Please critique my electrical architecture

[rdamazio]

Hi all.

I've started designing the electrical system for my -10, and I'd love feedback on it.

This is the overall architecture - it's a dual-battery dual-alternator system since I plan to use EFII's electronic ignition/injection (not a wiring diagram - CBs, actual pins, etc. not depicted, some switches are "virtual" inside the VP-X, busses are symbolic and each device may be connected to different VP-X pins, etc.):
Architecture diagram: https://drive.google.com/open?id=1fv...uL12FvRwGdhQzJ

And this is the load analysis I did - the numbers came back kinda high (yeah, I'm putting a lot of stuff in, but still...). (comment are enabled there, so feel free to comment there directly as well):
Load analysis: https://docs.google.com/spreadsheets...it?usp=sharing

I'm especially uncertain about the dual batteries and alternators setup, with two PPSs:

• Cross-feed: I'm understanding that the EFII Bus Manager handles cross-feeding for me - do I still need a cross-feed contactor in that case?
• External power: I connected the external power plug to only one battery, since it can feed both busses/systems (but did add a switch to isolate this)
• Dual VP PPSs: I ended up with two Vertical Power PPSs to avoid having a lot of solenoids all over the place, and as a way to implement EFII's two-battery starter selection. Is there an easier way?
• Battery overcharging: the plan was to use two EarthX ETX900s, but with two alternators (70A+30A) I'm afraid that, if the other battery somehow dies, both alternators will be (over)charging the single battery left with ~100A, which is above the EarthX's limit of 80A - what's the common way to prevent that? Does the voltage regulator prevent it automatically, or do I need some way to shut down the second alternator?
• Alternator capacity: I understand most people pick 60A alternators, I changed to 70 because that's what the load analysis (below) shows I may have >60A as max continuous - is that the right way to think about it? Or just I just assume that I'll be shedding load anyway if one of the alternators fails, and thus 60+30 (or even less) would be enough?
• Battery in the back/front: Was thinking of maybe installing one battery in the back and one on the cockpit side of the firewall (ETX900-VNT) - this also makes me have only one battery contactor on the back, since the PPS recommends a remote contactor if the battery is remote (which makes sense). Any arguments either way? (also, for anyone that has installed a ETX900-VNT - how did you route the vent lines?)
• EFII ignition: For the ignition redundancy (since that's the only component that's not dual in the EFII system), I placed a relay to switch it between power sources - have others done the same?

Any comments or suggestions are welcome. (how badly am I overthinking this? )

Thanks!
Rodrigo

P.S.: I've read the AEC but won't promise to remember everything in there.

[Carl Froehlich]

At first blush my thought is you should look to simplify. My guess is you have crossed that magical like between ?enhanced capabilities? and ?complexity induced risk?. I clearly understand the option to use two VPX boxes over one as you plan on a very electrically dependent engine, but the other option to achieve the same capability is to not use the VPX at all.

I also recommend you sharpen your pencil on load analysis. For example your lighting load numbers are way high if like most you plan on LED lighting, and you don?t included Pitot Heat.

My experience is a full up dual screen EFIS system with a TSO navigator will draw about 20 amps. Add to that intermittent loads like Pitot Heat, nav, strobes and Landing lights and you are well below the capacity of a standard 60 amp alternator. Your electrically dependent engine may push your numbers a little higher.

I also offer that you should consider the 20 amp B&C standby alternator instead of the 30 amp unit. It will fit much better and it is really all you need for power to extend battery capacity until fuel exhaustion. The standy alternator is just along for the ride if the primary alternator is still working (as in no current output), so your EarthX charging current limits are not an issue.

For W&B, unless you are adding AC or such mount both batteries in the normal aft position. I did a simple mod to the standard Van?s battery mount to accommodate two PC-625 batteries, two master solenoids and four 30 amp relays that feed the panel (two primary and two backup). If you use the EarthX batteries you may need to carry lead or other ballast in the back if flying with no back seat passengers or luggage - but your W&B will tell you all this.

