# Spark redundancy.

### Help Support Homebuilt Aircraft & Kit Plane Forum:

#### Mike0101

##### Active Member
Sorry if I sounded crabby. I'm having some health issues (and eating some lovely strong drugs ) but that's no excuse for my being uncivil.
It's all good. I generally have a self imposed X beer cutoff for posting LOL.

I agree about micros - that's one of my criticisms of Speeduino.
That plus the fact that I've seen soo much amateurish Arduino code, particularly at the low-level use of hardware, timers and interrupts.
I've only looked at Speeduino a few times, and that was at crank decode and speed density (SD) portions only. Crank decode AS IS, is not capable of detecting misfire due to lack of velocity change. SD is close but technically not correct (but neither is MegaSquirt).

Safety critical hardware / firmware requires an attitude and approach that most programmers / engineers are not familiar with.
I haven't had a project that didn't include functional safety in over a decade. It's not that bad, but safety cannot be an after thought. As levels increase, so does complexity and cost.

As an example, we've had issues where an opto-coupler fails at around 10 years old, and the (original) code went into a watchdog reboot loop waiting for a signal that never arrived.
Actually managed to "brick" an automotive ECU due to messing with DBW throttle opening rate (100% my fault). Safety coprocessor said, no your not and reset main processor. Nice endless loop, in which I couldn't download a calibration to undo condition. You'd be amazed at the number of watchdogs in newer ECUs, even some low-side drivers expect a SPI specific response (can be up to challenge-authenticate) or they simply disable outputs.

As a aside, every time our micro supplier gets a new sales rep, they come out to see us and try to persuade us to switch to their latest and greatest ARM chip range.
The ARM processors at embedded level are certainly dominating market. It's a shame, as a lot of micros are disappearing due to, it's easier to license a core vs developing. With that comes allot of clones, that are trying to compete on price and not features.

The only ARM that I'm aware of that would make a good engine ECU foundation with appropriate safety is TI TMS570 ( TI Link ). Could be more, as I said, it's a hobby.

Once you get out of ARM, then there are 4 or more manufacturers that make great automotive grade micros with correct hardware subsystems to get the job done. Sourcing them in low quantity is a different story.

When I ask about ESD, vibration, and extreme temperatures, I may as well be talking Martian.
Some things off the top of my head. ESD +/- 4000V vs +/- 2000V (human body), some intelligent drivers will not shutdown until 175C (die temperature) is exceeded, injection current (a simple series current limiting resistor and no need for 12V to 5V level translators), adjustable digital output current limit, or how hard do you need to drive output... This is why AEC-Q100 grade micro is 2-5x price of say a generic industrial grade ARM.

Last edited:

#### dwalker

##### Well-Known Member
I have a fair amount of experience operating various types of auto engines at high power levels over extended periods of time, and we rarely if ever have failures these days with the electronics or components. A 24hr road race now is a sprint race, because the cars will be as good at the finish as they were at the start.
That said, when there is a failure it is, in general, poor practice in the wiring or installation that is the issue, not the component parts.
With specific regard to the ignition components, and as an example case, in one type of car we ran, we never, even once, had a coil failure using OEM coils in an application where we operated at high RPM and load for extended periods of time, Others did experience failure, but it is a benign failure in that caused a slight miss, not unlike a fouled plug, instead of the loss of a cylinder.

#### Bigshu

##### Well-Known Member
While it's true that in 'big bore' a/c engines having dual plugs helps achieve complete combustion, I'm not so sure that was the original reason for dual ignition systems.

An O-200 has the same bore diameter as a GM 327 cu in V-8 and it's only slightly larger than the GM straight sixes that were developed in the same time frame as the O200. And it turns at half the rpm (or less). The auto engines from that day ran fine on one ignition; just not bullet-proof reliably.

I don't recall ever having a carb or fuel injection experience a hard fail on me, but I've had numerous mag failures (bullet-proof reliability wasn't a requirement on the tractors they were sourced from)...

Charlie
I've had one. The throttle body on a 2010 Mercury Mariner failed somehow, and the car went into "limp mode". Very disconcerting on the highway! got replaced under warranty, case closed. That's the only ignition system failure I've had in almost 50 years of driving. Had fuel injection trouble on a VW once, but the wasn't a car, it was a Volkswagon! I agree with the sentiment that a well maintained automotive system is very reliable.

