. The pilot will not be able to see details such as fences, small ditches, and gentle slopes with the 1 !/2-minute flare from 2000 feet. These details must be observed with the second, or approach and landing, flare. Since much of the available altitude will have been expended in selecting a field and maneuvering toward it, the second flare will be released at a lower altitude. Now there is a reversal of the previous problem. Because the area of illumination decreases quite rapidly as the 16 flare approaches the ground, the exact positioning of the second flare becomes extremely important. Exact positioning, however, is difficult to accomplish due to the uncontrollable element of drift. With increased release altitudes, exact positioning becomes less critical since a greater area will be illuminated. Illustration # 1 shows these principles. The pilot's first reaction in attempting to determine the release alti- tude for the second, or approach and landing, flare is to multiply the burning time by the rate of descent. A glance at illustration * 1 will show why this will not work. If the touchdown is timed to occur at exactly the time the flare extinguishes, the flare itself would also be on the ground, and, unless it happened to drift very near the touchdown point, the flare would be of no value as a means of illumination. This factor contributed to the failure of the first three attempts to obtain assistance from flare illumination on the final approach, the landing, and the landing roll. The flares, in two cases, were still burning but were too low and too far out of position to be of any value to the pilot. Experimentation and discussion on this point resulted in a decision to fire the second flare at an altitude which would position the flare at 400-500 feet altitude when the touchdown was made. This gave a reason- able area of very intense illumination and still permitted a margin of error for drift estimation in the resulting position. Thus, multiplying the burning time by the rate of descent plus 400-500 feet will give the rec- ommended altitude for releasing flare #2 for the approach and landing. Using this formula, the lV^-minute flare should be fired at an approxi- mate altitude of 1200 feet and the 1 -minute flare should be fired at 1000 feet above the ground. Summarizing considerations of illumination, drift, and rate of descent, the following three important points become apparent: 1. Flares should be released upwind of the desired area of illumina- tion to permit a 180-degree or 360-degree overhead approach to a landing; 2. The maximum release altitude for the 1 ^-minute flares should be 2000-2300 feet above ground, and 1500-1800 feet above the ground for the 1 -minute flare; 3. The minimum release altitude for a reasonable area of illumination for the lV2-minute flare should be 1200 feet, and for the 1 -minute flare 1000 feet above ground. SPECIAL PROBLEMS In discussing problems encountered with specific flares the pilot or owner must consider the electrical and pistol flares individually. There are particular problems of installation and methods of fire peculiar to each type. 17 a t 0) ( ELECTRICAL FLARES Electrical rack-mounted Hares must be permanently installed in the airplane. Experience indicates that these flares should be installed by a competent airframe mechanic who meticulously follows the instructions provided by the manufacturer. The structure supporting the flare must be capable of withstanding a momentary thrust of 150 lbs. recoil when the flare is fired. Since the electrical wiring is not a simple task, it should be done only by a qualified person. Entire hangars have burned to the ground as a result of the accidental firing of flares due to faulty wiring. The installation of electrical flares should not be taken lightly. It is possible that a slight dent in the projector tube will render the flare useless — this happened to one of the 12 electrical flares fired during the experiment. Photograph #4 shows the possible results of such a dent in the projector tube. This misfire occurred on flare #19002, manufactured and installed in a University of Illinois aircraft in Feb- ruary 1953. This flare failed to expel properly from the projector tube. The age of the powder and the dent in the tube were contributing factors. The pilot of the airplane was unable to detect any unusual Photo No. 4. Flare #19002, showing the location of the open parachute during flight. The flare was fired at 1200 feet above ground. The pilot was unaware of any unusual flight characteristics and did not know the flare had not left the tube until he disem- barked. 19 flight characteristics despite the fact that the open parachute created a tremendous drag. This flare proved also the effectiveness of the safety device, shown in detail in photograph #3, which prevents the electrical flare from igniting until it has been expelled from the projector tube mounted in the airplane. The pilot of this airplane was unaware of the flare condition in photograph #4 until he disembarked. (Note the slit torn in the aluminum tube by the parachute cable. Photo #5.) One electrical flare, #17421, manufactured in November 1950 and fired May 2, 1955, was ejected properly from the airplane but failed to expel from the projectile tube. This prevented the parachute from open- ing, and the entire projectile plummeted directly to the ground. The remains of this flare are shown in photograph #6. The experience with this flare reaffirms the recommendations made by the C.A.A. in Aviation Safety Release #321, and by the manufacturer, that all electrical and Photo No. 5. Flare #19002, manufactured and installed Febraury, 1953, and fired on May 20, 1955. A small dent in the projector tube plus ageing of black powder projector charge probably combined to produce the results shown. Note the slit cut in the projec- tile tube by the wire cable attached to the parachute. 