Why you should care about the new Part 23 proposal

There has been a lot of talk lately (perhaps too much?) about aviation issues in Washington: Air Traffic Control privatization, the third class medical, and user fees to name just three. Somewhat obscured by these Capitol Hill battles is a more complicated but also arguably more important legislative issue: aircraft certification reform.

Often called the Part 23 rewrite, since it deals with FAR Part 23, this is an effort to address the way light aircraft are designed, certified, and upgraded. Like improving American schools, it’s one of those ideas that everyone seems to support, but no one can implement. That may be changing finally, with the recent release of a Notice of Proposed Rulemaking. Far from being some obscure technical proposal, the FAA’s plan is bold and (mostly) sensible. If adopted, it would affect every general aviation pilot and aircraft owner, so it’s worth taking the time to understand what it says.

The history

Michael Huerta
Administrator Michael Huerta has pushed hard to change the FAA’s mindset.

This reform effort has been a long time coming. The general aviation industry, led by the General Aviation Manufacturers Association (GAMA), has been suggesting, pleading, cajoling and demanding for decades that the FAA update its 1950s approach to certifying aircraft and avionics. This push has taken on new urgency over the past five years, as the FAA has worked to change its culture into one that emphasizes a “compliance philosophy” and a “safety continuum.”

These may sound like fancy bureaucrat language that doesn’t really mean anything, but that cynicism is misplaced. Both of these concepts are welcome – and long overdue – responses to a changing aviation industry and our evolving understanding of risk. The first one is a shift in focus from enforcement (yanking certificates when a pilot screws up) to compliance (proactively encouraging safe flying before an incident). Many pilots claim they’ll believe it when they see it, but it’s at least a start.

The safety continuum philosophy may be even more important. In the FAA’s words, it recognizes that:

one level of safety may not be appropriate for all aviation. The FAA accepts higher levels of risk, with correspondingly fewer requirements for the demonstration of compliance, when aircraft are used for personal transportation.

That may not sound like a revolution to anyone who’s logged more than 25 hours in an airplane, but it’s a thoughtful and powerful summation of what general aviation really means. As a community, we know how to fly with almost zero accidents if we want to (the airlines show the way), but we choose to accept some risk because twin turbine engines, multi-crew cockpits and rigid rules are not practical for our operations. The FAA is explicitly recognizing this tradeoff.

Beyond the philosophical change of heart, the renewed interest in updating certification rules is a reflection of the sometimes-embarrassing state of the general aviation fleet. Experimental airplanes are often better equipped than certified ones, with powerful glass cockpits and advanced autopilots that are impractical or unaffordable for a Cessna (that goes the exact same speed, with the same number of seats). Likewise, portable technology that is exempt from certification is moving much faster than certified avionics. In the FAA’s eyes, neither of these trends are good for long term safety.

But while there is broad recognition that the current system is broken, in typical FAA style, the process to change has been deliberate. After years of study and debate, an Aviation Rulemaking Committee (ARC) was formed in 2011 to propose specific standards and actions. This group was made up of major general aviation airframe manufacturers, avionics companies and pilot organizations. After years of study and debate, and after months of review by the FAA, the official NPRM is finally here. What does it say?

What the proposal says

At heart, the new approach favors standards-based certification. This means that, rather than writing prescriptive, step-by-step rules for each component of the airplane, the FAA will use a goal-oriented approach that emphasizes the outcome of certification – not the process. They will demand reliability and safety, but they won’t describe what the specific technology looks like. Many of those specifics will live in a separate document, outside the actual regulations, commonly called industry accepted standards. The hope is that such an environment would, in the FAA’s words, “hasten the adoption of safety enhancing technology in type-certified products while reducing regulatory time and cost burdens.” In other words, have our cake and eat it too.

The new rule includes some lessons learned from LSAs.