#### dwalker

##### Well-Known Member
I've had one. The throttle body on a 2010 Mercury Mariner failed somehow, and the car went into "limo mode". Very disconcerting on the highway! got replaced under warranty, case closed. That's the only ignition system failure I've had in almost 50 years of driving. Had fuel injection trouble on a VW once, but the wasn't a car, it was a Volkswagon! I agree with the sentiment that a well maintained automotive system is very reliable.
That seems to be more a fuel/air delivery system failure than an ignition system failure. It does suck to go into "limp mode". When we were running against the Volvo's in Touring Car they would go into limp mode based on a variety of different things. The engineer on the cars asked me how we defeated all the systems in our cars, and what still worked to put the car into limp mode- too much wheelspin, too many cornering G, door sensor not working, the list was very long- as we had never had an issue. It was very simple, my cars had the ECU's redone at the OEM level to turn off anything that would put the car into limp mode with the exception of actually disconnecting the throttle pedal from the wiring harness, which apparently is the one box that could not be checked.

#### rv7charlie

##### Well-Known Member
My reference to never having a carb or fuel injection 'hard fail' referred to a/c carbs and a/c fuel injection, vs multiple a/c magneto hard fails. Some a/c magneto failure modes can actually damage the engine before the actual stoppage.

Automotive full-vehicle controllers are a completely different animal, as I and others have pointed out earlier in the thread.

Sorry for the confusion, but the interwebs are a tough place to communicate clearly. ;-)

#### pfarber

##### Well-Known Member
I drove my 1942 Ford GPW for a few years. It was a blast. Manual steering, distributor, 3 speed, manual hydraulic brakes , manual wipers, carb, road tube... it really was fun. Took it on and off road.

Constant vapor lock, hit the brakes the steering wheel slammed to the left (the pitman arm was mounted to the axle), starting it was more luck than anything and for every hour you drove it, you needed an hour to tinker/check everything before the next trip.

Dammit if I didn't look cool using the hand crank to start... because the battery didn't charge cause the mechanical cut out regulator stuck.

The worst 'modern' vehicle (EFI/ECU/ABS/etc) is my 2004 F-150 and the ONLY problem with that was the factory plugs needed a special tool becuase they broke when changing them. At 17 years and 154,000 miles that's the only mechanical issue... and that was a factory one (and Ford got their butts sued off for it).

Yeah, I'll take a MODERN efi/ecu car (with an aftermarket ECU) an not even think twice about it.

Even now, lycausur still won't offer fully electronic ignitions even though they are perfectly fine and STCs have had almost no issues. Why? Becuase they don't care about GA. They know the FAA will add whatever layers of regulations are needed to protect Lyco and Cont. from that EASY engine money.

#### dwalker

##### Well-Known Member
Yeah, I'll take a MODERN efi/ecu car (with an aftermarket ECU) an not even think twice about it.

Even now, lycausur still won't offer fully electronic ignitions even though they are perfectly fine and STCs have had almost no issues. Why? Becuase they don't care about GA. They know the FAA will add whatever layers of regulations are needed to protect Lyco and Cont. from that EASY engine money.

If I am honest, Certified aircraft engines scare the hell out of me after reading a bunch of accident/incident reports. Rounded off cams? In the 2000's???? What the hell man! Pistons seizing in washed down cylinders because the mechanical "fuel injection" had a leaky nozzle? Valves burning or simply wearing out?
Just a few common and easily correctable issues that would not be tolerated in todays automotive market, especially if the costs were as high.
And yet, if you start a conversation amongst airplane owners, even Experimental aircraft owners, along the lines of:

"I am thinking of putting a fuel injected turbocharged VW in my Varieze instead of the usual Conti 0-200, because not only is that engine an antique its NLA from Conti and I dont really want to pay 15,000 for a reconditioned long block when I can have a complete ready to bolt to firewall VW based motor for the same money with better GPH and power".

OR

"I am going to do a Mazda 13B rotary in my Long EZ/Cozy IV/Kitfox/RV-anything/etc. instead of a Lycosaurus, because I think its cool"

You WILL be looked as if you are a complete idiot and told how you just need to build the aircraft as "intended".