Carl

[Mich48041]

Just because an alternator is capable of putting out 60 amps does not mean that it will. An alternator puts out voltage and current determined by the load. It is up to loads to take whatever current they want. Loads include batteries. Suppose that an aircraft is flying with all electrical loads shut off. Then the pilot turns on an indicator lamp rated at 0.1 amps. A 60 amp alternator will not force 60 amps through that little lamp and burn it out. The current is limited by the lamp resistance. The lamp will take whatever current that it wants: 0.1 amps. Batteries take what they want too.
A load will only take current from the source with the highest voltage. If two alternators are running simultaneously, they will not have equal voltage. The one with the highest voltage will supply all of the loads, the other alternator with lower voltage will supply none. There is an exception to this. If the main alternator is overloaded so much that its voltage sags, then when the voltage drops to that of the second alternator, then both alternators will supply current because their output voltages are then equal.
Listen to Carl.

[DanH]

I'll refrain from specific comment on components and arrangement.

I would like to compliment your method, i.e. the concept of mapping out architecture before proceeding to detailed wiring.

A minor detail; not everyone is conversant in Garminese, that strange language in which components are named in letter and number combinations. If you want review, it would be better to use descriptive labels. Some of us are way too lazy to look up the Garmin names.

[johnbright]

Originally Posted by rdamazio

1234567890

• ... EFII Bus Manager...
• ...two alternators (70A+30A)
• EFII ignition:... For the ignition redundancy (since that's the only component that's not dual in the EFII system)...

Any comments or suggestions are welcome. (how badly am I overthinking this? )

Thanks!
Rodrigo

P.S.: I've read the AEC but won't promise to remember everything in there.

I lost faith in Bus Manager when I saw the fuel pump circuit has single points of failure in the wires and circuit breaker from the essential bus to the pump relay and in the relay itself. Are you showing two separate fuel pump relays? Buss Manager shows a single SPDT relay and if you connect it's common to both of your essential buses they will be tied together. Of course you could use a single DPDT relay but it would still be a single point of failure.

Auxiliary alternator capacity: For instance the B and C "20A" alternator puts out 32 amps at 3500 alternator rpm which is 2700 engine rpm on an O-360; I suppose O-540 has the same 1.3 alternator to crank rpm; BUT, B&C says their standby regulator will not work with VP-X. In any case but especially with the EarthX battery it's critical to have overvoltage protection; Vertical Power says the VP-X has dual alternator overvoltage protection.

Everybody has an opinion and one of mine is to keep it simple and go with something close to a Bob Nuckols circuit with no electronic circuit breakers. Bob's Z-14 is his dual battery dual alternator solution but you would want an essential bus so the injectors would be powered with both battery contactors de-energized.

Personally I'm planning a modification of Z-13/8 with a B and C 20A backup alternator. Z13 allows the backup alternator to charge the battery with the master contactor open. Full disclosure... Bob is opposed to this.

I'm installing dual SDS EFI who's ECU EFII uses until EFII's System 32 comes out. There are some differences in SDS's and EFII's implementation but for my SDS EM-5 system:

• ECU current draw is .03 up to .13A.
• There are two ignition coils controlled indepently by their respective ECUs.
• Four cylinder coils draw 1.1A at cruise so I'm gonna guess six cylinder coils draw 1.7A at cruise.
• One thing there is one of is not ignition coils but the injector set. The ECU does not power the injectors; they are tied to B+ and the ECU switches the other side to ground. I calculated injector current draw using the injector manufacturer's published fuel flow rate and coil resistance and engine fuel flow rate and came up with .31A per injector at cruise and .5A at full power; this for an O-360 which has 90 CI cylinders, same as the O-540. I'm doing a couple things: 1) Fusing each injector independently. 2) Connecting them to a Bob Nuckols style essential bus so they have dual routes to power and are powered with the master switch off.

[johnbright]

An FYI, doesn't affect me because I will have fuses.

