Forum made EFI?

HomeBuiltAirplanes.com

Help Support HomeBuiltAirplanes.com:

stanislavz

Well-Known Member
HBA Supporter
Joined
Aug 21, 2016
Messages
487
Location
Lt
So, the reasons we need more fuel per rev at higher rpm are:
We need more fuel, when we throttle down.

You may google for volumetric efficiency map vs torque curve. Ir vs fuel consuption/efficiency.

Just - combustion is very close on its efficiency, around aircraft used rpm range.

Unless we move throttle - throttling down, reduces compression ratio. I have seen many engines, all of them at 100% tps have close to maximum duty per rpm. And peak was on max torque pint, where we do achieve maximum volumetric efficiency.

Good point of no throttle sensor if loaded with same prop. Some temperature compensation/altitude - and all is done. Even manually.
 

Vigilant1

Well-Known Member
Lifetime Supporter
Joined
Jan 24, 2011
Messages
4,830
Location
US
We need more fuel, when we throttle down.
Yes, we generally need more fuel >per HP< when we move higher or lower than peak torque.
It appears, though, at least for a fixed-pitch prop on the O-360 in the area between 28% power 100% power, that absolute fuel required per revolution never goes up as we reduce the throttle. At least in this area, the changes in volumetric efficiency (and everything else wrapped up in BSFC) isn't enough to reverse the fuel-required-per-rev curve (though it changes the slope). In the present case, we care most about fuel-per-rev because it directly affects the pulsewidth of our injector.
 

stanislavz

Well-Known Member
HBA Supporter
Joined
Aug 21, 2016
Messages
487
Location
Lt
Yes, we generally need more fuel >per HP< when we move higher or lower than peak torque.
Around 10-20% in general , depends on how highly tuned engine. On full throttle.
it follows torque curve.

One more ad on to hp per prop rpm - it missing on more x in equation - the speed. You will need less hp at the same rpm, when you reach you cruise speed. So you still need an tps or map sensor.
 

Vigilant1

Well-Known Member
Lifetime Supporter
Joined
Jan 24, 2011
Messages
4,830
Location
US
One more ad on to hp per prop rpm - it missing on more x in equation - the speed. You will need less hp at the same rpm, when you reach you cruise speed. So you still need an tps or map sensor.
Yes, probably. Or just manual leaning in cruise (as we do with a carburetor). And a similar thing for a long descent.
Using the MAP sensor with the 555 analog EFI might be tough, we'll see. In a single cylinder runner, the MAP sensor will be in dead air and seeing ambient pressure (i.e. not dependent on throttle position) for about 75% of the time, and for at least 1 of the two injection events per intake event.
A TPS input would be simple--one sensor for all cylinders. Looking at the Lycoming graph, it's clear that the throttle will move a lot as we reduce throttle from WOT. A good addition to manual leaning and better than nothing, but it won't provide the altitude compensation we'd get from MAF.
In any case, without a real programmable EFI with a fuel map, the pilot will need to manually adjust mixture.
 

Vigilant1

Well-Known Member
Lifetime Supporter
Joined
Jan 24, 2011
Messages
4,830
Location
US
About programmable EFIs:
1) I learned that the Speeduino program has several MAP smoothing/averaging modes. If desired, you can use the "lowest MAP" reading that was seen over the preceding period. That might be an appropriate mode for getting a more accurate idea of the true charge density if you've got a "lumpy" MAP situation (e.g. single cylinder "manifold.").
2) Surely not a new idea: If programming your own EFI software, consider designing a cruise mode that automates the process of setting, say 50 degrees LOP for each cylinder. This would, in effect, be a closed-loop EFI but without need for an lambda sensor (using EGT instead). Punch the button when in cruise and the EFI adjusts the mixture to find peak EGT, then leans it to 50 deg LOP for each jug. Easy to think of, maybe harder to implement, I don't know.
 

stanislavz

Well-Known Member
HBA Supporter
Joined
Aug 21, 2016
Messages
487
Location
Lt
Yes, probably. Or just manual leaning in cruise (as we do with a carburetor). And a similar thing for a long descent.
Using the MAP sensor with the 555 analog EFI might be tough, we'll see. In a single cylinder runner, the MAP sensor will be in dead air and seeing ambient pressure (i.e. not dependent on throttle position) for about 75% of the time, and for at least 1 of the two injection events per intake event.
1592819031255.png
All mine discussion is about top few rows. Altitude compensation till 3000 ft is not necessary i think - only 10% pressure change with attitude. Ghetto style for ultralight is fully doable using only rpm and temperature sensor :) Will try source old book with 555 ic circuit variants.

