What not efi like cars

Its weight is 584 pounds. The 300-hp IO-520 it replaces weighs 428 pounds. How will they balance something like a C185 that already has its battery in the tailcone? Other airplanes like 210s or 206s might be balanced by moving the battery aft. Maybe a composite prop, such as used on the 182's SMA diesel, would help some.

I wish they had shown the PSRU gear ratio. The engine redlines at 3880, so I'd expect around 1.5:1 or more to get better prop efficiency and lower noise levels.

There is no TCDS that I can find, so I don't think it's certified yet. "FAA validation pending," they say. And yet, according to Wiki, the Diamond DA-50 has been certified with that engine in it. Go figure.

It is a different diesel than Continental's TD-300, 230 HP, an obvious adaptation of SMA's SR305-230.
 
Long body Mooneys with big bore Continentals get two batteries in the tail cone.
 
kyleb said:
EFI is available in the experimental world, but there are limitations due to physics and chemistry. First, because knock sensors have a hard time with our large air-cooled engines, so we can't use them to control ignition timing. Second, because lead contaminates oxygen sensors, which are used to optimize the fuel/air ratio.

There are workarounds for both issues, but at this time you can't just yank a system off of your Toyota and bolt it to your Lycoming...
Click to expand...
No and no.

EFI is available in either closed or open loop systems. Closed loop uses the O2 sensor to determine the fuel/air ratio whereas open loop does not and uses a simple lookup table to determine how long to pulse the injector.

Closed loop is way more efficient, but open loop runs fine.

As for knock… your airplane engine is knocking all the time, its just so noisy you never hear it… a knock sensor is more art than science but not difficult.

And yes, you can yank the system off your Toyota as long as the injectors can supply enough fuel and the displacement is close. Now the ECU may not like that there are emissions signals missing, but that can be faked.
 
Klaus M said:
You mean like this outfit?

https://www.flyefii.com/

Don't hold your breath waiting for the FAA to certify EFI. I recommend everyone trade their certified aircraft in for a Homebuilt. ;)
Click to expand...
You mean ‘wait for a company to pay to certify an EFI system’. All the diesel engines in AC now are EFI / FADEC so its being done… just not as a retrofit.

Look at the nonsense the FAA put unleaded fuel through. How much money do you think current FI manufacturers will pay to lobby the FAA not to allow a very cheap and simple EFI system?
 
pfarber said:
No and no.

EFI is available in either closed or open loop systems. Closed loop uses the O2 sensor to determine the fuel/air ratio whereas open loop does not and uses a simple lookup table to determine how long to pulse the injector.

Closed loop is way more efficient, but open loop runs fine.

As for knock… your airplane engine is knocking all the time, its just so noisy you never hear it… a knock sensor is more art than science but not difficult.

And yes, you can yank the system off your Toyota as long as the injectors can supply enough fuel and the displacement is close. Now the ECU may not like that there are emissions signals missing, but that can be faked.
Click to expand...
Wait, so are you saying you disagree with the larger portion of my post, or with the last part of the post which said there are workarounds? Not sure you can have it both ways...
 
pfarber said:
As for knock… your airplane engine is knocking all the time, its just so noisy you never hear it… a knock sensor is more art than science but not difficult.
Click to expand...
It's not detonating all the time, which is what a knock sensor looks for. If it was, it would quickly disassemble itself.
 
EFI will almost certainly appear on new airplanes to be introduced using the proposed new LS rules. It seems the FAA wants to divorce itself from private GA. That’s a big win for private owners.

There are some really good EFI systems used in experimentals. They don’t use knock sensors and the advantages are real. LOP ops are far better with EFI.

My exp Cub uses state of the art mechanical injection (API) and self-generating Pmags. I was reluctant to be electric dependent. If I was doing it over? I’d have a full EFII injection and ignition system.
 
Stewartb said:
EFI will almost certainly appear on new airplanes to be introduced using the proposed new LS rules. It seems the FAA wants to divorce itself from private GA. That’s a big win for private owners.

There are some really good EFI systems used in experimentals. They don’t use knock sensors and the advantages are real. LOP ops are far better with EFI.

My exp Cub uses state of the art mechanical injection (API) and self-generating Pmags. I was reluctant to be electric dependent. If I was doing it over? I’d have a full EFII injection and ignition system.
Click to expand...
This....I wouldn't hold your breath on a certified unit until there is an EFI/ignition system that works well for E-AB. Once that gets some time under its belt....it will migrate over just like SkyView did with Dynon.
 
