OBJECT of flight: To prove the practical application of scientific navigation to aircraft for transoceanic flying.
Our reason for not carrying radio equipment or pontoons was to prove that navigation by dead reckoning and celestial observations could be successfully applied in a practical way without the aid of radio, or being forced to land on the water in order to obtain a clear horizon for observations. Our original plan was to fly to Bermuda one day, land, and return the following day. Upon considering the problem of landing on the island, we concluded that a suitable place for that purpose did not exist there, and so decided upon a nonstop return flight.
Crew: Pilot, Roger Q. Williams; navigator, Lieutenant (J.G.) H. P. McLean Connor, U. S. Naval Reserve; co-pilot, Erroll Boyd.
The Columbia was fitted with a small compartment in the rear of the main tank which was used for the navigator’s compartment. An observation hatch was cut overhead and fitted with a removable windshield to permit taking observations. Without the aid of this windshield it wou)d have been impossible to hold a sextant steady against the tremendous air stream.
The Columbia was fitted with the usual motor instruments. We had three compasses, one earth inductor with dial, one cowling magnetic compass, and one cabin magnetic compass. These compasses were carefully checked for deviation and at no time were the compasses out of order.
The flying instruments consisted of one of each of the following: rate-of-climb indicator, turn-and-bank indicator, air-speed indicator, Daiber slow-bubble indicator, Sperry artificial horizon, thermometer, and tachometer.
The lighting system consisted of two separate systems: two dash lights conveniently arranged on dashboard with a rheostat control. Two flash lights (special) for emergency, also for energizing phosphorous material on flight instruments.
The navigational instruments Consisted of an octant with a bubble attachment and one aero chronometer and three hack watches. Where precision in celestial navigation is required, it is essential that chronometers and watches be accurate to the split second. We also carried one nautical almanac and nautical tables as well as one copy of Dreisonstok tables for quick calculations. The aircraft plotting sheet, published by the United States Hydrographic Office, which is specially divided into ten-mile squares, for quick determination of latitude and longitude, was used to good advantage on our flight, with celluloid aircraft plotter, parallel rulers, and dividers. For a chart table I used a ply board 2 by 3 feet, which fitted conveniently on top of the main tank in the after end of the ship, and made an excellent chart table.
Graduated on the stabilizer were drift angle lines for every 5° from 0° to 30° on each side of the fin in order to permit taking accurate drift “sights.”
We carried six Coston calcium water lights for checking drift at night. Of these we used only two.
The observations I obtained of the sun were calculated by line-of-position methods, each position obtained being carefully verified and used in conjunction with dead reckoning navigation.
The earth inductor compass dial and artificial horizon were separated from the dashboard by a special rubber mounting to absorb vibrations.
For sanitation purposes the crankcase breathers on the motor were extended by a rubber hose through the cowling to permit gases to escape into the air, thereby keeping fresh air in the cabin. The air vent on the large tank was extended from the roof of the cabin to prevent fumes, and thereby eliminate the possibility of the crew becoming air sick. At intervals of half an hour all windows were opened to permit change of air.
As an aid to night flying and to prevent the exhaust gases from annoying the pilot while flying blind, the exhaust manifolds of the top cylinders were shaped in such a manner as to bring the exhaust flames below the pilot’s line of vision.
The right rudder pedal was equipped with a series of elastic bands of different tensile strength, thereby permitting the pilot to fly “hands and feet off” with ease. This ship usually carried right rudder.
As a full measure of safety, a special drain for the carburetor strainer was installed in the cockpit, permitting the pilot to drain it while in flight.
The motor functioned perfectly throughout the entire flight. We carried 333 gallons of gasoline, 12 gallons of oil for lubricating purposes, in addition to 3 single gallons in cans for an emergency. Useful load 2,668 pounds including weight of fuel, three men, and instruments—weight of ship in light condition 2,100 pounds, total weight at take-off 4,768 pounds.
The ground speed of the motor upon warming up was 1,600 r.p.m., with an oil temperature of 40° Centigrade. The takeoff was made with a north wind, five to eight miles per hour, barometer 30.14 inches, temperature 69° Fahrenheit.
