|
-
POSTING RULES

-
Donate yearly (please).
-
Advertise in here!
-
Today's Posts
|
Insert Pics
|

08-08-2016, 11:08 AM
|
|
|
|
Join Date: Nov 2006
Location: Colorado Springs
Posts: 27
|
|
Maybe this write up will help someone else...
About a year ago during VFR flight, one of the dual AHRS rebooted resulting in EFISes displaying divergent flight information. I can totally see event could really become a fatal distraction if it ever happened at just the wrong time.
My panel includes 3 GRT HX EFISes driven by Dual AHRS and dual magnetometers. The electrical system is Z14 architecture (dual alternators and dual batteries). EFIS displays can be swapped but are normally set so that EFIS1 shows Primary Flight information, EFIS2 Map, and EFIS3 Engine. AHRS1 goes to the Primary flight EFIS1 display and AHRS2 goes to the EFIS2 display.
. . .
The first symptom that something was not right was "Dual AHRS" indication replaced by "AHRS x" which means something took a powder and only one AHRS was on line.
The Primary flight EFIS FROZE showing pitch up and roll left (takes several minutes to come back on line). But would you really trust it afterwards?
Ok, no problem. Just switch the backup EFIS2 (with AHRS2) to display Primary flight information. It showed good (?) data but it still disagreed with EFIS1.
It was SO TEMPTING to troubleshoot. The glaringly obvious problem was that neither EFIS could be trusted, so you REALLY need a tie breaker. Thankfully my panel includes backup AS, Alt and Attitude steam gauges. It wasn't difficult to switch over to steam gauges, but the EFISes were a huge distraction.
I'm very thankful this happened in VFR conditions and I learned a few valuable lessons.
1) Assume something will break and you will never be disappointed
Although my panel is very capable, I had assumed that my fancy multi-display, dual everything gizmos would always be there. Not so much!
2) Practice
Backups are worthless if you don't practice using them. We have SO much packed into EFIS displays. It's really distracting if one fails and there's little time to troubleshoot if in the soup. Even problems in VFR conditions can be equally fatal if they happen at the wrong time.
3) Remove the distraction
Do EFISes include an OFF switch? Mine don't and I didn't add any. Note to self, if an EFIS appears to fail, switch it to display that shows something else, such as engine parameters, or simply cover it up so there's no temptation to fixate on it.
Conclusion
After a good deal of troubleshooting on the ground, the reboot issue was traced to an AHRS power crimp. I fixed the crimp and then added a SECOND power wire input to each AHRS.
I now regularly practice switching to backups in flight... and have adjusted expectations that something will break regardless of every effort to prevent problems. Oh, and a totally low tech wet compass is still your best friend in case everything if all else fails.
Regards,
Jay
RV10 N433RV
|

08-15-2016, 12:24 PM
|
|
|
|
Join Date: Mar 2012
Location: Wichita Falls, TX
Posts: 315
|
|
Why don't these systems incorporate the ability to compare 3 AHRSs? So that if 1 shows different, but 2 agrees, it will still give you an AHRS error message but will continue to follow the 2 that are matching.
__________________
Scooby
Harrumph!!
RV-8 IO-390 N788MT
|

08-16-2016, 11:15 AM
|
|
|
|
Join Date: Jan 2005
Location: Hazlehurst, GA
Posts: 1,359
|
|
In agreement with Vic!
IMHO.......
Independent redundancy in technology via different manufacturers along with a minimum of a VSI, airspeed indicator and an altimeter has always been my preference. Whether GRT/Trutrak/Garmin or Dynon/Trio/Garmin, as I've used in the last two RVs I've built, these 2.25" instruments were MEL items for me personally and always will be.
With a myriad of technology possibilities, training in the use of all features and, more importantly, understanding the partial failure modes and the partial failure operation options, are as important, if not more important than any hardware redundancy.
While information presentation is important, information overload seems to be an issue, as well. Decades before EFIS/MFD/PFD, my instructor would routinely cover portions of the panel, reminding me how pioneers flew the world. It was training and practice that made the difference, no matter the circumstance.
Prayers for the lost souls in the Lancair. I hope other lives are saved by the lessons of this tragedy.
__________________
IHN,
2020 Dues Paid
Robby Knox
THEM: Why do you always carry a knife?
ME: I can't open a bag of chips with my Glock!
|

