How Certification Rules Are Hurting Us

A Dynon PFD for Experimental Airplanes

For as long as I can remember pilots have been blaming unnecessarily complicated FAA certification rules for the high cost of personal airplanes and for the slow pace of technological development. The FAA has formed a committee to study FAR  Part 23—the rules that govern certification of airplanes weighing less than 12,500 pounds for takeoff—to search for ways to simplify approval of piston singles and other light airplanes.

But I don’t think there is a lot to be gained by overhauling the certification standards for basic airframe structural strength and flying qualities. The real benefits in cost savings and performance improvements can come from a new look at how electronics can be used in piston airplanes.

Every pilot I know is frustrated by the capability of consumer electronics that is available for a few hundred bucks and can do more than FAA approved avionics costing many thousands. I know an iPad doesn’t have the same level of testing and software documentation as the primary flight display (PFD) flat glass system certified for a piston airplane. But the iPad cost $300 to $500 and the PFD $10,000 for entry level. The cost gap, in this case, really can be blamed on FAA certification.

It’s easy to see the extra cost burden of certification when you look at the Garmin flat glass lineup. The G600 and G500 combined PFD/MFD systems look exactly the same, and with the same list of options, have the same capabilities. But the G600 costs more because some of its software is FAA approved to a higher level than in the G500. The FAA requires the higher level of software security and testing in airplanes weighing more than 6,000 pounds so Garmin has to charge more for the G600 to perform the same functions the as the G500.

And Garmin makes a PFD for experimental airplanes that requires no FAA certification and costs a fraction of the price of the certified systems. So from one company you can trace the cost of certification without really seeing a benefit in performance and features, and I don’t think much, if any, improvement in overall reliability.

The cost difference is very dramatic when you look at avionics such as systems from Dynon, Grand Rapids Technologies, Advanced Flight Systems and others that are uncertified and intended for amateur built aircraft. These systems have amazing capabilities for a fraction of the price of an FAA certified system.

The price of avionics is a valid measure of certification costs because basic electronic components like glass displays, circuit boards, transistors and whatever come from the same pool of suppliers used to make consumer electronics like phones and computers. To certify a piece of avionics equipment these essentially standard off-the-shelf components have to be tracked and tested at a very high level. Software to operate the FAA certified system needs much more documentation and segregation than may be used in consumer electronics. And the certified system must be tested using standards and techniques that may not make all that much difference in terms of reliability, but does add cost.

Strict avionics certification standards make sense to me in larger airplanes, particularly turbine powered airplanes. These airplanes have multiple levels of power and system redundancy so you want the avionics to be as reliable and predictable as possible. Essential avionics in a transport airplane should be held to the highest possible standard.

But in piston airplanes we accept a much lower level of redundancy and basic performance in exchange for keeping cost in check. Most piston airplanes only have one electrical system, for example, and the big majority have only one engine. Demanding a higher level of testing, redundancy and performance of avionics—as certification standards typically do—than the entire airplane offers makes no sense.

Really big avionics gains are waiting to be made in autopilots if the FAA will just modify the rules. Right now an autopilot is costly and time consuming to certify, and a certification is only good for a specific airplane type. However, a “smart” autopilot could be built right now that would learn how to fly each airplane on its own and store that information in its memory without requirement for many hours of flight testing.

An autothrottle that could maintain target airspeed as well as cruise control does in our cars would be easy and inexpensive to make without complicated certification rules. Think how much safer it would be if the autothrottle maintained the ideal airspeed on landing approach.

How about automatic flight to the nearest airport in an emergency? The electronic capability is available right now for an autopilot and navigation system to jump in and automatically fly the airplane to the nearest suitable runway in an emergency such as engine failure. In fact, Vertical Power is selling such a system for experimental airplanes. But certification of this type of automated safety device under today’s rules is almost impossible.

Just look at what electronics are doing for us in all aspects of our lives. And think what could be done in our airplanes. In this case, it is out-of-date certification standards and attitudes that are keeping the potential safety and utility of modern electronics out of the cockpit, or adding so much cost that most pilots can’t afford to buy them.

This entry was posted in Mac Clellan's Left Seat Blog. Bookmark the permalink.

28 Responses to How Certification Rules Are Hurting Us

  1. Cary Alburn says:

    Can’t disagree in principle. I just spent $15K 6 months ago to upgrade my 50 year old airplane’s panel to include a Garmin 430W and a PSE audio panel–and that’s it, not even state of the art, no dual redundancy (I do have a good working Narco nav/com as No. 2). Much of the other things I’ve had done to my airplane over the years has been outrageously costly, not because my mechanic has over-charged (in fact, I think at times he’s under-charged), but because it’s a certificated airplane, there are rules, regulations, and the requirement for FAA approval for things that just don’t benefit safety, but which apply to certificated aircraft and not non-certificated, and those all raise the cost significantly.