If this sounds like an outgrowth of the ASTM process that has been in use by Light Sport Aircraft (LSA) manufacturers, it is: ASTM standards will be a key path for certifying new airplanes under the revised rules. (You can read the proposed standards here for a preview.) But notably, ASTM would not be the only acceptable standards; manufacturers would be free to offer their own as well. Plus, the current prescriptive certification standards would be grandfathered in so that current designs don’t have to be redesigned. Once FAA calls out a standard, it stays in place, which is critical for protecting long-term investments by manufacturers.

That’s a minor but significant point. Some LSA owners have been disappointed to learn that almost all maintenance decisions on their airplanes are driven by the aircraft manufacturer, not the FAA. This is a big change from Part 23 certified airplanes, with sometimes bizarre consequences. The aviation industry seems to have learned from the LSA rule: the same maintenance process as today will be in place.

The other major advantage of this system – where laws remain vague but reference more detailed standards documents – is that it’s far easier to keep up to date. If some exciting new technology comes along that was never imagined by today’s lawmakers, ASTM groups can write new standards and fix it in a matter of months, not years. This future-proofs the new Part 23 in a way that the old law never was.

Such standards will cover a wide variety of airplanes. The sweeping NPRM would replace airworthiness standards for normal, utility, acrobatic and commuter categories, covering all airplanes with fewer than 20 passenger seats and with a maximum takeoff weight of 19,000 lbs. or less. That means everything from two place trainers to mid-size Citation jets would fall under this process.

Instead of the arbitrary and outdated weight classes, there would be four levels of certification based on aircraft performance and risk. Factors determining which category an airplane would fall under includes stall speed, VFR or IFR operations, and whether the airplane is pressurized or not. The NPRM outlines suggested breakpoints for each, based on passenger seats:

  • Level 1 – no more than one passenger
  • Level 2 – 2-6 passengers
  • Level 3 – 7 to 9 passengers
  • Level 4 – 10 to 19 passengers

There’s also the potential for a “simple airplane” standard that the FAA wants more feedback on. The suggestion is that such a simple airplane would have a stall speed of 45 knots or less, and be approved only for VFR flight. This is very similar to EASA’s Very Light Aeroplanes (CS-VLSA) standard, and it would be included in Level 1. It’s not clear if this will make it into the final rule or not.

Focus areas

Reading through the proposal, it’s clear that a few topics were top of mind for the FAA. In particular, three focus areas stand out: loss of control, minimum controllable airspeed in twins and supercooled large droplet (SLD) icing.

Loss of control is still the most common cause of fatal accidents, so it’s a natural focus area for the FAA. The proposal argues that “the significant reduction in inadvertent stall-related departures from controlled flight” is the “largest, single safety gain expected from this rulemaking action,” so a lot of time is spent on the topic. In a nod to reality, the proposal admits that most loss of control scenarios happen too low to the ground (often in the traffic pattern) to be recoverable. That means even the best spin training is worthless if the airplane stalls at 300 feet.

Icon A5
ICON’s nearly stall-proof wing is one enhancement the FAA clearly supports.

Instead of focusing on pilot skills, the FAA argues that airplanes should be designed “to use new design approaches and technologies to improve airplane stall characteristics and pilot situational awareness.” The FAA is essentially saying the industry needs to forget spin training and take the pilot out of the picture; better to make an airplane unspinnable than to continue beating the drum about training. In fact, the requirement to demonstrate spin recovery has been removed entirely, perhaps pointing to the Cirrus SR-22 certification as a test run on this idea.

Of course, how to do this is a critical question. The FAA is studiously vague on this topic, offering a few suggestions but mostly asking for ideas from airplane manufacturers. The ASTM’s proposed spec offers more details, but still leaves a lot to be decided. At least three ideas come to mind here: angle of attack indicators, new wing designs, and electronic systems to prevent stalls.

AOA systems have been a pet project for the FAA over the last few years, and it’s likely we’ll see more of them in new airplanes. These are address the pilot situational awareness issue (and we’ve debated them endlessly), but they don’t do much to remove pilot error from the equation.