At 5:01 a.m., 60th meridian time, we slowly started down the 3,800-foot runway. The take-off was in a cross wind, and after running nearly the entire length of the runway, 3,000 feet, we lifted in twenty-four seconds and slowly attained an altitude of 200 feet, with an oil temperature of 48° Centigrade, and turning a maximum r.p.m. of 1,850 in the air.
We set our course for Ambrose Light Vessel, from which we took our departure at 5:17 a.m. Williams throttled the motor back to 1,640 r.p.m. in order to economize on fuel consumption. As we passed over Ambrose Light Vessel, we set our course to steer 138° true, which would take us to Gibbs Hill Lighthouse in Hamilton Harbor, Bermuda, a total distance of 657 nautical miles, or 755.56 statute miles. With a northeast wind, force three to four, we were making 11° drift to the right. The weather was fine, with hazy horizon.
After an hour’s run, the wind gradually became northwest with light breezes. We were averaging eighty knots, and obtaining our position by observations of the sun. With the change of wind, naturally we had to make an allowance for drift, and we were making our course good.
At 6:00 a.m. we calculated we would pick up the SS. Fort St. George on the track, which was southbound, at about 9:25 a.m. About 8:00 a.m. we passed over the northern limit of the Gulf Stream, in latitude 37° 50' north, 70° 53' west. There was no noticeable change in the air except for a slight bumpiness at times. The temperature of the air was 83°. A cloud formation of false cirrus and cumulo-stratus. Visibility twenty miles.
At 9:25 a.m. we sighted the SS. Fort St. George, and at 9:35 circled her and exchanged courtesies. She saluted us with three blasts of the siren and we returned this salute by three dips of our port wing. As we passed her we sighted the motor ship Bermuda, southbound, ten miles to port, or to the left of our course and abeam of us.
We circled the Fort St. George twice and then headed directly over her fore-and-aft line. By this method we checked our compasses on the vessel’s heading, and as we left her astern, I obtained another “drift sight,” which showed that we were making 8° drift to the right.
By this time the wind had again veered east, force three to four. With an average r.p.m. of 1,600 to 1,625, we were averaging seventy-one knots ground speed. The airspeed indicator registered ninety to ninety- one statute miles. The weather continued fine with some cumulus clouds.
At 10:39 a.m., or 14:39 Greenwich civil time, a line of position by the sun showed us to be twelve miles to the right of our course. Upon calculating our position and verifying it, I changed the course to 133° true (5° to the left) in order to bring us on our track. The fact that we had got off our course twelve miles in such a short space of time served to indicate to me, with even very careful dead reckoning, the necessity for obtaining good observations. We continued on this 133° true course until 1:00 p.m., when we were on our original track line, and then hauled the Columbia to 137° true, the course to take us to Hamilton Harbor.
By this time the wind had veered from east to southerly and south-southwest, force three to four. This naturally reduced our speed, and we were averaging only about sixty-eight knots. The sky became cloudy and partly overcast with a hazy horizon. The cloud formation was cumulus and cumulo-nimbus, with some flying scud. We were then flying at an altitude of 400 feet, with a visibility of about ten miles.
As we approached the island about 2:00 p.m., the visibility became less and the ceiling lower. The sky was overcast, and we were unable to obtain observations. Thick nimbus clouds were ahead of us, and we commenced to encounter rain squalls, the wind backing to southerly, force four to six, which was just as Dr. Kimball, of the New York office of the United States Weather Bureau, had predicted before the take-off; that is, that we could expect overcast weather with showers around Bermuda in the early afternoon.
By 2:10 p.m. the visibility was less than two miles, and at 2:23 p.m. we sighted the radio towers at Hamilton Harbor. At 2:24 p.m. we arrived over the harbor in a heavy downpour, with a visibility of less than half a mile. Flying between 50 and 150 feet altitude we made a complete circle of the islands. From a twenty-minute aerial survey of the island of Bermuda it was evident that there was no suitable field for landing purposes as there was not a long enough straightaway. For seaplanes the logical anchorage will be Hamilton Harbor, which is a landlocked harbor with a buoyed channel. At 2:43 p.m. Williams dropped the mail sack in the residential section near Hamilton. We later learned that the mail bag was delivered promptly to the postmaster. At 2:44 p.m. we headed for New York.