20 Photo No. 6. Flare #17421, manufactured and installed in November, 1950, and fired on May 2, 1955. Deterioration of the black powder expellant charge due to age caused this flare to plummet directly to earth where it was extinguished with two buckets of sand. pistol-operated flares should be renovated at least once every three years. This flare failure was probably caused by the aging of the black powder. A third 1 ^-minute flare failed to ignite properly. This particular flare, #18925, was manufactured in December 1952 and was fired on July 6, 1955. The flare was not recovered until July 22, 1955. The projectile was exposed to several heavy rains during the period between firing and recovery. An examination was made to determine the reason for the igni- tion failure. Due to accumulated moisture it was possible only to surmise that the age, 2Vi years, and possible moisture absorption of the black powder had lessened its sensitivity to ignition. It should be noted that this was the only flare within the three-year recommended renovation period which failed to function properly. The three-year renovation period recommended by the manufacturer and endorsed by the G.A.A. in Aviation Safety Release #321 is based on the factors of age and moisture which have a deteriorating effect on the black powder charge and which weaken the explosive force. A condition of overage could definitely contribute to improper propulsion from the projector tube, or improper expulsion from the projectile case. The candle also is affected by moisture and age. These factors cause the candle to harden and could result in a shorter burning time. A minor problem was encountered with the "Flare Master Circuit- 21 KJ § 9 p$ Photo No. 7. Flare switching arrangement in the circuit-breaker panel of a Beechcraft C-35 Bonanza. Breaker Switch" installed in one of the four Bonanzas used. This switch- ing arrangement is not the standard switch box supplied by the flare manufacturer. The master switch in the Bonanza incorporates a circuit breaker which is designed to throw the switch automatically back to the "off" position if an overload exists in the flare electrical circuit. If no such overload exists this switch will remain "on" when moved to that position. It was found that the master circuit-breaker switch in one of the four Bonanzas was defective. When this switch was moved to the "on" position and released, it immediately returned to the "off" position although no overload existed in the circuit. This defective switch made it necessary to reverse the firing instructions provided by the flare manu- facturer; i.e., the individual flare switch was first thrown "on" and the flare was then fired by turning on the master circuit-breaker switch. This process had to be repeated for each flare fired since the defect in the circuit-breaker portion caused the master switch to react like a spring- loaded toggle switch. It would appear, from this experience, that the entire electrical circuit should be checked carefully every three years. This particular switch was installed in August 1951, and was used for the first time on April 8, 1955. 22 Photo No. 8. Interior view of 3 rack-mounted 1 V^-minute electrical flares. Located di- rectly aft the baggage compartment in a Beechcraft C-35 Bonanza. The standard flare switch box provided by the flare manufacturer incorporates no such circuit breaker in the master switch, and thus could not present the problem discussed in the foregoing paragraph. The man- ufacturer's switch box includes four two-position "on-off" toggle switches. Each switch has a mechanical locking device to prevent accidental oper- ation. The lock must be disengaged before any of the switches can be operated. The pilot should make a careful study of the flare switch box, or panel, to determine which type is installed in his particular airplane. He can thus be prepared to minimize any possible delays arising as a result of defective switches. Photographs #7 and #8 show the electrical switch panel and the flare rack arrangement installed in a Beechcraft C-35 Bonanza. In summary, the following recommendations regarding l!/2-minute electrical flares were validated by experiences encountered during the project: 1. Have all flares installed by a competent airframe mechanic; 2. Have the entire installation reinspected at least once every three years; 3. Return all flares to the manufacturer for complete renovation at least once every three years. 23 Photo No. 9. Portable flare and pistol case in the Cessna 140 & 170, Piper Tri-Pacer, and Stinson 1 50. PISTOL FLARES Since portability is the principal advantage of the pistol flare, the first problem to solve is the packaging of the flares and the pistol. The manufacturer offers a solution to this problem in the form of a light fibre case which will hold five flares and the pistol. The Institute of Aviation had already solved this problem by design- ing a portable case for two flares and the pistol with a lanyard attached. The lanyard should be long enough to provide full arm-length movement of the pistol, when hung on a convenient point in the cockpit. The purpose of the lanyard is to prevent accidental loss of the pistol while firing. Photograph #9 illustrates the easy portability of the case designed by the Institute. This flare case, when resting in its customary position on the baggage compartment shelf in the airplane, was not readily accessible for emer- gency use. In this position the pilot averaged 24 seconds in getting his hands on the case, extracting the pistol, loading it, and firing one flare. The problem was solved by requiring the pilot to prepare his flare case and pistol immediately after clearing the ground on takeoff. This con- sisted of standing the case, bottom down, on the floor between the pilot's legs. The pistol was removed from its slot and the lanyard placed around the control yoke. The pistol was then inserted handle upward on one of the two flares. In this position the equipment was ready for firing with a minimum loss of time. By bringing his knees together the pilot could easily hold the case in place while using one hand for the pistol and the other hand for controlling the airplane. The testing pilots first attempted to fire the pistol with the right hand. They found