Checkout_my_Six said:
This....I wouldn't hold your breath on a certified unit until there is an EFI/ignition system that works well for E-AB. Once that gets some time under its belt....it will migrate over just like SkyView did with Dynon.
Click to expand...
There are a couple, but they are NOT system feedback driven. In the EAB world, they generally come with baseline settings which are fairly conservative. You get to produce your own optimized map for your engine. Getting that right requires a lot of testing, but can be done. The bigger issue is EFI pushes you to an electrically dependent design, and you've gotta be very careful there. Not saying it can't be or hasn't been done, but there have been numerous crashes of certified and experimental aircraft where the redundant, fail safe, electrically robust system failed and left someone riding a glider to terra firma.
 
Yes, there’s been a learning curve among my friends. An EFI electric failure in a friend’s plane is what steered me to Pmags and mechanical injection. That plane had one injector short and it killed the system. It got rebuilt using dedicated breakers for each injector. All the EFI guys I know converted to the same. No problems since. We learn from our peers as long as peers continue the share honest info.
 
kyleb said:
There are a couple, but they are NOT system feedback driven. In the EAB world, they generally come with baseline settings which are fairly conservative. You get to produce your own optimized map for your engine. Getting that right requires a lot of testing, but can be done. The bigger issue is EFI pushes you to an electrically dependent design, and you've gotta be very careful there. Not saying it can't be or hasn't been done, but there have been numerous crashes of certified and experimental aircraft where the redundant, fail safe, electrically robust system failed and left someone riding a glider to terra firma.
Click to expand...
Guess what happens when you add complexity?...Reliability goes down....until lots of time goes by... And learning over that time. One can try and use design standards from the ANSI/SAE or what ever design standards you like to get a jump start....but time and learning is needed in this "new" operating environment.
 
EFII is pretty sophisticated. They use an O2 sensor, temp sensor, and MP sensor to automatically adjust and maintain mixture. The “problem” for me was EFI systems require a fuel return and accumulator tank. More plumbing, more electrics. Bendix mechanical injection doesn’t require a fuel return while Continental FI does. There are advantages and disadvantages to both but from the builder perspective? No accumulator tank is easier. My Airflow Performance servo has zero leak-down so is an improvement over Bendix but gives up the fine tuning control of EFI. To retrofit to EFI? Not an easy task. That alone explains why not many guys are using it. Designing/building the plane for EFI is better done from the beginning.
 
kyleb said:
There are a couple, but they are NOT system feedback driven. In the EAB world, they generally come with baseline settings which are fairly conservative. You get to produce your own optimized map for your engine. Getting that right requires a lot of testing, but can be done. The bigger issue is EFI pushes you to an electrically dependent design, and you've gotta be very careful there. Not saying it can't be or hasn't been done, but there have been numerous crashes of certified and experimental aircraft where the redundant, fail safe, electrically robust system failed and left someone riding a glider to terra firma.
Click to expand...

You can still have a carbureted backup system in an EFI equipped plane, you just need to plan for it. I’m in the middle of converting my sonex over to EFI but am still leaving the aerocarb fully hooked up. The aerocarb doesn’t have a float bowl and can be used as the throttle body for the EFI system. If the EFI fails all I need to do is shut the power off to it and open the mixture to the carb and the engine will run as normal. The mags are also being left on as a secondary ignition system if the electronic ignition fails as well. I’m not sure why more aren’t designing a backup system like this for EFI equipped planes rather than a fully redundant electrical system. There is of course a weight penalty but it’s still less weight than a secondary alternator or battery. The joys of the experimental world I guess.
 
Stewartb said:
Bendix mechanical injection doesn’t require a fuel return while Continental FI does.
Click to expand...
Bendix (now RSA) is used on the newer Cessnas. There were problems with the lack of a return system, like air reaching the fuel servo, or vapor locking, so they fixed that. From the T206H AFM:

1691427328528.png

First 361 airplanes had headers (reservoirs), no fuel return. And then they added fuel returns:

1691427508363.png

The T182T has fuel return, too.

1691427669899.png

The 172R had a header but no return.

1691427861182.png

The 172S models got the return line later, too, and retrofit kits were offered for the earlier non-return-line airplanes.

1691428045869.png

1691428568034.png
 
Dan Thomas said:
It's not detonating all the time, which is what a knock sensor looks for. If it was, it would quickly disassemble itself.
Click to expand...
Yes, it is. You can’t hear it. It’s not a destructive detonation but it is happening. Any questions? I’m more than happy to help or point you to resources that can explain it to you. Here’s Mike Busch’s explanation if you want a second opinion:

“
In an automobile, we can usually hear detonation in the form of audible “knock.” In an airplane, we can’t – there’s too much noise – but we can observe it on an engine monitor in the form of excessive CHT and depressed EGT.

