FMC FAILURE

[emerydc8]

Thanks for the info, John. I was thinking more along the lines of an engine failure and not a probe heat failure. The probe heat system is sounding as bad as the LLCCAFR system though!

Hi Hardy. The Atlas/Polar FCOM says:

If a second engine fails, the page title changes to ACT 2 E/O LRC CRZ and the FMC calculates and displays two-engine out maximum/optimum altitudes and performance data.

[emerydc8]

Hi Peter,

Can you get your hands on the 3-engine ferry procedure for BA? Maybe it's something similar to ours.  It looks like VNAV will still work on our aircraft after a second engine fails.
Thanks,
Jon D.

[Britjet]

Hi Jon,

I just had a look through the current manual, and there is no checklist as such, and no mention of VNAV etc.
It merely says that in the event of a two-engine situation that the FMC will give two-engines inoperative data.

Just looking at your checklist - I am wondering if the page you show is talking about two separate things - the top section being about a further failure, and the bottom section being about handling the single failure...?
It almost looks like a poorly-placed cut-and-paste to me?

Peter

[emerydc8]

Hi Peter,

Yes, I see how it could be construed that way. It probably is a bad cut- and-paste job. I would note, though, that it doesn't say anything about VNAV not being available with a subsequent engine failure. I'll keep researching this. Thanks.

Jon

[Hardy Heinlin]

[ I moved the "one engine inoperative ferry" discussion to: http://aerowinx.com/board/index.php?topic=3697.0 ]

[emerydc8]

Hi Hardy,

I am revisiting this FMC failure issue, because I want to do a video on it. I just wanted to ask a few questions and, since Peter took his video down, I can't compare PSX with the big sim. In PSX, when the left (master) FMC fails during a VNAV climb, there is a time delay of about 30-seconds before the ND goes blank and the boxed MAP/VTK appears on the ND, followed about 10 seconds later by the right CDU going from RESYNCHING OTHER FMC to SINGLE FMC OPERATION, and then FMC L appears on the EICAS a few seconds after that. Does this duplicate the big sim? It has been so long since I saw his video I can't remember, and I recall that you were in the process of making some changes to PSX as a result of the video. Thanks.

Jon

[Hardy Heinlin]

The resync delay has been observed in the BA sim, I think, and agrees with the manuals. And the ND symbols and delays have been observed on the real aircraft on the ground, and agree with engineering manuals as well.


|-|ardy

[emerydc8]

Okay. I think I found the answers in John's chart below. After 35 seconds and no resync (and no pilot action), it goes to SINGLE FMC OPERATION followed by an EICAS FMC L. Is the RESYNCING and SINGLE FMC OPERATION reversed in John's chart? In one place, it says SINGLE FMC OPERATION then RESYNCING OTHER FMC for the right CDU scratchpad message.

I ask about the ND (VTK/MAP) delay because I know in our sim, there is no delay in displaying VTK after an FMC fail. In fact, they teach that this is usually the first sign of an FMC failure. I also noticed that the Bulfer book seems to support this (p.22.1 Jan. 1999). It would be nice to see the video again.

FMC FAIL
FMC FAIL and HSI MAP FAIL and VTK messages will be displayed continuously during the thirty second resynching process.

[Hardy Heinlin]

I ask about the ND (VTK/MAP) delay because I know in our sim, there is no delay in displaying VTK and MAP after an FMC fail. In fact, they teach that this is usually the first sign of an FMC failure. I also noticed that the Bulfer book seems to support this (p.22.1 Jan. 1999). It would be nice to see the video again.

I missed this detail, or something was different on the real aircraft. I can't remember.

I'll try to modify it this month ...

[emerydc8]

Thanks, Hardy.

As far as the message in the right CDU scratchpad, should it be like John's chart where it says SINGLE FMC OPERATION first and then RESYNCING OTHER FMC after that? It looks like the chart may be reversed. I don't remember if Peter got this part in his video.

[Hardy Heinlin]

They appear nearly at the same time, but RESYNCING OTHER FMC overwrites the other messages; it suppresses the SINGLE FMC OPERATION message, otherwise, during single FMC failure detection, you would never see the RESYNCING OTHER FMC message.

