What's new
Van's Air Force

Don't miss anything! Register now for full access to the definitive RV support community.

Power system architecture for EFI

A good reminder before first flight, or the next flight.

There was a fatal accident with our EFI installed on an Egg 6 engine a number of years back, which I helped the TSB up here investigate since they were unfamiliar with these engines or EFI. After going over the obvious, I found the cause in about 15 minutes for the power loss. Injector power was mis-wired to the wrong breaker value (adjacent breaker). In this case a 3 amp one.

It did not show up in extensive ground running as the rpms did not get quite as high as they did with the prop unloaded a bit in flight (current draw a bit higher at higher rpms). Engine stopped at about 300 feet after takeoff and a landing was attempted on the crossing runway. Stall/spin and my friend test flying, was killed.

As Dan said, this is vitally important stuff.

For Vertical Power users, please read our warnings about setting breakers values on these devices when using our EFI.

During checkout I found a cross wired breaker on my professionally wired panel. It can happen, this is a good point for the commissioning of a new aircraft, that is to check every breaker to ensure it is properly connected as designed and expected.
 
I previously asked about wiring diagrams (post 93):

In the meantime, it sure would be nice to see a complete wiring diagram for a dual EM-5. This one seems to be missing all the details for injector switching:

http://www.sdsefi.com/dualecu4.pdf

John Bright's notes state that injector switching is different for SDS and EFII installations. Robert doesn't seem to have published a diagram for his version of the EM, or for the System 32.

The SDS diagram linked above is partial, and doesn't cover the 6-cyl models. As noted, there does not seem to be any published diagram for EFII versions, which were supplied with a different harness. Both vendors supply a marked harness, and builders are expected to connect it as indicated on the tags.

Ultimately the builder is responsible for how the aircraft is wired. No available diagrams makes wiring difficult to evaluate, prior to purchase, or during installation when changes might be considered.

Considering only injectors for the moment, we can see individual circuit protection for the injectors on the SDS diagram. We have learned that some significant number of EFII installations are wired with a single 10 amp breaker for two coils and four injectors, but the vendor is or was evaluating individual supplies.

What we don't see is the ground control side of the injector circuits. Each injector's ground is opened and closed by the ECU to establish pulse width. So, a question. Does the supplied harness combine (junction) all the injector grounds, then run a single wire to the ECU (via switching relays, if dual ECU), or does the harness provide individual individual grounds all the way to the relays or ECU?
 
I previously asked about wiring diagrams (post 93):

Considering only injectors for the moment, we can see individual circuit protection for the injectors on the SDS diagram. We have learned that some significant number of EFII installations are wired with a single 10 amp breaker for two coils and four injectors, but the vendor is or was evaluating individual supplies.

What we don't see is the ground control side of the injector circuits. Each injector's ground is opened and closed by the ECU to establish pulse width. So, a question. Does the supplied harness combine (junction) all the injector grounds, then run a single wire to the ECU (via switching relays, if dual ECU), or does the harness provide individual individual grounds all the way to the relays or ECU?

On aircraft installations, we supply wiring with twin ground path connections to the ECU (pins 8 and 16 on white ECU connector). These should go to separate terminals and either one is sufficient to run all circuits concerned.

On 4 cylinder dual ECU setups, the injector relays have their own ground. Relay is normally closed for primary ECU control so no ground or power is required in that mode.
 
Both vendors supply a marked harness, and builders are expected to connect it as indicated on the tags.

Ultimately the builder is responsible for how the aircraft is wired.

Exactly. Here at SteinAir, we get many requests to modify the wiring scheme from what Ross and Robert Paisley specify in their instructions, and we, as a matter of policy, don't do it. If we do, and it doesn't work as it should, we assume some responsibility for it, which we are not willing to do. There are a number of ways to insure an uninterrupted power supply to the system, and it is indeed up to our customers to decide how and to provide it.

The bottom line for us: Alternative ignition and fuel delivery systems are here to stay. We gladly accommodate them for our customers, but we must insist the designers' instructions be followed.
 
