The notice of proposed rulemaking (NPRM) for the proposed revision of Part 23, published by the FAA in March, is a remarkable document. During an election campaign in which the inability of Washington to get anything done is a central theme, the NPRM is a libertarian’s fever dream of deregulation. It’s hard to imagine how bureaucrats, notoriously more timid than hermit crabs, were able to emerge from the crusty shelters of decades to stretch their limbs like awakening Olympians. The NPRM is not just a superficial rewrite; it’s a fundamental rethinking of one of the FAA’s principal roles.
Part 23 of the Code of Federal Regulations has been the basis of small-airplane certification since the 1970s. Part 23 itself is a reorganization of an earlier set of rules, CAR 3, which came into being in 1949. Naturally, standards devised so long ago have had to adapt, through modifications, additions, reinterpretations and waivers, to changing norms of design and performance and novel types of aircraft and equipment. In the process of adaptation, the rules have become tangled and illogical, with different requirements applying to categories that seem increasingly arbitrary. To put it simply, the assumptions undergirding Part 23 no longer apply.
A few years ago, the FAA set out to overhaul its creaky rulebook. In 2013, it received a helpful kick in the pants from Congress, which passed legislation requiring the FAA to streamline and facilitate the certification process. The gauntlet that any new product had to run to obtain certification was portrayed as needlessly complex, bureaucratic and obstructive, and an impediment to progress and profits. This was, not accidentally, the view of certification widely held in industry.
Congress was explicit about the types of changes it wanted to see, and the legislative language, which sounded as though it had been drafted by teams of lobbyists, is echoed in the introduction to the NPRM. The purpose of the revision is to “remove current prescriptive design requirements” in favor of “performance-based” standards. “Weight and propulsion-based” categories will be replaced by “performance and risk-based” ones.
A number of ideas we have long taken for granted are going out the window. For instance, no more Utility and Acrobatic categories. No more 6,000- and 12,500-pound weight cutoffs. Instead, the Normal category — the only survivor of the old trio — will be divided into four levels by the number of passengers carried, respectively zero to one, two to six, seven to nine and 10 to 19. Two performance levels are proposed, based on maximum cruising speed; the break is at 250 kcas or Mach 0.6. A special place is reserved for “simple” airplanes, defined as VFR-only Level 1 airplanes, under 45 kcas stalling speed (and under 250 kcas cruise, though the simple class seems to be intended for LSA-type airplanes, not some as-yet-undesigned device that stalls at 44 knots and cruises at 249). The type of propulsion is no longer a factor.
Not everything is simplification and pruning. There will be new requirements relating to stalling behavior and icing protection, and also long-overdue attention to the engine-out handling of twin-engine airplanes. By and large, however, the guiding principle has been to strip Part 23 of most of the language that says how things are to be done and replace that with language saying what is to be accomplished, leaving it to the applicant — the person or firm seeking certification for a new product — to demonstrate that the proposed design achieves its ends while ensuring a suitable level of safety.
Here is an example. Section 23.1353 of the present Part 23 concerns “storage battery design and installation.” It has eight subparagraphs (a) through (h), two of which have sub-subparagraphs (1), (2) and (3). Most of these concern overheating, or the potential discharge of noxious gases or acids. Paragraph (h) requires that the battery capacity be sufficient to support essential loads for 30 minutes after a generating failure. (Part 25, by the way, ups the requirement to 60 minutes for airplanes with ceilings greater than 25,000 feet.) Some commonsense precautions are not articulated; for instance, there is no explicit requirement that the battery be secured to the airframe.
The new Part 23 would be considerably more concise and leave a good deal more to the imagination. It retains the 30- and 60-minute capacity requirements but makes no mention of overheating or other battery-related hazards. It is content to require that batteries supply sufficient power for “likely” needs, and that they be reasonably free of single points of failure.
Another example: The current Section 23.1361 is a six-paragraph description of how a master switch is to be designed. Once you have complied with its requirements, you might as well have just bought somebody else’s master switch, since virtually no deviation from a standard pattern is possible. In the revised Part 23, all detail requirements — the “prescriptive” parts — have been jettisoned, leaving behind only a general demand that all airplane systems “perform their intended function” and “meet the level of safety applicable to the certification 
and performance level of the airplane.”
“The intent of this NPRM,” says the FAA, “is to reduce regulatory barriers by establishing a system based on safety-focused performance requirements and FAA acceptance — as a means of compliance — of consensus standards … The FAA believes these changes would allow 
the FAA to provide appropriate oversight based on the safety continuum and would restore a simple and cost-effective certification process based on proven engineering practices.”
“Consensus standards” is a key concept: It means standards developed and accepted by industry rather than by government. Such standards already exist in many other industries and are already applied to some types of aircraft here and abroad. Indeed, one of the stated purposes of the revision is to harmonize U.S. certification requirements with those of other airplane-manufacturing nations. The Part 23 rewrite involved much international consultation, and several other countries’ aviation authorities are in the process of revising their certification standards 
as well.
Fortunately, even after the current Part 23 has been laid to rest, all copies of it will not be collected and burned. For one thing, compliance with the old Part 23 will be one accepted means of demonstrating compliance with the new. For another, Part 23 is a wonderful instruction book for designing an airplane. I, for one, could not have done without it. I will feel its passing like, if not the death, at least the retirement of an old and valued colleague.
