Please satisfy my curiosity...

david0tey

david0tey

Line Up and Wait
Joined
Mar 2, 2012
Messages
545
Location
Virginia
Display Name
Display name:
Fox-Three
I don't know why I thought of this but it has been bugging me recently. Why don't cars have any way of controlling the mixture manually? I realize that altitude changes make the availability of mixture control in an airplane essential, but cars change altitude too (specifically in mountainous terrain). That may be a stupid question, but that's why I asked it on the internet, and not to my CFI.:redface:
 
david0tey said:
I don't know why I thought of this but it has been bugging me recently. Why don't cars have any way of controlling the mixture manually? I realize that altitude changes make the availability of mixture control in an airplane essential, but cars change altitude too (specifically in mountainous terrain). That may be a stupid question, but that's why I asked it on the internet, and not to my CFI.:redface:
Click to expand...

Back in the '50s, my parents took us kids cross country in the car. The car was way too rich in the mountains 'cause we didn't have "high altitude jets" installed in the carburetor. Today, the cars have computers and air mass measuring devices to make sure the mixture is appropriate.

OK, our plane engines are 1950s technology. The modern cars take care of this automatically so the driver can remain dumb and happy.

-Skip
 
Modern cars have electronic ignition and fuel injection that is computer controlled that makes those adjustments for you.
 
In that case, it's interesting that the newer model piston engines in airplanes don't function that way.
 
Skip Miller said:
Back in the '50s, my parents took us kids cross country in the car. The car was way too rich in the mountains 'cause we didn't have "high altitude jets" installed in the carburetor. Today, the cars have computers and air mass measuring devices to make sure the mixture is appropriate.

OK, our plane engines are 1950s technology. The modern cars take care of this automatically so the driver can remain dumb and happy.

-Skip
Click to expand...

Basically, modern cars have FADEC. A computer controls it all.
 
david0tey said:
I don't know why I thought of this but it has been bugging me recently. Why don't cars have any way of controlling the mixture manually? I realize that altitude changes make the availability of mixture control in an airplane essential, but cars change altitude too (specifically in mountainous terrain). That may be a stupid question, but that's why I asked it on the internet, and not to my CFI.:redface:
Click to expand...

Modern fuel injected auto engines do change mixture, it's just not done by the driver which the EPA would never allow. The computer is adjusting fuel mixture, ignition timing, and who knows what else based on crank position, cam position, inlet air temp, air density, etc. etc.
 
Rotax ULS doen't have a mixture control. Bing Carbs adjust automagically. 912ils adds fuel injection which allows a 100 HP engine to run on 3.5 gph. Older engine manufactures just haven't caught up to the latest technology yet.
 
There is an app for that. Bluetooth dangle for the OBDII port and a phone app, you set your mixture however you want. And the hotrod guys used to say you will never be able to tune them 'puter motors.
 
david0tey said:
In that case, it's interesting that the newer model piston engines in airplanes don't function that way.
Click to expand...

The reason is because of the high cost of getting anything certified for a factory built airplane. Electronic ignition has been around for years and you can put it on a few certified airplanes with a STC. The Porsche Mooney had a true fadec controlled engine with a single lever throttle, mixture, prop control. There are several fadec engines and add on systems for Lycomings and Continentals for the experimental side of aviation. I'm building a Zenith 750 for a friend and it has a UL 350 engine with fuel injection and electronic ignition fadec control. Don
 
deltafox said:
Rotax ULS don't have a mixture control. Bing Carbs adjust automagically.
Click to expand...
I've heard that, but studying the carburetor itself, I don't see it. It looks no different to me than any other constant vacuum carburetor.

A modern car will never have a manual mixture control. I spent much of my career working on automobile air fuel control - getting the right air fuel is critical for catalyst efficiency and emission controls. No human being would ever be able to keep up with what needs to be done.

There is a strong resistance to that sort of thing in aircraft. This topic has come up more than once and many pilots have stated that they would not trust any sort of electronic fuel injection - anything electronic is not reliable enough for an airplane. They also claim that they can do a better job of controlling the mixture. I have yet to see data to support that claim.

I assume that those same pilots would never ride in a passenger airliner because of all the electronics that are in control of the engines and flight control surfaces.
 
snoboy said:
Modern fuel injected auto engines do change mixture, it's just not done by the driver which the EPA would never allow. The computer is adjusting fuel mixture, ignition timing, and who knows what else based on crank position, cam position, inlet air temp, air density, etc. etc.
Click to expand...
beyond that it's constantly sweeping mixture rich-lean of stoich, back and forth. Your car never has a static "mixture". If it did the 3-way cat wouldn't work.
 
