You cannot optimize an engine for best economy, since that is dependent on aircraft speed, winds, and all kinds of things. An engine has best efficiency and max power. Max power is a function of how much air and fuel you can push into it. There is no upper limit until the engine disintegrate due to temp/forces, given you can pack enough air into it. Efficiency is another matter, but there is no good reasons not to operate at best efficiency point at any given RPM unless you want more power than the best efficiency will give you at that RPM.
The way I understand this is as follows (for modern aero angines as Rotax iS and ULPower among others). You measure inlet air temp and air pressure (which combined gives inlet air density, a more accurate form of “MP”), together with throttle position and RPM. Then you measure torque and fuel flow. Keeping the inlet air density, rpm and throttle position constant, varying only fuel flow, you will get an efficiency curve and a torque curve as function of fuel flow. The efficiency will have a peak point at one certain fuel flow, and the torque will have a peak point at another slightly higher fuel flow. You also vary the timing of the ignition to optimize efficiency and peak torque even more. So you have two points, best efficiency and peak power (peak torque) with the corresponding timing advances. You want the engine to operate somewhere in that band. You do that for a whole bunch of combinations, and you get a full map of the engine, a lookup table. EGT and CHT are irrelevant, they become what they become. Just as you get a whole bunch of peak power points, all different from each other, you also get a whole bunch of best efficiency points all different from each other.
In the aircraft, the ECU measures air temp, air pressure, rpm and throttle position, then use the lookup table and clever algorithms to inject the correct amount of fuel and set the timing advance. What’s a bit hard to grasp is how the ECU knows if the pilot want max power or max efficiency. Does it really need to know? I’m not so sure. Max power is one single point: max torque at max RPM, at max throttle setting, but it may depend on the propeller? It certainly depends on the alt. But, if you are not running at max (overall) power, you and the engine are better off running at best efficiency IMO.
A fixed pitch propeller: Max power can simply be implemented as the last 5-10% throttle movement. From 90-95% movement it will start to go rich (more power). Best economy is a simple matter of monitoring fuel flow.
A CS prop: The same, but could additionally be augmented by rpm.
Single lever control (and CS prop): Probably the simplest because you want a 1 to 1 relation between throttle position and power, and 100% is always max RPM and max torque at any given inlet air density. I imagine the actual implementation can be done in a number of ways though.
When Rotax first released the iS, they got complaints from all over because it used the same amount of fuel as the Bing carbureted ULS. No improvement in fuel consumption, how could that be with a FADEC engine? A year or two later (don’t remember exactly) they upgraded the engine to the current version. Huge improvements in fuel consumption only with a software update.
No matter how this is implemented, I see no reason why there should be a “battle” between best efficiency and max power. 100% throttle position is max power per def, only for the non single lever CS prop it could be adjusted by RPM. If the RPM isn’t also max, then go for best efficiency.
Wouldn’t this simply use gated throttle positions like a jet? Max available power, Max continuous power then pull it back for a percentage power?
I’ve come across FADEC engines in several incarnations. First one was the Saab 2000, where FADEC control basically meant it could be flown in most situations like a jet. No condition levers (apart from fuel on/off) no prop control, just a manual feather button if autofeather failed. Power was set in power units (%). No autothrottle. I found it very easy to use and very easy to understand what it was doing. As I have no experience in other turboprops, I had no comparison.
Airbus family vs SE210 Caravelle. Same genes, lots of left over design features when you came from the Caravelle to an Airbus (or, for the few who got it to Dassault Mercure which is at least indirectly the direct ancestor of the A320).
To start the engine in the Caravelle was a page long procedure, I remember it in parts so I go from memory. APU had to be on to deliver pneumatic and electricity (10B3), APU bleed needed to be provided (controls behind copilots seat at a FE position), starter button pressed (overhead) wait for N2 to reach a certain value (20% IIRR), put fuel cut off lever to on (center panel) wait for Fuel flow, wait for EGT rising (Call out “light up”) and check for temp rise not to exceed redline, keep N1 and N2 in view rising. Wait for start valve to close and call “engine stabilized”.
At take off and GA power needed to be set by hand and carefully. firewall the engines and you need two new ones in the best case.
In flight, fully manual control, check temps, FL, and set EPR according to a large table.
A320:
APU ON, APU bleed on, Engine lever “Start”. Monotor the engine instruments during start up and voila.
Throttle has 3 detents: TOGA Power, Flex Power and Climb. TOGA Power will set Max power. Flex power will set take off power as provided by flex procedures on the FMC (in theory you can do without it) and Climb sets initially climb power but then will act as an ATHR position where the throttle remains unti 20 ft off the runway on landing when the “Retard” command means you have to put them in idle. Anything betwen Idle and Climb position the throttle acts as a manual throttle. Limits, e.t.c are all taken care of by FADEC.
While doing the procedures in the Caravelle were fun of sorts, the Airbus procedures are much easier and also kinder to the engines.
In GA? Start up of our Mooney’s O360, bog standard. Master on, Mixture rich, Fuel Pump on, prime (depending on temp and conditions between 3 and 10 times, get it wrong and you get a carburettor fire). Wait 10 seconds. Turn ignition key to start and push in to engage starter. With luck, it will start on the first time, if not, be prepared to repeat with variants in priming e.t.c. Throttle and Mixture positions and procedures, books have been written about it. Take a M20E, F or J in a hot start (injected) and libraries have been written yet many wait for the engine to cool down..
