Forum made EFI?

HomeBuiltAirplanes.com

Help Support HomeBuiltAirplanes.com:

TiPi

Well-Known Member
Log Member
Joined
Aug 25, 2019
Messages
292
Location
Mackay (AUS)
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.
These simple systems seem to have some basic map but rely heavily on the O2 signal under load. This means the engine is running at Lambda 1 (stochiometric) most of the time. The downside is loss of peak power and loss of economy at lower loads. The manual is vague on the operating modes.
here is a link to my B&S EFI manual and the EFI parts listings for the 49 and 61: Manuals
 

blane.c

Well-Known Member
HBA Supporter
Joined
Jun 27, 2015
Messages
3,906
Location
capital district NY
These simple systems seem to have some basic map but rely heavily on the O2 signal under load. This means the engine is running at Lambda 1 (stochiometric) most of the time. The downside is loss of peak power and loss of economy at lower loads. The manual is vague on the operating modes.
here is a link to my B&S EFI manual and the EFI parts listings for the 49 and 61: Manuals
I doubt if these or any engines are running near
stochiometric because of the simple reason that the exhaust valve normally lives in hell and stochiometric is a special part of hell reserved for operations lasting a few seconds (dragsters) or with special equipment for additional cooling (water injection) a few minutes but no matter what you do from a practical and longevity sense is a exhaust valve subjected to the blow-torch like extreme hellish temperatures of stochiometric going to last very long and neither is the top ring land of the piston and soon after much of the rest of the engine.

Engines do very well at best power 12.5lbs air per 1lb fuel and 80% power and under with 16lbs air per 1lb of fuel down to somewhere around idle range were the mixture needs to richen again. Stochiometric is 14.7lbs of air to 1lb of fuel and without elaborate precautions and constant monitoring is to normally be avoided except to transition through momentarily while leaning or enrichening.

Additionally stochiometric changes the peak pressure rise of the expanding and burning gasses to closer to the TDC Than richer or leaner mixtures causing the pressure to push through the piston, rod and crankpin at a much more straight up and down angle and further aggravating an already stressed condition.

Also in addition to that the elevated temperatures of stochiometric push the head temperatures higher than recommended and cause the aluminum to weaken at the same time you are overheating the exhaust valve and detonating the combustion chamber.

So someone tell me all the good news about running at stochiometric. I don't have any.
 

TiPi

Well-Known Member
Log Member
Joined
Aug 25, 2019
Messages
292
Location
Mackay (AUS)
I'm sorry to have to contradict you blane.c, every engine fitted with a catalytic converter is running stochiometric (lambda 1) for the vast majority of its running time (usually except at very high to full load). Otherwise the converter doesn't work.
The standard O2 sensor can only sense "rich" and "lean", to make the system work the ECU is pendling at a high frequency between "a little bit rich" and "a little bit lean" (0.98 to 1.02). Millions of cars can't be wrong :)
 

blane.c

Well-Known Member
HBA Supporter
Joined
Jun 27, 2015
Messages
3,906
Location
capital district NY
It is idiocy.

The idea that you would run stochiometric when you have alternative.

Why would you want to?

I am sure that in some 80 years since Pratt and Whitney figured out some basic parameters for running engines that the modern engineers have figured you can change the ignition timing to change were the pressure peak occurs and that with enough instrumentation feeding at nanosecond intervals into a computer you can adjust things to keep it all under control temperature wise but you are still using a pound of fuel for 14.7 pounds of air and you could be using a pound of fuel for 16 pounds of air and make the same power.
 

akwrencher

Well-Known Member
HBA Supporter
Joined
Oct 16, 2012
Messages
1,211
Location
Gustavus, AK
It is idiocy.

The idea that you would run stochiometric when you have alternative.

Why would you want to?

I am sure that in some 80 years since Pratt and Whitney figured out some basic parameters for running engines that the modern engineers have figured you can change the ignition timing to change were the pressure peak occurs and that with enough instrumentation feeding at nanosecond intervals into a computer you can adjust things to keep it all under control temperature wise but you are still using a pound of fuel for 14.7 pounds of air and you could be using a pound of fuel for 16 pounds of air and make the same power.
Don't forget that an awful lot of the tuning of modern car engines is focused on emissions control, and then economy. The two are sometimes mutually exclusive......
 

jbiplane

Well-Known Member
Joined
Jan 29, 2014
Messages
231
Location
Krasnoyarsk, Russia
Iso.jpgTop.jpg
Our company www.flame-power.com made new REAL TIME engine management units with ARM processor based on Open Sourse RUSEFI.
It so fast that can calculate ammount of fuel and ignition advance angle each RPM for 1...4 cylinder engines

CAPABILITIES

  • Primary VR or Hall input (configurable with few resistor changes)
  • Secondary Hall input
  • x4 analog thermistor (temperature) inputs
  • x10 analog voltage inputs (0-5v)
  • x4 high-Z injector outputs
  • x2 high-current low side outputs for IAC/VVT/other solenoids
  • Dedicated main relay control output
  • x4 low-current low side outputs for relays or warning lights
  • x4 5v logic-level ignition outputs
  • x2 5v/12v configurable logic level outputs (requires resistor changes)
  • Electronic throttle body (drive by wire)
  • CAN connectivity on the plug
  • USB connectivity on the plug
 

pwood66889

Well-Known Member
Joined
Feb 10, 2007
Messages
1,633
Location
Sopchoppy, Florida, USA
"It so fast that can calculate amount of fuel and ignition advance angle each RPM"
Is there an RPM range? Have 2-cyl., 4-stroke I'm working with. Спасиба!
 

