Air Navigation: State of the Art in 1937
| “The factors which contribute to inaccuracies in surface navigation--currents other than anticipated or estimated, lack of sights, inaccurate radio bearings, etc.--are all encountered in aerial navigation and commonly in intensified form.”
Fred J. Noonan, Pan Am Memo, Subect: Navigation: Hawaiian Flight of NR 823-M, April 29, 1935
Aviation’s early decades saw the continuing development of aircraft with more powerful, reliable engines enabling aviators to venture far out of sight of land. The need became apparent for improved air navigational capabilities--cockpit instrumentation, charts of appropriate scale with special symbols and features, more comprehensive wind and weather forecasting, faster celestial procedures, sextants with artificial horizons, two way radio communication and radio navigational aids among them. Aviation was the benefactor of maritime navigational technology and procedural “hand-me-downs”, often not easily adapted for aircraft use. Bulky, heavy marine equipment was ill-suited for the limited lift of air machines. Even the slowest aircraft cruised at speeds in the magnitude of 10 times that of ships. Navigational positions needed to be determined more rapidly; equipment needed to be easier to use in a small cockpit by pilots who often wore gloves. Aviators, like Charles A. Lindbergh, coped with navigational requirements while piloting the aircraft, but it soon became apparent that for extended flights, an additional crew member would be useful to allow the pilot to devote his full attention to controlling the aircraft. Aerial navigation began, however, with the maritime legacy as its progenitor.
Ocean navigation began to develop during the European Renaissance with the advent of the magnetic compass. It was the compass that enabled ships to steer and hold a magnetic heading to their intended destination. The compass card defined 32 points known as winds. “Portolan” charts, depicted routes based on these winds and had no latitude and longitude reference. These charts were used by the vessel’s pilot, in conjunction with the “rutter” (from French “routier”), or sailing instructions, in order to follow a given magnetic heading to the destination. Navigation was done exclusively by dead reckoning--experienced pilots gauged the speed of their vessels by the wind in the sails, the wake, and the sounds and feel of their ships. Charted distances were often given in days of sailing time.  By the Columbian period, dead reckoning had become highly developed and quite reliable. Using pilotage, the master took his ship along the coast to a point at the same latitude as his destination, then followed the compass due east or west to destination. This technique was called “easting” or “westing”. A confidant of Columbus’ wrote, “A good pilot or master is not considered such if, in traveling over a great distance from land to land, out in the open sea with no indication of any land, he is off by ten leagues even when the trip is a thousand leagues long.” An empirical reconstruction of Columbus’ first voyage in 1492 was made using his logged magnetic headings and distances sailed for the 136 legs of the voyage, and demonstrated that Columbus’ dead reckoning provided 99.7% accuracy from departure to destination and return. 
Little effort was made to integrate the “New Navigation” (celestial) with the proven dead reckoning, as celestial positioning was the province of the learned mathematicians and cosmographers, not the vessels’ pilots. Celestial navigation was not reliable on a rolling vessel with instruments of the period; sight reduction was immensely time consuming and very challenging mathematically. Celestial observations were primarily used on land by the cosmographers along on the voyage to determine the latitude at destination. Its purpose was to update the Master Chart for the Crown in order to validate claims of sovereignty over new territories, not for navigation.  Significant advances in reliable celestial navigation did not occur until the eighteenth century following the invention of the marine sextant (octant) and John Harrison’s development of a chronometer capable of keeping accurate time aboard a ship under sail.
By the 19th century, balloon pilots were largely restricted to map reading as a method of navigation, pointing up the need for international cooperation for procedures to cross borders and to create dedicated aeronautical charts—ones which included symbols and information of obstructions, rail lines, roads, town and village shapes, forests and bodies of water. The need for more frequent and accurate wind forecasts became apparent. Although balloon pilots attempted to use celestial navigation, balloons rotated, making it difficult; cramped space made tedious computation and plotting impractical, even though balloonists experimented with bubble horizons on sextants and tabulated reduction methods. Not until lighter-than-air craft were powered by engines could they maintain a course for which the magnetic compass was useful. Development of the zeppelin and dirigible permitted longer voyages off-shore and they soon used adaptations of the mariner’s navigational equipment and techniques.