On aircraft installations, we supply wiring with twin ground path connections to the ECU (pins 8 and 16 on white ECU connector). These should go to separate terminals and either one is sufficient to run all circuits concerned.

Thanks, I was thinking about the injector leads. Found it in the manual.

There are a number of ways to insure an uninterrupted power supply to the system, and it is indeed up to our customers to decide how and to provide it.

The bottom line for us: Alternative ignition and fuel delivery systems are here to stay. We gladly accommodate them for our customers, but we must insist the designers' instructions be followed.

I don't think anyone blames the panel guys for the SQ-12 engine out. They did what they should, and probably what you would do...they got advice in writing, and built it.
 
wiring

Let's be very clear here:

We have published, recommended, and tested wiring layouts for our systems.
All of our harnesses are CLEARLY labelled with the recommended breaker values on each power leg. There are people who decide to implement their own versions of wiring in their 'experimental' aircraft. This is something we can not control.

This in no way means that we recommend custom installations, that we have tested these installations or that they are in any way known to work or not work. Once you start re-inventing the wheel, it's on you to test it to make sure it works. We can only control what happens inside our shop, not what happens in the rest of the world. We get into conversations with customers all the time about their custom wiring installations. This doesn't turn a custom installation into recommended or tested wiring.

Follow manufacture recommendations! Your manufacturers know more about their systems than anyone else. Be very careful about following the advise or conclusions of Internet trolls - typing 'Google' does not give someone experience in a subject, and it does not make them an expert.

Robert
 
Once you start re-inventing the wheel, it's on you to test it to make sure it works.

It is always the builder's responsibility, even when installing exactly as the vendor suggests. There is no reason to assume anyone (vendor, troll, best buddy, or A&P-IA) is correct. Trust, but verify.

Robert, will the System 32 harness be supplied with a single circuit protection device for coils and injectors, like past harness, or individual circuit protection for each device?
 
S32 harness

System32 harnesses have fusible links in the harness on all 12v supply wires.
Each injector and each coil pack power lead is individually protected.
The breaker requirement does not change. If a coil pack were to short out, it would pop its fusible link and not affect anything else. If an injector power lead were to short out, it would pop the fusible link for only that one injector.
The fusible links are located in the engine side circular connector backshell.

Robert
 
System32 harnesses have fusible links in the harness on all 12v supply wires.
Each injector and each coil pack power lead is individually protected.
The breaker requirement does not change. If a coil pack were to short out, it would pop its fusible link and not affect anything else. If an injector power lead were to short out, it would pop the fusible link for only that one injector.
The fusible links are located in the engine side circular connector backshell.
Robert

Thank you Robert. A tidy solution; my compliments sir.
 
So I'm getting closer to being able to start my systems installation and hoping to make some progress on my electrical diagram while I'm laid up next week. I've been looking over some thoughts on electrical systems and trying to figure out how best to go about providing reliable power to an SDS EFI system on an IFR-equipped aircraft.

So Bob (OP),

For those like me coming to this and reading through it over a long evening, with it?s twists and turns, waiting for the ending...

What did you decide?

Would you be able to share your final architecture with us?
 
So Bob (OP),

For those like me coming to this and reading through it over a long evening, with it?s twists and turns, waiting for the ending...

What did you decide?

Would you be able to share your final architecture with us?

I?ve settled on a system pretty much the same as the one in this post:
http://www.vansairforce.com/community/showpost.php?p=1261072&postcount=123


My remaining decision is whether to use relays/contactors for the engine bus, or go directly through high-current switches. I?m not sure if I like the idea of having all the engine power running up to a pair of switches on the panel and then back forward even if it is protected with high-current fuses. I?m planning an automotive relay for my pitot heat which would be the other high-current item.

The part of me that says ?simple? says just use the switch; it just goes against what I?d normally do for bus control.

Been working on the actual wire diagrams and think I?ve figured out the supplied wiring for the SDS system from the documentation. But I?m using prep and paint of interior components to stall and delay having to buy more parts right now.
 
