Quantcast

Adding redundant fuel and spark to auto/sled/motorcycle conversions.

HomeBuiltAirplanes.com

Help Support HomeBuiltAirplanes.com:

Vigilant1

Well-Known Member
Lifetime Supporter
Joined
Jan 24, 2011
Messages
5,365
Location
US
Why not just mount the ecu in the cockpit, if it goes tits up swap it for a new one?
In the cases above, the ECU isn't malfunctioning, it is doing what it was programmed to do. A different ECU box would react in the same way.
The limp-home mode in factory ECUs is designed to protect the engine when a sensor input is out of limits or missing. In a car, it is an inconvenience. In an airplane, the available power under limp-home mode might not be enough to remain airborne. So, maybe it should be called "limp into the trees" mode in our case.
 
Last edited:

blane.c

Well-Known Member
HBA Supporter
Joined
Jun 27, 2015
Messages
4,152
Location
capital district NY
Could the ECU be bypassed or reprogrammed so the engine is provided with enough fuel for say 75% power and the pilot places the throttle control in the correct spot and monitors temperatures (more or less air required)?
 

Vigilant1

Well-Known Member
Lifetime Supporter
Joined
Jan 24, 2011
Messages
5,365
Location
US
Could the ECU be bypassed or reprogrammed so the engine is provided with enough fuel for say 75% ....?
I dunno. The OEM versions are specifically designed to not be easily reprogrammed by others. Then, to modify the programming you'd have to understand it. I can barely debug programs I understand and wrote myself, modifying the work of another team isn't something I'd want to take on.
Seems easier and better to me to put in a stone-simple but maybe not as efficient backup system (sensors, computer, injector(s). Then I'm covered regardless of the reason the main system quit (sensors, ECU hardware, ECU software, wiring).
Check them both during engine runup, then run on the deluxe one until it quits, which could be a very long time. If I can't even assess the likelihood of a "main" EFI/EI problem and how it will manifest itself, but if I can build a "good enough" backup for a few hundred dollars and a weight of a few pounds, that sounds useful.
 
Last edited:

blane.c

Well-Known Member
HBA Supporter
Joined
Jun 27, 2015
Messages
4,152
Location
capital district NY
Isn't this were the 555 chip came into the discussion before? It was written basically that a person had a "drip valve" set up so it could be turned on in the intake and controlled in the cockpit for "drip" and then a rudimentary ignition system fired by a 555 timer?
 

Vigilant1

Well-Known Member
Lifetime Supporter
Joined
Jan 24, 2011
Messages
5,365
Location
US
Isn't this were the 555 chip came into the discussion before? It was written basically that a person had a "drip valve" set up so it could be turned on in the intake and controlled in the cockpit for "drip" and then a rudimentary ignition system fired by a 555 timer?
No, it is the other way around. The 555 chip is for a dirt simple EFI system, it doesn't provide spark. See post 49 in this thread, and the links there.
 

Vigilant1

Well-Known Member
Lifetime Supporter
Joined
Jan 24, 2011
Messages
5,365
Location
US
We are testing solution which manages 2 separate third side ECU including Magasquirt 2 extra.
Thanks, that will be interesting to see. I'm surprised you are using Megasquirt instead of the alternative less expensive ECU that you'd found. When you are far enough along, please share more information on your approach (for instance: can the two ECUs be powered independently, do they share injectors or sensors, noise shielding you've tested, how does the switchover from one ECU to the other occur, and could failure of the components used to do the switch lead to failure of the ECUs, etc).
Let us know more as your development continues.
 

pfarber

Well-Known Member
HBA Supporter
Joined
Feb 21, 2019
Messages
611
Location
Pennsylvania
My only concern about an auto ECU is failure modes - eg. If it's getting a bit warm, you don't want it to suddenly go into limp mode.
Assuming issues like that can be solved in the ECU programming, I think it will be fine.

Note: Several recent BMW owners have foung they go into limp mode after a couple of fast laps, and I've had an Evo shutdown the gearbox on me and give nothing but neutral, so it can happen.
ECUs are mounted in the engine bay.. lots of heat/vibration. Have you seen an automotive ECU? These are hardened, sealed, industrial units. How hot does it get on your airplane?
 

pfarber

Well-Known Member
HBA Supporter
Joined
Feb 21, 2019
Messages
611
Location
Pennsylvania
Why not just mount the ecu in the cockpit, if it goes tits up swap it for a new one?
The connectors have dozens of small pins. Maybe if you went from a OEM standard plastic connector to an amphenol type connector that would be better.
 