VP-X electronic circuit breakers can trip on transients that don't affect fuses and circuit breakers. This can shut down the transient loads we're interested in here; coilpacks and injectors. It's advisable to set the VP-X breakers higher than if a fuse or circuit breaker is used. How high, who knows, and that's a problem because it calls for high RPM testing. And a question; if the electronic breaker is set high, does that mean you need a bigger wire?... or you could set the VP-X really high and install an inline fuse or fusible link at the VP-X.

[rdamazio]

Originally Posted by Carl Froehlich

At first blush my thought is you should look to simplify. My guess is you have crossed that magical like between ?enhanced capabilities? and ?complexity induced risk?. I clearly understand the option to use two VPX boxes over one as you plan on a very electrically dependent engine, but the other option to achieve the same capability is to not use the VPX at all.

It's two PPSs, not two VP-Xs (the PPS is another product from Vertical Power which essentially replaces the battery/starter/alternator contactors with solid state equivalents).
I agree about simplifying if possible (I keep thinking "could a pilot that didn't build this figure out and safely fly this airplane without some advanced training?"), it's the path to that that I'm not seeing clearly

Originally Posted by Carl Froehlich

I also recommend you sharpen your pencil on load analysis. For example your lighting load numbers are way high if like most you plan on LED lighting, and you don?t included Pitot Heat.

Pitot heat is there (GAP26, sorry I used part numbers, just added descriptive names there), and I assumed "max" current to be the 12A that Garmin specified for initial heatup at -40˚C, but only 7A for typical as it draws less current once heated up.

The light numbers are the ones that I have some confidence in because I already have the lights - these are the Aveo Ziptips, and they do pull a surprising amount of current - a tad less than I put in (I put in the official numbers from their diagram), but still a lot.

Do you see anything in particular that seems wrong in the analysis?

Originally Posted by Carl Froehlich

My experience is a full up dual screen EFIS system with a TSO navigator will draw about 20 amps. Add to that intermittent loads like Pitot Heat, nav, strobes and Landing lights and you are well below the capacity of a standard 60 amp alternator. Your electrically dependent engine may push your numbers a little higher.

Interesting, you'd consider that strobes, landing and nav lights are intermittent? I thought anything that *could* be on for an extended period should be accounted for as continuous, even if it won't strictly be on all the time. The strobes do have a certain duty cycle which I didn't consider.

On the 20A estimate - that looks about right - consider that from my load analysis it's roughly:
11A for the EFII
About 12A continuous for the three-screen G3X plus idle GTN750
5A for the alt. field
About 6A for misc. avionics
That's ~34A typical current. A lot of the rest is lights, pitot heat, etc.

Originally Posted by Carl Froehlich

I also offer that you should consider the 20 amp B&C standby alternator instead of the 30 amp unit. It will fit much better and it is really all you need for power to extend battery capacity until fuel exhaustion. The standy alternator is just along for the ride if the primary alternator is still working (as in no current output), so your EarthX charging current limits are not an issue.

Thanks, that's a great hint about 20 vs 30, I haven't done careful consideration of mounting yet (other than seeing that I clearly will need to mod the rear battery mount in some way).

Originally Posted by Carl Froehlich

For W&B, unless you are adding AC or such mount both batteries in the normal aft position. I did a simple mod to the standard Van?s battery mount to accommodate two PC-625 batteries, two master solenoids and four 30 amp relays that feed the panel (two primary and two backup). If you use the EarthX batteries you may need to carry lead or other ballast in the back if flying with no back seat passengers or luggage - but your W&B will tell you all this.

Good point. I guess with two ETXs the tail weight will be only a few pounds below the standard (plus I'm mounting transponder and etc. back there as well).

Originally Posted by Mich48041

Just because an alternator is capable of putting out 60 amps does not mean that it will. An alternator puts out voltage and current determined by the load. It is up to loads to take whatever current they want. Loads include batteries. Suppose that an aircraft is flying with all electrical loads shut off. Then the pilot turns on an indicator lamp rated at 0.1 amps. A 60 amp alternator will not force 60 amps through that little lamp and burn it out. The current is limited by the lamp resistance. The lamp will take whatever current that it wants: 0.1 amps. Batteries take what they want too.
A load will only take current from the source with the highest voltage. If two alternators are running simultaneously, they will not have equal voltage. The one with the highest voltage will supply all of the loads, the other alternator with lower voltage will supply none. There is an exception to this. If the main alternator is overloaded so much that its voltage sags, then when the voltage drops to that of the second alternator, then both alternators will supply current because their output voltages are then equal.
Listen to Carl.