I don't remember there being much, if any, discussion about possible failures downstream of an engine bus. On the Aeroelectric website, Bob has some videos discussing an accident where he points out that a failure downstream of the engine bus can blow the fuses on both paths to the engine bus (although it sounds like this was determined to not be the cause of the accident).

The videos can be found here if anyone is interested: http://www.aeroelectric.com/Reference_Docs/Accidents/N811HB_Feb2008_LA-IVp/
The 3rd video is where he shows this happen but all of the videos have good information in them.

Perhaps proper component selection can prevent this but I figured it was worth mentioning.
 
Bob has some videos discussing an accident where he points out that a failure downstream of the engine bus can blow the fuses on both paths to the engine bus

I do recall reading about an accident like that he analyzed in depth. If its the one I recall, the builder had put fuses on the feeds of both sides of the engine bus. They were the same size as the individual circuit breakers. So if there was a short to ground failure on any part of the engine bus, it would blow one fuse followed quickly by the other.

He made it through phase 1 with no issues then crashed on one of his first flights with passengers.

0reBmwK.png
 
Last edited:
I do recall reading about an accident like that he analyzed in depth. If its the one I recall, the builder had put fuses on the feeds of both sides of the engine bus. They were the same size as the individual circuit breakers. So if there was a short to ground failure on any part of the engine bus, it would blow one fuse followed quickly by the other.

He made it through phase 1 with no issues then crashed on one of his first flights with passengers.

0reBmwK.png

It scares me that a DAR or FSDO inspector signed off on an airplane with such a glaring failure waiting to happen.

Lesson learned (again) - if you must have an electrically dependent engine you better know what you are doing. Do not just copy what looks pretty on paper and repeat other peoples mistakes.

Carl
 
I do recall reading about an accident like that he analyzed in depth. If its the one I recall, the builder had put fuses on the feeds of both sides of the engine bus. They were the same size as the individual circuit breakers. So if there was a short to ground failure on any part of the engine bus, it would blow one fuse followed quickly by the other.

He made it through phase 1 with no issues then crashed on one of his first flights with passengers.

0reBmwK.png

Correct.
It has to do with the blow time differences between fuses and breakers.

The breakers are slow when compared to fuses. So when you have breakers protecting the ignitions, a fault on either one of them caused both the upstream fuses to blow prior to the breakers ever popping which resulted in loss of power to the ignition bus.
 
It scares me that a DAR or FSDO inspector signed off on an airplane with such a glaring failure waiting to happen.

Lesson learned (again) - if you must have an electrically dependent engine you better know what you are doing. Do not just copy what looks pretty on paper and repeat other peoples mistakes.

Carl

Do you really expect an FAA guy or DAR to analyze your system architecture? If you do, you're in serious trouble.

This also brings up another point, if the builder of a aircraft with an elaborate electrical system, electrically dependent engine, electronic ignition, fuel injection etc. ever sells the plane or isn't around to take care of it.. you may have a bit of trouble finding someone to maintain it or help fix it.
(if you don't think this stuff fails you haven't been around very long)

When you build one of these complex birds you're pretty much on your own.
 
Last edited:
Do you really expect an FAA guy or DAR to analyze your system architecture? If you do, you're in serious trouble.

No - I do not rely on a DAR or FSDO inspector and I would not expect either to do a detailed analysis. I would however expect that an inspector (or any trusted agent) doing a cursory glance at a power diagram having fuses/breakers in parallel to a load should have alarm bells going off in his head.

I say again - if you must have an electrically dependent engine you better know what you are doing.

Carl
 
No - I do not rely on a DAR or FSDO inspector and I would not expect either to do a detailed analysis. I would however expect that an inspector (or any trusted agent) doing a cursory glance at a power diagram having fuses/breakers in parallel to a load should have alarm bells going off in his head.

I say again - if you must have an electrically dependent engine you better know what you are doing.

Carl

Are you kidding? The feds/dar is there to make sure you complied with the regs for building an EAB aircraft, nowhere in their job description does it say they are responsible for looking at those things, heck you could put a briggs and straton in the thing and they would sign it off if you did all the paperwork!
 