 
Last edited:

Vigilant1

Well-Known Member
Lifetime Supporter
Joined
Jan 24, 2011
Messages
4,830
Location
US
View attachment 98357
All mine discussion is about top few rows. Altitude compensation till 3000 ft is not necessary i think - only 10% pressure change with attitude. Ghetto style for ultralight is fully doable using only rpm and temperature sensor :) Will try source old book with 555 ic circuit variants.
I suspect we'll live our lives pretty much in something similar to the area shown in yellow, unless the throttle is moved rapidly in or out. If we are using that B&S 810cc and are at 3600 RPM with our fixed pitch prop designed to get all the the thrust possible at that (max) RPM, the throttle will probably be pretty close to fully open. And if we are making 50% power (which requires about 2800 RPM), then the throttle will be close to half open. The throttle might be on the move and the RPM will follow close behind. ( Hopefully! )

1592875655032.png
 

stanislavz

Well-Known Member
HBA Supporter
Joined
Aug 21, 2016
Messages
487
Location
Lt

blane.c

Well-Known Member
HBA Supporter
Joined
Jun 27, 2015
Messages
3,882
Location
capital district NY
One reason for fuel injection for me personally is economy of operation. I can get less than 0.48lbs per hp per hour at around 50% power out of most carburetors. Fuel injection does not make any sense to me unless it is closer to around 1/3lbs per hp per hour at around 50% power. Without that increase in efficiency what is the point really?

From the D-Motor website; LF 26

They are advertising 16 liters fuel at 2800 RPM at PS 88.78 which converts to approximately 25.38lbs fuel at 87.53hp (imperial) and is about .29lbs per hp per hour.
 

Vigilant1

Well-Known Member
Lifetime Supporter
Joined
Jan 24, 2011
Messages
4,830
Location
US
One reason for fuel injection for me personally is economy of operation. I can get less than 0.48lbs per hp per hour at around 50% power out of most carburetors. Fuel injection does not make any sense to me unless it is closer to around 1/3lbs per hp per hour at around 50% power. Without that increase in efficiency what is the point really?

From the D-Motor website; LF 26

They are advertising 16 liters fuel at 2800 RPM at PS 88.78 which converts to approximately 25.38lbs fuel at 87.53hp (imperial) and is about .29lbs per hp per hour.
The D-Motor is liquid cooled, it's apples-to-oranges to compare it to an air-cooled engine. Liquid cooling allows tighter engine tolerances and reduces the hot spots that exist in an air cooled engine (that limit CRs, etc). A Lycoming O-235 with a carb has a BSFC of .452, and that's according to Lycoming. The factory Lycomings with fuel injection get about .40, it's likely possible to do better than that in an air cooled engine with a more sophisticated EFI (SDS, etc).

I'd be highly skeptical of any claims that the D-motor achieves .29 lb/hp/hr, even with the advantages offrred by liquid cooling. That's diesel territory. The very sophisticated and designed-for-fuel-efficiency engines like those in the Prius achieve about .37. Toyota et al would kill to get a BSFC of .29 with a spark-ignition engine, and if they could do it with an engine as simple as the D-Motor, that's what they'd put in their gas mileage champs.

Fuel injection does not make any sense to me unless it is closer to around 1/3lbs per hp per hour at around 50% power.
I think this will require a liquid cooled engine, or maybe some type of turbocompounding, etc.