New wing designs, on the other hand, offer promise. The Icon A5 amphibian is a notable example of an airplane that is designed to be very difficult to stall. A variety of aerodynamic tricks are used so the wings keep flying through even aggressive maneuvers. The airplane just mushes along, with no sharp break or wing drop. This is proven technology, but pilots usually pay the price in cruise speed for such forgiving manners.

A higher tech approach is found in new Cirrus airplanes, in the form of Garmin’s Electronic Stability and Protection (ESP) system. This is like a passive autopilot, always on the in the background, waiting to step in and gently nudge the controls back to straight and level if the bank gets too aggressive. In many ways, it’s a simplified version of sophisticated fly-by-wire envelope protection systems found on large airliners. It’s not cheap or easy – it requires autopilot servos for one – but it works.

Will the FAA get its wish and encourage the development of airplanes that don’t stall? It’s too soon to tell, but it’s certainly a bold idea.

On the subject of minimum controllable airspeed in twins (Vmc), the proposal offers another simple and somewhat radical idea. It calls for certifying Level 1 and 2 twins with a Vmc that is equal to or less than stall speed, essentially designing out the possibility of a Vmc rollover event. If the airplane gets too slow on one engine, the airplane will buffet and stall – a maneuver pilots are well-trained to recover from. There will be no need to practice Vmc avoidance because it can’t happen. Again, it’s an intriguing idea, but one that will need the close eye of some engineers to determine how practical it is.

In-flight icing is an area that never seems settled. SLD is the latest development, brought to light by the crash of a turboprop airliner in Roselawn, Indiana in 1994. Currently, there are no airplanes approved for flight in SLD conditions, a serious type of icing that can rapidly coat the wing far beyond leading edge anti-ice equipment. These new standards attempt to close this blind spot, requiring manufacturers to either specifically demonstrate safe operations in SLD or, if flying in SLD is prohibited, offer a means of detecting it and proving that the airplane can exit such conditions. It seems like a sensible proposal, although one that probably won’t affect most general aviation airplanes.

Other changes

Another issue, unwritten but clearly considered, is electric propulsion. Current standards offer no realistic avenue for certifying an airplane with an electric motor, which has left many innovative airplane designs stillborn. The new goal-oriented certification approach is an attempt to leave plenty of openings for such technology, and whatever follows it. As long as the propulsion system meets the performance and reliability standards set out by ASTM, it can run on electricity, dead dinosaurs or whatever else we dream up.

Pipistrel Alpha Electro
Electric airplanes have been nearly impossible to certify under current rules.

However, there’s one additional change that would make electric airplanes even more cost effective. That’s the concept of certifying the propulsion system as a part of the airframe, instead of the current system that certifies both a Lycoming engine and a Cessna airplane as completely independent parts. There’s precedent for this change to a single certification: Europe’s VLA aircraft don’t have to have a certified engine or propeller, and EASA’s proposed reforms for larger airplanes suggest they will allow the engine and propeller to be certified as part of the airframe.

At this point, the FAA has only proposed such a change for simple aircraft, but if extended to Level 1 or 2 airplanes, it would be especially important for electric airplanes. Type certifying an electric motor is often overkill, because many clean sheet electric airplane designs feature motors that are deeply integrated into the airplane (maybe as many as 12 of them). Plus, such modern designs require custom motors that are very airframe-specific. A series of small electric motors across the front of a wing is not at all like a big piston engine hanging on the front. Expect GAMA to push for this change to allow for a single certification.

There’s also a refreshing recognition that so-called “non-required safety equipment,” things like portable weather displays, carbon monoxide detectors, clocks, and fire extinguishers, should be easier to certify. As the FAA says, the failure of such a device would not significantly affect the safety of the flight, so the current stringent standards only serve to prevent use of such technology. The fact that this was included in the proposal was a pleasant surprise – GAMA and others are supportive but doubted it would happen.