Last edited:
It is not a good idea to put over-current devices in series. Both could blow or trip, even if they have different current ratings.
 
Having read the whole thread here, I was wondering what happens if the “engine bus” shorted.

[Please be kind to a half thought out idea here]
Wouldn’t it be better to split that bus in two? Most all designs so far I’ve seen have one engine bus with 2 feeds. Why not 2 seperate busses each feeding half the engine required items. I.e. one feeding ign 1, one feeding ign 2 etc. pri fuel pump on 1, sec pump on 2 etc. you’d add up all the amps of the planned items on that bus, and cable it and breaker that bus feed appropriately. Then each item has a fuse sized for the wire to the item.

For your pri pfd and pri nav/com these would have their pri power from one bus and their secondary power from the other. Pitot head off one and landing light off the other. Cabin lights fed from both?

Everything else is on a main bus which gets shed when the master goes off.

If your 2 alts are not the same size, then a diode feed from the engine bus with the higher capacity to the lower one.

What is the consensus on this?
 
Last edited:
Trent,

That's a perfectly reasonable thing to do, *if* the engine-related 'stuff' is completely isolated, like a pair of electronic ignitions. But Everything's a compromise; in doing that, you also complicate the wiring architecture, and likely, complicate operational tasks, as well.

With full fledged engine controllers, even ones with redundant control modules, it gets more complicated. Most (notice that I used the word 'most') controllers may fire separate sets of plugs with separate coils, but the injector side; not as likely. Most (<<) use one set of injectors and supply a way to switch between controller injector drivers.

The builder in question just made some really bad wiring architecture decisions; likely through lack of education or really bad advice. Proper circuit protections would have almost certainly avoided the disaster. Now, something dropping a wrench on the bus itself... you really can't fix stupid. We all tend to focus on doubling up on electrical stuff, but remember, virtually all a/c engines have only one fuel path to the engine, and one fuel metering device. The ones with two (Rotax 91x) will likely just get you an additional 40 feet into the crash scene if one side fails. :)
 
A good design has no "Oh ****" switch to keep the engine running. Engine electronics need dual power or auto switching.

The airplane is at 200 feet, the departure end just went under the wing, and the engine quits. Right then, nobody has enough spare brainpower to diagnose an electrical problem.

I agree completely! Also, given the high power-density batteries now available, I don?t think dual alternators make sense anymore. Better a symmetrical, more capacious, dual-battery system with one of each redundant systems pair like electronic ignitions operating off each battery as SOP. Easy to do these days.- Otis
 
Otis,

This might seem to be a semantics discussion, but terms can be important. High power density guarantees nothing for duration. You *might* get high total energy in the bargain, but it's not a given; many high power starting batteries have minimal total energy.

A backup battery is by definition a limited, finite source of electrical power. And as a pragmatic reality, any battery loses *energy density* as it ages. Everyone gets to pick their own comfort zone; if you're comfortable that battery power will always carry you to fuel exhaustion, or if you're comfortable with it having just enough capacity to get you to the nearest airport, that's fine. Just hoping that you're making fully informed decisions.

My comfort zone is a bit different. For my electrically dependent engine, the mix of much lower cost, minimal weight penalty (actually lighter than an SLA battery), effectively unlimited supply (duration of fuel in the tanks), and the ease of mounting a 2nd identical alternator on my engine (meaning no need for load shedding), drove me to pick dual alts over dual batteries. I may yet revise this to dual 'both', but for now, dual alts floats my boat. :)

Charlie
 
I?m going with a two battery, one alternator setup for my electrically dependent -7 build. The alternator is a 60A B&C. Batteries will be Odyssey 625 for primary power and back power will come from EarthX and power management/distribution will be handled by the EFII bus manager.
 