In real-world ops, a sophisticated EFI will best a carburetor in fuel consumption in an air cooled engine. I suspect a less-than- sophisticated EFI setup can also beat a typical carb in fuel burn. To me, the major advantages would be the ability to individually adjust the mixture going to each jug (more important with some manifolds/engines than others) which reduces other maintenance/reliability issues. Also, potentially less fiddling around under the cowling to get things to run well, and potentially better in-flight reliability (if we count carb ice and are willing to have a near-MIL-spec EFI setup or one that is truly redundant). But, a carb also has a lot of things in its favor.
 
Last edited:

stanislavz

Well-Known Member
HBA Supporter
Joined
Aug 21, 2016
Messages
487
Location
Lt
Fuel injection will not add more economy for constant load, than carb.
But - if you want to have more economy on cruise - raise compression to 1:15, and control throttle to achieve 0.7-0.78 absolute pressure at injection manifold at any attitude. Bonus - you will have same power on any atitude.

Old motor suffers most on fuel economy due to lowe compression, not carb think..
 

Chris Matheny

Well-Known Member
HBA Supporter
Joined
Nov 26, 2019
Messages
100
Location
St. Paris OH
I think the lack of de-fueling on a Lycon type engine comes from fixed timing. When you go leaner on a mixture it needs a hotter and earlier spark to burn it efficiently. This is where the main advantage comes in on cars that operate at a low throttle opening with EFI.
 

blane.c

Well-Known Member
HBA Supporter
Joined
Jun 27, 2015
Messages
3,882
Location
capital district NY
Fuel consumption at power settings around 50% power and 0.35lbs per hp per hour in air cooled engines are realistically achievable when you consider displacement to power ratio. When the power to displacement is high like around 0.63hp per ci (0.63x49ci=31hp) then you need extra fuel for cooling and fuel consumption will be near 0.5lbs per hp per hour more or less. When power is less like around 0.37 hp per ci (0.37x49ci=18hp) then the total calories needed to burn creates much less heat and a mixture that is on the lean side of peak egt can be safely utilized, you can use excess air to cool instead of excess fuel to cool so the fuel to air ratio goes from around 12.5lbs air to a lb of fuel at high power to around 16lbs air to a lb of fuel at reduced power. So 0.5lbs fuel/16lbs air x 12.5 lbs air = 0.39lbs fuel per hp hour and 0.48lbs fuel/16lbs air x 12.5lbs air = 0.36lbs fuel per hp hour. These are carbureted numbers and fuel injection can be more precisely metered and so numbers around and about 0.35lbs per hp per hour are realistically achievable when you factor in all the decimal points.

PROOF;

From the following link scroll down past the propeller stuff somewhere between 1/3 and 1/2 way down to the article "Pelicans Perch #18: MIXTURE MAGIC"
https://www.advancedpilot.com/downloads/prep.pdf
 

blane.c

Well-Known Member
HBA Supporter
Joined
Jun 27, 2015
Messages
3,882
Location
capital district NY
The reason I wanted EFI isn't because I couldn't achieve near or the same results with a carburetor because most anyone can with a carb set up to properly lean the mixture, it is because I wanted to make it easier (less pilot workload) for multiengine aircraft. It isn't so big a deal to lean one engine but the more engines you add the more of a conductor you need to be with multiple carburetors and with EFI it can be automatic so no conductor needed. This is safer because you can spend less time on engine management and more time on flying.
 

Vigilant1

Well-Known Member
Lifetime Supporter
Joined
Jan 24, 2011
Messages
4,830
Location
US
The reason I wanted EFI isn't because I couldn't achieve near or the same results with a carburetor because most anyone can with a carb set up to properly lean the mixture, it is because I wanted to make it easier (less pilot workload) for multiengine aircraft.
If you are still thinking about using 3 industrial engines, then just getting ones with the OEM EFI system might be the best route for you. Already engineered, comparatively cheap, and nothing to fiddle with in flight. As long as you can live with their degraded/limp home modes (whatever they are), the 100LL issues, getting no more than the OEM claimed HP, and running at stoichiometric mixture, you should be good to go. The price is right, and the road is simple(r). With a triple, it's hardly a problem if one engine rolls back to 60% power (or whatever) due to loss of the O2 sensor, and even a complete shutdown isn't a tragedy. If you'll be paying the weight/cost to include 3 engines in a design, it makes sense to reap any advantages you can. Just be sure that 2 will let you climb (i.e. enough power at stoichiometric) and not overheat (can't cool with fuel) when running at the mixture dictated by the stock EFI.