The United States did not have a significant military aviation capability before its entry into the World War. The role of World War I aircraft was primarily that of artillery spotting, reconnaissance, and aerial pursuit which required only pilotage, or map reading. Blimps and longer range aircraft were used to a lesser degree for patrolling off shore or for bombing behind enemy lines but had little additional navigational capability or crews trained in navigation. The Great War saw the development of the drift meter, more for use as a bomb sight than for navigation. The evolution of the compass accelerated during this period, as the inherent characteristics of the airborne compass detracted from its usefulness and had to be addressed. Rapid turns and attitude changes caused the compass card to swing excessively. Movement of the rudder pedals and control stick, if made from ferrous metal, caused the compass reading to change, as did the expenditure of bombs and ammunition, which had high iron content. Improvements developed by the end of the war included a remote indicating compass with the sensor near the tail, and another, having a lighter card with less inertia called an” aperiodic” compass. A later, development was the earth inductor compass which proved to be more stable, but difficulty in keeping the sensing component level during flight was never resolved. Night navigation was facilitated by providing instrument lighting. Two-way radios preceded the experimentation and development of rudimentary directional radio navigation using ground based stations, even though engine magneto interference, long antennas and oversized equipment were problematic.
The dead reckoning computer was another important development, with each country's air arm finding a different solution. The French favored a large plotting board used in conjunction with a chart. The U.S. and Britain used a circular slide rule, subsequently integrated with a graphic plotting surface which was placed on the reverse side. Later improved by Ensign Philip Dalton and called the Dalton computer, it eventually became commonly known as the E-6B. Instruments which contained optics, like the drift meter, and sextant were adversely affected by aircraft attitude changes, and their use required the pilot to maintain a level, stable platform. Sperry’s early gyroscope was imperfect, though it improved after the war and gyroscopic stabilization provided significant improvement for a number of navigational instruments. Other instruments that saw successful development were the altimeter with ranges expanding to 20,000 feet, the turn and bank indicator and the artificial horizon. Harold Gatty developed a method of wind determination called the “double drift maneuver” in the U. S. and the “wind star” in Britain,  to calculate winds aloft by taking drift readings on two different headings, then calculating the wind trigonometrically by plotting vectors on a graph. (Wind could alternatively be determined by reading the drift on a single heading, at two different airspeeds, but the former was the preferred method.)
World War to World Flight
Commercial Aviation: The period following World War I saw the emergence of commercial aviation and expansion of air routes across both oceans. Planes which made the early transatlantic crossings were generally capable of carrying only the fuel required for the trip, but by the 1930’s several countries were using larger craft to experiment with passenger service between continents. France and Germany had experimental routes to South America. In 1931 the Graff Zeppelin offered service to Rio de Janeiro which lasted until 1937. By 1934 Deutsche Lufthansa began service across the South Atlantic followed by Air France in 1936. Pan Am had developed a network of routes in the Caribbean and South America, and in 1935 opened a Pacific Division carrying mail to Manila in November of that year. 
North Atlantic service developed more slowly, with political difficulties hindering international agreements. The Hindenburg made 10 round trips before exploding in 1937 during the first trip of that season. Pan Am and Imperial jointly developed the route from New York to Bermuda, but it was 1939 before Pan Am offered North Atlantic passenger service.
