Weight of EFI

Homebuilt Aircraft & Kit Plane Forum

Help Support Homebuilt Aircraft & Kit Plane Forum:

RJW

Well-Known Member
Joined
Feb 9, 2011
Messages
710
Location
Wisconsin and Kansas
Electronic fuel injection versus carburetion—there are innumerable arguments concluding one or the other is the “best” way to go for an aero engine. But one argument I’ve not seen is one comparing the weight of the two types. Any opinions? The reason I ask is because I’m converting a V6 from EFI to a carbureted motor (don’t ask why). I noticed while removing all the EFI stuff and chopping up the EFI intake manifold, that I had quite a big heavy pile of stuff on the ground when I finished. The pile didn’t include most of the wiring or the computer. I didn’t weigh all this stuff but it occurred to me that the EFI installation could be substantially heavier than a simple 2-barrel carb setup. Thoughts? Experience?

Rob
 

Hot Wings

Grumpy Cynic
HBA Supporter
Log Member
Joined
Nov 14, 2009
Messages
7,984
Location
Rocky Mountains
It's hard to make comparisons on weight since there can be considerable overlap due to the complexity of each system and how well it's packaged. A single updraft carburetor and and old automotive distributor system is naturally going to weigh less than a modern EFI with dual exhaust sensors and integrated coil on plug ignition. Swap a couple of heavy magnetos into this comparison and there may not be too much difference.

If you are looking for absolute minimum weight something like a single throttle body injector with a waste spark ignition would be my starting point. Component count and part/wiring weight could be minimal. The complete injection/ignition system on a Subaru Robin industrial engine weighs less than 2 pounds more than the carburated version. Other than the size of the components needed to deliver the volume for the Hp requirements of an aircraft engine an EFI system could be this simple/light and function. It would not be my first choice for an aircraft system due to failure modes without some redundancy.
 

nerobro

Well-Known Member
Log Member
Joined
Nov 6, 2011
Messages
1,114
Location
Northern Illinois
A gravity fed carburetor will almost always be lighter than a fuel injection system. the on engine weight might be reduced, if you go with a plastic intake manifold, and that sort of thing. But once you add the ancillary systems, fuel injection will be heavier. (pumps, computer, wiring harness, sensors..)
 

Jay Kempf

Curmudgeon in Training (CIT)
Lifetime Supporter
Joined
Apr 13, 2009
Messages
4,352
Location
Warren, VT USA
A gravity fed carburetor will almost always be lighter than a fuel injection system. the on engine weight might be reduced, if you go with a plastic intake manifold, and that sort of thing. But once you add the ancillary systems, fuel injection will be heavier. (pumps, computer, wiring harness, sensors..)
Normally if you build a custom tubular aluminum intake it's a wash over stock cast one piece manifolds and all that heavy auto stuff. A simple MAF, tubular aluminum manifold (which is more efficient anyway), ford mustang injector and a rudimentary harness and you have it. There is no real big difference in pumps, carb vs MAF and throttle body, etc... The new mini version of the megasquirt computer is a few ounces.
 

rv6ejguy

Well-Known Member
HBA Supporter
Joined
Jun 26, 2012
Messages
4,574
Location
Calgary, Alberta, Canada
Many variables to compare. A big Marvel carb is a heavy beast as are some magnetos. Many automotive EFI intakes are way heavy and can be replaced by a lightweight 6061 tubing design saving several pounds in some cases.

Here are some actual component weights:

SDS ECU 1 lb.
Walbro pump 1 lb.
Throttle body 12-17oz
Wiring harness 1-1.3 lbs depending on lengths
Regulator 5-8 oz
MAP and temp sensors 4 oz
Pico injectors and bosses 4 oz/ea
Hall Effect sensor 2 oz
4 Cyl. coil pack(Subaru EJ) 2 lbs.
Fuel lines/ rails 1 lb. (4 cyl.)

We find on certified engines, the EFI often weighs a bit less than the big Marvel carb, mechanical pump and mags.
On the Rotax 912ULS, the EFI is also lighter than the 2 Bing carbs, S airbox and mechanical fuel pump.

OEM auto EFI stuff is often very heavy.