A big difference between car engines and airplane engines is a car engine is constantly changing power, rpm, manifold pressure, etc... whereas an airplane engine more or less runs at one setting all day long. FADEC controls in piston airplanes would be great but they don't buy you all that much.
 
It's because as pilots we like knobs and switches, the more the better.
 
Capt. Geoffrey Thorpe said:
They also claim that they can do a better job of controlling the mixture. I have yet to see data to support that claim.
Click to expand...

I'm an EE, and also a very cheap sort of guy. So, the combo led me to investigate the BSFC of the various mixtures I was running. The EGT kit allows me to see basically the same as a O2 sensor on a car. In fact, since I have a 6 probe EGT, it could be said that I have an even greater advantage since I can see the products of combustion efficiency on each cylinder.

Now, I'm prolly one of those guys that thinks he can do better than the computer when it comes to mixture control. I have the BSFC curves for my engine, and found where the factory(back in the 40s) set their data points, and where the best efficiency is found.

Now, based on three flights where I very carefully calculated the fuel consumed by replacement, and the engine run times, and the RPM, and MP for this engine, I actually worked at a lower BSFC by controlling the mixture than the factory numbers indicate. Of course, the engine was running lean of peak, and that I had the plane barely running smooth but yes I think I beat the system.

Now we have nothing to measure against because I don't know of anyone who has a FADEC system working on my engine. Could they do better? Well, what is 'better'? When I need full rich for take off, I select full rich. When I start into cruise, I lean carefully to maximize efficiency. When I'm in cruise, I run lean of peak.

The only thing I'm missing is a detonation gauge. But then again, a single detonation event can be pretty bad on a plane engine, so I don't want to operate that close to the detonation margin. I would think, but don't know that any kind of computer driven aviation mixture would take into account to avoid detonation margin by a significant amount. If that's the case, I can see where I could do as well, or better than an automated system.
 
And that knock sensor is a critical part of the spark timing management. Without it, you cannot do better. Period. Because your airplane has no way to retard timing and no knock sensor, you have no chance of arresting detonation or even knowing it's going on. If the engine is rough, it might be, but it's hard to distinguish that from a lean miss. EGT would be a hint. In automotive engines, mixture oscillates lean/rich several times per second to keep cylinder temperatures under control while still maintaining good power (and to feed both beds in the cat). Spark timing is a critical part of this; it is advanced to the detonation limit, detected with a knock sensor, then retarded slightly.

As an EE, you must understand the relation between complexity and reliability. Automatic systems are always more complex than equivalent manual systems. Automotive engines are ridiculously complex, and despite continual mass production on an enormous scale for 100 years, they still get serviced under TSBs and occasionally safety recalls with astonishing regularity. That's the reason it's done manually. Not to "beat the computer."
 
Average kid now cant even drive stick, you really want to give him mixture control? I can smell the burnt valves from here!
 
MAKG1 said:
And that knock sensor is a critical part of the spark timing management. Without it, you cannot do better. Period. Because your airplane has no way to retard timing and no knock sensor, you have no chance of arresting detonation or even knowing it's going on. If the engine is rough, it might be, but it's hard to distinguish that from a lean miss. EGT would be a hint. In automotive engines, mixture oscillates lean/rich several times per second to keep cylinder temperatures under control while still maintaining good power (and to feed both beds in the cat). Spark timing is a critical part of this; it is advanced to the detonation limit, detected with a knock sensor, then retarded slightly.

As an EE, you must understand the relation between complexity and reliability. Automatic systems are always more complex than equivalent manual systems. Automotive engines are ridiculously complex, and despite continual mass production on an enormous scale for 100 years, they still get serviced under TSBs and occasionally safety recalls with astonishing regularity. That's the reason it's done manually. Not to "beat the computer."
Click to expand...

Actually, as I said I have already 'beaten' the factory, and there is no computer controlled piston engine to compare against. I also mentioned that since I don't have knock sensors, that I was limited to purely mixture control. If there ever is a computer management of mixture and timing with a knock sensor and closed loop control, I don't know if I could beat it, but I know I'd get real close. The aviation management computer is going to be calibrated to avoid detonation at all costs, and that is not the case in the auto world. Due to the low ave power delivery of the auto engine, they can allow a limited amount of knock/detonation before retarding the spark or increasing the mixture or both.
 
Same reason cars don't have ailerons or rudders.

Cars are not planes. Sure you sit in them and can turn left and right...but that's about the end of similarities.

Why don't we do mag checks in a car? Why don't we use cross controls when driving in a car? Why don't we check the oil ever drive in a car? Why don't we pre-drive a car? Why doesn't the car have a log book? Why doesn't the FAA care about my car? Why are my car's tires so cheap? Why doesn't my care have a compass? Why don't we have to do a VOR check in my car? Why doesn't my car have any VORs?
 