Wouldn?t it be better to split that bus in two? Most all designs so far I?ve seen have one engine bus with 2 feeds. Why not 2 seperate busses each feeding half the engine required items. I.e. one feeding ign 1, one feeding ign 2 etc. pri fuel pump on 1, sec pump on 2 etc. you?d add up all the amps of the planned items on that bus, and cable it and breaker that bus feed appropriately. Then each item has a fuse sized for the wire to the item.

Most (notice that I used the word 'most') controllers may fire separate sets of plugs with separate coils, but the injector side; not as likely. Most (<<) use one set of injectors and supply a way to switch between controller injector drivers.

Yep, this is the issue. As I posted earlier:

The kicker is the injectors; each one is powered individually (rather than by the ECU driving it) and there's only one per cylinder, so one way or another power needs to be able to get to them from both "sides". Running on two of four injectors is not good. So at that point, if I need to feed power from both sides to make the injectors work, why not do it for all of the engine components? That's how I went from the original plan down to these later concepts, trying to address that issue.

Also, as I was playing with wire diagrams and got better details on the EFI system I discovered some failure modes with that approach that would have had me throwing switches after one failure, or even just losing a bit of power (like losing one mag).

This question of protecting the engine bus feeds ties into the switch vs. contactor question for the engine bus feeds. There are conflicting desires--minimizing complexity, minimizing chances of something killing both feeds, and not having an unprotected always-hot high-current wire running to the panel and back to the bus (typically regarded as not a good idea).

The way I see it there are two ways of dealing with this.

The first is to go battery --> overcurrent protection --> panel-mounted switch --> engine bus
where overcurrent protection is something like an ANL fuse (e.g. size the fuse and wire for the engine bus for 40A or so). But as others have pointed out with the accident noted, having non-resettable and/or quick-blowing current protection on all of the power feed can bite you.

The second is
battery --> contactor or relay --> engine bus with short wires
This keeps the power feed for the engine bus shorter so you can do away with overcurrent protection on the feed, but adds the slight extra complexity of a contactor.

I'm leaning towards contactors/relays for the engine bus feed. The standard Spruce-supplied contactor sucks a lot of power to stay on, so I'm looking at some sealed industrial units or even some heavy-duty automotive relays.
 
I do recall reading about an accident like that he analyzed in depth. If its the one I recall, the builder had put fuses on the feeds of both sides of the engine bus. They were the same size as the individual circuit breakers. So if there was a short to ground failure on any part of the engine bus, it would blow one fuse followed quickly by the other.

He made it through phase 1 with no issues then crashed on one of his first flights with passengers.

0reBmwK.png

For learning purposes and to clarify the teaching moment, would this unfortunate incident have been mitigated by having a much larger breaker and appropriately sized diode fed wire running between the battery and the bus?
 
For learning purposes and to clarify the teaching moment, would this unfortunate incident have been mitigated by having a much larger breaker and appropriately sized diode fed wire running between the battery and the bus?

I don?t know what the deal is on having two power feeds to a ?dual ignition bus?. Why not have one ignition on each battery - no ignition bus or diodes?

Carl
 
I don?t know what the deal is on having two power feeds to a ?dual ignition bus?. Why not have one ignition on each battery - no ignition bus or diodes?

Carl

Certainly not needed for what is drawn.

One thing that I've seen is with full EFII there are some components that aren't able to be redundantly fed without doing some OR'ing scheme. For example, the injectors. There is only 1 per cylinder and 1 power feed to each. If you were to run half of them on 1 battery and the other half on the other, if something happens, you have an engine that wants to shake itself apart.

I've then seen that concept expanded from well if you have to OR for some components, why not OR for all of the EFII components.

I know for what I am planning, I plan to have each battery bus drive each ECU, coil, and fuel pump separately and then only OR for minimum components needed (i.e. the injectors).
 
I don?t know what the deal is on having two power feeds to a ?dual ignition bus?. Why not have one ignition on each battery - no ignition bus or diodes?

Carl

I'm won't argue against that as an option, however, I'd still like to understand exactly how this particular builder could have prevented the problem that ultimately killed him with the bus that he had set up. Just learning the pitfalls.....
 
