"One of the reasons for its success is that science has built-in, error-correcting machinery at its very heart. Some may consider this an overboard characterization, but to me every time we exercise self criticism, every time we test our ideas against the outside world, we are doing science. When we are self-indulgent, and uncritical, when we confuse hopes and facts, we slide into pseudoscience and superstition."

- Carl Sagan 1934 - 1996

The electrical system is generally the least understood of all airplane systems. Although any certified aircraft came from the factory with an adequate electrical system design, there are always compromises. In our current fleet of aircraft, these compromises were made with 20 or more-year-old technology, many with a "fail safe" idea mindset. We can do much better today.

The enclosed proposal is a series of redesigns of the Grumman Tiger's factory electrical wiring system, striving to design and fabricate a "failure tolerant" system.

"Failure tolerance" is a strange concept in our society. Intuitively, we know it's a faulty concept, a concept where any system should be designed with failures expected and tolerated. Suppose we relax our expectations a little and say, "some things are going to break, and it's okay..." This comes a bit closer to reality, because in reality many "certified" things break a lot. Unfortunately, we usually end up replacing them with the same old technology, a "certified" part that will likely give us the same poor reliability.

Of all the electrical system components on an aircraft, some are critical to safe completion of the flight, some are helpful to the flight, and others are simply conveniences. There is no reason why an efficient system cannot be designed with the goal that the failure of any single fuse, wire, regulator, or any other part of the alternator system should be nothing more than a simple maintenance issue. No single failure of any component should create a hazard to flight or a tense pilot (well, OK, but engine failure is outside the scope of this discussion…)

Reliability discussions often include matters of cost, convenience, and perceptions founded on incomplete or inaccurate data. A reliable flight system is defined as one that permits comfortable termination of flight (preferably at the intended destination) irrespective of any of these single failures of a system component.

When designing or maintaining any part of our airplanes, we should operate on two progressions of thought. The first involves doing a quality job on a task at hand. The second is "what if this part fails?" We go over the ways in which the part may fail and deduce whether or not any failure presents a hazard to successful termination of flight. We analyze how the failure will manifest itself to the pilot (handling qualities, strange noises, engine roughness, dead radio, etc. etc.) And finally, we decide if the failure pre-flight detectable.

While designing products for the big guys, engineers have expended hundreds of hours going over these points. The fancy name for this procedure is Failure Mode Effects Analysis or FMEA. If any failure does present a hazard, the best means for dealing with it is determined and specified. Re-design may be in order. Is the failure pre-flight detectable, is the item buried too deep to visually inspect or simply not on the checklist?

The primary goals for the proposed modifications to the Grumman Tiger are:

1. to encourage proper and effective FMEA studies,
2. to relieve pressure to purchase the most expensive components in an effort to improve "reliability", and
3. to encourage design for failure tolerance (inexpensive) instead of striving for failure proof or "fail-safe" (very expensive and nearly impossible).

Properly implemented, these techniques will reduce to near zero the probability that any electrical problem will ruin the pilot's day.

These proposals are offered in small pieces, to better implement the plans and to afford a higher probability of getting the FAA approvals when required.