I wonder if anyone has experienced similar and what the resolution was?
We have dual Garmin GI275’s installed along with the GMU11 magnetometer to provide heading. We have noticed that the heading will “drop out” from time to time with HDG being suppressed with a red X. This is normally for 5 seconds or less and recovers quickly to an accurate reading. This seems to be most prevalent during turns through greater than 90 degrees
Most significantly, if you continue to make turns in various directions heading does not then recover and this will eventually cause the PFD/ADI to declare it’s AHRS solution as “bad” and switch to reversionary mode. (At least we know this works!)
This doesn’t sound a loose connection type issue as the circumstances are quite specific. The GMU11 is mounted sensibly on the opposite wind to the pitot, landing and taxi lights.
Just a WAG but is there a big ferrous balance weight in the aileron? ISTR Garmin specifies >1m away from ferrous structure up to 1kg and >2.5m for >1kg.
Raiz wrote:
Just a WAG but is there a big ferrous balance weight in the aileron? ISTR Garmin specifies >1m away from ferrous structure up to 1kg and >2.5m for >1kg.
That is definitely something check. Logically if that were the case it would be possible to duplicate the issue while stationery by making full control deflections. No one has spotted anything anomalous during “full & free” checks so probably not the issue, but will go and try it in case it just hasn’t been noticed.
I would suggest to report it to Garmin.
BR
Luis
not sure if you have a certified version, and who performed the install. In any case, a magnetometer calibration could be performed to try to TS the system.
As @Raiz mentioned, there could be some interference, though it should have been discovered during the initial calibration.
Also, from my exp G5 manual:
Sending Troubleshooting Data to Garmin
To assist with troubleshooting an issue, it may be helpful to send pictures and/or the log files stored on the microSDTM card to Garmin. This data can be especially useful to troubleshoot autopilot, attitude, and air data related issues. This data can be sent to [email protected] for troubleshooting assistance.
Thanks all. I was already coming to the conclusion that a recalibration might be a good idea. I can’t find any evidence that the engine was running during initial calibration. Not sure if this would result in what we are seeing, but would seem contrary to the specified procedure.
That’s funny. My Sandel SG102 started doing this, a year or two ago, so I swapped in another unit. The one which was doing it still does it on the bench but it is hard to reproduce, and happens rarely in the aircraft.
I think these “low end” AHRS systems just have weird vulnerabilities in them. For example I reckon that if the aircraft is lined up when last parked precisely on a 180 heading, the AHRS may not power up 
I suspect you are right Peter. There is an SB that we need to apply to the units which improves the AHRS model/logic but seems to be more specifically referencing loss of pitot input, however one will never know what else was revised. Unfortunately the first attempt to apply this failed and is waiting on a solution from Garmin. In the meantime I wanted to try and get to the bottom of this curiosity before throwing in another variable of an update.