I'm won't argue against that as an option, however, I'd still like to understand exactly how this particular builder could have prevented the problem that ultimately killed him with the bus that he had set up. Just learning the pitfalls.....

Hey Randy,

From what I recall, is that it has to do with having fuses upstream of breakers.
Breakers blow much slower than fuses. A fault on one of the ignitions caused both upstream fuses to blow quickly and way before the breakers that were attempting to protect the wires for the individual ignitions could take effect. Both fuses blowing caused both ignitions to lose power.

The basic premise is to only have one protection device in any given circuit and don't serialize them especially don't serialize two different types of protection devices that have different blow characteristics.

Hopefully that explained it well enough.
 
Last edited:
I don?t know what the deal is on having two power feeds to a ?dual ignition bus?. Why not have one ignition on each battery - no ignition bus or diodes?

Carl

Exactly what I'm doing on my SDS equipped RV-10. Dual bus, dual alternator, dual battery. Yes, there is a cross-tie relay, but it will only be used for start. One bus is dedicated to a single ECU, fuel pump and ignition.

Maybe a bit overkill and obviously adds weight, but for my first electrically dependent airplane, I'm trying to be conservative (at least as far as ensuring electron flow is concerned). I know some will say conservative means magnetos and mechanical fuel injection.
 
Exactly what I'm doing on my SDS equipped RV-10. Dual bus, dual alternator, dual battery. Yes, there is a cross-tie relay, but it will only be used for start. One bus is dedicated to a single ECU, fuel pump and ignition.

Maybe a bit overkill and obviously adds weight, but for my first electrically dependent airplane, I'm trying to be conservative (at least as far as ensuring electron flow is concerned). I know some will say conservative means magnetos and mechanical fuel injection.

I'm going to go with what Ross has suggested for system architecture. If each circuit off of the bus bar is fuse or c/b protected, and the bus bar has two sources of potential power input, why is that better other than having two physical bus bars? I don't think bus bars themselves are prone to failure. I think this a rehashing of the discussion much earlier in the thread, at least how I understand it.

Edit: After reviewing the Nuckolls' videos referenced in post 281 I'm going to reconsider what I just said.....
 
Last edited:
Hey Randy,

From what I recall, is that it has to do with having fuses upstream of breakers.
Breakers blow much slower than fuses. A fault on one of the ignitions caused both upstream fuses to blow quickly and way before the breakers that were attempting to protect the wires for the individual ignitions could take effect. Both fuses blowing caused both ignitions to lose power.

The basic premise is to only have one protection device in any given circuit and don't serialize them especially don't serialize two different types of protection devices that have different blow characteristics.

Hopefully that explained it well enough.

Thanks for that Adam. So I understand correctly, the main problem is that both the injectors breaker and the upstream fuse were rated the same and for the reason that you outlined, the fuses from both battery sources blew first when the injector took a dump. It seems strange to me that the wiring and fuse from the battery was the same amperage of the injector which tripped. That wire should be powering the whole bus, not just one component. That implies that it should have been rated significantly higher If that had been a 30 amp breaker with appropriately sized wire, this shouldn't have happened. That is the way I am understanding it.

Edit: Reviewing the videos referenced in #281 answers this well.
 
Last edited:
Breaker Sizes!!!

I've gone back a ways in this discussion to the schematic in the links posted by Dan of the accident aircraft.

The 10A fuse recommended in the schematic for the 2 coil packs and 4 injectors is plain scary.

We know that each coil pack draws around 2A and the injectors about an amp each in round figures- AVERAGE current. That's around 8 amps.

However the momentary surge current on these devices is MUCH higher. The coils can draw 10A each and the injectors almost 2 amps each for a few milliseconds or so. This is one reason why we have big cautions for breaker sizes when using a VPX which responds to PEAK current, not average current as traditional breakers and fuses do.

SDS sizes fuses and breakers at at least double the average current of the device if it doesn't have high inductance and around triple for things like fuel pumps, injectors and coils which do have high inductance and high surge current.