We set our course to steer 322° true from Gibbs Hill Lighthouse in order to intercept the motor ship Bermuda, which was southbound, on the 137°- 317° true steamship track. By 3:00 p.m. we had run out of the bad weather. We continued on this course until 4:06 p.m., or 20:06 Greenwich civil time, when an observation showed us to be slightly to the left of this course. About 4:30 p.m. we passed over the southern limit of the Gulf Stream and, as with our experience in the morning, there was no appreciable change in the air. The wind was east- southeast, force three to four, and we were allowing 6° to 7° for drift. The weather was fine with cirrus and cirro-cumulus cloud formation.
I neglected to add that on commencing our homeward-bound trip, Williams stepped her up to 1,675 r.p.m., and, with the aid of favorable winds, we were averaging ninety- five knots ground speed, or between 108 and 110 statute miles per hour.
At 5:00 p.m. we sighted the Bermuda slightly on our starboard bow as we had expected, and at 5 :06 p.m. circled her, accompanied by the usual exchange of courtesies and salutes. Similarly as with the Fort St. George, we headed directly over her fore- and-aft line and obtained a good check on our compasses. As we were then on the regular direct line, we changed our course to 317° true, which would bring us into Ambrose Light Vessel. Subsequent observations showed us to be making this course good, and accompanied by easterly and southeasterly winds, force three to four, we were stepping right along at ninety-five to ninety-six knots ground speed.
We passed over the northern limit of the Gulf Stream about 7:30 p.m. and the temperature of the air gradually dropped from 78° to 74°. We were flying at an altitude between 500 and 800 feet.
At 7:56 p.m. we sighted the SS. Fort St. George dead ahead and at 8:04 p.m. passed her to starboard. This time we did not circle the vessel as night was coming on and we were anxious to avoid as much “blind flying” as possible. We passed the vessel on her port side with her siren blowing three blasts in salute and Williams giving the Columbia three quick dips of our port wing in return. In each instance that we circled the vessels we flew alongside the boat deck and were able plainly to see the passengers.
The sun set about 8:30 p.m. and with cumulus clouds around the horizon, the horizon gradually melted. We climbed to 2,500 feet altitude as a safety precaution for “blind flying.”
As darkness approached, the artificial horizon rendered good work and aided us considerably. This instrument is an invaluable aid to night flying over the ocean as it records the plane’s actual altitude in flight relative to the horizon, and enables the pilot to maintain normal flight.
A low haze covered the sea and at 9:40 p.m. we sighted Navesink Lighthouse off the Highlands. We could not distinguish Ambrose Light Vessel until we were right over it. At 9:44 p.m. our sea passage ended and we made a bee line for Roosevelt Field. Thq visibility then was less than five miles, and a ground fog was rolling in from the sea. We were able to discern the lights of Long Island in places and at 9:55 arrived over Roosevelt Field. Williams “jazzed” the motor as a signal for the attendants to put on the flood lights. We were unable to make out the lights on the water tower which is 200 feet high, due to the ground fog, and so we decided it was unwise to land under these conditions.
We headed for Curtiss Field, Valley Stream, where fortunately the flood lights happened to be on, and at 10:03 p.m. Williams made a perfect landing, successfully ending our nonstop flight from Roosevelt Field to Valley Stream, Long Island, via Bermuda, in seventeen hours and two minutes.
Summing up, our average speed southbound was 72.2 knots ground speed, or 83 miles per hour, and northbound, 93.86 knots ground speed, or 108 miles per hour. The air during the whole flight was exceptionally smooth. As far as the flight was concerned, it seemed little different to me from sailing in a fast surface vessel.
Our average fuel consumption southbound was 12 gallons per hour; northbound, 14 gallons per hour. We landed at Curtiss Field with a reserve of 120 gallons of gasoline (equivalent to a weight of 720 pounds), sufficient to remain aloft another twelve hours if necessary.
The total round trip distance from Roosevelt Field was 1,358 nautical miles, or 1,561 statute miles. We obtained photographs of each vessel we passed and one good “shot” of Bermuda.