AdrianS

Well-Known Member
Joined
Jul 5, 2014
Messages
572
Location
Australia
ECUs are mounted in the engine bay.. lots of heat/vibration. Have you seen an automotive ECU? These are hardened, sealed, industrial units. How hot does it get on your airplane?
I'm not talking about ECU temps (obviously), I'm referring to the ECU going into limp mode due to oil / coolant temp.

I don't know why you're so aggressive: all modern auto ECUs will go into limp mode under certain circumstances, to protect the drivetrain. Those limits are pretty conservative.
Being aware of the parameters that can trigger limp mode is pretty important.

I'm not saying you need redundancy, but you do need to know what could make the fan stop turning.
 

rv7charlie

Well-Known Member
Joined
Nov 17, 2014
Messages
1,016
Location
Jackson
Could the ECU be bypassed or reprogrammed so the engine is provided with enough fuel for say 75% power and the pilot places the throttle control in the correct spot and monitors temperatures (more or less air required)?
Bluntly, no.
Modern car ecus' firmware is so complex and intertwined with the rest of the car, there's no way to know if you found all the ' protection' modes. If one kills you, you wouldn't be the first.
 

pfarber

Well-Known Member
HBA Supporter
Joined
Feb 21, 2019
Messages
611
Location
Pennsylvania
I'm not talking about ECU temps (obviously), I'm referring to the ECU going into limp mode due to oil / coolant temp.

I don't know why you're so aggressive: all modern auto ECUs will go into limp mode under certain circumstances, to protect the drivetrain. Those limits are pretty conservative.
Being aware of the parameters that can trigger limp mode is pretty important.

I'm not saying you need redundancy, but you do need to know what could make the fan stop turning.
I think you are chasing your tail. You want to override a critical motor incident why?

Using an off the shelf ECU from a production motor, with the transmission set to manual (thereby removing all need for speed, lock up, temp and other transmission issues) solves those issues.

Security can be disabled.

Most cars use a power/fuse center in the engine bay, but most of the required circuits (like starter and fuel pump, cooling fans etc) can be bypassed taking that out of the equation.

VVT can be disabled, so those sensors are gone.

The only thing left that you might need to fake out (and again, some of might be configurable in software) are emissions (fuel tank pressure, fuel vapor purge solenoid). This is done by using a 'start box' to simulate these signals.

The only thing left are engine operational sensors... CPS/TPS/spark/O2, temps and pressures. All of these are pretty vital to the operation of the engine.

If you are looking for an 'emergency override' then you can add a small box to simulate just about every one of those sensors save the CPS.

I get that there is no list of 'what causes limp home mode'. But my position is that if you shouldn't ignore them to begin with. I mean no oil is bad. Oil over temp is bad. Many things will stop the motor from working. Its up to each person to decide whether you want a complete, catastrophic event or a pre-catastrophic shut down.

I'd think a 'pre-shutdown' is safer then snapping a crank, losing a rod or valve etc. Something that could put oil and fluid in a position to ignite or cover the windscreen etc.

Honestly, how many times have you had a car go into ACTUAL limp home mode? Not just a CEL, but an actual limited operation? I've never seen it in 20 years of driving.
 

pfarber

Well-Known Member
HBA Supporter
Joined
Feb 21, 2019
Messages
611
Location
Pennsylvania
Bluntly, no.
Modern car ecus' firmware is so complex and intertwined with the rest of the car, there's no way to know if you found all the ' protection' modes. If one kills you, you wouldn't be the first.
Not really. The ECU talks over the CAN bus or whatever network is used, but they are not critical to the ECUs operation. The ECU will send messages like RPM and speed, temps and data to the instrument cluster, and the BCM would talk to the ECU to increase RPM for things like AC on.. I don't believe that they are required.

Emissions is a little more in depth, but can be fooled. Many models have a way to disable ignition based security.

Would an LS1/3 be the easiest to buy? But for other motors its doable. Just be willing to make the time investment to figure it out.
 

rv7charlie

Well-Known Member
Joined
Nov 17, 2014
Messages
1,016
Location
Jackson
Honestly, belief systems belong on some other forum. That's especially true when belief and facts disagree. People have literally died proving your beliefs on this subject incorrect. No doubt there will be more, because more will 'believe' that it won't happen again, even though the facts haven't changed.
 

AdrianS

Well-Known Member
Joined
Jul 5, 2014
Messages
572
Location
Australia
I think you are chasing your tail. You want to override a critical motor incident why?

Using an off the shelf ECU from a production motor, with the transmission set to manual (thereby removing all need for speed, lock up, temp and other transmission issues) solves those issues.

Security can be disabled.

Most cars use a power/fuse center in the engine bay, but most of the required circuits (like starter and fuel pump, cooling fans etc) can be bypassed taking that out of the equation.

VVT can be disabled, so those sensors are gone.