Airlines were very concerned with the safety of their passengers and the regularity of their service. They adopted near-universal policies regarding navigation—most stipulated that all forms of navigation available would be used at all times, with a comparison made of the results of each. If the navigational data was consistent, there was reasonable assurance of the aircraft’s position. If, not, the navigator attempted to verify the position with additional data. Navigation was sometimes called an “art”, as the navigator’s evaluation of the data used in interpolating a position was subjective based on his assessment of each element’s reliability. (Fred Noonan discussed some limitations of navigational methods, radio equipment, and the affects of personal errors in an internal Pan Am memo written following the Alameda to Honolulu Clipper route surveys of April 1935.) Aside from dead reckoning, celestial navigation and radio bearings were the only overwater navigational methods available during this period. When flying in conditions of overcast skies or when in the clouds, radio alone could be relied upon and compared with a dead reckoning position. 
Instrument Technology: Instruments and equipment continued to improve during the decade of the 1930’s. The 1-5% instrument errors of the 1920’s gave way to substantial improvements in the 1930’s. Pitot and venturi systems were improved and relocated away from turbulent areas of the aircraft. Sperry, after a 10 year effort, developed a successful air bearing for the gyroscope (steel bearings affected the magnetic compass) enabling the gyro compass and autopilot to become reliable instruments. Combining the magnetic compass with the gyro was still some time away, but using the compass to reset the gyro was a workable alternative and was used aboard Earhart’s Electra. Aircraft stability with the use of the gyro compass and autopilot also increased the accuracy of celestial observations. Companies like Kollsman, Pioneer, Sperry Gyroscope, Link Aviation, and Aera of Paris continued to make many small incremental improvements in instruments, which together greatly increased their function and reliability.
There were two additional instruments under development, that didn’t appear until 1938, after the World Flight. One was the absolute or radio altimeter, necessary for accurate ground speed timing made with the drift meter; the other was the gyro-stabilization of the drift meter optics to permit more accurate drift determination in turbulent air. A method sometimes used to determine absolute altitude over oceans was to descend to sea level and reset the altimeter. The downside was the increased fuel consumption for the climb back to altitude. German zeppelin navigators were extremely interested in the atmospheric pressure system’s usefulness in wind determination and lowered an aneroid sensor on a tether to the ocean’s surface to determine the sea level atmospheric pressure.
Celestial Navigation: Bubble chambers in sextants continued to be problematic. P. V. H. Weems, probably the period’s foremost proponent of celestial air navigation, exhorted sextant manufacturers to produce a more durable, reliable bubble chamber for the sextant. Weems also experimented by taking a series of celestial observations which were averaged to mitigate the acceleration errors induced in the bubble by aircraft axes motions. He made eleven groups of ten observations each, and even though one was 128 miles in error, he found that by averaging the observations, the overall error was 3 miles. The error using 10 observations, was 5 miles. Though none were available for the World Flight, several companies had mechanical sextant averagers under development with some being evaluated by the airlines in 1938.
Sight Reduction: Sight reduction is the mathematical solution of the spherical celestial triangle which provides the navigator with a geographical line of position from a celestial observation. As done by the mariner, it was a twenty minute procedure, adequate for slow moving ships, but not suitable for aircraft. In 1874, a French naval officer, Captain (later Admiral) Marcq St. Hilaire, devised an iterative procedure, now called the intercept method. This concept was based upon an assumed position and calculated with haversines, resulting in a line of position, or as some knew it, a “Sumner line”. Philip Van Horn Weems took the next giant step in speeding up the celestial sight reduction process to five minutes or less. Weems challenged the hallowed methods of the Navy and set out to simplify calculations using the Moon for celestial navigation. He developed “The Lunar Ephemeris for Aviators” which worked so well, he applied it to stars, planets and the sun. It was published as the “Air Almanac” in 1933, though it was discontinued in 1934 by the Nautical Almanac Office. An enthusiastic endorsement of his Air Almanac by the British led them to publish it every year since 1937; the U.S Nautical Almanac Office resumed publication in 1941. Weems encouraged fellow Naval officers Dreisonstok and Ageton who developed popular tabulated “short methods” in very concise formats, well suited for aviation. Not satisfied, Weems simplified the reduction process even further with his pre-computed “Star Altitude Curves” for pre-selected stars capable of producing good “cuts”. The navigator had only to enter a graph with the appropriate arguments and extract the data to plot his fix.  In his Air Navigation (1938) Weems wrote: “The weak link in celestial navigation at the present time is that altitudes cannot be observed with extreme accuracy with the present aircraft sextant. When, however, accurate altitudes can be observed, the recent methods (of sight reduction) give positions with great speed and accuracy.”  In 1939 it was discovered celestial observations were in error due to acceleration of the bubble from the coriolis affect. This error of up to several miles varied with ground speed and latitude, and was uncompensated for at the time of the World Flight. 