My old Gopher engine in the Navion had an automatic mixture control on the carb. You never touched it below 8000' or so. Little aneroid on the side of the carb.

I remember disabling the automatic choke on my old Ford 302 V8. Darn thing kept sticking. I connected it to the knob marked "RIGHT AIR".
 
I flew the Mooney PFM and liked it.
Came close to ordering one but I'm too darn cheap to ever spend that much on a plane.
I was sad that it did not work out in the market place.
 
docmirror said:
I'm an EE, and also a very cheap sort of guy. So, the combo led me to investigate the BSFC of the various mixtures I was running. The EGT kit allows me to see basically the same as a O2 sensor on a car. In fact, since I have a 6 probe EGT, it could be said that I have an even greater advantage since I can see the products of combustion efficiency on each cylinder.

Now, I'm prolly one of those guys that thinks he can do better than the computer when it comes to mixture control. I have the BSFC curves for my engine, and found where the factory(back in the 40s) set their data points, and where the best efficiency is found.

Now, based on three flights where I very carefully calculated the fuel consumed by replacement, and the engine run times, and the RPM, and MP for this engine, I actually worked at a lower BSFC by controlling the mixture than the factory numbers indicate. Of course, the engine was running lean of peak, and that I had the plane barely running smooth but yes I think I beat the system.

Now we have nothing to measure against because I don't know of anyone who has a FADEC system working on my engine. Could they do better? Well, what is 'better'? When I need full rich for take off, I select full rich. When I start into cruise, I lean carefully to maximize efficiency. When I'm in cruise, I run lean of peak.

The only thing I'm missing is a detonation gauge. But then again, a single detonation event can be pretty bad on a plane engine, so I don't want to operate that close to the detonation margin. I would think, but don't know that any kind of computer driven aviation mixture would take into account to avoid detonation margin by a significant amount. If that's the case, I can see where I could do as well, or better than an automated system.
Click to expand...

If you are working from dyno data, then that puts you worlds ahead of what a lot of pilots seem to be doing (based on what they post). But I would like to point out that some pilots compare fuel consumption at constant manifold pressure / engine speed without correcting for the reduction in actual crankshaft delivered power as you go lean. It makes the perceived benefit a lot larger than actual.

To actually be better than a well designed automatic system a pilot would need to either
a) Have information not available to the automatic system (and the engineer that designed it),
b) have a control not available to the automatic system, or
c) violate limits built into the design of the automatic system.
​
If I had a system similar to what is found in todays automobiles (with EGT in place of O2 sensors), I could do things like trim each cylinder individually and adjust the timing to match the actual air fuel ratio. But that wouldn't exactly result in ground shaking results. The typical intake / cylinder / ignition designs of most aircraft engines are not optimized for running lean so the potential is limited. The big BSFC benefit comes as you move from pig rich to peak - beyond that you are looking at about 2% - 3% at best. Then there is the issue of limited pressure at altitude which means that you have no choice but to reduce the actual power delivered as you go lean - which is good from a total fuel consumption perspective but does slow you down.

The bottom line, from a practical standpoint, if you aggressively lean during cruise, and lean to max power during climb (except as required for cooling) based on actual dynamometer data, then you are going to get purd near the same results as a well designed automatic system. Even just pulling the red knob as far as you can without running rough during cruise gets you most of the way there.
 
docmirror said:
I don't know if I could beat it, but I know I'd get real close.
Click to expand...

I think with care you could get pretty close, but I doubt you could beat it - any algorithm you use could just as easily be designed by an engineer. Also, you'd have to recheck your measurements as the engine ages, while a computer would be monitoring current conditions. That chip doesn't have to scan for traffic - it is just doing nothing but monitoring engine performance data.

I was reading up on my car - it isn't expensive but has direct injection and apparently it wildly varies the mixture any time you move the pedal. It tends to run rich under load, and run super-lean on steady cruise. The pedal almost certainly just operates a transducer and not a throttle valve, so the computer knows how much power is requested before the engine knows - the injectors don't do anything that the computer hasn't already figured out already. It certainly can be done, but of course aircraft do not have the same production volumes, and you do need appropriate levels of safety/etc especially with only one engine. I'm sure a modern jet (which burns which volumes of fuel) is engineered to be as efficient as a modern car - they're already crippled without electrical.
 
Dr. O said:
I flew the Mooney PFM and liked it.
Came close to ordering one but I'm too darn cheap to ever spend that much on a plane.
I was sad that it did not work out in the market place.
Click to expand...

Hey, I forgot about the PFM! Now there - that was a gas hog. As I recall it used significantly more fuel than the comparable Lycosaurus.
 
You must log in or register to reply here.
Back
Top