This is a complex subject and the change is very subtle, but it would have a major impact. In the past it was illegal to produce any item intended for installation on an aircraft without cumbersome certification processes, like a PMA. But now there is an important addition: “Any other manner approved by the FAA.” This would allow the FAA to create a policy, outside of the legislative process, that makes it much easier to add something like an AirGizmo mount to the panel. One important detail to work out here is where the line is drawn. What exactly is non-required? Expect a debate.

Finally, there’s a decidedly international perspective here: the NPRM tries to stay in formation with EASA’s recently-announced CS23 reform plan. Keeping US rules in sync with other countries should make it easier for aircraft manufacturers to design once and achieve faster certification in many countries. This would be a significant improvement over the current environment, which often forces manufacturers to satisfy numerous, sometimes-contradictory local standards. The Cessna sold in France may not match the Cessna sold in the US. So even for pilots who never fly outside the US, such a change would save money; it instantly makes the market for a potential airplane 30% bigger. That drives up volume and drives down cost.

All this is great news for new airplane designs, but what about the hundreds of thousands of airplanes that make up the majority of the piston fleet? At less than 1,000 per year, new aircraft – no matter how safe – won’t have much of an impact on the safety record. The good news is that the revised Part 23 would cover both Supplemental Type Certificates (STCs) and amended Type Certificates (TCs), the two main ways to upgrade an airplane. The wording is vague, but the goal is to streamline both, requiring fewer special conditions and exemptions.

What comes next

The publishing of an NPRM starts the clock on another step in the long, winding process towards becoming law. First there is a comment period where pilots and manufacturers can offer suggestions and complaints, then a review of those comments. After that, the proposal must wind its way through the FAA Administrator’s office, then a review at the Department of Transportation, and finally the Office of Management and Budget. As Churchill said, we are not at the end, merely at “the end of the beginning.”

But there’s a real sense of urgency to get this done. EASA has publicly stated that its final proposal will be published in May and a final decision is coming by the end of the year. That puts a stake in the ground for the FAA. There’s also a great desire by both the FAA and the aviation industry to finalize the new law before a new President comes into office. This inevitably leads to disruption and schedule changes, no matter who it is. That’s an ambitious schedule, but it is possible. As an example, the comment period for this NPRM ends on May 13, a relatively short window that keeps the process moving.

Does it deserve the support of general aviation pilots and companies?

Greg Bowles is a pilot and former Cessna engineer who now leads the GAMA effort to update Part 23. He says the proposal has “80-90% of what we were hoping to see.” AOPA, EAA and NBAA have all welcomed the new rules, and have been involved in the process from the beginning.

Flight Design C4
Flight Design’s proposed C4 airplane depends on a new certification process. Will there be more?

If the proposal does make it into law, the first result would probably be the introduction of new avionics and airframe upgrades for existing airplanes. This could be in the form of less expense glass cockpits or new autopilots with stability systems. It could also include less flashy products like new seat belts or panel mounts.

After that, there is the hope of new aircraft designs. Flight Design, for one, is betting their sleek C4 airplane on a new Part 23. It has been designed from scratch in anticipation of more flexible options for avionics and airframe technology. Others would surely follow, although such new airplane projects are notoriously time-consuming.

Beyond aircraft certification, Bowles says there’s a need for similar reform efforts in maintenance and pilot licensing as well. Together, these three areas define almost all of the ways that general aviation works. Some tentative steps have been taken in these areas, notably the move from Practical Test Standards to Airman Certification Standards for knowledge and flight tests. This update recognizes that a move from prescriptive, dated standards to a more flexible, goal-oriented approach is good for safety and cost – just like the Part 23 effort. But there’s much more work to be done here.

Whatever the final outcome, this reform process will have a major impact on almost all parts of the general aviation industry. Just like ADS-B over the past ten years, it’s a subject that cannot be ignored. Get informed, submit comments if you have them, and let your aviation organizations know you care.

A new aircraft certification law is, at the end of the day, nothing more than a process. But it’s one that transfers a significant amount of control from government to the general aviation industry. It’s up to us to take that offer and run with it.