The only thing left that you might need to fake out (and again, some of might be configurable in software) are emissions (fuel tank pressure, fuel vapor purge solenoid). This is done by using a 'start box' to simulate these signals.

The only thing left are engine operational sensors... CPS/TPS/spark/O2, temps and pressures. All of these are pretty vital to the operation of the engine.

If you are looking for an 'emergency override' then you can add a small box to simulate just about every one of those sensors save the CPS.

I get that there is no list of 'what causes limp home mode'. But my position is that if you shouldn't ignore them to begin with. I mean no oil is bad. Oil over temp is bad. Many things will stop the motor from working. Its up to each person to decide whether you want a complete, catastrophic event or a pre-catastrophic shut down.

I'd think a 'pre-shutdown' is safer then snapping a crank, losing a rod or valve etc. Something that could put oil and fluid in a position to ignite or cover the windscreen etc.

Honestly, how many times have you had a car go into ACTUAL limp home mode? Not just a CEL, but an actual limited operation? I've never seen it in 20 years of driving.
I work in the dyno world, and have a lot of experience making car engines run on a dyno, with no gearbox or body control ecu.
I've seen several engines go into limp mode.
I've spent a week working with the Japanese engine manufacturer trying to sort out why.
 

pfarber

Well-Known Member
HBA Supporter
Joined
Feb 21, 2019
Messages
611
Location
Pennsylvania
I work in the dyno world, and have a lot of experience making car engines run on a dyno, with no gearbox or body control ecu.
I've seen several engines go into limp mode.
I've spent a week working with the Japanese engine manufacturer trying to sort out why.
On your dyno are you tuning the motor/ECU for max performance?

You say they go into limp mode... ok, why and what happens? Is the emergency mode justified? Over RPM? Over temp? What motor? Japanese motor as in 'made in Japan' or 'only available in Japan'?

See, you can't just say something without why. I don't know anything about you or what you do, and your 3 lines of 'well I triggered limp home mode on a dyno' has little weight. I don't think a complete resume is required.. but say something other than doing a drive by post with an opposite point of view.

My position is that the ECU monitoring the engine sensors and running into limp home mode is a much safer option than blindly running the engine until complete failure.
 
Last edited:

AdrianS

Well-Known Member
Joined
Jul 5, 2014
Messages
572
Location
Australia
You're not listening, just trying to have an argument I can't be bothered with.
Good luck.
 

pfarber

Well-Known Member
HBA Supporter
Joined
Feb 21, 2019
Messages
611
Location
Pennsylvania
You're not listening, just trying to have an argument I can't be bothered with.
Good luck.
You make some claims and don't provide any information. All I'm trying to do is figure out what you are saying, but you speak in vague terms.

What caused the Japanese engine on the dyno to trigger limp mode?
Were you performance tuning the engine?

I think these are logical questions to your previous post. IF you don't feel like responding, OK, but don't claim I am being argumentative. You are one person who claimed to trigger limp mode in a semi-controlled environment... of course I would like to hear more about it.
 

AdrianS

Well-Known Member
Joined
Jul 5, 2014
Messages
572
Location
Australia
It's some years ago, and we were doing endurance testing for an OEM, on totally standard engines. They wouldn't tell us why the first two standalone ECU's / looms went into limp mode : something to do with them not coding sensors out properly.

The point I'm trying to make is that OEM ECUs are stunningly reliable within their designed environment, but it's worth checking what they'll do outside that.

The BMW issue I referred to earlier is, I believe, due to oil temperature: above a certain temperature the engine cuts power savagely.
At the very least, I'd want to know exactly what that limiting temperature is, so as to have a gauge with a red line well below that.
 

TiPi

Well-Known Member
Log Member
Joined
Aug 25, 2019
Messages
337
Location
Peeramon (AUS)
an automotive ECU is a poor choice as the programmed parameters are trying to keep the engine in the fuel efficient or emission compliant mode and they will constantly need a good O2 sensor signal (amongst all the others). In an aircraft application, the demand is different: full power (rich mixture) on take-off and climb-out and then lean for cruise if you care about the fuel burn. An automotive ECU might not give you full rich at the level and the time needed. In cruise, it will go into environment mode (lambda =1) so that the catalytic converter works. Now don't forget that the latest ECUs have 2 O2 sensors, 1 before and 1 after the catalytic converter. Both of those signals need to correspond to the fuel injection volume variations (the air-fuel ratio is not steady, it fluctuates ever so slightly either side of Lambda 1). If those signals don't correspond to each other, the yellow light will come on and it is anyones guess what action the ECU will take after some time.

Maybe go back to 2015 and read up on the VW diesel emission scandal, it provides some insight into the programming of these ECUs.

I would use any or all of the sensors but not an OEM ECU. They are not just a can of worms, they are a drum full of worms.
 
Top