that this procedure was not satisfactory because the pilot normally occupied the left seat and consequently had to fire to the left. Right-handed firing under these conditions meant that the elbow, and probably the wrist, could not be straight. One of the pilots sustained 24 Photo No. 10. Line of fire from a Cessna 140 with no window modification. Note straight wrist and elbow. Left-handed fire allows shoulder to absorb recoil. minor cuts when the recoil shoved his arm and hand back into the air- plane. A decision was then made to fire the pistol with the left hand, since this method permitted a straight wrist and elbow and allowed the shoulder to take up most of the recoil. Pistol recoil was the first question raised by the testing pilots. As pre- viously noted, an attempt was made to give them a concrete answer while on the ground by allowing each one to fire, with one hand, a 30-second distress flare. All were quite impressed with this experience, and one of the pilots sustained a sprained wrist during the firing. This was not entirely surprising considering that the manufacturer recom- mends using both hands when firing this particular distress flare. The pilots agreed that the 1 -minute parachute flare had considerable less recoil than did the distress flare, and when quizzed about recoil after the simulated night forced landing most of them stated they noticed none. It is quite likely that the left-handed technique, permitting a straight wrist and elbow, was partially responsible for this. The line of fire for a Cessna 140 or 170 is illustrated by photograph #10. In firing a pistol flare, the pilot's first consideration must be to avoid hitting any portion of the airplane with the projectile; his second 25 consideration must be that of getting the line of fire as nearly parallel to the horizontal as possible. In some airplanes it is possible to fire somewhat forward and downward. In the Cessna 140 and 170 it is best to employ a backward and downward line of fire. Only one airplane presented the problem of adequate space for the pilot to get the pistol flare outside. This was the Tri-Pacer, which has only a small pilot window on the left side. This problem was solved by removing the hinge pin at the window base and replacing it with a removable pin which would allow the window to be opened completely by the pilot during the flight. The pin, illustrated in photograph #11, was designed to prevent accidental removal while still allowing the pilot to remove it quite easily. The line of fire obtained in the Tri-Pacer, after this modification, is shown in photograph #12. The same modification was adopted for the Cessna 140 and 170. It was discovered that the open window allowed a more nearly horizontal line of fire and also acted as a makeshift stall-warning indicator. At normal-approach speeds the window would remain entirely open, within one or two inches of the underside of the wing. As the airplane slowed to below normal-approach speed, the window slowly lowered to the closed position. In firing the pistol flares it is desirable to slow the airplane to normal- approach speed before attempting to shove the flare-loaded pistol and Photo No. 11. Tri-Pacer window hinge pin removed allowing the window to open com- pletely. This same modification could be used in other airplanes. 26 Photo No. 12. Line of fire from Tri-Pacer with window completely open. Note straight wrist, elbow, and left-handed fire. arm outside the window. Wind resistance to the flare, the gun, and the pilot's arm creates a problem in holding a steady aim while firing. The lower the speed, the lower the resistance. In firing 16, 1 -minute pistol flares, no misfires were experienced. It is vitally important to check the pistol barrel for roundness and for easy fit of the flare. This should be done before takeoff. A too-snug fit can easily be remedied by using a small piece of emery cloth on the barrel or the shell. In any case, do not attempt to force the shell portion of the flare into the pistol, for, although it may be possible to force a too-snug fit in this manner before firing, the shell may expand slightly after firing. This will make it difficult to remove the expended shell and will waste valu- able time when every second counts. Even with an easy fit between pistol barrel and shell the pilots occasionally found it necessary to use both hands in removing an expended shell, and, although there were no serious problems of this nature, it was obvious that such misfits could have resulted in fouling the entire simulated emergency. The 37-mm. pistol used for firing the 1 -minute flare is equipped with a quick-release device for expelling the shells. This device can be oper- ated with the thumb or a finger of the same hand holding the pistol. It solves the problem of what to do with possible "hangfires." 27 Assume that the trigger and the safety have been squeezed and that the resounding "elick" of the hammer was heard but the flare failed to fire. Under no circumstances should the pilot draw his arm, the pistol, and the flare back into the airplane. A hangfirc, or delayed primer action, could result in setting the airplane and the pilot on fire. This is where the quick-release ejector on the pistol pays off. Simply press the ejector and drop the unfired flare overboard, and, of course, reload im- mediately. Hangfires are remote possibilities, but they do occur occasion- ally, usually as a result of faulty action of the center fire primer. There were no such experiences with 28 pistol-fired flares, but the pilot should realize that a hangfire is a potentially dangerous possibility. In summary, the special recommendations for using pistol flares are: 1. Check pistol barrel and shell for easy fit prior to takeoff; 2. Check the airplane carefully for adequate openings through which the pistol flare can be fired; 3. Check the potential line of fire by sitting in the airplane and actually putting the loaded gun through the opening; 4. Have the flares and pistol in readiness at all times, and keep the pistol loaded — it won't go off unless the trigger AND the