From the Rolls-Royce experimental archive: a quarter of a million communications from Rolls-Royce, 1906 to 1960's. Documents from the Sir Henry Royce Memorial Foundation (SHRMF).
Investigation into fire hazards from airplane ignition systems and the development of fire-proofing methods for magnetos.
| Identifier | ExFiles\Box 37\1\ scan 163 | |
| Date | 1st July 1927 | |
| Vol. XXI July, 1927 No. 1 32 THE JOURNAL OF THE SOCIETY OF AUTOMOTIVE ENGINEERS the generator, must be shielded, and all brace wires must be carefully bonded together to prevent interference. This is particularly true with battery ignition, in which the primary-coil oscillations and the commutator ripple are fed back into the entire low-tension system. This may prove to be a point in favor of magneto ignition, for it may not be necessary to shield the lighting and the starting circuits, if no battery ignition is used. In an ideal ignition system for radio-shielding, the secondary cables would be carried in passages integral with the cylinder-block or valve-housing, and the magneto or distributor would be completely enclosed in an integral metal-housing with a single-flanged opening leading to a corresponding opening in the cylinder-block. Such an installation would give inherent radio-shielding, protect the cables and magneto from oil, water and mechanical injury, and materially reduce the fire hazard. FIRE HAZARD A thorough investigation was recently undertaken by the Experimental Engineering Section of the Materiel Division to determine the cause and means of preventing fires in airplanes, both during flight and after accidents in landing. Obviously, if no gasoline, or other combustible substance, gets near the exhaust stacks, magnetos or other possible sources of ignition, no fires will occur; but so long as gasoline is piped to the carbureters, the engine is liable to be accidentally sprayed with gasoline or surrounded with gasoline vapors. This being the case, the possible sources of igniting these vapors should, so far as possible, be eliminated. With this in mind, various types of magneto and distributor were tested to determine a suitable means of preventing them from starting a fire when they were sprayed with gasoline or surrounded with gasoline vapors. It has already been pointed out that some ventilation of the air-gap distributor is required, so that tightly closing the magneto is out of the question. In testing the magnetos and distributors for fire hazard, they were set up on a test-stand and driven by a motor, as shown in Fig. 3, and gasoline was sprayed over the entire magneto from a blow-torch. The fires that did occur were the result of a sharp explosion inside the magneto that blew flames out of the vent-holes and ignited the gasoline outside the magneto. However, it often required many trials before a fire could be started. It was a great surprise to find that a 40-mesh brass-screen over a 1/8-in. vent-hole was entirely ineffective in stopping the flame. After carefully considering these tests and conditions during flight, it was decided that when any ignition-apparatus is sprayed with gasoline, enough gasoline will creep into the interior to form a combustible mixture that can be ignited by either the sparks at the breaker-contacts or the sparks in the distributor. Since such explosions may occur, it is necessary to design the magneto and the necessary vents in such a manner that no flame will be blown outside the magneto. The type of vent shown in Fig. 4 was finally developed as a simple means of providing ventilation and drainage and, at the same time, of quenching the flame by cooling and diffusing the hot gases, in case of an internal explosion. The present test-method consists of introducing gasoline into the interior of the magneto in the proper proportions to produce a sharp explosion when the ignition is switched on. All the spark leads are connected in the usual way to protected spark-gaps, and the entire magneto and wires are sprayed with aviation gasoline before the spark is turned on. At least 10 such internal explosions shall be obtained without resulting in a fire. This usually means at least 50 explosions of varying intensity. The magneto is then run continuously at normal speed for 30 min. and, at frequent intervals, gasoline is sprayed over the entire magneto. After each spraying, the gasoline is allowed to evaporate before the next spraying, thus going through all stages of vapor density in and about the magneto. If no fire results from these tests, the magneto is considered fire-proof. Another possible source of fire is the loose connections between the spark-plug cables and the distributor, or the terminals of the plugs, which allow small sparks to occur as the cables vibrate. A safety-lock terminal has been developed that positively locks the terminal on the spark-plug and assures a tight contact, at the same time allowing the requisite flexibility. It is simple to operate and FIG. 4—VENTILATING AND FLAME-QUENCHING VENT This Type of Vent Was Developed To Provide Ventilation and Drainage and To Quench Flame from Internal Explosions by Cooling and Diffusing the Ignited Gases FIG. 3—SET-UP FOR TESTING FIRE HAZARD OF MAGNETOS AND DISTRIBUTORS The Test Consisted of Mounting the Magneto and Distributor on the Test-Stand, Driving Them by an Electric Motor and Spraying Gasoline All over the Magneto from a Blow-Torch. Many Trials Were Required Before a Fire Could Be Started. Such as Did Occur Resulted from a Sharp Explosion inside the Magneto That Blew Flames Out of the Vent-Holes. A 40-Mesh Brass-Screen over a 1/8-in. Vent-Hole Did Not Stop the Flame. When Gasoline Is Sprayed over Any Ignition Apparatus, Enough Will Creep into the Interior To Cause a Combustible Mixture That Can Be Ignited | ||