The navigation of aircraft is fundamentally the same as that of surface vessels, the difference being the greater speed of the former, and the wind drift or leeway to which an airplane is subjected, making it necessary to apply slightly different methods in which accuracy and speed in calculating one’s position are essential.
In ocean flying a thorough knowledge of meteorology and seamanship is also necessary. This, of course, is obtained only from years of experience at sea.
Contrary to the general opinion among even experienced navigators that aerial navigation need not be as accurate as surface vessel navigation, it has been my experience that aerial navigation should be as accurate, if not more accurate, than nautical navigation, especially when endeavoring to hit your point of destination “on the nose.” Those men who have flown in low visibility will appreciate these facts.
The observations of the sun were taken by the bubble octant. Briefly, I applied nautical navigation in a practical way on this flight. This was the ultimate result of some three years of experimentation in applying navigation in aircraft (both dead reckoning and navigation by astronomy). The average time of taking each observation, including opening of hatch, noting time, and altitude obtained, was one minute, ten seconds.
I was most impressed by the realization of how disaster overtakes inexperienced flyers over the water when I realized that the most dangerous time of the flight is the first hour after sunset, as the darkness compels the pilot to rely solely on his instruments to maintain normal flight. In this connection only pilots who have had a large amount of instrument flying should be permitted to attempt ocean flights where so-called “blind flying” is involved.
The success or failure of any venture depends entirely on careful and minute preparation. These preparations include careful selection of crew, plane, motor, as well as exact and reliable instruments, and supervision by experienced mechanics and ground crew.
I want to point out the necessity for perfect team work and cooperation of each member of the crew, which in our case made our flight successful. The careful steering of the compass courses by the pilots was necessary in order to aid in making our courses good, and I wish to pay a warm tribute to Roger Williams and Erroll Boyd for their perfect cooperation.
On our (Williams’ and mine) numerous cross-country flights in which we applied navigation, we frequently arrived at our destination within the short period of thirty seconds of our estimated time.
These facts, coupled with our Bermuda, flight, we believe put aerial navigation on a par with nautical navigation for accuracy and dependability. We have further experiments to conduct in perfecting the practical application of navigation to aircraft, for the use of experienced ocean navigators in applying their talents to aerial navigation.
I recommend either the use of H.O. Publication Number 208 or Weems’ method of navigation for calculating lines of position. Suggested improvement in H.O. Publication Number 208: That a table containing sun’s and moon’s refraction and parallax combined be inserted in this publication for use of the navigator using a bubble octant with reference to the artificial horizon. (The present altitude correction is a total correction with an allowance for the height of the eye, given only for use of the visible horizon.)
Obviously for a regular transoceanic air line, large flying boats with seaworthy hulls will be the practical ship of the future. These vessels, of course, will be equipped with radio for communication purposes and as a safety measure. It is my opinion that radio has not been developed to a point where it can supersede, or where it can be anything more than an aid to navigation in the event of emergency. The ideal set for transoceanic flying ships would be an intermediary, or short-wave set, for two-way communication between plane and shore. With a short wave and high frequency set, the radio operator would be able to obtain radio bearings from a radiocompass station on the ground or surface vessels, but due to the contour of the land and the curvature of the earth and varying atmospheric conditions, these circumstances must be taken into consideration when using this means as an aid for navigation. The practical use of the radio will be to send out position reports and to obtain weather data for safety precautions.
The present radio beacons in use over the land have a range of not over 150 miles, and this distance only in good atmospheric conditions. One of these installed at Hamilton Harbor and another at or near Sandy Hook, New Jersey, would be of some assistance for New York to Bermuda air lines, but should not be relied upon implicitly.
I understand that experiments are now being conducted for a direction finder to be installed in aircraft, similar to those used in naval vessels and large ocean liners. But as in the case of surface craft, they should be used only as an aid to navigation.
In connection with this I have conferred with Radio Corporation of America officials and they confirm this opinion with regard to the use of radio for transoceanic flying. As an aid to the future enhancement of aviation, it is suggested that there should be government supervision of contemplated ocean flights and that for the present only navigators who have had a number of years’ service as line officers in the Navy or Merchant Marine be permitted to act in such capacity.
|