Forum made EFI?

Homebuilt Aircraft & Kit Plane Forum

Help Support Homebuilt Aircraft & Kit Plane Forum:

Vigilant1

Well-Known Member
Lifetime Supporter
Joined
Jan 24, 2011
Messages
6,956
Location
US
Valery,
Thanks for the update. It seems you are concentrating on 2 stroke engines for the moment.
Looking at your previous post and video, this appears to be using the "alpha-n" type of EFI logic (I.e. only throttle position and RPM are used to choose the cell on the fuel map). Obviously, your hardware can also use the more sophisticated speed-density mode (which accounts for a lot other factors, especially air density).
It is good to know that alpha-n is available. It doesn't give super accurate fuel delivery, but if it can be easily selected in flight and if a manual mixture knob is available it could be a good backup mode in case of the failure of other sensors (air temp, MAP, etc) or other problems.
Mark
 

jbiplane

Well-Known Member
Joined
Jan 29, 2014
Messages
327
Location
Krasnoyarsk, Russia
alpha-n is available
Our Alpha-n implementation take into account air density and temperature. We as well enrich fuel mixture if cylinder head overheat.

I have tenth of engine and fuel injection ideas, but after desperate 5+ years attempts got any grant support from Russian funds or firm government orders finally we negotiate with Chinese company to launch serial production of 2 my engines and have development budget for other projects.
 
Last edited:

Vigilant1

Well-Known Member
Lifetime Supporter
Joined
Jan 24, 2011
Messages
6,956
Location
US

jbiplane

Well-Known Member
Joined
Jan 29, 2014
Messages
327
Location
Krasnoyarsk, Russia
Paramotor instalation kit prepared for tests in South Africa. All together below 1kg.
EFI KIT в ЮАР.jpg
1. Crankshaft position sensor (says of Toyota 1zz)
2. ECU fixed on throttle body on place of throttle position sensor
3. Throttle body 28mm for 24...32hp of chinese moto
4. Wire harness with connectors
5. Adapter to place kit instead of carburetor Walbro WB37
6. Fuel nozzle of 1300...1600cc car (140cc of Bosch or Siemens)
7. BLDC generator. This one I got just fot 30USD, but exist much better alternates directly on shaft, but ~100+ USD.
8. Rectifier which give 12V DC from 5...48 AC
9. Fuel pump assembly in tank
10. Smart coil Denso of Nissan or VW

industrial Capacitors
Electrolytic capacitors
We use ceramic high end capacitors for customer who ask for, it add 50USD.
Possible use processor which can work at -50..+120C, but it would add 120+ $
 
Last edited:

Armilite

Well-Known Member
Joined
Sep 5, 2011
Messages
3,643
Location
AMES, IA USA
Can this forum come up with a reasonable EFI for small engines? So software, ardino or appropriate, and common source the bits throttle body, injector, airmass flow sensor, temp, etc. for small engines. It seems a better alternative to have a fuel injection system that compensates for density altitude automatically as a normal function than trying to cob up a carburetor that kinda' works. Maybe the files for an EFI could be filed in a manner for easy access.
============================

Ecotrons or Mega Squirt. There are many EFI Throttle bodies out there if you look. SFS makes a Square Throttle Body that can be Adapted to Small Engines. Weber makes some nice ones also.

Some drawings I did for adapting a 40mm SFS Throttle body for a 277, 377, 447, 503 Intake and an old Skidoo 335 Single.
 

Attachments

Vigilant1

Well-Known Member
Lifetime Supporter
Joined
Jan 24, 2011
Messages
6,956
Location
US
============================

Ecotrons or Mega Squirt.
Ecotrons: Experienced people who have tried to get a reliable Ecotrons system working have found the hardware to be crummy and the software/tuning to be opaque and difficult. Do you know of any Ecotrons systems that are flying?

MegaSquirt: One version (MicroSquirt) is used by Aeromomentum. The hardware is not cheap and the software is quite large and complicated (with plenty of "should be explored" what-ifs that apply to aircraft but not cars). For some airplane applications, maybe it can be made to work well, but I'd want to understand its response to every type of errant sensor data (no data, erratic data, failed at high reading, failed on reading, etc) for every sensor.
To weed through all of that and to understand the hardware and software code of MegaSquirt could be quite an undertaking. It is 'fat' with all types of features (VVT, sequential ignition, closed loop operation, etc) that I wouldn't want or need in my planned aircraft installation. It's fine for "tuners' who want to experiment and can pull over if things stop working. I'm not sure it is great for an airborne system that must be bulletproof.
 

rv7charlie

Well-Known Member
Joined
Nov 17, 2014
Messages
2,154
Location
Pocahontas MS
On the other hand, with the MS stuff you have control over (and can see) all those inputs and outputs. You can buy a JimStim simulator for them to test the action of every in/out, working, failed, and disabled, on the bench, without burning an ounce of fuel.

With any engine controller not purpose-built for an aviation engine, the builder should explore every failure mode of every sensor and output (open, shorted, out of range on either end, etc etc before flying the controller. I'd want to do that even with an aviation controller, if it didn't have a long-proven track record. With adapted factory automotive controllers, all one can do is to keep bypassing known failure functions and hope (not a plan) that no more will crop up. Even the sled/watercraft controllers can have some weird unexpected 'hooks', as have been documented on some youtube videos.
 