Radio: Aircraft radio developed rapidly during the 1930’s, analogous to the fast-changing computer technology of today. A leader in the field of radio communications, the Western Electric division of AT&T, manufactured the Model 13C radio transmitter and the Model 20B receiver which were selected for the Earhart “Flying Laboratory”, but it was already three-year- old technology at the time of the World Flight. The Western Electric radios were adequate, but couldn’t be considered “top of the line”. Model 13C was 50 Watt, three-frequency transmitter operating in the 2500-6500 KHz range and was factory modified to operate on the maritime distress frequency of 500 KHz. As installed, it was capable of transmitting a Morse code signal as well as phone (voice).
Radio direction finding circuitry had been developed in the 1920’s. Ships and ground stations operated direction finding radio for years, but weight had been a limiting factor for aircraft. The first radio direction finder, or radio compass, designed for aircraft became available in the early 1930’s, and newer, upgraded technology developed by several companies was available by the mid thirties. The Earhart Lockheed was equipped with a new generation radio direction finder in 1936 (which became known as the automatic direction finder or ADF), designed by Frederick J. Hooven, Chief Engineer and Vice President of the Radio Products Division of the Bendix Aviation Corp. However, Earhart removed this state of the art Hooven Radio Compass and installed one of lesser capability with older technology having a manually rotated loop antenna, ostensibly to save 30 pounds of weight. 
The World Flight
Navigational Equipment: The Electra’s navigator station was in the aft cabin, behind the internal fuselage fuel tanks. Communications between the navigator and pilot were by written notes passed with a bamboo stick. Following the Luke Field, Hawaii takeoff accident, equipment for the first World Flight attempt was inventoried by U.S. Army personnel who shipped the Electra to Lockheed for repairs. This inventory gives us a glimpse of the equipment available for the second attempt of the World Flight.
Navigation Equipment Taken from the Luke Field Inventory
|30||12||Ea.||Aircraft Water Lights|
|31||7||Ea.||Aluminum Direction Bombs|
|60||1||Ea.||Base Plate for speed and drift meter|
|89||1||Book||Radio Aids, Navigation|
|91||1||Book||List of Broadcasting stations|
|92||2||Books||American Nautical Almanac 1937|
|93||1||Book||List of Coast Stations & Ship Stations|
|94||1||Book||List of Aeronautical Stations and aircraft stations|
|95||1||Book||List of Stations performing special services|
|96||2||Navigation tables for Mariners and Aviators|
|97||1||Ea.||Envelope containing miscellaneous navigation papers|
|106||1||Pkg.||Navigation Charts and airplane log|
|107||1||Ea.||Speed & drift indicator, type D-270, with handbook|
|112||3||Folders with maps|
|115||1||Pencil type flashlight|
|117||4||Clocks, Start & Stop "Omega"|
|118||1||Airspeed Indicator "Pioneer"|
|119||1||Gage Air Temp. Model 602|
|121||1||Altimeter, Kohlsman, 0 to 20,000|
|122||1||Pelorus drift sight, MK II B with extra base|
|123||1||Straight flight compass|
A bubble octant is not listed in the inventory. Harry Manning had signed for Navy Pioneer Bubble Octant, Serial No. 12-36 from the Naval Air Station in San Diego and retained possession of it rather than have it shipped back with the damaged aircraft. Noonan then signed a receipt for the octant on Matson Line stationery and gave it to Manning following the Luke Field accident. We do not know whether Noonan used that instrument on the second attempt or whether he returned it and used something else.  
