I challenge you to create an electronic fuel injection system that is as reliable as these older-tech aircraft systems, without a whole bunch of redundancy for backup. You can't. EFI relies 100% on a supply of electricity at adequate voltage and current, and in airplanes where that alternator is running at twice or three or four times as fast as in the car (we've argued THAT before) but not often getting the inspections they should, their brushes wear out and fail and the battery has to keep radios and lights and some instruments working until a landing can be safely made. Electronic fuel injection absorbs a lot of power, too. The old mechanical systems just keep on working when the electrics fail, like magnetos will do. The FAA knows this and so designing and certifying an engine with EI and EFI is expensive.
I'll be vague, because some of this stuff is patent-worthy and I hope to have enough free time at some point to put it all in an application and submit it to the USPTO.
An electronic fuel injection system, fully sequential and with optional closed-loop mixture control (if we manage to get rid of the lead, although even with the lead a sensor should last enough to make it to the next oil change) could be created from commercial, off-the-shelf parts, with about 10-15% added for redundancy to what it would be needed for a single, non-redundant system.
You just have to design smart. Sometimes it's a lot cheaper to duplicate components than to go through the reliability analysis required to prove one of them is enough.
In this day and age, even back-up power is not an issue. You mentioned alternators and the complete lack of proper maintenance on them. Valid point. But we don't need brushes anymore. Most FADEC systems use a PMA (permanent magnet alternator) to keep the FADEC alive if all primary power is lost. It's literally a magneto without the points and the high voltage coil and distributor. Cheap in the grand scheme of things.
Certification is a problem really just in terms of cost, and the market not being committed and unified enough to make it viable.
Here's what I mean by that. An EFI system today costs about $1000-1500 in parts, including ECU, sensors, injectors and ignition coils. Add maybe another $500 if you include parts required for redundancy. We're not going to discuss installation costs, as those can vary and would be outside the scope of a company selling the retrofit kits. Of course, if I make this there will be some profit margin tacked on top. I'm not a charity and I expect to be sued, even if the guy crashed because he ran out of fuel.
But to certify this, between the design work, creating drawings, manuals, certification documentation, ground and flight testing and everything else required for certification, you need, optimistically (again, assumes a smart design with a good understanding and mitigation of the hazardous/catastrophic failure modes) a minimum of $250k, with a more realistic number being probably closer to $500k.
Now, if I show up in July at OSH with an experimental aircraft, equipped with an O-360 and a full EFI system, and I tell everyone that it could be on their aircraft in a couple years if they give me a $500 deposit (balance due at delivery), how many deposits do you think I'm going to walk away from OSH with? Based on what I've seen so far, I think I know what that number will be. 0
Too much "I'm perfectly happy with my system that I can barely get parts for", "Why change it if it works?", "Too new for me, I don't understand electricity" out there to have a viable market.
A bit of a tangent here, but if I were some big exec at AOPA, this is the kind of stuff I would focus on. Put out a request for proposals for a given system (EFI in our case), have people submit preliminary designs, get them vetted, select the best, work with the proponent and the FAA to establish a certification path and then fund the certification. Get this thing to market and use my marketing might to convince people to go for it. We don't need two dozen competing systems, all eating each other alive in too small of a market for all of them to be viable. You need one or two solutions that work and cover 80-90% of the existing fleet. Stuff like that will keep the certified fleet going and potentially economically viable. Pick a project every year. Skytec is gouging people on starters? RFP for a new starter option. And so on.