It is really interesting how few people experiment in Experimental aviation.

#### BJC

##### Well-Known Member
Even now, lycausur still won't offer fully electronic ignitions even though they are perfectly fine and STCs have had almost no issues. Why? Becuase they don't care about GA.
See Lycoming’s TC electronic ignition here LYCOMING EIS

BJC

#### speedracer

##### Well-Known Member
In 26 years/2,000 + hours in my two Long EZ's I've had 6 (six) magneto failures. I've always run one mag, one Lightspeed Engineering EI. I've had no EI failures. After the 6th one I threw the mag in the garbage can (after I took the high priced drive gear off) and ordered a second LSE EI. So.... the big question: If I'd been running two mags would I have had eight failures?

#### Daleandee

##### Well-Known Member
My Corvair conversion is dual ignition with points on one side of the distributor and electronic pickup on the other. Fuel is supplied by a gravity fed Marvel Schebler carb. Been flawless for over nine years ...

#### rv6ejguy

##### Well-Known Member
We have in excess of 750,000 flight hours on our EIs over 26 years. Only one unexplained controller failure in that time and the guy came back to base on the backup.

#### Dan Thomas

##### Well-Known Member
Even now, lycausur still won't offer fully electronic ignitions even though they are perfectly fine and STCs have had almost no issues. Why? Becuase they don't care about GA. They know the FAA will add whatever layers of regulations are needed to protect Lyco and Cont. from that EASY engine money.
Not true at all. Their iE2 engine has been out for about 12 years already. Fully electronic, no resemblance to magnetos or mechanical fuel injection at all. Pilots are just too cheap to buy it. iE2 Integrated Electronic Engine

#### Map

##### Well-Known Member
Way back when the Corvalis was Lancair Columbia we tried using FADEC engines. There were two, one 350 (normally aspirated) and one 400 (turbocharged). We had so many problems with those engines, the system just was not ready for customers, so they were removed.

#### pfarber

##### Well-Known Member
See Lycoming’s TC electronic ignition here LYCOMING EIS

BJC

Which they didn't design.. they just licensed from smarter people.

"The Lycoming EIS was developed by SureFly Partners, Ltd. with input from Lycoming’s engineering teams to meet Lycoming’s exacting specifications. "

I've posted before, the FAA does not require dual ignition. Only 'reliable' and 60+ years ago when mags were still hot garbage the 'reliable' solution was bolt a complete, second ignition to the motor because mags are hot garbage.

If anyone says modern EI is not reliable just walk away from them.

Last edited:

#### BJC

##### Well-Known Member
Even now, lycausur still won't offer fully electronic ignitions even though they are perfectly fine and STCs have had almost no issues. Why? Becuase they don't care about GA.
See Lycoming’s TC electronic ignition here LYCOMING EIS
Which they didn't design.. they just licensed from smarter people.
Well, p, you were wrong.

If you don't want to fly behind a Lycoming, by all means power your HBA with something else. Meanwhile, thousands of us are actually flying E-AB aircraft powered by Lycoming and or Lycoming clones. Whining will get you nowhere.

BJC

Last edited:

#### rmeyers

##### Active Member
Technically pfarber is correct. However from the FAA point of view it is a lot more complicated. Part 33 says in part:
§33.37 Ignition system.
Each spark ignition engine must have a dual ignition system with at least two spark plugs for each cylinder and two separate electric circuits with separate sources of electrical energy, or have an ignition system of equivalent in-flight reliability (emphasis added)

I was part of an engine certification project when this very question came up. The FAA's stance was that if you could should show that your system was as good as or better than the mandated equivalent in-flight reliability that they would have no objection. When asked how that could be shown they responded 'fly your system for about 100 years and then come back to us'. Point taken.

#### rv7charlie

##### Well-Known Member
I haven't read every post in this thread, but in reference to 'equivalent in-flight reliability'... Mfgrs are obviously at the mercy of the FAA on that. But if they were allowed pick an existing certified system to compare to, they'd just need to beat a Slick's 600 hr overhaul requirement. Or Bendix mags with their original coils. I once had consecutive coil failures within a couple of flight hours of each other; they both could easily have happened on the same flight.

#### dwalker

##### Well-Known Member
Corvairs have been flying for 40+ years without twin spark plugs per cyl..