Also remember than the EFI is likely to cut the fuel burn so you'd be able to carry less weight in fuel. On the Jabiru we see about a 10+% reduction in fuel flow at the same TAS, Rotax 5-8% due to the much improved mixture distribution. On an O-200, people are seeing 15-20% reductions in FF because the carb/ intake is so poor on these engines.
 
Last edited:

Autodidact

Well-Known Member
Joined
Oct 21, 2009
Messages
4,513
Location
Oklahoma
It seems to me that EFI is dependent on power source for electricity for it's ability to operate and so for redundancy (and therefore a high level of reliability) you would need two alternators, two fuel pumps and two controller computers. Can the digital ignition/fuel delivery still be lighter with two of everything? It would seem that it could, but in practice? Two alternators at the very least, because if you loose your current, then nothing will work, and a battery isn't going to keep it going very long with the need for all of that electricity. At least a points type distributor(s) will keep running for quite a while as long as there is a mechanical fuel pump (or gravity feed) and fuel metering. I'm not against electronic engine control, I just like to see high levels of redundancy and it seems like carbs and distributors are easier to do that with. On the other hand, electrical devices are more reliable than they have ever been, but two completely separate electrical systems would almost be failsafe - that would be nice...
 

Vigilant1

Well-Known Member
Lifetime Supporter
Joined
Jan 24, 2011
Messages
6,834
Location
US
. . . and a battery isn't going to keep it going very long with the need for all of that electricity.
IIRC, a reasonably-sized battery will keep the typical EFI and ignition going for a considerable period of time. Not long enough for long overwater flights, but 45 minutes is do-able.
It's also possible to integrate a carb with EFI for redundancy purposes. Below is a picture (from EAA Experimenter) of a RevFlow carburetor fitted with a throttle position sensor and injector nozzle.



The above photo is of an experimental version of the RevFlow we found on Joe’s desk the day we conducted our interview. This unit has been fitted with an electronic fuel injection nozzle and a throttle position sensor. When mated with an oxygen sensor in the exhaust system and an engine control unit (ECU,) optimal mixture can be established automatically. The real beauty of this system is that in the event of a computer malfunction or other electrical issues, if the carb is plumbed to gravity feed (in addition to the high pressure needed by the injector), opening the fuel feed line to the carb will get the engine running again. This carb can also be fitted to most any multi-port fuel injection system (MP-EFI,) functioning as the throttle body. If connected to gravity feed as described above, it too can act as a backup for the MP-EFI system in the event of a failure.
 

rv6ejguy

Well-Known Member
HBA Supporter
Joined
Jun 26, 2012
Messages
4,574
Location
Calgary, Alberta, Canada
The standard 25 amp/hr battery will usually provide at least 90 minutes of running the pump, ECU, coils and injectors before falling below a critical 8-9 volts. You are looking at around 10 amps collectively here. A backup battery around 12 amps hours would give you close to another hour at about 8 lbs. weight penalty. If you are running a Denso alternator, it's very doubtful in the first place that it would fail. I've never seen one fail in 35 years of working with them in the auto repair business unless the brushes wore down to nothing- at something over 5000 hours.

A Rotax 912 only has one fuel pump and a Jabiru only has one carb but people don't seem to worry much about that. Carbs have no redundancy either. We do recommend twin fuel pumps for most applications but again, the Walbro pumps we use are very reliable. I've sold hundreds of them and never had one sent back yet.

Twin ECUs? They simply don't fail (ours at least) unless they get wet. The MTBF is far higher than the engine they are attached to. We have 20 years and over 100,000 flight hours to back that statement up with along with 24 million hours on ground based applications. One of our bench test ECUs has run continuously for over 135,000 hours now.

For the unconvinced, you can retain the stock carb as a backup, just leave it in idle cutoff and retain one mag as well.
 
Last edited:

Autodidact

Well-Known Member
Joined
Oct 21, 2009
Messages
4,513
Location
Oklahoma
The standard 25 amp/hr battery will usually provide at least 90 minutes of running the pump, ECU, coils and injectors before falling below a critical 8-9 volts.
What kind of fuel savings do you get with EFI?
 
Top