  2. Kayak Jack says:

    It hurts us in many ways. I just had to replace an ignition switch on my 50 year old Cessna 172 – $350 for just an ignition switch!

    I tried to get LED strobes for my wing tips. Daytime visibility is an important step to avoid mid air collisions and near misses. Easy and cheap for an experimental plane. Expensive and nearly impossible for a certified plane. Here, FAA stands in the way of safety, effectively blocking access to proven technology.

    On the other hand, FAA does some good too – I guess, somewhere, at some time, maybe, I’ve heard.

  3. Brent says:

    Exactly Mac! This is an example of the bureaucracy inpeding safety.
    As pilot who flies an experimental with advanced avionics as well as certified light planes and part 25 corporate jets. There are dozens of examples out there and it seems like no one is listening.

  4. Marc Rodstein says:

    Great article Mac. Spot on. This is why iPads have been a runaway hit in the cockpit. For under $700 they offer capabilities that even $20,000 certified systems cannot match. The biggest reason is certification costs.

  5. David Sucher says:

    Yout remarks about automatic and semi-automatic airplanes are fascinating. My sense is that there is going to be a HUGE upsurge in light plane flying once those systems are available at the mass market.

    What sort of timeframe do you see as possible? Is it all politics? Bureaucratic inertia? Five years? Ten years?

  6. Mac, let be point out two key things that bear on the subject of your blog post. This is sort of a “good news, bad news” thing…

    The good news is that the FAA has elected to “look the other way” with respect to the use of what they call PEDs (portable electronic devices) in our airplanes. The basic unwritten rule is that if a device is not structurally attached to the airplane (Velcro and wingnuts are okay) and if it doesn’t send data to any certificated avionics (it’s allowed to receive data from certificated avionics), then the FAA treats it as baggage, and so we can use any bleeding-edge technology we want without any certification hassle. This is a wonderful thing. The FAA didn’t have to take this liberal position on PEDS (and so far as I know it’s not written down anywhere), but much to their credit they did.

    The bad news is that the FAA has never recognized the concept of a “backup instrument,” and IMHO they really should do that. For example, I can understand why the FAA would be reluctant to let me install (say) an uncertificated Dynon EFIS in my certificated Cessna 310 as primary flight instrumentation. But logic would suggest that if I already have all the necessary primary flight instrumentation required by the FAA, it should be okay to install the Dynon EFIS as a “supplemental instrument” as a minor alteration with much-relaxed regulatory requirements (e.g., so long as it meets the flammability requirements of CAR 3 and is properly fused or circuit breakered so that it can “do no harm”). Unfortunately, there’s no such provision in the regs. An instrument is an instrument; there’s no such notion as “primary” and “supplemental.” In my view, a simple regulatory change greatly relaxing what can be installed on a supplemental basis could go a long way to getting us into the 21st century.

    In the meantime, we’ll just have to put up with Velcro.

  7. Jon Carlson says:

    I don’t think conceptually the problems created by certification are limited to avionics. Engine components could become much more reliable, electronic engine controls more common, magnetos replaced, etc except that most of this stuff can’t be certified for more than a single airframe at a time – which means it can’t get enough traction to make a market out of it. Same with minor engine variants – Continental could release new variants of its engines, maybe, with minor improvements to intake systems, exhaust systems, etc. Except that they can’t certify them on enough airframes to make upgrades viable.

    It is not just avionics.

  8. Jim Hiatt says:

    This is the conversation that has been going on in my hangar for years. At the current price point for a certified WAS GPS, an MFD to display all the nav information and the trafic and weather from the ADSB boxes a lot of us will be required to spend at least 50% of the value of our planes – and that is not going to happen! Not in 2020, not ever. If it comes to that I will just scrap my Comanche 250 and my Aztec and but a boat. They will not need as much FAA after the new requirements finish off pricing us out of aviation!

  9. Don Heitzmann says:

    Can’t disagree with Mac’s arguments, as he makes them. But, he largely discusses these new avionics boxes as hardware. Speaking as someone who spent his career in the software business, I’d like to say that it is very easy to underestimate the importance of testing for the software part of these avionics boxes. It is often noted that a vast majority of GA accidents aren’t caused by the aircraft itself failing (a “hardware” failure) but rather by the pilot making bad decisions or otherwise not doing what needs to be done (a “software” failure). So it is with an avionics box: the hardware itself is probably pretty reliable, but since these newest systems must involve 100′s of thousands of lines of programming code, built by humans, how do we know that they will not fail in some obscure, but important, situation? The answer is that they must be rigorously tested in as many “corners” of flight conditions as possible.