Detonation is something that occurs near the peak pressure point in the combustion event, after the air-fuel charge has been ignited normally by the spark plugs. It is characterized by abnormal pressure spikes near the peak pressure point, caused by spontaneous combustion of end gas due to excessive temperature and pressure.

Contrary to what your CFI or A&P may have told you, detonation is not necessarily harmful. Many engines operate in light detonation quite regularly, and some can withstand moderate detonation for extended periods of time without damage. Detonation is not an optimum situation, but it’s not necessarily destructive. The higher the specific output of the engine, the more likely it is to sustain detonation damage. An engine that produces 0.5 hp/in3 (horsepower per cubic inch of displacement) – as is typical of most carburated aircraft engines – can usually sustain moderate levels of detonation without damage, but highly-boosted turbocharged engines rated at 0.625 hp/in3 or more can be damaged fairly quickly by detonation.”
 
Checkout_my_Six said:
Guess what happens when you add complexity?...Reliability goes down....until lots of time goes by... And learning over that time. One can try and use design standards from the ANSI/SAE or what ever design standards you like to get a jump start....but time and learning is needed in this "new" operating environment.
Click to expand...
A turbine engine has just entered the chat.

Variable inlet geometry, variable turbine inlet guide vanes, electromechanical FADEC, thrust reversers, bleed air… whomever said a modern jet engine isn’t full or moving parts hasen’t seen a modern jet engine. And trust me, jet engine accessories break all the time. That’s why there are THREE of everything (don’t forget about that little APU in the back).

I’ll agree that the engine itself is robust, but jets also have CONSTANT maintenance on them.
 
kyleb said:
Wait, so are you saying you disagree with the larger portion of my post, or with the last part of the post which said there are workarounds? Not sure you can have it both ways...
Click to expand...
Yes I am saying most of your post was incorrect. Running off a fuel map is NOT a workaround. The O2 sensor was added later on when the technology was developed. Lots of early EFI was basically a map of RPM and air density.

Is there anything specific that you think I am incorrect about?
 
pfarber said:
Contrary to what your CFI or A&P may have told you, detonation is not necessarily harmful. Many engines operate in light detonation quite regularly, and some can withstand moderate detonation for extended periods of time without damage. Detonation is not an optimum situation, but it’s not necessarily destructive. The higher the specific output of the engine, the more likely it is to sustain detonation damage.
Click to expand...
As in old cars, light detonation won't blow it up anytime soon. But in an airplane, you don't hear ANY detonation and you have no idea how close you are to ruining the engine. Like Busch says, high CHTs and low EGTs are all we have, mostly, and how many pilots are trained to interpret those to avoid detonation?

Busch is covering himself there, if you can see it. When an engine fails due to detonation, it's not Mike Busch that has to deal with the angry owner or the relatives of the dead pilot. It's Lycoming or Continental or the last rebuilder.

The POH/AFM tables are provided to give the pilot safe operating margins. Lycoming and Continental both publish Operator's Manuals for their engines, and there is advice and published limits there, too.

In my aircraft systems class I had a bunch of aircraft and engine parts. One of them was the cylinder of an IO-520; its head was blown off. The aluminum had fractured all the way around. Detonation did that, and a pilot caused the detonation somehow.

It's not something to dismiss lightly.

Once more: What are your pilot and aircraft maintenance credentials?
 
Aircraft engine ignition timing is set to produce the best cylinder pressure for the power stroke while maintaining a margin of safety for pre-ignition. Standard mags are fixed at 25° or 20° advance (per engine data tag) and use impulse couplings to retard timing for engine start. Electronic ignitions use manifold pressure to vary spark advance into the mid and high 30° advance range. Systems like EFII and SGS allow the pilot to change fuel and ignition settings to squeeze out more power or more economy. EFII’s ignition sparks for 32° of crank angle, too. No unburned fuel is the goal. My own IO-390 has Pmags with A-B curve switching and the guys at Emag advised me to disable the more advanced B curve for my 10-1 compression or risk detonation. I’ve also retarded a couple of degrees to promote safer starts, but that’s another story.