Earhart’s aircraft was equipped with the Sperry “AutoGyro Automatic Pilot”, providing a more stable platform for the navigator’s celestial observations increasing their accuracy. Also on the Electra was a Mk IIB Pelorus drift sight with an extra base to facilitate its use on either side of the aircraft. Adequate for drift readings during stable flight, it was difficult to obtain accurate drift readings in any amount turbulence. The inventory also listed a second “Speed and Drift Indicator, type D-270” with a base. A New York Herald Tribune article included in Amelia, My Courageous Sister, by Earhart’s sister Muriel Morrissey, described how the drift sight was used. “An arrangement has been devised to open the cabin door about four inches, where it is held rigidly in place. A Pioneer drift indicator is mounted for use looking down through this aperture to check wind drift on the earth or sea below. For this work flares are used at night over water, smoke bombs in daylight.”  Drift bombs, used in lieu of the less visible smoke bombs, were ceramic or glass bomblets filled with either bronze or aluminum shavings. They broke on impact with the water, and created a spreading reflective surface on the water that could be tracked with the drift sight. Magnesium water lights were used similarly during darkness.
The Lockheed Electra 10E was fitted with low distortion windows in several positions to minimize refraction errors during celestial observations.  (Military navigators were later cautioned to avoid using windows for celestial observations, and not to observe celestial bodies below 11 degrees, due to the excessive refraction error.) 
While the Earhart Electra 10E Special was initially equipped with adequate navigational equipment, modifications to communications equipment made following the Luke Field accident were not well thought out, and were accomplished by technicians who may not have been fully aware of the nuances of radio wave propagation. The trailing antenna system had been a victim of the accident and was not replaced. A modification to lengthen the dorsal “V” antenna inadequately compensated for the trailing wire antenna’s loss, and resulted in degraded radio performance on all frequencies.
Noonan’s Navigational Procedures: In a letter to Weems, Fred Noonan described his navigational procedures during a 1935 Pan Am Pacific flight, stating that he used a “Pioneer octant with a mariner’s sextant as a “preventer”. Noonan also described carrying marine general, coastwise, and harbor charts, as well as aviation strip charts of the California coast. During the Pan Am flights overwater navigation was done on VP-3 and VP-4 plotting sheets (blank charts for a band of latitudes with user defined longitude lines). These were reused by relabeling the longitude lines and transferring his position to continue at the appropriate latitude on the same chart. This allowed him to use only two charts for the overwater passage. In the letter, Noonan also told of his preference for Dreisonstok’s reduction tables, and the Dalton Mk VII dead reckoning computer. Pan Am navigational policy, which Noonan helped develop, dictated that celestial positions were to be taken hourly, or more frequently, day or night. All forms of navigation would be used, and positional data would be cross checked between the different methods. It was also Pan Am policy that at least two direction finding stations would “track” the aircraft at all times, and aircraft DF equipment would be used to take bearings. 
Original charts of the Oakland to Honolulu leg of the World Flight first attempt illustrated the use of the following navigational techniques:
- 7 radio bearings
- 14 star/planet LOP’s
- 9 navigational fixes
- 4 course corrections
The flight path proved to be consistent with subsequent meteorological patterns for the area, and that corrections were made when deviation from course became too great.  The archived chart for the June 7, 1937 Natal, Brazil to Dakar, Senegal, leg of the second attempt of the World Flight, showed that Noonan used 5 sun line LOP’s including a noon position. A portion of this chart is shown here to illustrate his chart work.
Training--A Missed Opportunity Commander PVH Weems offered to provide Earhart with celestial navigational and radio skills, including Morse code, which he felt was necessary for overwater flight. His May 14, 1937 letter was 5 days before Earhart’s Oakland departure on the second attempt of the World Flight and was politely parried by George Putnam. Neither Earhart nor Noonan possessed Morse code skills.
