DEFECTS

The DH-4 as first sent to the western front was unfit for practical use. It contained structural defects that practical tests readily revealed overseas. After thorough inspections and tests on the other side, many recommendations of changes were cabled to America. Most of the changes recommended were of a trivial character, some of them of importance, and most of them, but not all, readily over

come.

The running gear was found weak, resulting in crashes where the landing was not made under favorable circumstances. This defect was overcome by reinforcements made in France and by the use of larger and stronger materials in future machines sent from the United States. The same was true of the shock absorbers.

The radiator from the engine was too small, resulting in overheating in warm weather. This required enlargement in future machines and a supply of larger radiators from the United States. The strut fittings were too weak and had to be replaced by stronger ones. The tail supports were insufficient in strength and had to be reinforced until replaced.

Shipment from America was temporarily retarded while the above and other defects of the original machine were remedied in the United States. This resulted in a decreased shipment for the one month of August 1918.

Before the De Havilands were sent over the German lines they were tried out by the test pilots in France. Gen. Patrick refused to let them be used over the lines until defects were remedied (205).

LOCATION OF GAS TANKS

After correcting the above-named defects and some others of a less important character, the De Haviland was still subject to some substantial criticism. The other criticism was directed at the safety of the machine rather than to its performance in the air. One of the most substantial of these objections was the location of the gas tank back of the pilot, between the pilot and the observer, and under which arrangement the pilot was between the engine and the gas tank (3505). In case the machine crashed or turned turtle on landing, there was always the danger of the main support of the body giving way under the force of the contact, resulting in the crushing of the pilot between the engine and gas tank.

The DH-9 was really a re-formed DH-4, and the elimination of this danger to the pilot was one of its principal features. Since the war the Army (3832) and the Post Office Department have rebuilt a number of the De Havilands, placing the pilot behind the gas tank and strenghtening the support of the body. The landing gear has been moved forward about 7 inches. With this new arrangement the danger of the machine turning over on landing in rough places has been greatly lessened. The danger of jamming the pilot between the gas tank and the engine has been eliminated. The safety and steadiness of the machine has been thereby greatly improved (3832, 3854, 3871).

The comparative simplicity of the change in the location of the gas tank and the factor of safety realized thereby emphasizes the

unwisdom of the original construction. The practical effect of the old arrangement was perhaps not so much in the number of lives that may have been lost thereby as it was in the loss of morale of the pilot, who operated his machine under a constant apprehension in case of a crash or somersault in landing. It lessened his faith and self-confidence in the operation of the machine, particularly if he was an inexperienced pilot.

However, this criticism can not be directed alone against this American-made machine. Probably 50 per cent of the machines used on the western front had the gas tank in the rear of the pilot, including some of the Spads, the Sopwith-Camel, the Sop, and Snipe (3506, 2910-CI, 2645, 2452).

Looking back with the knowledge of after-acquired wisdom, it seems the practical thing to have done would have been to have tried the change of location in one of these machines in the early stages of production.

DANGER FROM FIRE

Another objection of a serious character commonly urged against the De Haviland was its danger from fire. The gasoline was supplied from a main tank under pressure feed. It seems to be conceded that the danger of fire is greater from a pressure tank when pierced by incendiary bullets than from a gravity-feed system (2910-CII).

Many of the machines used on the front, including the Spad, had pressure tanks (2628, 201).

Persistent complaint has been made of a fire risk occurring in the De Haviland on account of the lack of a protected tank. There were no protected tanks in use up to the time the De Haviland plane was adopted by the United States. In the spring and summer of 1918 the Salmson, a French machine, came out with a cover over the gas tank, consisting of a combination of rubber, a soapy combination, and wire gauze to hold the same together, which, when pierced by a bullet, would expand and close up the hole, thus preventing the escape of the gasoline and tending to prevent ignition. On the 7th of July 1918, this tank was installed on the French Breguet.