For injectors, we recommend a 5A breaker or fuse for EACH injector
For coil packs, we recommend a 10A breaker or fuse for EACH coil pack
For fuel pumps, we recommend a 15A breaker or fuse for EACH fuel pump (this value is stamped right into the Walbro pump bodies BTW)

Please remember that fuses and breakers are just there to protect the wiring from going up in smoke. Generally they won't protect the device. As in most things to do with electronics, running a component near its max rating is likely to shorten its lifespan.

A marginally sized breaker or fuse will run hot and may eventually oxidize and fail after a bunch of hours. There is no good reason to use marginally sized circuit protection here. The results can and have been fatal as I mentioned earlier in this post. I helped the TSB here investigate that fatal accident and found the cause for them pretty quickly- wrong breaker size used to feed the same components we are talking about here.

Lots of people in this thread who have no experience (and some who apparently do) with EFI and its components, have posted their opinions here while ignoring the most fundamental aspects of reliable circuit design- fusing. Bluntly, I caution people about taking the advice of people with no experience in this field. In this case, bad advice can kill you...
 
Last edited:
I've gone back a ways in this discussion to the schematic in the links posted by Dan.

The 10A fuse recommended in the schematic for the 2 coil packs and 4 injectors is plain scary.

We know that each coil pack draws around 2A and the injectors about an amp each in round figures- AVERAGE current. That's around 8 amps.

However the momentary surge current on these devices is MUCH higher. The coils can draw 10A each and the injectors almost 2 amps each for a few milliseconds or so. This is one reason why we have big cautions for breaker sizes when using a VPX which responds to PEAK current, not average current as traditional breakers and fuses do.

SDS sizes fuses and breakers at at least double the average current of the device if it doesn't have high inductance and around triple for things like fuel pumps, injectors and coils which do have high inductance and high surge current.

For injectors, we recommend a 5A breaker or fuse for EACH injector
For coil packs, we recommend a 10A breaker or fuse for EACH coil pack
For fuel pumps, we recommend a 15A breaker or fuse for EACH fuel pump (this value is stamped right into the Walbro pump bodies BTW)

Please remember that fuses and breakers are just there to protect the wiring from going up in smoke. Generally they won't protect the device. As in most things to do with electronics, running a component near its max rating is likely to shorten its lifespan.

A marginally sized breaker or fuse will run hot and may eventually oxidize and fail after a bunch of hours. There is no good reason to use marginally sized circuit protection here. The results can and have been fatal as I mentioned earlier in this post. I helped the TSB here investigate that fatal accident and found the cause for them pretty quickly- wrong breaker size used to feed the same components we are talking about here.

Lots of people in this thread who have no experience (and some who apparently do) with EFI and its components, have posted their opinions here while ignoring the most fundamental aspects of reliable circuit design- fusing. Bluntly, I caution people about taking the advice of people with no experience in this field. In this case, bad advice can kill you...


Ross, in post 281 there is a link to five very short videos done by Nuckolls that seem to indicate that there are dangers associated with using one single bus, even appropriately fused for each component. The main jist is that there is a potential difference in amps from each of the battery sources that could blow fuses or breakers. I don't quite understand why the current draws he shows are so different but I'm have a request in to an electrical engineer buddy of mine (also an RV10 builder (Charrois here on VAF)) to review and help clarify for me. If you get the chance to look at them, your feedback would be extremely valuable.
 
Last edited:
Exactly what I'm doing on my SDS equipped RV-10. Dual bus, dual alternator, dual battery. Yes, there is a cross-tie relay, but it will only be used for start. One bus is dedicated to a single ECU, fuel pump and ignition.

Maybe a bit overkill and obviously adds weight, but for my first electrically dependent airplane, I'm trying to be conservative (at least as far as ensuring electron flow is concerned). I know some will say conservative means magnetos and mechanical fuel injection.
So for those who are running their system off of split, independently powered busses, what do you plan for the fuel pumps and the injectors? Do you keep each bus running one fuel pump (ie: both pumps on all of the time)? How about the injectors? If one bus goes down then you lose half of your injectors. How will the engine run in that situation?
 
