Auto Racing Articles: Spark Plugs

Spark Plug
Article courtesy of http://RacingSecrets.com

In December, 1916, the National Advisory Committee for Aeronautics requested the Bureau of Standards to undertake a laboratory study of the spark-plug problem which had arisen because of the very severe conditions of temperature and pressure existing in modern aviation engines. The work has proved to be of much wider scope than was originally anticipated. It includes an investigation of the various properties of materials entering into the construction of spark plugs with a view to improving their construction and design, and a study of various related problems in ignition systems. Connection with this work, tests have been devised for specifications, and routine acceptance tests made on plugs purchased by the Government.

The work has also involved numerous conferences with representatives of all the more important manufacturers of airplane spark plugs, and also with members of the foreign scientific missions.

As a. result of these conferences it appears that the sources of trouble in spark plugs may be roughly grouped as follows:

I. "Cut-out," or failure of the spark, due to short circuit resulting from the electrical conductivity of the body or the surface of the insulating material of the plug at high temperature.

2. Preignition due to excessive heating of parts of the plug.

3. Fouling of the insulating parts by carbon.

4. Cracking due to differential thermal expansion.

5. Excessive gas leakage.

6. Minor causes, such as warping of electrodes, bridging of the gap with oxides, carbon, or oil, direct puncture of the material, and mechanical breakage.

Some of these troubles, such as mechanical breakage, bridging, etc., are essentially accidental and can be studied satisfactorily only from statistical data. Others, such as "cut-out," fouling, and preignition, lend themselves more readily to experimental study and most of the work thus far has been. aimed at these troubles. An understanding of the causes of these failures requires primarily a knowledge of the conditions existing in the cylinders and spark plugs of a modern aviation engine. The first two also require a knowledge of the ability of ignition apparatus to deliver a spark from an imperfectly insulated plug, and the first is intimately connected with the electrical resistance of the insulating material at high temperatures.

The gas leakage tough a spark plugb though not in itself enough to affect the operation of the engine, is exceedingly detrimental since it rapidly heath the body of the insulator and brings on other types of failure frequently causing the complete destruction of the plug. The work thus far accomplished will therefore, be discussed under the headings of "Engine conditions," "gnition systems," "Gas tightness," and' Electrical conductivity at high temperatures."

ENGINE CONDITIONS.

To indicate the conditions under which spark plugs operate, a study of the sparking voltage and temperature conditions in a gasoline-engine cylinder has been undertaken. Using a pressure bomb which could be heated in an electric furnace, measurements have been made of the sparking voltage at 60 cycles across each of three different air gaps. This was done at temperatures up to 430°C. and at pressures up to 7.5 atmospheres. The data confirm the conclusions of other observers at lower temperatures, that the breakdown voltage is proportions] to the density of the gas and is hidependant of pressure or temperature at constant density. It was noted that when the electrodes are heated as is the case in a plug firing steadily from a magneto, the gas in die gap is thereby warmed and the density diminished so that the sparking voltage is consequently reduced.

By using a Kenotron vacuum tube in series with a static voltmeter, it was found possible to measure the maximum voltage on an ignition circuit. The voltage across a plug with a gap adjusted to 0.020 inch in a Hall-Scott aviation engine running at full load was found to be about 6,000 volts. This value can be reconciled with that computed from the density of the gas in the cylinder at the instant of firing by making reasonable allowance for the heating effect of the hot electrodes on the gas near them.

A number of spark plugs have been drilled so that thermo couples can be inserted in the center electrode or in the porcelain inside the shell. Determinations of the temperature at these points have been made in several engines. Temperatures as high as 800° C. in the electrode and 400° C. in the porcelain have been obtained. It is highly desirable that further measurements of this kind be made on other types of plugs and engines particularly on engines which are known to be severe on plugs, such as the Hispano-Suisa..

IGNITION SYSTEMS.

As an aid in studying the behavior of ignition systems a number of instruments have been devised including a thermocouple cross for measuring the current, the peak voltmeter mentioned above, and a copper calorimeter for measuring the entire heat energy of the spark.

A collection of magnetos of different makes have been submitted by the Signal Corps for an exhaustive series of tests. The energy per spark from each- of these has been measured at several speeds and found to vary from 0.05 to 0.12 joule per spark. Also a determination has been made of the resistance which, put in parallel with the magneto, will draw so much current that the magneto fails to ea spark at about 6,000 volts. The value of this resistance has been found to be of the order of 80 000 ohms. Further measure-ments with an oscillograph will be taken up as soon as more assistants can be obtained.