In September 1918, a call from the front came for a protected tank for the De Haviland. Orders for the same were let on the 18th of October and 25th of October. An order was let on October 20 for 500, and on November 2 an order for 1,000 was given. Covers began to arrive about the date of the armistice (2910-CII-CIV).

The lack of a cover increased the fire danger from incendiary bullets. It also increased the danger from any penetration of the gas tank, because experience shows that frequently the gas escaping from the penetrated tank will be ignited by the exhaust from the engine. The same thing occurs in a crash on landing when the gas tank is ruptured and throws the gas over the heated engine (3679).

The Germans and the English did not use covers on their gas tanks (2910-CIII, 182). The machines furnished by France for the use of the American flyers were protected in the same way that similar type machines used by themselves were protected after the use of the covering was introduced.

Many misleading statements have been made about the fire risk of the De Haviland. Unquestionably a fire risk existed. It was not,

however, a risk peculiar to the De Haviland, but it was a risk that was shared by every airplane that went over the enemy's lines. A machine protected by the gas-tank covering finally introduced to a degree lessened the fire risk, but did not eliminate it (201, 3624, 3626, 3680).

The danger of fire was not peculiar to the De Haviland, but practically common to all airplanes (3680, 3681).

It is the opinion of Gen. Patrick that no larger percentage of De Haviland 4's came down in flames than other types of machines (205).

The casualty list due to operation of the De Havilands in battle, over the battle lines, does not indicate any unusual loss, so far as there is a basis for comparison with other machines. Fourteen and fifteen hundredths per cent of the total airplanes lost by the American squadrons were De Havilands, or 11.38 per cent of the total airplanes lost by the Americans, including machines attached to the service of the Allies.

Only 38 were killed in De Haviland 4's due to the enemy; 8 were wounded; 10 missing; and 20 taken prisoners, making total casualties of 76. It is obvious that only a small number perished in the flames of their ships.

The total airplane casualties on the front up to November 11 were 516, in addition to 42 men killed by accident (2910-CIV-CV).

Only 33 De Haviland 4's were lost over the lines in fighting (3306) out of 417 that went over.

The De Haviland was an excellent machine when selected for manufacture by the Americans. Before it reached the front, it had lost its relative position and was inferior to later machines which had preceded it to the front. As time has advanced since then it has retrograded, due to superior advantages of later machines. It performed a useful service in the war.

Criticisms directed against it after the time that it was placed over the battle lines were for reasons that were not peculiar to the De Haviland alone, but were more or less common to contemporaneous machines of all belligerents.

Most of the flyers who appeared before the committee and criticized the De Haviland 4 were men who had little or no experience with that machine. If experienced flyers at all, their experience was in single-seater pursuit planes, which are an entirely different grade of machine from the De Haviland, small in size and built for maneuver ability. No information is obtained by comparing the De Haviland with single-pursuit combat planes, which are intended for entirely different purposes and were used from different bases of operation.

Lieut. Maynard, who was test pilot for the Americans in 17 different kinds of machines in France, and had visited many of the front line stations and practically all the training fields in France, and had flown the De Haviland for over 600 hours, and with it performed practically every test to which a machine is subjected, candidly admitted defects in the De Haviland, but declared that in talking with many American pilots he found the principal criticism came from those who had little familiarity with its practical operation, and that he regarded the De Haviland as good as the other machines of similar type at the front during the war (3503, 3504, 175).

Lieut. Maynard had flown 10 times as many hours in the DH–4 as the combined hours of flight of all the other witnesses who appeared before the committee and criticized the DH-4.

Mr. Keyes, of the Curtiss Co., says that the DH-4, notwithstanding its defects, had been "a good machine; it was a good

performer" (3457).

*

*

The DH-4 was the chief observation plane of the Army and enabled the United States forces to constantly be informed of the operations of the enemy behind the lines. For more than three months preceding the end of the war bombing operations were constantly carried on over and beyond the German lines.

No unusual loss of life resulted to the American pilots, and the result of its work was highly beneficial to the American forces. To declare it an utterly useless machine is to ignore plain facts. Such declarations, instead of establishing the uselessness of the DH-4, rather suggest the utter uselessness of investigations that result in such reports.