When RESYNCING OTHER FMC disappears from the message stack, the next message in the message history becomes visible, which in this case must be SINGLE FMC OPERATION for obvious reasons. As the single operation continues, this message can only be cleared by the CLR key. RESYNCING OTHER FMC cannot be cleared by the CLR key. It clears automatically when the resync attempt is done.

The chart agrees with PSX: In PSX, SINGLE FMC OPERATION is visible for ca. 2 seconds, then RESYNCING OTHER FMC appears which cannot be cleared (all FMC keys are inop). When the resync is done, RESYNCING OTHER FMC disappears. If there are still single FMC operation conditions, SINGLE FMC OPERATION re-appears (no longer overwritten by RESYNCING OTHER FMC).

[emerydc8]

Got it. Thanks! I missed the initial appearance of SINGLE FMC OPERATION.

I was just looking at the Smith FMS manual for the 737-900. I know they are different animals than the 744 but I thought it was interesting that VTK doesn't appear at the time the ND map blanks. I don't remember if the map blanked right away on our -400 sim, but I do remember seeing VTK immediately.

If the annunciator is lit due to an FMC failure, the following additional indications are provided:
− Dual configuration:
− Right FMC fails:
• The VTK flag is displayed on right ND and the map is blanked after approximately 30 seconds.

If the annunciator is lit due to an FMC failure, the following additional indications are provided:
− Dual configuration:
− Right FMC fails:
• The VTK flag is displayed on right ND and the map is blanked after approximately 30 seconds.
• SINGLE FMC OPERATION message is displayed in both FMCS display scratchpad areas.
− Left FMC fails:
• The VTK flag is displayed on left ND and the map is blanked after approximately 30 seconds.
• The FMCS display FAIL annunciator is lit and FMC is displayed in the screen (CDU Only).
• The FMCS display displays the MENU page with the FMC prompt (LSK 1L) blanked (MCDU and FANS
MCDU).

[Britjet]

We often have to reset the FMCs in the big sim after a lot of repositionings (the sim sometimes gets confused).
In this case, after pulling the relevant FMC CB, the Amber map indication on the ND takes quite a while to appear (probably 30 seconds), and only then do we reset the breaker. The sim engineers told me that unless you wait for the MAP message the thing hasn't totally shut down.
HTH.
I will see if I still have the video, but I had a clear out recently!
Peter.

[Hardy Heinlin]

Just to be clear: I won't remove the 30 sec delay for the MAP. I'll just remove the delay for the VTK flag and the active waypoint's ETA & DTG -- those will disappear immediately.

These 30 seconds refer to the volatile FMC memory (INIT REF stuff, i.e. POS REF, PERF etc.) and the volatile CDU memory (standby nav). The data will vanish after 30 seconds. (While the nonvolatile magnetic-bubble memory stores the databases and routes only).


|-|ardy

[Britjet]

I think my memory is volatile and has quite a lot of bubbles...

[emerydc8]

Just to be clear: I won't remove the 30 sec delay for the MAP. I'll just remove the delay for the VTK flag and the active waypoint's ETA & DTG -- those will disappear immediately.

Thanks, Hardy. That would comport with John's chart above too. The VTK flag appears, plus the ETA and DTG for the active waypoint disappears. If there is no resync after 35 seconds, then you get the MAP flag. I presume that's when the map (magenta line) disappears too?

Peter, thanks for checking on the video. I hope you can find it.

Jon

[Hardy Heinlin]

I presume that's when the map (magenta line) disappears too?

Yes.

[emerydc8]

Thanks!

[Hardy Heinlin]

In PSX update 10.1.1, in case of an FMC failure, the amber VTK flag and the removal of the active waypoint's ETA and DTG will now occur after 1 second instead of 30 seconds:

http://aerowinx.com/board/index.php?topic=4020.0

(Refer to item 1.1.009.)


Regards,

|-|ardy

[emerydc8]

Thanks Hardy!

Jon