  • John,

    Thanks very very much for this post!

    Reinventing the world of certified aircraft, both new manufacture and legacy, is a huge and complicated deal. The FAA was very slow to release their draft rulemaking (it was due for release last year by law), but the good news is that this was not just a pro forma for-show attempt to comply with Federal law that compelled Part 23 reform. The NPRM certainly is not perfect, and won’t ever be perfect as a first cut … but getting both the experts and us regular owner/pilots involved in commenting will help make this first attempt better.

    Several areas ripe for comments back to FAA:

    1) Making the rule vague and allowing for future advancements is huge … great way to set it up! I’m shocked that the FAA would agree, but hey, pigs may learn to fly after all!

    2) There should be a review process after the final rule goes into effect – perhaps after three years after implementation there should be an “after action report” that focuses on what is working well and what part of the rule needs further tweaking

    3) We need to make sure that we don’t just jump from the frying pan (FAA over-regulation) to the fire (manufacturer over-regulation of maintenance and mods that leads to grossly-expensive costs of maintaining our birds). The latter is a fact in the LSA industry, where lack of maintenance rules via Airworthiness Directives (that are at least required by law to consider compliance costs) was supplanted with “manufacturer recommendations” that are overly prescriptive and always fail on the side of protecting manufacturers from product liability, cost be damned. Mike Busch has a great post on that subject this week on AOPA.org’s “Opinion Leaders” blog.

    To all our pilot friends – comment on whatever concerns you most … this is our shot to get our two cents worth into the system at a time when the rulemakers are open to change. The entire rule is a change, so at least they’re open to more changes.

    Thanks again, John, and Richard too!

    • Duane, you make a fine point about LSA and maintenance, but it does appear that the NPRM (at least tries to) solve that. By keeping the existing standards in place, it should avoid the iron fist of manufacturers like we have with LSA. Implementation is key, but so far I’m optimistic that this scenario will be avoided.

  • John – Some questions for you or anyone else who is knowledgeable:

    How does the NPRM mesh with the existing LSA rules and Sport Pilot operator rules? Will the “simple aircraft” and Category 1 cert rules replace LSA including SLSA? What kind of pilot certification will apply to each of the new categories?

    How might all of this work with the pilot medical reform rules that are mandated in the current Senate FAA reauthorization bill?

    Pilot certification rules and aircraft certification rules are, of course two separate sets of rules, but they obviously correspond to each other.

    Maybe that subject is worthy of comments back to FAA on Part 23 reform.

  • ” It calls for certifying Level 1 and 2 twins with a Vmc that is equal to or less than stall speed, essentially designing out the possibility of a Vmc rollover event.”

    I have a degree in aerospace engineering and a commercial multi-engine ticket. Have owned and flown a Baron 55 for 15 years. This sounds like a request for underpowered twins and or higher stall speeds. Neither of those sounds like a good idea.

    Just my 2 cents as a somewhat educated observer.

  • For an R-44 pilot, I would have expected you to mention how this affects the rotorcraft manufacturers and owners. The focus of the ARC seems quite “airplane-centric” (LoC, twin Vmc) despite Robinson Helicopters producing more new GA aircraft than most of the airplane manufacturers combined. The quotes from the NPRM use the term “aircraft” but the rest of the article tends to use “airplane”.

    This is a concern because rotorcraft were specifically excluded from the LSA category.

    • Chip, this proposal specifically deals with Part 23, so it applies to airplanes. As you probably know, light helicopters are certified under Part 27, which is a separate process. For now, helicopters will remain the same. However, there is an industry group working on a new Part 27, taking some of the lessons of this most recent reform effort. Hopefully we will see some progress there as well.

  • The new Part 23 is interesting, but I wouldn’t expect a huge aviation safety/cost revolution to come from it. There are twins out there with the same stall and VMC speed (the Seminole comes to mind and don’t forget the Cessna Skymaster); no huge safety improvement. Then how about the Ercoupe? No stall, no spin, no safety improvement either.

