Multiple failures question

[Will]

This has been discussed on this forum a few times over the years, but it's still an interesting topic. I was flying in some unstable air last night, with the forecast showing temporary thunderstorms. I was using weather injected from RealTraffic, which supplies real world weather. Visuals were courtesy of MSFS 2024. It was night.

On final, at about the final approach fix on the ILS, there was a weather update (I assume) and a thunderstorm cropped up right around me. I could still see the runway. The MSFS visuals were depicting maybe 7 miles visibility, with a ceiling above me at maybe 4000 feet. Then all at once, an engine failed and then BAM! I got hit by lightning. The engine recovered.

The lighting strike tripped all 4 generators off line. As a solo pilot, I was pretty overwhelmed. The EICAS messages kept coming, the autopilot and autothrottle disconnected, and the F/O's instruments went blank.

After practicing the scenario, I now know the steps to getting things back under control:

- One by one, cycle GEN CONT and BUS TIE
- Engage appropriate lateral and vertical navigation modes
- Engage autopilot

But in the moment I was overwhelmed by the sheer number of EICAS messages. So I decided to land, as I couldn't discern any problems with the primary flight instruments, the flight controls, or the engines (after recovery from the flameout). It was a nice landing, on centerline, I'm proud to say.

I checked the QRH that I have, and there isn't anything for "Lightning strike" which I guess is obvious, because there's little consistency with what lightning strikes will do to an aircraft.

In retrospect, I would defend my decision to land, because I had no troubles with the flight controls or the engines, and I could see the runway the whole time. I'm curious if any chief pilots or safety officers out there would have a different opinion. The bigger question for me is how a crew would approach this from an analytical perspective. It seems to me you wouldn't solve the problem if you simply tried to run checklists according to what was displayed on the EICAS. You'd need to know that it was an AC electrical problem at its core, and then prioritize the AC ELEC BUS captions over all the others. Is that something that's taught? If so, does it make its way into QRH materials somewhere?

[FlyItLikeYouStoleIt]

If you have the situ it would be fun to share!

[Jeroen Hoppenbrouwers]

I think this is one of those situations that you want two trained pilots aboard that know both their airplane systems and the environment. And not a link to somebody out there, or even two canned brains in replay mode aboard.

[Britjet]

A cascade list of items appearing can be very confusing. On short final it might indeed be best to continue, provided that you haven't lost hydraulics.
As a good example of the consequences try switching off one of the AC buses and watch the fun! (This happened to me once for real over the south Atlantic when a change of FO after in-flight rest meant he accidentally pulled the AC bus 1 CB as he put his flight bag beside his seat!).

The listing unfortunately appears in the order that it occurred, rather than importance, but less important items will be indented one space, so that's a good start.

In your case the ELEC AC BUS 1234 checklist is the one to go for, to quickly attempt to reinstate the AC buses, but you don't reset the bus-tie switches in this case, because if there is an earth fault on one of the buses you will lose all the buses again! Also, this isn't a recall checklist, so you can't really do this on short final.

The bottom line, as you suggest, is "fly the aircraft" - which would be a delegated task in multi-crew, but not so easy on your own.

Speaking personally, I would have landed the aircraft. I'd probably have finished up in the office in front of the pen-pushers, though..

[Bluestar]

On final with less than 5 miles I would maintain visual flight conditions and land.  Once I go visual I normally disco the A/P and A/T.  There is no way I'm climbing and risking going IMC at night in bad weather with an electrical problem. 

[Will]

Britjet, thanks for the reminder about the Bus Tie switches!

FlyItLikeYouStoleIt, I don't have a situ, but if you want to recreate this, all you need to do is fly a normal approach to the airport of your choice, and then at the FAF just hit Activate on the "Lightning Strike" malfunction. It's the last item on Instructor > Situation > Malfunctions > Elec.

[Jeroen Hoppenbrouwers]

A cascade list of items appearing can be very confusing. [...]
The listing unfortunately appears in the order that it occurred, rather than importance [...].

Now my systems engineer brain gets interested.

To order active items in order of priority (a typical Airbus approach), the system needs to have a way to, well, decide what is more important to fix first, and/or what is just a consequence of something else.

Largely this can be pre-determined by system evaluation, design, and other rational engineering that happens on safe grounds and without too much time pressure.

However there is a limit to this, as for example the QF25 A380 incident demonstrated. When something happens that the engineering team did not anticipate, the presentation order may be (totally) inadequate.

What would the pro pilots here prefer? A list that is basically unordered and that requires the PIC to evaluate it as just another piece of information together with everything else, or a list that heavily suggests a work order, which may change constantly as things progress, and that by design cannot be absolutely correct?


Hoppie

[Hardy Heinlin]

I can imagine if every message had a priority number and the list would always be sorted by these numbers while a cascade is growing, the display would be even more unsteady. Certain messages would jump higher on the list, then back down, then swap with another one, and so on.

The current solution is a compromise, I guess.

W-messages are very rare and always at the top.
C-messages are rare and always below W.
A-messages are always below C.

So the chances are very high that you see the most important message in any case, even if they are sorted by time-stamp within their categories. The categories are sorted by importance. So that's the compromise: The importance is not ignored, nor is it set in each message, but it's set in the category layout.

[Bluestar]

I prefer to see messages in the order they are received/occurred.

The like the Boeing philosophy - "if the light comes on, push the button."