Thank you for the explanation, much appreciated.
That makes sense, now that I'm thinking about generator theory.

Originally Posted by DanH

I would like to compliment your method, i.e. the concept of mapping out architecture before proceeding to detailed wiring.

Thanks. I guess doing it that way made it simpler in my mind.

Originally Posted by DanH

A minor detail; not everyone is conversant in Garminese, that strange language in which components are named in letter and number combinations. If you want review, it would be better to use descriptive labels. Some of us are way too lazy to look up the Garmin names.

Sorry, I realize that now - I've just updated the spreadsheet with descriptive names, hope that helps.

Originally Posted by johnbright

I lost faith in Bus Manager when I saw the fuel pump circuit has single points of failure in the wires and circuit breaker from the essential bus to the pump relay and in the relay itself. Are you showing two separate fuel pump relays? Buss Manager shows a single SPDT relay and if you connect it's common to both of your essential buses they will be tied together. Of course you could use a single DPDT relay but it would still be a single point of failure.

It's supposed to be a single relay, but I haven't settled on the specific way to wire it. My thought there was that, if the Bus Manager completely loses power, the pump on the other bus (on the VP-X) would be turned on. The single relay technically resists the failure of the Bus Manager as well, but possibly not a failure of the relay itself - with separate busses and separate ways to switch them, I'm hoping to eliminate that SPoF. Likewise with the ignition coils, if the Bus Manager essential bus loses power, it'll automatically switch to the other bus. The ECUs are supposed to be "smart" and work ok if the other dies, so no tricks needed there.
(but again, maybe I'm overthinking this ).

Originally Posted by johnbright

Auxiliary alternator capacity: For instance the B and C "20A" alternator puts out 32 amps at 3500 alternator rpm which is 2700 engine rpm on an O-360; I suppose O-540 has the same 1.3 alternator to crank rpm. In any case but especially with the EarthX battery I would use external voltage regulators with crowbar overvoltage protection.

Oh, of course - looking at some alternators now, it seems that the "60A" B&C (BC462) only outputs 41A at cruise, and the BC460 gives 60A only at 9000 alt rpm (no data on the ratio?). Also can't find data on the Plane Power output vs RPM.

Originally Posted by johnbright

Everybody has an opinion and one of mine is to keep it simple and go with something close to a Bob Nuckols circuit with no electronic circuit breakers. Bob's Z-14 is his dual battery dual alternator solution but you would want an essential bus so the injectors would be powered with both battery contactors de-energized.

Personally I'm planning a modification of Z-13/8 with a B and C 20A backup alternator. Z13 allows the backup alternator to charge the battery with the master contactor open. Full disclosure... Bob is opposed to this.

I'm installing dual SDS EFI who's ECU EFII uses until EFII's System 32 comes out. There are some differences in SDS's and EFII's implementation but for my SDS EM-5 system:

I'm probably going with the EFII System32 once it's out (I won't be done building before then anyway) - the data available on it is still partial, but I know it pulls a little more power: 11A total at high RPM, with 4.5 being the fuel pump. I don't have data on how much of the rest is the ECU vs the coils and injectors, so I took a wild guess 1A for the ECU and the rest for the coils.

Yes, I got some inspiration from Z-14, hence my essential busses, one connected to each power source (and half the EFII connected to each).

Originally Posted by johnbright

VP-X electronic circuit breakers can trip on transients that don't affect fuses and circuit breakers. This can shut down the transient loads we're interested in here; coilpacks and injectors. It's advisable to set the VP-X breakers higher than if a fuse or circuit breaker is used. How high, who knows, and that's a problem because it calls for high RPM testing. And a question; if the electronic breaker is set high, does that mean you need a bigger wire?... or you could set the VP-X really high and install an inline fuse or fusible link at the VP-X.