    On the twins, I believe it would be more beneficial to increase the single engine climb gradient requirements. Everybody just increases gross weight till they just barely meet the minimums. But when you’re climbing toward the trees at 200 ft/min after normally seeing 2000 ft/min on the dial, they look pretty tall; so you pull back in an attempt to make them smaller.

    Really, the current Part 23 does not prevent one from certifying a simple airplane with fairly minimal requirements (spin testing is an exception here). The problem is that nobody wants a simple airplane. They want 200+ kt, IFR, through ice: One can make the regs as simple as they want, but it takes a fair amount of effort to design and build that capability and prove to yourself (if not the Feds) that it is ok to unleash it on the unsuspecting public. And yes, it does take a lot more thought than just bolting that uncertified Dynon screen to the panel.

  • This has the potential to benefit many in the GA fleet and, as noted, we need to keep the problems that the LSA group live with away. However, I do see this will also slow down the installation of the required ADS-B equipment as many, including myself, are holding out to see what huge technological changes are going to be made in the next 4 years. With making the certification process easier, the prices should drop and the equipment features should skyrocket.

  • How about the upcoming fuel crisis? Nobody seems to be talking about the fact that the entire GA fleet is going to need a major overhaul once 100LL is outlawed by the EPA, which is something that is going to happen sooner rather than later. Is the FAA taking any regulatory steps to help with this, or is 90% of the fleet just going to end up junked because most owners simply don’t have the capital to repower their aircraft? And what about diesel?

    All of this Part 23 stuff really seems like whistling past the graveyard at the moment.

    • Has anyone considered making piston engines that run on jet fuel and eliminate AVGAS altogether? All aircraft would be required to switch over to the new jet fuel capable piston engines when they reach TBO or by a date when AVGAS is to be phased out.

      The Diamond DA42 already has piston engines that use jet fuel, so, there is no reason the rest of the GA piston fleet cannot be transitioned to jet fuel based piston engines. Using jet fuel costs less than AVGAS, burns more reliability, and would simply the fuel system.

      • The problem is that diesel engines (like the Diamond) are heavy and lower horsepower than the typical Continental IO-550. When we have lighter weight diesel engines that can go to 2000 hours it could be a great transition. But we aren’t there yet, even after lots of work by Continental, Lycoming and others.

        • John – with respect to diesel aircraft engines, you have to consider the weight of the entire propulsion fuel system, not just the weight of the engine itself. The superior fuel efficiency of the latest generation electronically-controlled diesels such as the Continentals and Austros that are already flying, as well as new models still seeking certification such as the Graflight EPS 350, provide a much lower fuel weight to go the same distance as the equivalent gas opposed aircooled engines long used in light aircraft.

          Fuel weight is of course a large proportion of useful load on any aircraft.

          And then, of course, not only does the superior economy of the diesel greatly extend range and/or flexibility in trading fuel for payload, it also significantly reduces the hourly cost of operation, with fuel cost per hour typically accounting for more than half the operating cost of most light aircraft.

          And as for horsepower, there’s no natural limit to power for diesels any more than for gas engines. The Graflight V-8 is anticipated to turn out 350 to 420 horsepower, higher than any TIO-550. Also consider that even if a diesel is rated at lower power than some other gas engine of equal displacement, diesels being far more durable can easily operate at 80 to 100% of power continuously, which few to no gas engines can do … and give that all diesels are turbocharged, they can operate at much higher power levels at high altitudes where fuel efficiency is greatest, giving the pilot more “useful power” than an equivalent rated power on most gas engines.

  • Great article, John. Just a quick note – you refer to the legislative process a couple of times, but actually most of what you’re talking about happens outside of legislation already, in the Administrative Procedure Act’s (APA) rulemaking process (through the NPRM).

    That said, APA is still slow, and the point about referencing standards documents is still a great one – and a really exciting development for GA.

Leave a Reply

Your email address will not be published. Required fields are marked *