Bottom line, ignition timing is all about max cylinder pressure on the power stroke.
 
well....you certainly don't want max cylinder pressure at the wrong time. :rolleyes:
 
Dan Thomas said:
As in old cars, light detonation won't blow it up anytime soon. But in an airplane, you don't hear ANY detonation and you have no idea how close you are to ruining the engine. Like Busch says, high CHTs and low EGTs are all we have, mostly, and how many pilots are trained to interpret those to avoid detonation?

Busch is covering himself there, if you can see it. When an engine fails due to detonation, it's not Mike Busch that has to deal with the angry owner or the relatives of the dead pilot. It's Lycoming or Continental or the last rebuilder.

The POH/AFM tables are provided to give the pilot safe operating margins. Lycoming and Continental both publish Operator's Manuals for their engines, and there is advice and published limits there, too.

In my aircraft systems class I had a bunch of aircraft and engine parts. One of them was the cylinder of an IO-520; its head was blown off. The aluminum had fractured all the way around. Detonation did that, and a pilot caused the detonation somehow.

It's not something to dismiss lightly.

Once more: What are your pilot and aircraft maintenance credentials?
Click to expand...
You just ignored the physics because 'you had a class once'? Detonation happens ALL THE TIME. Certainly some of them are serious enough to cause a failure but the vast majority of the time its not sever enough to be an issue.

Please read this (its got all the charts and graphs to explain it) https://resources.savvyaviation.com...es_eaa/EAA_2010-09_destructive-detonation.pdf

Here's more if you think 'Busch is covering himself" (I must of been in the bathroom when they talked about Mr. Busch's liability concerns in anp school.)


"However, it is normally considered impossible to hear detonation occurring in an aircraft because of the
high noise level. As a result, it is possible to suffer mild detonation and not be aware."


Engines, especially the old tech in airplane engines, is nowhere near as good as most people think it is. Ever wonder why, even after a rebuild, many folks need to replace a few jugs? Or why mags are rebuild every 500 hours? Or why oil only lasts a few dozen hours? Its becuase aviation is stuck in the 1940's and sadly, the old heads passing on information are not up to speed on how and why things are they way they are... they just repeat 80 year old bad information.
 
Last edited:
pfarber said:
You just ignored the physics because 'you had a class once'?
Click to expand...
I taught Aircraft Systems in a College for 12 years. Engines, airframes, everything. What are your credentials and experience? I have asked you that about six times and never get an answer.
 
pfarber said:
Detonation happens ALL THE TIME.
Click to expand...
Only in aircraft not properly maintained or operated improperly. However, based on your other replies on PoA I can see why detonation happens "all the time" on your aircraft.:rolleyes:
pfarber said:
Ever wonder why, even after a rebuild, many folks need to replace a few jugs? Or why mags are rebuild every 500 hours? Or why oil only lasts a few dozen hours?
Click to expand...
It seems only on aircraft you are around that needs cylinders replaced after rebuild and mags rebuilt at 500 hours. I've never seen that trend in my 40+ years in mx. Never. And show me any air-cooled engine where the oil stays clean after 50 hours.

pfarber said:
the old heads passing on information are not up to speed on how
Click to expand...
The only "old head" I've seen consistently pass on bad information is yourself. In some cases I actually ask myself how you ever got your A&P certificate, if in fact you have one. Regardless, there's zero wrong with the 1940s tech currently used on aircraft and if maintained properly (key words) will exceed overhaul times and provide long term dependability. When its not maintained properly then all bets especially when you think is normal for an engine to be in detonation all the time. Simply unbelievable.
 
What is the definition of detonation? I think pre-ignition is more correct. A properly timed aircraft engine using properly maintained fuel and ignition components doesn’t suffer from pre-ignition even though the spark induces combustion well before TDC. That’s because fuel takes time to burn. My old 0-320 with parallel valve heads fired at 25° BTC. My IO-390 with angle valves and hemi heads is spec’d to fire at 20° BTC because the angle valve heads are more efficient. My EI fires at up to 34° BTC and works great. Actually I have my mags retarded a couple of degrees for safety, so 32° is more accurate. I figure that’s evidence that factory 20° timing is pretty conservative to avoid any pre-ignition potential. Given my CHTs are quite cool I don’t think my 32° advance is causing any unnecessary stress.
 
Pre-ignition is ignition that starts before the spark plug fires. A real engine killer. Quick, but not merciful.
Knock (the common form of detonation) is spontanious combustion in the end gas region ahead of the flame front - mild knock is not usually a big deal. Excessive knock can damage / destroy engines. Holes in pistons are not uncommon when it gets bad. Sometimes excessive knock can lead to hot spots that result in pre-ignition.
 