ETA: Here's a listing for a Vanguard 810cc 28HP engine with EFI for $1170 USD. That engine with an EFI system costs about $200 more than an engine with a carb, it is quite a deal IF it meets the requirements of the application. To check: The listed model (49E-877) is equipped with EFI, but the listing says it has no muffler, and I don't see one in the photo. AFAIK, the B&S EFI engines always come with a muffler, since that's where the lambda sensor goes.
 
Last edited:

blane.c

Well-Known Member
HBA Supporter
Joined
Jun 27, 2015
Messages
3,882
Location
capital district NY
If you are still thinking about using 3 industrial engines, then just getting ones with the OEM EFI system might be the best route for you. Already engineered, comparatively cheap, and nothing to fiddle with in flight. As long as you can live with their degraded/limp home modes (whatever they are), the 100LL issues, getting no more than the OEM claimed HP, and running at stoichiometric mixture, you should be good to go. The price is right, and the road is simple(r). With a triple, it's hardly a problem if one engine rolls back to 60% power (or whatever) due to loss of the O2 sensor, and even a complete shutdown isn't a tragedy. If you'll be paying the weight/cost to include 3 engines in a design, it makes sense to reap any advantages you can. Just be sure that 2 will let you climb (i.e. enough power at stoichiometric) and not overheat (can't cool with fuel) when running at the mixture dictated by the stock EFI.

ETA: Here's a listing for a Vanguard 810cc 28HP engine with EFI for $1170 USD. That engine with an EFI system costs less than $200 more than an engine with a carb, it is quite a deal IF it meets the requirements of the application. To check: The listed model (49E-877) is equipped with EFI, but the listing says it has no muffler, and I don't see one in the photo. AFAIK, the B&S EFI engines always come with a muffler, since that's where the lambda sensor goes.
The thing is the stock 810cc EFI supports 28hp (each engine) and modified the engines are supposed to go to 35hp (waiting for TiPi to finish his results) and it is not clear to me if the stock 28hp injection setup will support 25% more hp properly. Maybe the way to go is to get the injection setup off the larger displacement B&S 993cc engine? The larger engine injection setup would be optimized for the increased airflow necessary for the increased power of the modified 810cc engines?
 

Vigilant1

Well-Known Member
Lifetime Supporter
Joined
Jan 24, 2011
Messages
4,830
Location
US
The thing is the stock 810cc EFI supports 28hp (each engine) and modified the engines are supposed to go to 35hp (waiting for TiPi to finish his results) and it is not clear to me if the stock 28hp injection setup will support 25% more hp properly. Maybe the way to go is to get the injection setup off the larger displacement B&S 993cc engine? The larger engine injection setup would be optimized for the increased airflow necessary for the increased power of the modified 810cc engines?
You'd have to learn more about the setup of those OEM EFI systems. My guess is that they are designed to run at/near "best economy" mixture as determined by the readings from the O2 sensor. That won't give as much HP as the "best power" mixture that Tipi will probably use in getting his higher power outputs. And so getting that additional HP will required a new fuel map and modified treatment of the O2 sensor input for any proprietary EFI you try to fit (whether the one for the 810cc or the 993 cc). Gonna take some coding skills to do that, it ain't designed for user modification.
For your application (a three engine airplane), is that extra HP above 28 worth it? Obviously, you'll have to decide for yourself, but in your boots I'd probably take a hard look at whether 28 cheap, easy HP per engine would be enough rather than develop an entirely new fuel supply system, shave heads, rework the ports and valves, etc, etc to get a few more (that, hopefully, you'll almost never use). If the plane could be made to fly safely on 56 HP while dragging a stopped prop, going with the stock EFI might be very appealing if, as you say, your goal is to reduce pilot workload. Those guys mowing the parks ain't fiddling with the mixture and worrying about EGTs. Their engines work hard and live long lives.
 