08-16-2016, 02:11 PM
|
|
|
|
Join Date: Dec 2011
Location: Livermore, CA
Posts: 6,797
|
|
Quote:
Originally Posted by BHunt
Why don't these systems incorporate the ability to compare 3 AHRSs? So that if 1 shows different, but 2 agrees, it will still give you an AHRS error message but will continue to follow the 2 that are matching.
|
Because, in the case of the OP, there were only two AHRS systems, not three.
As the OP pointed out, he had very good round dial backups; you just have to be willing to use them, and include them in your scan.
Last edited by BobTurner : 08-16-2016 at 02:15 PM.
|

08-17-2016, 01:08 AM
|
|
|
|
Join Date: Mar 2012
Location: Wichita Falls, TX
Posts: 315
|
|
Quote:
Originally Posted by BobTurner
Because, in the case of the OP, there were only two AHRS systems, not three.
As the OP pointed out, he had very good round dial backups; you just have to be willing to use them, and include them in your scan.
|
Yup, I'm aware that it only had two. I was asking why there isn't a 3 vote logic to use 3. It seems (from my MINIMAL research) that EFISs will compare two, but not three
__________________
Scooby
Harrumph!!
RV-8 IO-390 N788MT
|

08-17-2016, 05:04 AM
|
 |
|
|
Join Date: Nov 2006
Location: Savannah, GA
Posts: 1,029
|
|
Quote:
Originally Posted by BHunt
Yup, I'm aware that it only had two. I was asking why there isn't a 3 vote logic to use 3. It seems (from my MINIMAL research) that EFISs will compare two, but not three
|
If I had to guess, they probably figure most builders are going to install an independent attitude unit vs. a third AHRS.
__________________
RV-7ER - finishing kit and systems installation
There are two kinds of fool in the world. The first says "this is old, and therefore good"; the second says "this is new, and therefore better".
|

08-17-2016, 06:13 AM
|
 |
|
|
Join Date: Aug 2005
Location: Dallas/Ft Worth, TX
Posts: 5,686
|
|
Quote:
Originally Posted by BHunt
Yup, I'm aware that it only had two. I was asking why there isn't a 3 vote logic to use 3. It seems (from my MINIMAL research) that EFISs will compare two, but not three
|
Bottom line is I don't think they have the technology to do it yet, at least in GA world, most likely due to cost.
I've asked the same question as many of todays EFIS system will support 3 ADAHRS.
__________________
Walt Aronow, DFW, TX (52F)
EXP Aircraft Services LLC
Specializing in RV Condition Inspections, Maintenance, Avionics Upgrades
Dynamic Prop Balancing, Pitot-Static Altmeter/Transponder Certification
FAA Certified Repair Station, AP/IA/FCC GROL, EAA Technical Counselor
Authorized Garmin G3X Dealer/Installer
RV7A built 2004, 1700+ hrs, New Titan IO-370, Bendix Mags
Website: ExpAircraft.com, Email: walt@expaircraft.com, Cell: 972-746-5154
|