Detonation is the sudden explosion of the entire mix in the cylinder. Its flame front travels at around 5000 feet per second as opposed to 100 fps or less for normal combustion. The combustion starts normally, but the increasing temperature and pressure ahead of the flame front causes the complex, detonation-resistant fuel molecules to break down into autoignitable molecules and the whole works lets go. Aircraft engines, with their higher operating temps, relatively low RPM, large bores and ability to lean the mixture, are much more prone to it than cars. It takes time, pressure and heat for those molecules to break down these engines have all that. It's the reason we still have leaded fuel. The new unleaded avgas has some fancy chemistry to discouage detonation.
 
My point? There is no consensus of what detonation is. I’d explain it as premature excessive pressure before TDC. It can be caused by a few things but in any event ignition and subsequent high cylinder pressure prior to TDC is destructive.
 
Mild detonation can raise CHTs and make serious detonation more likely.
 
Stewartb said:
My point? There is no consensus of what detonation is. I’d explain it as premature excessive pressure before TDC. It can be caused by a few things but in any event ignition and subsequent high cylinder pressure prior to TDC is destructive.
Click to expand...
Oh, there is consensus among the people who study it. It's been around a very long time. Tetraethyl lead was first used as an antiknock agent in 1921. 101 years ago.

It's just not covered in groundschools, and even the A&P texts have little to say about it. The chemistry is the least-known, I think.

Dan Horton used to post here, and he covered it well 15 or more years ago. Might have been on rec.aviation.piloting.
 
Last edited:
Personally I have no problem with three knobs. Combined with a brain they do a heck of a lot better than many people seem to realize. Why add a bunch of circuits and sensors for something you can already do yourself? The reason we're stuck with 40's tech is because it works.
 
Dan Thomas said:
I taught Aircraft Systems in a College for 12 years. Engines, airframes, everything. What are your credentials and experience? I have asked you that about six times and never get an answer.
Click to expand...
I've been through through two anp school and they are hilariously bad. Although I will say that my experience at Embry Riddle was somewhat informative you are still teaching 80 (well, 60 year old technology when at ERAU) year old technology. Magneto's belong on farm tractors and lawn mowers.

I've posted two authorative sources that state that yes, aviation engines are constantly detonating. Are you confused by the use of the word 'constantly'? Are you saying you can hear detonation in an avation engine in flight? That it never happens?

What specifically are you saying I got wrong?

Are you attacking my 'credentials'? lol attacks from a position of authority never end well. I can easily say I was an instructor for 13 years. One more than you, so I MUST know more lol. See how foolish that is?

How many sources do I need to show you besides my own personal knowledge do you need? Because two sources isn't enough? REALLY?

I'll throw in a third reference as a cherry on top:

Pelican's Perch #43:Detonation Myths

We've all been taught about detonation in piston aircraft engines. It's what occurs when combustion pressure and temperature get so high that the fuel/air mixture to explodes violently instead of burning smoothly, and it can destroy an engine in a matter of seconds. Right? Well, not exactly...
www.avweb.com www.avweb.com

"You won’t hear it in an airplane for a couple of reasons. First, there are no mufflers on airplanes (see below), and the high noise level masks the sound. Second, the audible “pitch” of the sound is directly related to the size of the cylinder bore, with “big-bore” aircraft engines emitting a much lower-pitched sound. That sound is far more likely to be lost in the noise of the engine itself"

Is there anything else I can help you with?

Ninja edit: I will refer you to NACA Technical Note 977 page 7

books.google.com

Technical Note - National Advisory Committee for Aeronautics

books.google.com


"in accord with common experience, which shows that engines will withstand detonation to a given intensity for a period depending on the spark plug used and the combustion chamber design."

Or is NACA also not right because no one there taught at a 147 school for 12 years? lol
 
Last edited:
@pfarber you're wasting your time trying to argue to data and facts on PoA, unfortunately. This would never stand on Beechtalk, where George would swoop in with actual data. Case in point:

Dan Thomas said:
Oh, there is consensus among the people who study it. It's been around a very long time. Tetraethyl lead was first used as an antiknock agent in 1921. 101 years ago.
Click to expand...

GAMI's testing of G100UL exposed that the "consensus" understanding of detonation in aircraft engines is flawed. For example: https://www.beechtalk.com/forums/viewtopic.php?f=37&t=133045&start=20

This happens often when 101 years of mythology comes up against actual science.
 
You must log in or register to reply here.
Back
Top