Last edited:

Vigilant1

Well-Known Member
Lifetime Supporter
Joined
Jan 24, 2011
Messages
4,830
Location
US
Analog EFI ("Again??? Still ???!!)
As described in a previous post, it looks like our "fuel per revolution" requirement will likely change over our RPM range, something like this (right column):

Lycoming O-360 RPM
RPM %
Lycoming O-360 HP (Prop Load)
% MAX HP
Lyc O-360 BSFC (Prop Load, Full Rich)
B&S 810cc RPM
B&S 810cc HP (est)
B&S 810cc GPH (at O-360 BSFC)
B&S 810cc fuel per rev (ml)
2700​
100.00%​
180​
100%​
0.55​
3600​
30.0​
2.8​
0.048​
2600​
96%​
161​
89%​
0.51​
3467​
26.8​
2.3​
0.042​
2500​
93%​
144​
80%​
0.48​
3333​
24.0​
1.9​
0.036​
2400​
89%​
128​
71%​
0.47​
3200​
21.3​
1.7​
0.033​
2300​
85%​
113​
63%​
0.47​
3067​
18.8​
1.5​
0.030​
2200​
81%​
100​
56%​
0.47​
2933​
16.7​
1.3​
0.028​
2100​
78%​
88​
49%​
0.48​
2800​
14.7​
1.2​
0.026​
2000​
74%​
73​
41%​
0.48​
2667​
12.2​
1.0​
0.023​
1900​
70%​
62​
34%​
0.53​
2533​
10.3​
0.9​
0.023​
1800​
67%​
50​
28%​
0.57​
2400​
8.3​
0.8​
0.021​

Given the relationship of RPM to HP with a fixed pitch prop, the fact that we need more fuel per revolution at higher RPMs isn't surprising. Also, the impact of varying BSFC (from the Lycoming chart) are also included in the calculations.

Here's how that change in fuel-per-rev looks when graphed against HP:

1593478033201.png It's fairly close to a linear relationship. That might be good news for anyone considering this analog EFI idea. For our fixed-pitch airplane engines, if throttle position is related to HP in a fairly direct way, then a simple potentiometer on the throttle/throttle linkage would go a long way toward keeping our mixture in the right neighborhood across the RPM/power range. The relationship needn't be linear (offset the potentiometer and use a sliding linkage to the throttle to get the approx ratio of throttle movement to electrical resistance required). Feed the resultant voltage into the 555 chip in the earlier schematic to vary injector pulsewidth with throttle position.

Does anyone happen to know about/have a graph of throttle angle/throttle opening vs power for an airplane engine turning a fixed-pitch prop?

Is using throttle position better than using MAP? Well, in the >real< (digital) EFI systems, MAP (together with RPM) is much more popular and considered much better than throttle position + RPM ("N-Alpha" logic). But, for a simple analog EFI system without a cam sensor (so no way to easily grab a MAP reading only during induction, where it counts), and given the erratic pressure in a single-cylinder induction runner, throttle position may have some advantages over MAP.
Other adjustments likely needed/desired:
-- Manual mixture knob on the instrument panel
-- For extra credit: automatic altitude/density compensation using an an absolute pressure sensor.

Again, there's no chance this analog approach could give the hands-off precision of a true digital EFI. But, maybe it would be sufficient, or a useful backup to something else.
 
Last edited:

TiPi

Well-Known Member
Log Member
Joined
Aug 25, 2019
Messages
276
Location
Mackay (AUS)
The thing is the stock 810cc EFI supports 28hp (each engine) and modified the engines are supposed to go to 35hp (waiting for TiPi to finish his results) and it is not clear to me if the stock 28hp injection setup will support 25% more hp properly. Maybe the way to go is to get the injection setup off the larger displacement B&S 993cc engine? The larger engine injection setup would be optimized for the increased airflow necessary for the increased power of the modified 810cc engines?
I had a quick look at the 49 and 61 EFI parts listings:
- The fuel injectors are the same
- The TMAP (air temp and MAP) sensor package is the same
- The fuel module (box with fuel pump & pressure regulator) is the same

Throttle bodies, air valves and control moduels are different.

Bottom line: the fuel injector and fuel pump/regulator is good for 40hp (marine version of the 61-series).
 
2
Top