ENGINES

During the war the Army acquired 37,375 aeronautical engines. (520.) Of these 30,630 were manufactured by the United States (3311) and 6,745 were delivered to the Army by the Allies (520).

Of the engines manufactured in the United States, 13,286 were produced prior to the 1st of July, 1918, leaving 17,344 furnished after that date.

The largest production of any one engine was the Liberty 12, of which 13,574 were manufactured during the war. The total production of the other principal engines was (3311).

Deliveries during the war increased from 28 in April, 1917, to 187 in August and 635 in December, with a total delivery of 2,431 for 1917. During 1918 deliveries increased from 743 in January to 5,329 in October, with a total to November 11 of 30,630.

The A-7-A reached a production of 28 per month in August, 1917; of the Curtiss OX45, 64 were produced in July, 1917; the Hispano-Suiza reached a production of 92 in November 1917; the Liberty reached a production of 21 in December 1917; of 40 in January 1918; 70 in February; 415 in April; 1,589 in July; and 3,878 in October.

During the war the Army delivered to the Navy 3,483 engines, all of which were Liberties except one.

The Navy secured all of its motors from the Army. In the early part of the war there was a partial distribution of the program between the Army and the Navy under which the Army was to furnish the engines, as was done (3517).

America's principal reliance for training-plane engines was upon the Curtiss OX-5, which was of demonstrated worth at the beginning of the war. Quantity production was shortly reached and, substantially speaking, the training-plane engine need was satisfactorily met and supplied (3581, 3582).

The experience of the war demonstrated the superior ability of America to produce engines in production quantities. It was easier for Europe to make planes in quantities than it was to make engines. The United States had greater success in quantity production of engines than of planes. European methods of engine manufacture, being largely based upon hand work as distinguished from machine work, do not result in quantity production, neither can their machines be readily adjusted to American methods of manufacture. Prior to the war the Gnome and the Hispano-Suiza had been introduced in this country for the purpose of manufacture. Contracts were made with Great Britain and France. Great difficulty was experienced in accommodating them to American methods. It required more than nine months to complete the Gnome engine, notwithstanding English experts were present in the contractor's plant and aided in its manufacture. French experts were likewise sent to assist in the manufacture of the Hispano-Suiza. It took 13 months to produce the first engine and another month before it could be delivered (2910-XXX-XXXI). The attempts were accompanied by costly experiments and the development of methods and tools requiring intricate operations. There were engines perfected by European practice before this attempt to conform their manufacture to American methods was made. Before these engines reached quantity production their qualities had been surpassed by engines subsequently introduced on the western front, notwithstanding the American manufacturer was finally successful (3582).

On recommendation of the Bolling Commission a Bugatti was sent to the United States for manufacture. The design and development of this engine created such difficulties that it had not reached quantity production at the time of the armistice, 14 months later.

The English Rolls-Royce was one of the best engines produced during the war. The Americans were favorably inclined toward its manufacture in the early part of the war. A sample engine was shipped to the United States. A royalty of $2,000 per engine was at first asked. Eventually a proposition was made to manufacture this engine in the United States, in which the number promised within over 11 months was 500. The Rolls-Royce had over 200 parts more than the Liberty engine. The American manufacture thereof was expensive and difficult of adjustment to American methods, it being largely a hand-made engine, therefore, the purpose of manufacturing in America was abandoned (3526, 2910–XXXIII). The experience during the war, of the Allies as well as ourselves, demonstrated that from the starting of a plane or engine to production required from one to two years (3475, 3457, 3339). Each engine and plane on the front had its inception many months previously (3550).

THE LIBERTY ENGINES

Three years of the European war failed to solve the engine problem for airplanes, so far as a powerful engine required for heavy planes, that could be put in quantity production, was concerned. Soon after we entered the war it was realized that what America and the Allies needed was not necessarily the finest engines in the world, but there was an extreme need of a powerful, first-class engine that could be placed in quantity production. While the Allies