On Automotive Fuel Gauges HE FOLLOWING interchange of letters between TerT rence W. Algeo, (EAA 13108) of 183 Larchdale Crescent in Winnipeg, Manitoba, Canada and the StewartWarner Corp. is significant for the authoritative technical light beamed on a matter of great interest to the amateur builder. The letters are reproduced following.

Chief Engineer Engineering Department Autocar, White Truck Division Cleveland, Ohio Dear Sir:

Five of us have purchased Fuel Tank Gauge Assembly No. 626869 (12 volt sending unit) for installation in aircraft fuel tanks. These are for homebuilt aircraft of modern design and construction which do not require DOT certification although we insist they be just as airworthy and at least as safe. As an electrical engineer, I cannot assure myself and others that an open type rheostat is safe in a fuel tank (80-87 octane gasoline). The tank is vented and the rheostat will be submerged until the tank is partly emptied. I have spoken to the local White Truck Service manager and have been assured that no trucks have "blown up," so it must be safe but we are not quite sure why. I realize the electrical potential will be very low, but I have seen rheostats, which look similar, burn in areas. Some rheostats used for similar purposes seem to be enclosed or sealed. Assy. 626869 is excellent for our application, so we would very much appreciate a positive explanation regarding the safety of this type of installation. Yours very truly, Terrence W. Algeo, P. Eng. Registrar ***** Dear Mr. Algeo: I have read with interest your letter to Autocar, White Truck Division, concerning the safety of open rheostat assemblies in gasoline fuel tanks and, as an engineer, I can understand your concern. Perhaps I can best allay your fears by offering the following evidence attesting to the safety of our fuel senders. First, of course, is the historical argument with which you are already somewhat familiar. The fuel tank guage assembly, White Truck part No. 626869, is a member of a large family of fuel tank senders which have been designed in many thousands of applications for use in many vehicle markets which include over-the-road trucks, off-the-road farm and construction equipment, marine, and light aircraft. I believe every single-engine private aircraft manufactured in this country using reciprocating engines utilizes a rheostat assembly of this type manufactured either by Stewart-Warner, AC Spark Plug, or other manufacturers. In addition, the passenger car automobile which, of course, is manufactured by the millions, has for 20 to 30 years been using a similar unit and, for the past ten years at least, the octane ratings for passenger automobile gasolines have equalled or exceeded the 80 to 87 octane gasoline to which 30

MARCH 1971

you refer. Never in all the history of this usage am I aware of a suggestion that any disaster, fire, or explosion has occurred as a result of the normal usage or failure of one of these units. I believe the reason that this type of assembly is safe, meaning that a combustible situation cannot occur, is for the following reasons: 1. A gasoline tank, even though vented, always contains a sufficient amount of unremovable gasoline such that the mixture of gasoline vapor to air is too rich to be in the combustible range. That is, without means for actually pumping fresh air into the tank or very large venting, a sufficiently balanced mixture of air or gasoline vapor does not occur. 2. Even if such a mixture were present, as in the case of a badly damaged gasoline tank, the spark that could be generated by the electric fuel system does not contain enough energy to initiate combustion. Consider the following circumstances: The fuel sender system with which you are concerned is designed to be used with a fuel indicator that places a minimum of 120 ohms in series with the rheostat assembly. The rheostat assembly itself is designed for a minimum resistance at full position of 33 ohms and a maximum resistance at empty position of 240 ohms. Under worst case conditions with as much as 15 volts applied to the system, interrupting the contact on the rheostat strip would not produce a visible spark nor, I believe, could ignite an air/gasoline mixture, as history has indicated. After all, in all the millions of applications, such a condition must have occurred many times. Consider the design of a system in which combustion is intended, such as in the cylinder of an internal combustion engine. In the first place, the system could not function at such low voltages. All ignition systems are designed to generate at least three orders of magnitude more voltage than the 15 volts just described, a minimum of 15,000 volts. In addition, the air/gas mixture is preheated by compression of the gas in the engine with a compression ratio of eight or nine being typical. You mentioned in your letter that you had seen rheostat assemblies similar to the one in question which were burned in some area, which I can assure you can occur if the systems engineer in applying this component does not take precaution to utilize the fuel tank sender within its design capacity and guard against system faults which may in some manner overstress the component. Let me illustrate my point. Our fuel tank sender, at its minimum resistance condition of 33 ohms, will not bear application of a full 12 to 15 voltage, directly applied. This would cause a power dissipation in the sender of seven watts or more and the sender is just not designed to dissipate that much power. It is, therefore, the system engineer's responsibility to see that such a condition cannot occur. In an application such as you describe, I for one would give serious consideration to fusing the fuel tank sender line with, say, a quarter amp slow-blow fuse if there was any conceivable possibility of a short occurring in the wiring system which might cause excess voltage to be applied to the sender, although I am not certain that even this condition could produce a fire or an explosion. As a matter of fact, I doubt that it could, but I would want to be sure. I hope that the remarks that I have made will help you to wrestle with your problem and are assuring enough to convince you that with proper system design the fuel tank sender which you would like to use is perfectly safe. Good luck!

Very truly yours, STEWART-WARNER CORPORATION

M. J. Johnson Chief Engineer Instrument Division