08-17-2016, 12:18 PM
|
|
|
|
Join Date: Oct 2014
Location: Cedar Rapids, IA
Posts: 232
|
|
3 sources
I work on the big jets, and there are cases where we have 3 sources of the same data (think air data, inertial, and others). Sometimes even 4. To handle that correctly takes a lot of system engineering expertise. It seems like a simple task, but the devil is definitely in the details here.
To give you an idea, think about the different ways it can fail. If 1 signal disagrees with the other two, who agree with each other, thats pretty easy, just kick that one out. But what if they all disagree but by a small amount? Sometimes you get a very brief but high magnitude spike in 1 signal, those should be ignored. But what about a failure that looks like a slow ramp, a fast ramp, or oscillating at any number of frequencies/magnitudes? How do you determine what the acceptable level of disagreement is? A lot of things play into that like sensor accuracy, data transmission rates, comparison monitor execution rates, etc.
This is just a taste of signal management. People build careers on this topic alone.
|

08-17-2016, 01:54 PM
|
|
|
|
Join Date: May 2014
Location: Fort Worth, Texas
Posts: 436
|
|
Quote:
Originally Posted by unitink72
I work on the big jets, and there are cases where we have 3 sources of the same data (think air data, inertial, and others). Sometimes even 4. To handle that correctly takes a lot of system engineering expertise. It seems like a simple task, but the devil is definitely in the details here.
To give you an idea, think about the different ways it can fail. If 1 signal disagrees with the other two, who agree with each other, thats pretty easy, just kick that one out. But what if they all disagree but by a small amount? Sometimes you get a very brief but high magnitude spike in 1 signal, those should be ignored. But what about a failure that looks like a slow ramp, a fast ramp, or oscillating at any number of frequencies/magnitudes? How do you determine what the acceptable level of disagreement is? A lot of things play into that like sensor accuracy, data transmission rates, comparison monitor execution rates, etc.
This is just a taste of signal management. People build careers on this topic alone.
|
And I fly them (777) and you are exactly correct.
The system monitors so many different failure points, our manuals don't even cover/address most of them. When something happens, it's recognized and you get an alert and (usually) the system has already configured a workaround for you.
Those are the easy ones...
The hard ones are when the system can't figure out the source of the bad data and dumps it back into your lap for the decision...which then leaves it up to the pilot to look at the 3 different attitude/air data sources and determine which one is bad.
Hopefully, you aren't already inverted/stalled when that happens.
The books say while sorting these kinds of problems out, to try to remain in visual conditions while doing this if an attitude problem...or set a known pitch attitude/power setting if an airspeed one...
As we all realize, you don't get to pick when these failures occur...
Additionally, you're trying to sort all this out while usually getting inundated with EICAS messages/stall warnings/overspeed warnings/autopilot disconnects/etc.
It's a very challenging situation, even in air transport with a professional crew at the controls...as several very high profile accidents have shown.
We are getting MUCH more emphasis (and training) flying and troubleshooting these types of problems, but I am under no illusion that it would be anything other than one of the most difficult inflight situations I could ever be presented with...
As a single pilot operating IFR, that challenge is even greater...and we need to consider what level of risk we are comfortable with, potential failure modes, and train accordingly...
Rob S.
__________________
Rob Schroer
RV-7/N75WV
YIO-360-M1B
New Braunfels, Texas (KBAZ)
VAF Monthly Donor
|

08-17-2016, 01:56 PM
|
|
|
|
Join Date: Oct 2007
Location: Sacramento, CA
Posts: 3,351
|
|
Quote:
Originally Posted by jaybob
2) Practice
Backups are worthless if you don't practice using them. We have SO much packed into EFIS displays. It's really distracting if one fails and there's little time to troubleshoot if in the soup. Even problems in VFR conditions can be equally fatal if they happen at the wrong time.
RV10 N433RV
|
Could not agree more with this recommendation and my main reason against a different brand EIFS as secondary. One will not know it nearly as well in the time that really needs it. Steam gauges are my best option for redundancy, no matter how good one's electrical system is.
__________________
Mehrdad
N825SM RV7A - IO360M1B - SOLD
N825MS RV14A - IO390 - Flying
Dues paid
|
Posting Rules
|
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts
HTML code is Off
|
|
|
All times are GMT -6. The time now is 12:23 AM.
|