Last edited:
Yes, in my case each fuel pump will be powered by its respective hot battery bus, circuit breaker protected and controlled by a Milspec Honeywell locking toggle. I would anticipate both pumps on during critical phases of flight and one off the rest of the time. Probably alternate even/odd by calendar day.

The injector power is another issue. Since normally half the injectors (on a 6 cylinder) are controlled (but not powered) by one ECU, each bank of three would be powered its respective bus. If an ECU fails, there is a switch to place all injectors on one ECU. You could switch injector power to that one bus at the same time or you could have a diode protected “injector bus” fed simultaneously from both aircraft busses.

There are concerns using either approach and we are working our way through that to arrive at an acceptable solution.
 
Last edited:
Research...

If you are using the EFII system, you may want to do a bit of research...the stated ECU logic and fuel pump logic may not be what you think...
 
EFII

I’m using SDS. I’m interested in what you mean by “not what you think” though.

A previous post suggested running both fuel pumps at once from separate busses. The EFII pumps are in a manifold and the mfg does not recommend trying to run two pumps at once.

The injectors on the EFII are controlled by one ECU. You can switch between ECU injector control.
 
A previous post suggested running both fuel pumps at once from separate busses. The EFII pumps are in a manifold and the mfg does not recommend trying to run two pumps at once.

The injectors on the EFII are controlled by one ECU. You can switch between ECU injector control.

Hopefully Ross or Barry from SDS will weigh in shortly and correct my post below -

Unless I was misinformed, the recommendation from SDS is to run both pumps during critical phases of flight. Similar to turning on your electric pump on an airplane equipped with a mechanical pump - you want a backup in case one or the other pump fails.

On SDS one ECU controls half of the 6 injectors and the other ECU controls the other half. There is a switch to flip if one ECU fails so the working ECU controls all injectors - but you lose some functions.

EFII System 32 may be different.
 
Yes, in my case each fuel pump will be powered by its respective hot battery bus, circuit breaker protected and controlled by a Milspec Honeywell locking toggle. I would anticipate both pumps on during critical phases of flight and one off the rest of the time. Probably alternate even/odd by calendar day.

The injector power is another issue. Since normally half the injectors (on a 6 cylinder) are controlled (but not powered) by one ECU, each bank of three would be powered its respective bus. If an ECU fails, there is a switch to place all injectors on one ECU. You could switch injector power to that one bus at the same time or you could have a diode protected “injector bus” fed simultaneously from both aircraft busses.

There are concerns using either approach and we are working our way through that to arrive at an acceptable solution.

Krea I would be interested in seeing the solution you come up with for the injectors in particular. I had difficulty resolving that problem without introducing a level of complexity that made the overall system more complex and less reliable. For me at least there was no free lunch. The way I set up mine the only real danger is the unlikely event of shorting the entire to bus to ground, a risk that easily can be mitigated.
 
On SDS one ECU controls half of the 6 injectors and the other ECU controls the other half. There is a switch to flip if one ECU fails so the working ECU controls all injectors - but you lose some functions.

I think you've got the injectors mixed up with the spark plugs. My understanding from the installation manual is that the injectors are all controlled by only one ECU at a time, but each ECU only controls half of the spark plugs.

From the SDS installation manual:
Fuel ECU Switch
When running dual ECU boards, this toggle switch activates relays which switch operation of the
injectors from one ECU board to the other. In normal operation, one board is always firing the top
spark plugs, the other fires the bottom plugs. The ECU select switch only switches the injector
connection over to the other board. If one ECU board fails, you?ll lose one set of plugs but the engine
should continue to run.
 
Probably true on 4 cylinder SDS

I think you've got the injectors mixed up with the spark plugs. My understanding from the installation manual is that the injectors are all controlled by only one ECU at a time, but each ECU only controls half of the spark plugs.

From the SDS installation manual:

From the 6 cylinder dual ECU supplement -

?In NORMAL position both ECU?s will operate and each ECU will control 3 injectors.?
 
Back
Top