G3X went haywire!

[PeytonM]

On a flight yesterday, about 10 minutes into the flight with the GFC500 flying, climbing to and level at 7000 for about 3 minutes, my G3X generated error messages, showed chevrons to climb, A/P beeped at me, but did not disconnect. I was level at 7000. I disconnected and reconnected A/P a couple times with no change. I finally pulled the PFD CB, waited a few seconds, reset it.  All good for the remaining 1 hr+ of the flight. I landed and took off, everything worked perfectly. No turbulence. Pics below. 

[rklems]

System is showing attitude & heading miscompare errors, and in the second picture there is a significant attitude difference between the G3X display and the G5 display (there might be in the #1 & #5 pictures too, but they don't show the G5 to compare to). In that situation, which one appeared to be correct? There are essentially 3 AHRS in the system, the GSU25 with the G3X, the internal one in the G5 and the one in the GMC507 that is used for cross checking. It looks like you might have had a data card in the G3X, do you have the data recording enabled?

[Rick Junkin]

27 minutes ago, PeytonM said:

On a flight yesterday, about 10 minutes into the flight with the GFC500 flying, climbing to and level at 7000 for about 3 minutes, my G3X generated error messages, showed chevrons to climb, A/P beeped at me, but did not disconnect. I was level at 7000. I disconnected and reconnected A/P a couple times with no change. I finally pulled the PFD CB, waited a few seconds, reset it.  All good for the remaining 1 hr+ of the flight. I landed and took off, everything worked perfectly. No turbulence. Pics below. 

It appears your ADAHRS (GSU 25) faulted. It strikes me as odd that your second picture is the only one that shows a mis-match between the G3X and G5 attitude displays yet there are no advisories displayed. Can we assume your G5 looked correct throughout?

The flight data file from your G3X will provide more data to help determine the nature of the fault. Garmin will want to see that first off.

[Rick Junkin]

@PeytonM Thinking through this a bit more. If it were to happen again and you can confirm the G5 is good you might consider pulling the breaker to the GSU 25. That would force a reversion to the G5 ADAHRS and keep your G3X display. The issue with doing this is that the GMU 11 (magnetometer) is on the same circuit as the GSU 25 so your heading would start to drift. Resetting the breaker of course will bring both back online.

Having said that, you had the AHRS REVERT advisory indicating the switch from the GSU 25 to the G5 had happened. Cycling the breaker for the GSU 25 would allow you to reset it while retaining the PFD with valid ADAHRS data from the G5..

 

[PT20J]

Not sure that the pictures are in the correct order. It appears that there was an issue with the GSU 25 as others have stated. The G3X continuously compares the GSU 25 AHRS with the G5 AHRS. The comparator flagged an attitude miscompare and reverted to the G5 AHRS. Eventually, it also flagged a heading miscompare. Reverting should cause the autopilot to disconnect -- at least it has always done that when I test failed the GSU 25 by pulling the CB. But perhaps it doesn't if the reversion was caused by a miscompare. It seems to have corrected itself after you rebooted the G3X. It may never happen again, or it may be the beginning of something failing, so I'd watch it. I would also pull the SD card and send the log file for the flight to Garmin and have them analyze it. One problem with a miscompare is that it may not always be obvious which AHRS is generating erroneous data and it's not clear that the logic in the G3X can always sort it out. This is why I added a AV-20S to my panel to have a tie breaker.

The yellow heading indication in the last picture indicates that the heading input is valid but may not be accurate. 

[PT20J]

23 hours ago, Rick Junkin said:

The issue with doing this is that the GMU 11 (magnetometer) is on the same circuit as the GSU 25 so your heading would start to drift.

If you pull the GSU 25 CB, or the GMU 11 fails, the G5 will revert from HDG to TRK. Interestingly, if the GMU 11 fails, the G3X does not revert and it's HDG indication will drift. I know this because we put my GMU 11 on a separate breaker so if the GSU breaker is pulled I still have the GMU.

[Ragsf15e]

6 hours ago, PT20J said:

If you pull the GSU 25 CB, or the GMU 11 fails, the G5 will revert from HDG to TRK. Interestingly, if the GMU 11 fails, the G3X does not revert and it's HDG indication will drift. I know this because we put my GMU 11 on a separate breaker so if the GSU breaker is pulled I still have the GMU.

Is there an additional adhrs in the ap control unit that is also compared against the g5 and gsu25 to help figure out which one is bad in a miscompare?

[PT20J]

1 hour ago, Ragsf15e said:

Is there an additional adhrs in the ap control unit that is also compared against the g5 and gsu25 to help figure out which one is bad in a miscompare?

The GMC 507 has a AHRS. As I understand it from an email from Trek Lawyer, it compares with the currently active AHRS (either the GSU or the G5) and if there is an internal miscompare it will disconnect the GFC 500. It does not annunciate the miscompare on the G3X screen, but the error will show up in the logs.

[Ragsf15e]

1 minute ago, PT20J said:

The GMC 507 has a AHRS. As I understand it from an email from Trek Lawyer, it compares with the currently active AHRS (either the GSU or the G5) and if there is an internal miscompare it will disconnect the GFC 500. It does not annunciate the miscompare on the G3X screen, but the error will show up in the logs.

Interesting.  If that’s the case, when miscompare is annuciated, you might not know which is bad in- g5 or g3x.  That makes your av20 even more important!

[Rick Junkin]

12 hours ago, Ragsf15e said:

That makes your av20 even more important!

Indeed. I put one in my panel as well to replicate the “peanut gauge” in the legacy fighter panels. Totally independent of the Garmin collective and roughly right in a pinch.