safety catch on the rear of the handle are squeezed; 5. Should an engine failure occur at night, slow the airplane down to normal-approach speed in order to reduce wind resistance when holding the gun outside the plane; 6. If flying from the left seat, fire the pistol with the left hand, hold- ing the wrist and elbow straight to allow the shoulder to take up the recoil; 7. Drop any hangfires OUTSIDE the airplane by using the quick- release device on the left side of the pistol. FLARE VS. AIRPLANE PERFORMANCE The most important airplane factor to consider is the aircraft's rate of descent. The rate of descent for each airplane used in the experiment was established by loading the airplane with a full complement of pas- sengers, climbing to 2000 feet actual, and then making a complete power- off descent and landing. This process was repeated five times while an observer timed the process with a stopwatch. The rate of descent was based on the average of five trials. In each case the pilot was instructed to maintain the approach airspeed carefully during the entire descent. Flaps were not used and, in the case of retractable gear, the gear was not lowered until the turn from base leg to final approach. Since the "maneuvering" rate of descent was desired, each trial was based on a 360-degree spiral followed by a 360-degree overhead approach to the landing. The stopwatch was started when the throttle was cut and was stopped when the wheels touched ground. 28 The airplane's rate of descent is important because it fixes the total amount of time available to complete a forced landing and it establishes procedures which must be followed to coordinate the airplane's perform- ance with the performance of the flare. This is especially true of the second flare; i.e., the approach and landing flare. For the Bonanza this flare is released at 1200 feet and will burn for VA> minutes. At a 740-fpm rate of descent the Bonanza would not be on the ground by the time the flare extinguished. In IV2 minutes the Bonanza normally descends only 1110 feet. To compensate for this difference in flare and airplane per- formance the pilot was instructed to dive off 150-200 feet immediately following the release of the second flare. This 150-200 foot dive achieved a twofold benefit. (1) It immediately put the airplane below the flair where best visibility is attained and (2) it allowed the airplane to land while the flare was still burning. Since the Tri-Pacer had an 870-fpm rate of descent, very little diving was necessary to acquire the above-mentioned twofold benefit when using the 1 -minute pistol flare. In the simulated forced landings, flaps were used only to compensate for errors in judgment of gliding distance. In most cases, the pilots had too much altitude on final approach and thus needed to dissipate the excess altitude by use of flaps. Only one airplane with retractable gear was used — the Beechcraft Bonanza. All landings during the experiment were made with the wheels down. The pilots were instructed to use the gear, together with the flaps, to compensate for excess altitude. If the airplane appeared to be too high, the gear was lowered early; if the airplane was too low, the gear lowering was delayed as long as possible. During the project the gear was lowered usually during the turn into the final approach. In an actual night forced landing it is reasonable to believe that the gear would not be lowered. In many instances a wheels-up landing is advantageous, especially when it cannot definitely be established that the selected field has no ditches, fences, stumps, etc. Should a wheels-down landing be selected the gear should not be lowered until it is certain that the airplane is within gliding distance of the field. The pilot should re- member also that the best illumination will occur during the final stages of the flare's burning time. All of these factors point to delaying the gear lowering until the final approach. The important airplane considerations such as approach speed, pas- senger load, use of gear and flaps, and the airplane's average rate of descent are summarized in Table #4. The information on those airplanes not listed in Table # 4 can easily be obtained by making several power- off descents to a landing from the altitude appropriate to the flare to be used. Take the average of at least five trials. Be sure the descent involves at least 720 degrees of maneuvering, since the rate of descent will vary 29 in direct relation to the amount of maneuvering. This procedure can be accomplished during daylight hours with an observer riding along to operate a stopwatch. If the aircraft has retractable gear, do not lower it until the turn is made to the final approach. TABLE NO. 4. RECOMMENDATIONS BASED ON FOURTEEN SIMULATED NIGHT FORCED LANDINGS USING SPECIFIC AIRPLANES AND SPECIFIC FLARES Flare Type Class 2 1 Vi -minute Electrical Class 3, 1-minute Pistol Flare Airplane Type Bonanza C-35 Piper Tri-pacer Cessna 170 Cessna 140 Stinson 150 Power setting used Idling Idling Idling Idling Idling Indicated airspeed mph 95 80 80 70 85 Passenger load including pilot 4 4 4 2 4 Landing gear Lower on base or final as necessary Down Down Down Down Flaps Lower on base or final as necessary Lower on base or final as necessary Lower on base or final as necessary Lower on base or final as necessary Lower on base or final as necessary Airplane's average rate of descent in feet per minute 740 870 670 640 860 Altitude above ground to release flare for field selection purposes 2000 1800 1600 1500 1800 Average time from release of first flare to altitude for second flare 67 seconds 55 seconds 57 seconds 59 seconds 55 seconds Altitude above ground to release second flare for landing purposes 1200 1000 1000 1000 1000 Position of airplane relative to selected field for release of second flare Extreme upwind Extreme upwind Extreme upwind Extreme upwind Extreme upwind Recommended approach pattern from second flare to landing. Overhead approach. 