ELECTRICAL CONDUCTIVITY

One very important property of the insulating material of a spark plug is its ability to maintain a high electrical resistance at the high temperatures at which a spark plus operates and a very considerable amount of work has been done m studying this property. Over 100 complete curves of resistivity vs. temperature have been obtained on specimens of various insulating materials. The ceramic laboratory of the Bureau has made up a large number of specimens of porcelain covering the whole quartz-kaolin-feldspar field, in the form of cups. which are convenient for resistivity measurements. Several porcelain manufacturers have also submitted samples of their material in this shape. It is also possible, though with much less accuracy to determine the resistivity of material in completed spark plugs and a number have been measured including not only the principal American porcelain plugs but also German and French porcelains, mica , "lava," steatite e, glass and quartz specimens.

In" all these substances the resistivity drops very rapidly with increase of temperature and roughly follows the exponential law.

Iog10p=c-bt.

Where p is the volume resistivity (ohm-centimeters) and t is the temperature in degrees C.

The following table are given the values of b and c for various materials, applicable in the range from 2500 to 600° C).; also in the column headed p^500 the resistivity of the material at 500° 0. in megohm-centimeters, and in the column headed Te the temperature at which the resistivity is 1 megohin-centimeter. In making these measurements it was found that if direct current were used the specimens showed a very considerable "polarization" effect. This caused the apparent resistance to increase considerably with the time of application of the measuring voltage, even for impressed voltages of several hundred volts. It was therefore found necessary to use alternating current in these measurements and with this method the resistance was found to be independent of the voltage, frequency, and previous applications. At the higher temperatures the resistance is so low that the heating of the specimen by the measuring current becomes appreciable. Most of the data given in the table were obtained with 500 volts at 60 cycles impressed on the specimen.

GAS LEAKAGE.

The volume of air leaking through the body of the plug, has been measured for a number of types of plug. This was done by screwing the plug into a container and applying an air pressure of 150 pounds per square inch. The entire container was then immersed in a tank of oil heated to 200° 0. This heating develops any leaks which result from differential thermal expansion and is very severe since the outer parts become heated first. A graduated glass tube, originally filled with oil, can be placed over the plug and the volume of air escaping thus collected and measured. The amount of this varies from zero up to several cubic centimeters per second, but a good plug should not show more than 0.2 c. c. per second.

SPECIFICATIONS.

In collaboration with the inspection division of the Signal Corps, their material specification No. 28004 was drawn u and is now being revised in several details. A copy of this specification is appended to this report. Some 25 different types of plug have been tested in accordance with these specifications. The principal tests so far developed are those for electrical resistance at high temperatures, gas leakage, dielectric strength and cracking under sudden changes in temperature.

The first two tests are based on the work described above the limits set at present being that the resistance of the plug shah be greater than 500,000 ohms at 400° 0. and that the gas leakage shall not exceed 0.2 c. c. per second of free air at 200° 0. and 150 pounds per square inch.

The cold dielectric, strength is measured by placing the insulators under oil and applying an alternating electromotive force, between the center electrode and a narrow metal strip around the center of the insulator. Good material should withstand 25 to 80 kilovolts, at which voltage flash over through the oil occurs.

The only test available for cracking consists in heating the porcelain insulators to 300° 0. and then plunging them into water atroom temperature. They are then subjected to an electrical breakdown test to detect any serious cracks. It has been found that soaking the porcelains in an alcoholic solution of eosin is a very sensitive test and will show cracks which would otherwise be entirely invisible. The quenching test at 300° followed by an application of voltage has been the subject of considerable adverse criticism and admittedly very drastic; but since materials are available which will withstand it, it seems best to retain it until a better test is devised. An attempt has been made to duplicate engine conditions with respect to temperature gradients by heating one end of the plugs with a surface combustion burner while the shells are screwed into a water jacket. It has not as yet been found practicable to obtain sufficient temperature gradient by this method.

FURTHER WORK.

Since the element of accident enters into so many of the causes of failure of spark plugs, it is of the highest importance that accurate statistical data be accumulated concerning the average life and ultimate nature of failure of a large number of plugs of various types used in a variety of engines in actual service. Such data can, of course, be best obtained at the flying fields, and the committee is planning to secure systematic statistical data by cooperation with the Signal Corps.

Other work which should be undertaken as soon as possible includes a study of the mechanical strength of the plugs, a more intimate investigation of the de,position of carbon and of preignition, and of the characteristics of ignition systems.

Send this article to a friend

Submit an Article

Webmasters: You may freely reprint this article and/or ebook on your website provided the following caption remains intact, along with a working link.

This article courtesy of http://RacingSecrets.com , your racing technology resource.