Take off: The Piper Seneca II’s turbos (or the Mooney 231) can be overboosted like nothing. So fiddling to find the spot where those MP’s are just below the max at a moment where you should be watching other things.
Engine failure (Twin) you need to reduce throttle, feather prop, put mixture idle cut off before the windmilling prop gets you into trouble (see the DC3 accident 2 weeks ago).
In flight? How many people don’t lean at all because they are too lazy, too afraid because of nerdy FI’s who don’t teach it because they themselfs don’t know how? How many pages of discussion have you read how to reach the actual TAS advertized in your airplane, what is the most efficient, the most speed, the most long range power setting available, 2600 or 2450 RMP, ram air on or off, ROP, Peak, LOP? While that is fun to read at times, it is NOT state of the art, not by a long run.
What would I expect of a FADEC? Engine Start done like I start a car. Throttle with detents for TOGA, 75%, 65%, 55%. Power selector for Best power or best economy which would regulate mixture and prop setting to either max performance or long range performance. Properly leaned engine at all time. Auto Feather on twins. Power set automatically and indicated in % Power. Finished.
I’d expect much more stress free engine handling, maybe slight fuel economy but primarily a much higher percentage of engines (particularly turbos) making TBO and beyond as finger trouble is out of it. Even autothrottle in connection with a nice system like the Aspen could be a thing in reach.
In other words, no more “goat sacrifice” to start the darn thing, no more page long threads about leaning them and constantly fouled spark plugs because someone forgets to lean on the ground, no more overboosts, do I need to carry on? And oh yea, that would be an STC I’d buy the day it becomes legal to install.
Mooney_Driver wrote:
What would I expect of a FADEC? Engine Start done like I start a car. Throttle with detents for TOGA, 75%, 65%, 55%. Power selector for Best power or best economy which would regulate mixture and prop setting to either max performance or long range performance. Properly leaned engine at all time. Auto Feather on twins. Power set automatically and indicated in % Power. Finished.
If you want that over a range of altitudes, running the engine at minimum rpm and maximum throttle opening to enhance efficiency, you need throttle and propeller by wire with no mechanical connections. Stepper motor to actuate throttle etc. Moving the power selector will then set minimum rpm for a given power demand, with that rpm varying with absolute altitude. That increases efficiently and decreases noise. Leaning can be mapped, or done with a combination of map and exhaust feedback like a car. There is no need for the pilot to select best power or best efficiency, the controls can figure out the right mixture (as well as rpm and MP) for any power demand in any operating condition, and change it when the lower demand changes. Percent power can be calculated and displayed, relative to both rated sea level power and power available in the current operating condition,
Me, I’ll stick with what I’ve got and I’ll be happier. I fly among other reasons to escape stuff like I’ve written above, use my senses and brain, move the controls myself, look at the view and enjoy it all. When I’m back on the ground and doing maintenance I smile with satisfaction at super simple parts, e.g. little cable operated single barrel carburetor, that can take me anywhere I want to go regardless. Starting is 100% reliable. I would go for self contained electronic ignition with manifold pressure based advance, because it’s simple to own and operate.
One would also need to bring CHT into it, which you probably don’t need to do with a watercooled engine (well assuming the cooling system is correctly sized).
Water cooled engines also need an engine temperature sensor for warm up, but once warmed the engine temperature does not vary as much which makes the mapping simpler, at least that part of it that is normally in use. One of the things you find on water cooled EFI motorcycle is that when the coolant temp gets hotter than the range the manufacturer really studied (i.e. in traffic) the mixture keeps leaning out which reduces the idle speed. The solution on the newest engines is the stepper motor actuated throttle which then maintains idle speed with the leaner than max power mixture.
My lesson learned is that all this stuff is possible but if you want it to work as well as a tuned-in human in all conditions you have to accept a lot of ‘stuff’ and that the human operator becomes a supervisor of the systems running the engine. I’m a supervisor like that at work and get an emotional recharge from being directly connected to anything. People fly for different reasons 
MedEwok wrote:
In any case I find the argument against adapting technology that has been used in cars for decades weak.
I offer you a joint venture to develop, certify and produce a complete aftermarket refit FADEC for Lycomings, Contis, Franklins and Rotaxes. You deliver the money, I deliver the tech and certification.
Even if you did it, almost nobody would buy it unless it is 100% standalone, like mags are. If it needs battery power etc etc, it’s a no-sell.
Racing cars have electronic engine management with O2 wide band sensor to deliver what is exactly needed, where CHT is not really a need, but anyway we can add this into the sw management loop. Turbos are even more complex and working very well. I don’t think it is a complexity issue but more a business plan from lyco and conti for the lineup.
I think the traditionnal engine makers just doesn’t want to spend money in R&D but just take the money that is just here below. They are not evolving since years… I mean thielert and now conti and Diamond are using ECU/Fadec and this is working great. Theses are electric planes, and especially DA40 are concerned by an alternator failure, but do we have data on these? When googling “DA40 alternator failure”, I have pretty much result about the lyco version…
DA42 have 2 of them (plus the 30m ECU battery), they are less exposed to a full electric failure. RAT would be a solution?
At the end, it is really simple plane to fly from an engine management perspective, never overheat, never shock cooled, no VP to handle.
Peter wrote:
Even if you did it, almost nobody would buy it unless it is 100% standalone, like mags are. If it needs battery power etc etc, it’s a no-sell.
FADEC in jets tends to have (after engine start) primary power from PMAs (permanent magnet alternators) which run off the accessory gearbox with ship’s power as standby.