180 or 360 degree 180 or 360 degree 180 or 360 degree 180 or 360 degree 180 or 360 degree Desirable airplane position in relation to flare to obtain best visibility Below Below Below Below Below Time from second flare to extinguishment 1 Vi min. 1 min. 1 min. 1 min. 1 min. Excess altitude necessary to dive off immediately following re- lease of second flare to land before flare extinguished 200 ft. 100 ft. 400 ft. 400 ft. 150 ft. Table # 3 summarizes the pertinent data and performance character- istics of both Class 2 and Class 3 flares. This table also gives the source of the particular item of information. Table # 4 shows the pertinent information for a particular flare used in a specific airplane type. This information is based upon 14 simulated night forced landings at the University of Illinois. A Simulated Night Forced Landing Let's summarize all of the facts by carrying out a hypothetical night forced landing in a Beechcraft Bonanza. We'll assume that you're cruis- ing at 4000 feet above ground over relatively unfamiliar terrain. We'll further assume that your engine has stopped and that you've already made corrective checks of gas selector valves, carburetor heat, and mags, to no avail. You're committed to a night forced landing. These recom- mendations are based on breaking the entire problem into three phases, from engine stoppage to completion of landing roll. PHASE I —GENERAL AREA SELECTION AND PREPARATION Let's try to think of our problem, all three phases, in terms of both time and altitude. Phase I starts when the engine stops and ends when you reach an altitude of 2000 feet above ground. If the engine failure occurs at 2000 feet or below, Phase I is omitted entirely and you begin the procedure with Phase II. If engine failure occurs at 4000 feet actual, there is no reason to fire a flare at that altitude because ground illumi- nation is practically nil. Based on your rate of descent of 740 fpm, you have 2000 feet or approximately 3 minutes to complete Phase I. First slow the airplane, with trim tab, to normal-approach speed; in the case of the Bonanza, 95 mph. This slowing process will add a few precious seconds to the available time, since excess airspeed can be con- verted into altitude. During this slowing-down period, your attention will be devoted to selecting the most desirable general area for the forced landing. Some ground objects are visible even on the darkest nights. At least, it is possible to distinguish cities and general areas of congestion by lights on the ground. If the cockpit lighting has been kept to a mini- mum, night adaptation of the eyes will usually allow the pilot to dis- tinguish dim silhouettes and shadows of prominent features on the ground. The decision may resolve itself into a choice of the lesser of two evils; however, the time spent in descending to 2000 feet can be put to best advantage by selecting a definite general area for the forced landing. Once selected the remaining alititude must be used in maneuvering to- ward the general area. If possible, the pilot should attempt to reach the upwind portion of the selected area, since the flares will drift downwind. Any spare time available while gliding toward the area should be used to open the circuit-breaker panel door and turn the master flare switch 31 to "on." You may as well turn the ignition system off at the same time, but do not turn off the battery system. It will be needed for the flares and other eleetrically-operated devices such as flaps and gear. You should, however, turn off by individual switches any unnecessary items which place a heavy drain on the battery; e.g., the radio, pitot heat, or auto pilot. Time permitting, you should check individual flare switches for the "off" position before turning the master switch on; you must also determine whether the master switch remains in the "on" position when released — if it does not, it will be necessary to hit the master switch each time AFTER turning on the individual flare switches. When you arrive over the selected general area, you should be headed into the wind, if possible, because ( 1 ) this establishes a base reference for the remainder of the problem, and (2) it will minimize the maneu- vering necessary to stay within the flare's cone of illumination. When you arrive at 2000 actual, ready or not, Phase I is ended and you must proceed to Phase II. PHASE II — SPECIFIC FIELD SELECTION AND POSITIONING Phase II, in the Bonanza, begins at 2000 feet above ground and ends at 1200 feet. Assuming that you fired flare #1 at 2000 feet above ground, it should still be burning when you reach 1200 feet. In terms of time it is slightly more than 1 minute; in flight distance, approximately IV2 miles. In terms of turning it means that one complete 360-degree turn at a double standard rate is possible. In the experimental simulated attempts, since the specific area was already known, Phase II was always started at 2000 feet, headed directly into the wind, at the upwind point of the area. Illustration #2 shows that this always resulted in a 360- degree spiral for Phase II. This 360-degree spiral could easily be modi- fied into a 180-degree turn followed by a 3 /4-mile straightaway if the most desirable field was the downwind extremity of the cone of illumination. Immediately following the firing of flare #1, you initiate a turn to the left (flare fired to left) to come back into the cone of illumination. Your best position for ground visibility is below the flare. Your field should be selected as soon as possible after the flare is fired. The sooner you select the field the more time there will remain to maneuver to it before reaching 1200 feet actual, or the end of Phase II. To complete a 180- degree turn at an airspeed of 95 mph and within 30 seconds, a bank of at least 30 degrees is needed. This same bank will allow you to complete a 360-degree turn in 1 minute; i.e., the total available time for Phase II. You must be at the upwind end of the selected field by the time you reach 1200 feet actual. Although it is desirable to reach this point headed directly into the wind, you can still complete Phase III successfully even though you arrive headed upwind, downwind, or crosswind. The impor- tant point to remember is that Phase II is completed, like it or not, at 1200 feet. 32 * o o "5 2 -O c "8 o E ^ "2 2 1 a) o C 9) o Q. 2 o 8 o 0) in .