    None of us wants to be flying along in IFR only to find that our primary flight devices have frozen, or decided to reboot. We’ve all had this happen with our personal computers, and we all would expect that the software in our airplane would be better than that. Requiring backup steam gauges might be part of the answer (as long as the software in the plane (the pilot) is current with them). But as much as I wish that I could put an all-glass cockpit into my Piper for something less than the cost of a luxury car, I also want those devices to be tested to a better standard than, say, .

    I’m not arguing with Mac’s basic point, which is that it seems that things should/could be less expensive than they are if regulations were brought into modern times. But I do think that we should want the FAA to insist on some sort of more rigorous testing for avionics software than might be expected for, say, the hand-held GPS I carry in my car. The (many) folks who put non-certified avionics in their experimental aircraft do so completely legally, but they are accepting an increased risk that something might fail. I would bet that because these systems are so impressive in their normal use, most folks don’t think about the risk of obscure software bugs causing great harm. But the risk exists. So I don’t think that we should ever expect certified glass-cockpit avionics devices to be as cheap as “similar” consumer devices.

    A good article; lots to discuss. Thanks.

  10. Don Heitzmann says:

    Sorry, the prior post got snagged with an errant html tag. The dangling sentence should have said “…. to a better standard than, say [I won't name them but put your favorite PC operating system here]. ”

    The italics in the last paragraph weren’t intended. Again, sorry!

  11. Michael Shaw says:

    Lots to agree with, but I think the iPad vs. installed avionics is less than thought through. Millions of iPads are sold vs thousands of installed avionics. Surely there are economies of scale in the iPads favour!

  12. Todd Parker says:

    While I agree with your points concerning avionics certification in the cockpit, I disagree with your opening statement concerning structure and flight characteristics not being an impedement to certification. If you read and understand the implications of these, then you see that these put huge burdens on the aircraft companies to show compliance through extensive test programs. The flight characteristics also put serious restrictions on the designs, which is one reason most certified aircraft have rather unspectacular performance numbers when compared with experimental aircraft. So, while avionics are a huge part of the cost of certified aircraft, so is certification of the airframe and powerplant(s), which have similar stringent qualification requirements.

    • Mac says:

      Hi Todd,
      One reason that E-AB airplanes often have higher cruise speeds than a standard airplane with the same available power and configuration is wing size. The rules require a certified single–piston or turbine–to stall no faster than 61 knots at maximum takeoff weight in the landing configuration. That requirement dictates a fairly large wing area plus effective flaps for most singles. E-AB builders are not bound by the 61 knot stall speed maximum so they can opt for much smaller wing area. A smaller wing creates less drag in cruise. This is not a new aerodynamic understanding. A quick look at any of the GB racers, for example, will show you how even 80 years ago designers knew that the smallest possible wing made for the fastest airplane.
      I, for one, think the 61 knot rule has served us well giving the pilot making a forced landing a good chance to survive even when crashing into difficult terrain.
      Of course, another big reason E-AB can be faster is weight control. Customers of standard airplanes demand plush leather interiors, lots of extras such as air conditioning and super soundproofing, and all of that adds weight. Nothing slows down an airplane–or we humans for that matter–more than increased weight. Light is fast and E-AB builders get to choose how to trade the weight of equipment and furnishings for extra speed. And if your airplane weighs less the wing can be smaller without increasing stalling speed. The small wing is faster and so on and so on.
      Mac Mc

  13. Dan Nelson says:

    I miss my Northstar Loran. It would work at voltages well below 5 volts if all electrical dumped. I could dial up airports, flight plans faster than my ride along could on his Garmin. Emergency airports was just a push of two buttons together. I could actually work the knobs and buttons in heavy turbulence,by hanging my fingers on the knob and chassis. Sure it had no moving map display, but my Mark 1 cranium computer could picture what the Northstar was indicating and a map in my lap backed it up along with VOR’s and a good DME. Updates were affordable for around $120. Repeatability was a loran feature that no GPS could duplicate. Being ground based the Northstar would bring me to the exact spot on terra firma without any deviation of footage. The Northstar never had any issues in the 20 odd years that I owned it. It may have burped when signals were squelched from heavy rains, but that was rare and it didn’t last more than a minute. It was totally immune to solar activity, and during military conflicts the signals were never turned off. The loran system was much cheaper to maintain being ground based. I guess it was too good and the GPS lobby had to kill it. And we bought into this new doo-dad stuff, so we are to blame for wanting the next new shiney gadget. Cough up the cash. More in the dash, less in the tanks and cut down on the flying time.

  14. Dov Elyada says:

    Autothrottle is a great idea, and I’ve been thinking of other, similarly great ideas, that only the extreme conservativism of the industry and authorities seems to hamper:
    (1) Automatic slip/skid control for cruise flight in aircraft lacking yaw trim tabs, that would disengage immediately upon the pilot applying any rudder pedal force.
    (2) Automatic carburetor heating system for GA/sport aircraft, that would automatically sense and respond to icing conditions within the carburetor — how many fatal accidents could such a system have prevented?