That's good to know.
I'm assuming no need for the thicker wire, since it's a transient - it'll have a bit more voltage drop at that time, but the transient would have to be really high for the voltage to drop enough that the equipment shuts down. As for thermal considerations, if it's a brief transient it shouldn't be enough to heat up the wire, I'd hope?

[rdamazio]

Also, any thoughts on whether a cross-feed switch/relay is needed?

[Carl Froehlich]

Originally Posted by rdamazio

Also, any thoughts on whether a cross-feed switch/relay is needed?

Look at running power from each battery, via a separate 30 amp relay, to half the panel. Have another relay on each battery to be the ?alternator? power for that half of the panel. That way either battery can run either side of the panel, or both sides.

With such a set up the immediate action on any electrical power problem is to open both master solenoids. This one action drops all non-IFR loads (lights, pitot heat, seat heaters, etc.), isolates most common failures and places the system in it?s most reliable state. Once stabile, then you can look to see what has happened and bring back loads as needed. Remember, you are still flying the plane.

This approach is more aligned with ?if needed, what can the existing components do besides normal operation? instead of ?what do I have to add if something goes wrong?. So to answer your question, my scheme has no cross feed relay, but it does have backup modes that can redirect power. My only real issue with a cross feed relay is that you may be connecting the remaining side of your panel with a fault on the other side.

One other tip - run the output of your standy alternator, via diodes, to each side of your panel, not to your battery master solenoid. No need to have the fault that took out your primary alternator do the same to your standby alternator. As one or both sides of the panel will be connected to one or both batteries, you always have at least one battery floating on the standby alternator. The isolation diodes provide another mitigation risk of one fault taking out both sides of your panel.

Carl

[rmartingt]

A couple random notes:

I'm provisionally planning on AeroLEDs and together, the nav, strobe, and landing lights draw 13.1 "continuous", which I assume is the average draw. 20 amps looks a little high. Aveo should publish an average draw figure too--size your power budget on the continuous number, but your wires and fuses/breakers off the peak (or as recommended by the manufacturer).

I too would skip the ECBs on critical engine components.

Originally Posted by johnbright

Everybody has an opinion and one of mine is to keep it simple and go with something close to a Bob Nuckols circuit with no electronic circuit breakers. Bob's Z-14 is his dual battery dual alternator solution but you would want an essential bus so the injectors would be powered with both battery contactors de-energized.

Personally I'm planning a modification of Z-13/8 with a B and C 20A backup alternator. Z13 allows the backup alternator to charge the battery with the master contactor open. Full disclosure... Bob is opposed to this.

Any idea why? Because I'm starting to look down this path as well...

One thing there is one of is not ignition coils but the injector set. The ECU does not power the injectors; they are tied to B+ and the ECU switches the other side to ground. I calculated injector current draw using the injector manufacturer's published fuel flow rate and coil resistance and engine fuel flow rate and came up with .31A per injector at cruise and .5A at full power; this for an O-360 which has 90 CI cylinders, same as the O-540. I'm doing a couple things: 1) Fusing each injector independently. 2) Connecting them to a Bob Nuckols style essential bus so they have dual routes to power and are powered with the master switch off.

I was about to chime in with this until I saw you said it already. This is the big thing that's been tripping me up--I'm planning on the SDS EFI system and I was intending a Z-14 based setup with each bus able to run the engine itself, but once I realized the injector power wasn't split I was having a lot of trouble figuring out how to power those injectors cleanly*. I started leaning towards an "engine bus" with two feeds through diodes, and "hard wiring" (through fuses of course) the injectors and injector relays.

That led me down the path of wondering if I really needed a true dual-bus system, and I started looking more at two alternators and a single big battery, as that might be simpler and thus more reliable, if offering less redundancy (an original "stretch goal" was to be VFR mission capable with one battery or the aux alternator out).


* "cleanly" meaning in a way that was simple to operate, and not vulnerable to failure.