§ x 1 I .- M .5 c § 1 1 1 £ ^ •B 3 c i PHASE III — APPROACH AND LANDING Phase III begins with the firing of the second flare at 1200 feet am ends when you step out of the airplane upon completion of the landing. In terms of altitude, time, and distance you have 1200 feet, \ x h minutes, and approximately 2!4 miles available. A third flare should be fired only in case #1 or #2 misfire. Illustration #3 is based on starting Phase III headed directly into the wind at 1200 feet over the upwind end of the selected field. Flare # 2 is fired at this point, followed immediately by a 135-degree turn (30-degree bank) to the left. If you arrived at this point going downwind you would fire the flare and follow with an immediate 45-degree turn to the right. From any heading it is important to remember only that the com- pletion of this original turn should result in a heading 45 degrees down- wind of the intended final approach heading. A left-handed approach is better because ( 1 ) the pilot has better visibility in the left seat, and (2) he is probably more used to left-hand rather than right-hand approaches. During the initial turn to the 45-degree downwind leg, you should dive ofT enough altitude to allow the average rate of descent to put you on the ground within the IV2 minutes of flare illumination. In the Bo- nanza with a rate of descent of 740 fpm, this dive should be approxi- mately 200 feet. From this point — the 45-degree downwind leg — you will be busy with the normal duties attending any forced landing. The illumination intensity on this leg is such that you will be able to distinguish details on the ground, and you should now make a definite decision whether to have wheels-up or wheels-down for landing. There is much to be said for the wheels-up landing; and if there is the slightest doubt about the length or the smoothness of the selected field the decision should defi- nitely be wheels-up. The actual gear lowering process can be saved for the base-leg or final approach, but it is best to give serious consideration to the question while outbound on the 45-degree leg. Your entire pattern will probably be somewhat closer in to the field than normal power-available approaches. Turns will definitely be steeper than usual. Previous practice and experience with complete power-off approaches, and the obstructions in and around the selected field, will dictate exactly where and when you turn to base and final approach. On the base leg, turn on your landing light, or lights, in case the flare extinguishes before you complete the landing. You should also con- sider where and when flaps are necessary and, if you have elected a wheels-down landing (only if absolutely certain of the terrain), allow time for the gear to come down. (In the Bonanza this is a minimum of 7 seconds or, to be safe, 200 feet above ground.) 34 2 g £■0 o .E -° I 2 S- > £ ■o * 6 o o T5 • £ *. a, W° c o 1 • o >- X II ^ -5 2 1 .E -o a> O m 00 c E .2 a, o -° * -D The 13 test pilots had a tendency to level off too high under the flare light, and when this tendency was called to their attention they went to the opposite extreme, allowing the airplane to touchdown before the leveling-off process was completed. The same factor was true of cases where the landing light only was available for the touchdown. Don't forget, the gear-up landing will mean a greater distance to drop after the airplane stalls. In the gear-down condition, the brakes should be applied as quickly as possible after touchdown. In some airplanes, the Cessna 195 for example, it is possible to land with the brakes locked and not nose over. If your timing has been perfect, you will come over the end of the field on the final approach with flare #2 almost directly overhead at 500-600 feet above ground, and the illumination will be sufficient for you to see and avoid such small things as boulders, ditches, and fences. If your timing is slow, the flare may extinguish while you are on the final approach and you will have to depend entirely on the landing lights. The landing lights will begin to pick up ground details at 20-30 feet above ground. Remember, you have had the advantage of the flare illumination all the way from 2000 feet down to this point and you are committed to this particular field. If you have used this illumination to the best advantage, you have already seen and planned for major objects such as trees, houses, and ravines. These couldn't be avoided anyway at this late stage, with or without flare illumination, because there is little maneu- vering potential left. When the landing roll ceases and you climb out intact, you will have a greater respect for the parachute flare and your own farsightedness in preparing, by reading and practicing, for just this eventuality. Now, let's summarize the important actions and decisions for each individual phase. Phase I 1. This phase begins with the actual engine stoppage and ends at 2000 feet above ground. If the engine stoppage occurs below 2000 feet, the entire phase is omitted. 2. Using night vision only, the most desirable general area is selected for the forced landing. 3. The airplane must arrive over the selected area not below 2000 feet. It is desirable, but not mandatory, to be headed into the wind. 4. Any additional time (altitude) should be devoted to preparing the airplane for the actual landing. This task will vary in different types of airplanes but should include such things as adjusting trim, cutting unnecessary switches, setting fuel selector valves, and bringing the propeller to a horizontal position with the starter. 36 5. If the airplane has retractable gear, all such adjustments should be based on the assumption that a gear-up landing will be made. Phase II 1. This phase begins at 2000 feet and ends at 1200 feet above the ground. 2. The first flare is fired at 2000 feet and is used to select a specific field within gliding distance (approximately IVz miles) of the point of the flare release. 