  15. Tom Miller says:

    One area that would be greatly benefited by simplified certification standards: ANTI-ICE AND DE-ICE SYSTEMS. There are some amazing new technologies out there (water repellent coatings, electro-mechanical deicing, and so on), but it will be years and years before anyone sees them in affordable light aircraft. Sad, really.

    • Mac says:

      Hi Tom,
      Actually, ice protection systems can be added without icing certification. If an anti-ice or deice system adds external elements one would need to demonstrate they do not alter the flying qualities, particularly slow speeed and stall behavior. But you don’t have to prove that the system qualifies for icing certification. There are a number of FAA approvals for TKS or deice boots that are not also approved for flight into icing.
      I have never been impressed by icing certification–flight into known icing certification–because there are icing conditions that occasionally exist that can defeat a certified system. Know ice approval to me would be like flight into thunderstorm approval. Yes, you can make it through some thunderstorms, and yes, a certified system can remove or prevent most ice, but not all. In fact, the FAA definition of severe icing is icing that an approved system cannot keep up with.
      So, if the anti-ice coatings, or versions of electro-mechanical impulse systems work and pilots will buy them, known ice certification won’t stop that progress.
      Mac Mc

  16. Craig Maiman says:

    Great article. The functionality you mention of the Vertical Power box that will automatically glide the plane to an airport in the case of an engine failure, is now available on the iPad. It’s called Xavion and it has synthetic vision display, back gps-based instruments and will show HITS boxes to glide through if the engine fails.

    The cost? $99. One-time charge.

    See here:

    The software for Xavion is the same software as in the Vertical Power box, because they were both written by Austin Meyer of X-Plane fame. I may sound like I work for him, but I don’t. I’m just very excited by this new app. A no-brainer.

    • Mac says:

      Hi Craig,
      The Xavion iPad app is exciting, but there is one big difference between it and the Vertical Power system and that is Vertical Power can couple to the autopilot and actually fly the airplane to the emergency runway. In some installations the Vertical Power system can even control wing flaps too add drag when needed. But, having an app that can guide you to the best available runway is still pretty darn good.
      Mac Mc

      • Craig Maiman says:

        Yes, the Vertical Power box looks great (for it’s other features too). I’m planning to build a plane and I’ll definitely be installing that.

        I’m the creator of and I’ve been talking to Vertical Power and Austin about being able to upload that database into the box as an option (there won’t always be an airport in gliding distance…). I’m hoping to have that option available within a year or so (need to get lots more people to enter “emergency” runways).

        The technology and much lower cost of that technology that you can put in an amateur built aircraft is incredibly compelling for me.

  17. Jeff Aryan says:

    All the comments have been great. They have all been talked about for years.
    What would be of real value, is hearing from the FAA side of the argument. Not the standard company line, but the real person and the supervisor who came up the rules. This in my opinion will offer greater credence as to why the rules have been made. It should then be looked at again by both sides and be determined it is not overkill or a good rule.
    The bottom line, is “Common Sense” being used. Not just numbers and statistics.


  18. SkyGuy says:

    Why are we chasing “apps”….that keep the eys inside….not outside…of the flown aircraft.
    VFR is VFR.

    • Michael Shaw says:

      Interesting Comment SkyGuy. When I learned to fly over 40 years ago, time was spent looking at a map in my lap and writing times as I passed landmarks. Then it was spent looking at a whiz wheel to calculate ground speed and estimate time enroute compared to fuel quantity. This does not include the moments of uncertainty, is that Lake whatever… Now I look at the GPS for a couple of seconds and all of this is done for me. I have more time to look out of the windows for traffic, not less.

      Another take on apps…

  19. SkyGuy says:

    - I agree….I just program the destination in…..and activate the autopilot…..then monitor progress to make sure nothing has failed.
    - No tablet or iPad required.
    - Paper maps are the backup.
    - Forgot how to run the EB-6 years ago.

  20. sid siddiqi says:

    Mac correctly targets the high cost that gets embedded by the certification process.
    We should keep in mind that the entire process of delivering airplanes in the hands of pilots and travelers needs to be leaned out not just the certification of avionics. The FAA and the manufacturing companies all constrained parts of the processes that make the whole a high cost edifice.

    In my opinion (others too have pointed this out) the survival of General rather than Elite aviation need to grow pilot starts to maintain critical mass; cost along with ease of access and use remain huge barriers. Mac’s blog shows how avionics certification expenses act as a huge inhibitor and if his suggestions are followed the ball can begin rolling to address the entire problem and preserve this marvelous General Aviation eco-system our ancestors from the Wright brothers onward have created – which is endangered by the stagnation.

  21. locate says:

    excellent points altogether, you just received a logo new reader. What could you suggest about your put up that you made some days ago? Any positive?|

Leave a Reply

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


You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>