Vigilant1

Well-Known Member
Lifetime Supporter
Joined
Jan 24, 2011
Messages
6,956
Location
US
On the other hand, with the MS stuff you have control over (and can see) all those inputs and outputs. You can buy a JimStim simulator for them to test the action of every in/out, working, failed, and disabled, on the bench, without burning an ounce of fuel.
True. And a person could, in this way, in theory, try out every one of the hundreds of combinations of relevant sensor anomalies. They could also download the software and even modify it to better meet aircraft needs. But, they can't buy or build their own hardware (to better meet aviation requirements) without risking a lawsuit. And the software is very fat feature-laden, which drives up complexity and hardware requirements.

Still, the MS route is probably easier than starting with a OEM ECU with locked-down/opaque software and unknown "limp" modes, optimized for emissions, EPA, and CAFE requirements, etc.

The Speeduino world isn't perfect, but the software and hardware is entirely open so experimenters can optimize both for their requirements. Jbiplane is working in that world now. If his project bears fruit and he can bring it in at his presently estimated price, then his system (ECU, sensors, fuel pump, throttle body, fuel pump, injectors, harness, etc) might be about 1/2 the price of the MS 3 EMS alone ($645, assembled). It's a tough challenge (esp with regard to reliability and ease of use), but I hope he can do it.
 
Last edited:

rv7charlie

Well-Known Member
Joined
Nov 17, 2014
Messages
2,154
Location
Pocahontas MS
The hardest thing for me is mentally getting past using a consumer grade Arduino for the core processor, instead of a purpose-designed engine control processor as used in the MS stuff. Engine control is a hostile electrical environment anyway, and various other a/c systems can make it even more hostile.

BTW, the Microsquirt seems to be about half the price of the cheapest Mega stuff, and is less 'feature rich' (simpler).
 

jbiplane

Well-Known Member
Joined
Jan 29, 2014
Messages
327
Location
Krasnoyarsk, Russia
consumer grade Arduino
Its very big difference to use Arduino or only its processor atmega2560
Processors we tested on our boards:
- atmega2560-16cu
- MC9S12C64 like used on Megasquirt or Microsquirt
- STM32F405xx STM32F407xx (10x10mm or 4.223x3.969 mm footprint) - still under testing
- military grade Russian processor Milandr (expensive and need export licensing)

We dont feel any big diference for end user if outside themperature is in -28...+50 range
The more important is processor supplier.
More than 50% is chinese counterfeight or "secord hand" and you dont sequre even if you buy in proven shops.
Motorolla chips long time on market and you can buy MC9S12C64 for 4 and 25USD and there are not any varancy that 25USD unit not the same as 4...
Even big companies like Japan OKI now often got in this trap. For Atmega and STM counterfeight possibilities much less.
We tested Atmega stuff with Speeduino firmware in thermobarocamera and awaiting when Speeduino will completelly support STM32.

We dis as well project of electronic carburetor which regulate main jet by fuzzy logic algorithms.
If all electronics go down it will continue work in carburetor mode. But this will be long time project cause we have 0 money on this.

Now we concentrate on small engines :)


Related ECU unit
At the moment our ECU is very universal. It can be integrated or not integrated to throttle body.
It is up to user which contacts use, which skip. Only "ordered" wires will be in round connector.
 

Attachments

Last edited:

Armilite

Well-Known Member
Joined
Sep 5, 2011
Messages
3,643
Location
AMES, IA USA
Ecotrons: Experienced people who have tried to get a reliable Ecotrons system working have found the hardware to be crummy and the software/tuning to be opaque and difficult. Do you know of any Ecotrons systems that are flying?

MegaSquirt: One version (MicroSquirt) is used by Aeromomentum. The hardware is not cheap and the software is quite large and complicated (with plenty of "should be explored" what-ifs that apply to aircraft but not cars). For some airplane applications, maybe it can be made to work well, but I'd want to understand its response to every type of errant sensor data (no data, erratic data, failed at high reading, failed on reading, etc) for every sensor.
To weed through all of that and to understand the hardware and software code of MegaSquirt could be quite an undertaking. It is 'fat' with all types of features (VVT, sequential ignition, closed loop operation, etc) that I wouldn't want or need in my planned aircraft installation. It's fine for "tuners' who want to experiment and can pull over if things stop working. I'm not sure it is great for an airborne system that must be bulletproof.
============================================

It's not my Job to keep track of what 8 Billion people in the World are doing let alone the % of them that are using EFI. If you can't afford it, or can't understand how to use it, then use the standard Carb setup.

Ecotron EFI kits have been around now for about 8+ years. They offer a nice kit and have good documentation on their EFI. Since most of the people on here don't even take the Time to learn How to Build their own Engines, I seriously doubt many take the Time to really learn about installing EFI & Learning How to Tune it, yourself included. It's not as Cheap today. Can you LOOK at the photos and probably put together a similar System much cheaper, Yes.

Yes, you have to be smarter than an 8th grader to Read & Understand the EFI Info. You also have to be able to Soldier and READ the Electronic Diagrams if you want to build your own ECU. You can Pay more for premade ECU's. You can be resourceful and get many EFI parts, off Bikes, Scooters, Sleds, Jet Skies, ATV's which can also save you money. You might actually have to do some research and READ that info.

Hooking Ecotrons kit to just about any Engine is the same. There are many Videos out on How Too!

Fuel Injected Rotax 503 on a Plane.

Ecotrons EFI Kit on Briggs & Stratton Junior 206 engine.

35 hp Briggs and Stratton with ECOTRONS EFI.

Install Ecotrons EFI kit on a Honda GX390 engine.

ECOTRONS EFI Tuning Software ProCAL Video Tutorial.avi
 
Top