3. The airplane must arrive at the extreme upwind end of the selected field by the time it has descended to 1200 feet above ground. 4. The field selection search should be made using a left turn and keeping the airplane within the flare's cone of illumination. 5. During the positioning process (maneuvering to the upwind ex- tremity of the selected field) the pilot should mentally note the type of overhead approach which the 1200-foot heading will necessitate. 6. In general, fields selected in the upwind portion of the cone of illumination will necessitate a 360-degree overhead, fields selected in the downwind sections of the cone of illumination will dictate a 180-degree overhead approach, and fields selected in the central portion of the illuminated areas will require combinations of the 180- and 360-degree overhead approaches. Phase III 1. This phase begins at 1200 feet and ends with the completion of the landing roll. 2. The second flare is fired at 1200 feet above ground and is used for the approach and landing. If the flare is fired above 1200 feet it will very likely extinguish before the landing is accomplished. Firing altitudes below 1200 feet are likely to result in the flare's being out of usable position for the final approach. 3. The triangular-shaped overhead approach is almost mandatory, since the approach and landing flare must be fired at the upwind extremity of the selected field. 4. The all-important decision on gear position must be made during Phase III with sufficient time to get the gear down if the pilot elects that configuration. 5. Landing lights should be turned on before the flare extinguishes, probably on base leg. Flare extinguishment before touchdown does not necessarily mean a complete failure on the problem. 6. The third, or reserve, flare should be fired in case of misfire of either of the two planned flares. 37 SPECIAL PROCEDURES FOR THE PISTOL FLARE The primary advantages of this flare are low cost, less weight, porta- bility, and, we believe, less possibility of misfire due to faulty installation. The pistol flare can be used to advantage, but more advance preparation and planning are required if good results are to be obtained. The same three-phase approach used with the ll^-minute flare is recommended for the 1 -minute flare, but, due to less candlepower, the altitudes become a maximum of 1800 feet for beginning Phase II and 1000 feet for begin- ning Phase III. Even under these circumstances the 1-minute flare may burn out before the airplane lands unless the timing and positioning are near-perfect. There were no misfires of 1-minute pistol flares during the entire experiment. However, as insurance against possible misfires, or hangfires, the pilot should keep at least one extra flare, or a total of three flares, in the airplane at all times. There will not be time to fire more than three 1-minute pistol flares from 1800 feet, control the airplane, and plan the approach. The pistol-fired flare (1-minute) does have two distinct disadvan- tages: (1) The already cramped emergency procedures are complicated with the additional duty of unloading, reloading, aiming, and firing; and (2) the shorter burning time allows less margin for error in positioning the flare and in timing the airplane's arrival at given points. PRACTICE IN A SPECIFIC AIRPLANE Personal confidence in this emergency procedure can come only through serious practice. Every one of the 13 pilots who participated in this project felt that they could now walk away from a night forced landing. You too can acquire this confidence by practicing the procedure in your own airplane. Preliminary practice can be done in daylight hours with a pilot friend who starts a stopwatch or photographic timer when you call "bang" at the flare firing points. He can tell you, or allow the timer bell to warn you, when the flare theoretically extinguishes. You will want to make two or three dry runs at night once you have the timing perfected. If you have any flares over three years old, and thus due for renovation, you'll find it relatively inexpensive to use them for a simulated night forced landing over your local airport. You'll need to enlist the cooperation of your local airport manager and you must also provide adequate ground fire protection for potential misfires. In 14 trials, only one of 40 flares failed to extinguish before striking the ground. The above information, seasoned with serious practice on the part of the reader, will go a long way toward putting potential night forced landings in the same mental category as similar daytime emergencies 38 Having mastered the technique and acquired the attendant confidence, a greater degree of safe utilization of the light airplane will follow. Trial Results The following table shows each simulated night forced landing at- tempt by date, pilot number, airplane type, and other pertinent data: TABLE NO. 5. SIMULATED NIGHT FORCED LANDING ATTEMPTS Pilot Airplane Flare Type &No. Moon Phase Official Weather Unexpected Experiences Landing Com- pleted by Flare or Landing Lite Total Time Flare #1 to Landing Landing Success- ful Date Ceiling and Visibility Wind 8/55 1 Bonanza IlrV 17875 17874 Dark Clear 15 + SSW12 Defective flare master switch Landing lite flare #2 drifted out of position 2:58 Yes 1 2/55 2 Bonanza 17420 17421 % E250(D15 NNW8 Flare #2 mal- functioned, no illumination Landing lite 2:32 Yes / 2/55 3 Bonanza U4' 17876 17422 % E250(D15 NNW10 #2 flare re- leased too late Landing lite flare #2 drifted out of position 3:03 Yes 9/55 4 Cessna 140 r 6804 6805 Dark 3O0E5O © 10 SSE10 Did not have gun loaded with #1 flare Flare 2:21 Yes / 9/55 3 Cessna 140 V 6802 6803 Dark 35 E50 © 10 SE11 #1 cartridge hard to remove Flare 2:20 Yes 10/55 5 Cessna 170 r 8336 8345 Dark 1O0E5O CD 8 N10 Right hand pattern, over- shot landing No landing No landing No no/ss 6 Cessna 170 V 6807 6806 Dark 1O0E5O (D6HR- N10 Fumbled un- loading and reloading pistol Flare 2:16 Yes 20/55 7 Piper Tri-pacer 1' 8334 8337 Dark 80 012 E10 Flare hypnosis. Failed to re- load pistol at proper time Flare 2:05 Yes , 