Vigilant1

Well-Known Member
Lifetime Supporter
Joined
Jan 24, 2011
Messages
4,950
Location
US
View attachment 98807View attachment 98808
Our company www.flame-power.com made new REAL TIME engine management units with ARM processor based on Open Sourse RUSEFI.
Thanks, and good to know. This might be appropriate to post in the Supplier and Manufacturer Announcements area. Also:
- Could you include a link directly to the page on your site that covers this product?
- What is the product's name? Is it the same or different from the ”Flame-power 4/4" EFI?
- Price? What is included?
- I suspect folks here will be most interested in the reliability/robustness aspects, comments on that would be appreciated. Sensor failure backup modes, power and sensor surge protection, EFI shielding, etc. In an airplane, the basic "must work" stuff is critical, the VVT, multiple temp sensors, etc probably won't be important, and might even be seen as potential problems/software hangup points.
Thanks
 
Last edited:

sming

Well-Known Member
Joined
Apr 10, 2019
Messages
69
There was talks for modern diesel support on the rusefi forum, is it something you could do? I have a market for you :)
 

jbiplane

Well-Known Member
Joined
Jan 29, 2014
Messages
231
Location
Krasnoyarsk, Russia
What is the product's name? Is it the same or different from the ”Flame-power 4/4" EFI?
Flame-Power relatively mature product with 50+ sucsessfull instalations and 2 user faults. We are unable at moment provide good remote support outside of Russia due weak logistic, unclear custom rates from both sides of boundary, time shift... Non experienced users messed up ignition and injection, doing other incredible thingies. Most likely we should sell this ECU as complete kits for definite engines cutting of any user adjustments.

RusEFI is nice and strange comunity. They developed very sloooooovly, use exotic difficult to find components, sell boards with 0 profit, but charge for support 90usd per hour. Its wrong marketing. We support each customer responding each 24 hours by e-mail till promlems resolved. At moment we make only 10 test boards with some modifications. Must test intensively before undestand its potential and our abilities.

There was talks for modern diesel support on the rusefi forum, is it something you could do?
Need clear tech specs to estimate realistic time and expences and if problem itself solvable.
 
Last edited:

blane.c

Well-Known Member
HBA Supporter
Joined
Jun 27, 2015
Messages
3,906
Location
capital district NY
Well I am still thinking about the lbs air to lbs fuel thing. And it should be stupid simple. If you desire more than 80% power you need 12.5lbs air to 1lb fuel, a computer should be able to do that no problem and if you want 80% power or less then you need 16lbs air to 1lb fuel and a computer should be able to do that easy as well. The problem that I see is what should be the determination for 100% power? After that the rest is easy. I bought the Pratt and Whitney 1955 Engine book they have all manner of charts and graphs that demonstrate the two ratios and show the problems inherent with the middle ground.
 

Vigilant1

Well-Known Member
Lifetime Supporter
Joined
Jan 24, 2011
Messages
4,950
Location
US
Well I am still thinking about the lbs air to lbs fuel thing. And it should be stupid simple. If you desire more than 80% power you need 12.5lbs air to 1lb fuel, a computer should be able to do that no problem and if you want 80% power or less then you need 16lbs air to 1lb fuel and a computer should be able to do that easy as well. The problem that I see is what should be the determination for 100% power?
It seems to me that a significant challenge that might sound simple is figuring accurately how much air went in--the mass of it. And if we aren't using a fuel table, we need a formula that captures a lot of different variables-- no such formula is spot-on for real engines in the real world.
Another real-world challenge: making sure all these electrical components and wires always work, 100 percent, in the hot, shaky, EMI-rife, dirty electric power environment of an airplane.
 
Last edited:

jbiplane

Well-Known Member
Joined
Jan 29, 2014
Messages
231
Location
Krasnoyarsk, Russia
making sure all these electrical components and wires always work
All existing EFI systems are very redundant. If they dont get signal from definite sensor than use values on default. If some signal missing then remaining sensors permanently make some required corrections. I saw many projects which start as simplified and finally become very similar to Megasquirt or whatelse. The price of full featured ECU board could be about 50USD if minimal order quantiny will be 20+. If use ARM processor possible fit to 95USD. But it will not include any support exept for instalation manual.
 

AdrianS

Well-Known Member
Joined
Jul 5, 2014
Messages
517
Location
Australia
All existing EFI systems are very redundant. If they dont get signal from definite sensor than use values on default. If some signal missing then remaining sensors permanently make some required corrections. I saw many projects which start as simplified and finally become very similar to Megasquirt or whatelse. The price of full featured ECU board could be about 50USD if minimal order quantiny will be 20+. If use ARM processor possible fit to 95USD. But it will not include any support exept for instalation manual.
I like the look of that.
Can you connect a CAN wideband O2 sensor?

Once I saw megasquirt get fancier and more expensive, I toyed with the idea of building my own (for my car) around the Motorola/Freescale/NXP HC12x processor we use at work.
It's a proper robust automotive CPU : probably outdated, but with all the peripherals you need, and proper high temperature ratings. And I know it backwards.

What put me off was the amount of boring configuration management I ended up forseeing.
 

rv7charlie

Well-Known Member
Joined
Nov 17, 2014
Messages
671
Location
Jackson
Define 'cheapest'. ;-)

The arduino based Speeduino stuff is likely cheapest, but you can probably find a Microsquirt model that's pre-configured for your GM engine. Setting up and tuning typically costs many times as much (in hours) as the initial cost of hardware. The Megasquirt models are quite a bit more expensive, and likely do dozens of things you don't really need in a boat.

Don't forget that any of them will likely require a major re-work of the fuel system, if it's currently carb'd.
 
Top