A Mark IIB pelorus drift sight like that installed on the Earhart Lockheed Electra 10E for the World Flight. A U.S. Army inventory made following the Luke Field accident included this type drift sight and an extra base, plausibly for mounting the drift sight on either side of the aircraft. (Courtesy: Chris Rudge www.warbirdsite.com)
- Sobel, Dava, Longitude: The True Story of a Lone Genius Who Solved the Greatest Scientific Problem of His Time, Walker and Company, New York, NY, 1995
- Wright, Monte Duane, Most Probable Position, The University Press of Kansas, Lawrence/ Manhattan/ Wichita, 1972, Ch. 3
- Wright, Monte Duane, Most Probable Position, The University Press of Kansas, Lawrence/ Manhattan/ Wichita, 1972, Ch. 6
- Wright, Monte Duane, Most Probable Position, The University Press of Kansas, Lawrence/ Manhattan/ Wichita, 1972, Ch 6
- Wright, Monte Duane, Most Probable Position, The University Press of Kansas, Lawrence/ Manhattan/ Wichita, 1972, p. 167
- Wright, Monte Duane, Most Probable Position, The University Press of Kansas, Lawrence/ Manhattan/ Wichita, 1972, pp. 142-144
- Weems, P.V.H., Air Navigation, McGraw-Hill Book Company, Inc., New York and London, 1938, p. 316
- Naval Oceanography Portal, History of the Air Almanac
- Emmot, N.W., "The Grand Old Man of Navigation”
- Weems, P.V.H., Air Navigation, McGraw-Hill Book Company, Inc., New York and London, 1938
- Wright, Monte Duane, Most Probable Position, University Press of Kansas, Lawrence/ Manhattan/ Wichita, 1972, p. 153
- Everette, Michael, A Technical Analysis of the Western Electric Radio Communications Equipment Installed on Board Lockheed Electra NR16020
- Hooven, Frederick J., The Hooven Report, 1982
- Long, Elgen M., Amelia Earhart, the Mystery Solved, Simon and Schuster, New York, N.Y., 2001, p. 73
- Morrissey, Muriel Earhart, Osborne, Carol L., Amelia, My Courageous Sister, Osborne Publisher, Inc., Santa Clara, 1987, p. 192
- Wright, Monte Duane, Most Probable Position, The University Press of Kansas, Lawrence/ Manhattan/ Wichita, 1972, p. 156
- Air Force Navigator Observer Association, "DR Ahead",Vol 27 No. 1, January 2011
- Weems, P.V.H., Air Navigation, McGraw-Hill Book Company, Inc., New York and London, 1938, pp. 423-425
- Wright, Monte Duane, Most Probable Position, The University Press of Kansas, Lawrence/ Manhattan/ Wichita, 1972, p. 167
- Jacobson, Randall S., Ph.D., The World Flight, First Attempt: Oakland to Honolulu, TIGHAR,2006
Ageton, Arthur A., Manual of Celestial Navigation, D. Van Nostrand Company, Inc., New York, 1942
Dreisonstok, J. Y., Navigation Tables for Mariners and Aviators, United States Government Printing Office, Washington, 1930
Eberle, William C. and Weems, P.V.H., Learning to Navigate, Pitman Publishing Corporation, New York and Chicago, 1939
Kells, Kern & Bland, Spherical Trigonometry with Naval and Military Applications, USNA, 1942.
Morrissey, Muriel Earhart, Osborne, Carol L., Amelia, My Courageous Sister, Osborne Publisher, Inc., Santa Clara, 1987
Sobel, Dava, Longitude: The True Story of a Lone Genius Who Solved the Greatest Scientific Problem of His Time, Walker and Company, New York, NY, 1995
Weems, P.V.H., Air Navigation, McGraw-Hill Book Company, Inc., New York and London, 1938
Wright, Monte Duane, Most Probable Position, The University Press of Kansas, Lawrence/ Manhattan/ Wichita, 1972