20/55 8 Bonanza V/ 2 ' 19003 19002 Dark 80 012 E10 #2 flare failed to leave plane Landing lite 2:28 Yes / 6/55 9 Bonanza W%' 18923 18924 Dark 8 0M22 © 15R- W15 None Flare 2:54 Yes ,20/55 10 Stinson 150 r 8341 8335 Dark 60 012 NW8 Difficulty re- moving first cartridge Flare 2:16 Yes / 24/55 11 Cessna 170 r 8338 8340 Dark E100CD 15 + SE7 Fumbled re- load. Flare re- leased wrong altitude and position Landing lite flare #2 extinguished on final 3:00 Yes 7 5/55 12 Cessna 170 T 8342 8344 Full E80 © 10 S6 Flare extin- guished just before landing Landing lite flare #2 extin- guished on final 2:47 Yes 7 5/55 13 Bonanza l'/2' 19001 18925 Full E80 © 10 S6 #2 flare failed to ignite Landing lite 2:59 Yes 3° Fourteen simulated night forced landings were attempted, using 13 different pilots. Thirteen of the attempts were carried to a successful landing within the specified 70-acre area. One overshoot occurred which necessitated a "go around." In this case the subject pilot arbitrarily modi- fied the previously-planned procedure to a right-hand instead of the normal left-hand approach. In all 14 cases the pilot was able to select the specified landing area by flare illumination only and simultaneously maneuver into a position appropriate for either a 180-degree overhead or 360-degree overhead approach. In six of 14 attempts the pilots were able to accomplish the approach and landing with the second flare still burning and in such a position that there was adequate illumination for the landing. On one attempt an overshoot occurred, and on the seven remaining attempts a successful landing was accomplished using the landing lights as the only means of illumination. In three cases where only the landing lights were available for illumi- nation, flare #2, or the approach and landing flare, malfunctioned. In two other cases the flare was either too far downwind of the selected site or too low to give adequate light for the landing. In the two remain- ing cases the flare candle burned out while the airplane was still on final approach. Conclusions 1. Both the 1 -minute pistol and the l!/2-minute electrical parachute flares can be used to distinct advantage for actual night forced landings by the lightplane pilot. The lM>-minute electrical flare has distinct advantages in this situation as compared to the 1 -minute pistol-fired flare. 2. The pilot's ability to make safe and efficient use of these flares varies in proportion to his knowledge of, and prior preparation in the use of, the particular flare and the airplane type to be used. 40 Bibliography 1. Emergency Flare Equipment for Aviation and Marine Use, Interna- tional Flare-Signal Division, Kilgore, Inc., Westerville, Ohio (no pub- lication date) . 2. Fire Hazard — Magnesium Flares, Ansul Bulletin, Ansul Chemical Co., Fire Extinguisher Division, Marinette, Wisconsin, August 10, 1951. 3. Irregular Air Carrier and Off-Route Rules, Part 42 — Civil Air Regu- lations, as amended to December 15, 1954, U. S. Government Printing Office, 1955. 4. Information Regarding the Dropping of Parachute Landing Flares from Aircraft, Their Inspection and Renovation, Department of Com- merce, Civil Aeronautics Administration, July 1, 1949. 5. Landing Flares, Technical Standard Order C24, Effective October 10, 1950, Department of Commerce, Office of the Administrator of Civil Aeronautics. 6. Miscellaneous Aircraft Equipment, TM 1-416, War Department, U. S. Government Printing Office, Washington, D. C, 1944. 7. Specification — Flares, Landing, NAS 805, National Aircraft Stand- ards Committee, Aircraft Industries Association of America, Inc., Washington, D. C, February 2, 1950. THE INSTITUTE OF AVIATION, established in 1945 as the Institute of Aeronautics, is operated as the administrative agency responsible for the fostering and correlation of the educational and research activities related to aviation in all parts of the University. Other functions include aca- demic instruction, flight training, management of the University of Illinois Airport, and aeronautical research. In connection with the latter function, the Institute issues two types of publications . . . first, a group of reports on research results, and second, a series of bulletins on aviation subjects of an extension-service nature to the citizens of the State. The following publications have been issued: bulletin ONE: Municipal Airport Management, Leslie A. Bryan, 1947. (Out of print) BULLETIN TWO: Landscape Planting for Airports, Florence B. Robinson, 1948. BULLETIN THREE: Labor Relations in the Air Transport Industry Under the Amended Railway Labor Act, E. B. McNatt, 1948. bulletin FOUR: Airport Zoning, J. Nelson Young, 1948. (Out of print) bulletin FIVE: Evaluation of the School Link as an Aid in Primary Flight Instruction, A. C. Williams, Jr. and Ralph E. Flexman, 1949. bulletin SIX: Lightplane Tires on Turf and Concrete, Leslie A. Bryan, 1949. bulletin seven: Light Aircraft Operating Costs, Leslie A. Bryan, 1949. bulletin eight: Evaluation of the School Link and Special Methods of Instruction in a Ten-Hour Private Pilot Flight-Training Program, Ralph E. Flexman, William G. Matheny, and Edward L. Brown, 1950. (Out of print) Flight by Periscope: I. Performing an Instrument Flight Pattern; the Influence of Screen Size and Image Magnification, Stanley N. Roscoe, 1951. Operating Costs of a Light Aircraft Fleet, Leslie A. Bryan, 1952. 180-Degree Turn Experiment, Leslie A. Bryan, Jesse W. Stonecipher, and Karl Aron, 1954. Aviation Ground School, Leslie A. Bryan, 1954. Organizing for Flight Operations, Leslie A. Bryan, 1954. Developing an Aircraft Maintenance Curriculum, Leslie A. Bryan/l955. Airport Shop Operations, Leslie A. Bryan, 1955. College and University Airport Management, Leslie A. Bryan, 1955. Parachute Flares as an Aid in Night Forced Landings, {esse W. Stonecipher, Miller R. Green, Richard E. Mankus, 1955. INSTITUTE OF AVIATION WILL BE SENT FREE OF CHARGE UPON REQUEST BULLETIN NINE: BULLETIN TEN: BULLETIN ELEVEN: BULLETIN TWELVE: BULLETIN THIRTEEN: BULLETIN FOURTEEN: BULLETIN FIFTEEN: BULLETIN SIXTEEN: BULLETIN SEVENTEEN PUBLICATIONS OF THE