Menu Sign In Contact FAQ
Welcome to our forums


I was flying some weeks ago (Cirrus SR22 G2 with Avidyne PFD and DFC90 autopilot) ; everything working OK in a climb using Auto Pilot while Autopilot went off with the normal loud beeps having this message in the PFD:

The preflight had been OK and the autopilot had engaged normally.
The PFD didn’t seem to fail. Everything kept working OK… it’s just like the Autopilot lost connection with the PFD AHRS data (DFC90 relies on the AHRS information from the PFD so if that is missing, the autopilot becomes unavailable), as the AHRS in the PFD was working OK: I tested that using the secondary analog instruments. Also, no red crosses or other messages. Just that AUTOPILOT INOP AHRS FAIL that is a direct message from the DFC90.

I cancelled my IR flight (it was a IR training session) and we went back to my home airfield… I turned everything off… restarted avionics… but even though the AHRS finished calibration normally, the message was still there so we couldn’t resume the flight.

I have access to a spare Avidyne PFD wich I asked to be installed in order to test everything some days later. After replacing the PFD, the failure didn’t occur so although at first it may look as it is a PFD problem, I have read similar issues in the Cirrus forum related to interferences and other components involved in this problem (like the magnetometer).
(Since then, problem has ocurred twice with one PFD, it never arises with the other)

First step will be performing a Magnetometer recalibration (that is a software thing to be done in the PFD, I have not re-calibrated yet after the issue) but I am a bit lost after that… Any suggestions on how to diagnose if it does not solve it? I have the feeling that the PFD is OK and sending it to Avidyne USA will not solve the problem. (At will be very costy).

(And yes… Avidyne dealers here have no clue about the problem. They will just suggest packing the PFD, charging 600€ for that, and sending it to USA)


LEMT, Spain

That’s a pain.

I have the same setup as you combined with 2 × IFD440.

I don’t have any new information to help you. I can only share my own experience with you so you can compare notes.

  • A few Cirrus instructors told me to always wait for the alignment to be complete (on startup) before pressing “Enter” on the IFD’s. This is, apparently, the “right order”. I’m not even sure what the connection is, to be honest.
  • The plane should not move while doing the AHRS initial checks
  • I believe I saw that message once on the ground while testing the AP. I cycled the systems, and all was ok.

Again, I’m pretty sure I’m not telling anything new… I hope you manage to find out without spending a small fortune. I’ve got the Avidyne insurance for situations like this.

EGSU, United Kingdom

That “order” does not surprise me. I have the SG102 AHRS (actually have done a lot of work with this box, having repaired a couple for ARINC429 development purposes) and if it isn’t sitting quite still during startup, the software gets into a mess from which it may never exit. See the “avionics internals” thread for more fun

Shoreham EGKA, United Kingdom

@Peter, the “Avionics internals” thread should receive a Grade 2 listed status :) I didn’t know about its existence until you mentioned it. It’s such a fascinating thread.

Your SG102 photos/post are really interesting. The EX5000 PFD displays a message asking the pilot not to taxi until the process is complete. I’m starting to understand why now.

Last Edited by Fernando at 21 Feb 11:41
EGSU, United Kingdom

Yes; the AHRS boxes use a classic 3D navigation approach, dating back to the Polaris missile, 1950s. 3 yaw gyros and 3 accelerometers, and you feed all this into an algorithm which produces various outputs like pitch, roll, and heading stabilisation for the fluxgate. High grade (low drift) versions (using fibre optic gyros, or mechanical gyros) give you full navigation in 3D by integrating acceleration to get velocity and then integrating it again to get distance. You can get really weird effects with these boxes if you move them too soon after startup, or even after startup if you move them “too fast”. Or move them in a way which confuses the algorithm.

The box outputs a “valid” signal after a warmup (some minutes) and it looks like whatever it is feeding is not able to deal with that being invalid if the aircraft is being moved.

Shoreham EGKA, United Kingdom

Avidyne do a terrific warranty – for my R9 it was about 5k for three years. Worth every penny. I’d highly recommend it.

If only I’d known that….
EG.., United Kingdom

5k for 3 years being terrific value must be the best example of Stockholm Syndrome in 2024 so far

This is also worth a read.

And this.

Also if 5k/3 years was good value, the kit must have spent a big chunk of its life broken.

Shoreham EGKA, United Kingdom

Well it works very well for me, but I guess I’m an anathema then. I like stuff to work when I want it to, and when it doesn’t I like others to fix it rapidly. It depends how you value your time.

If only I’d known that….
EG.., United Kingdom
8 Posts
Sign in to add your message

Back to Top