# Are solid-state attitude indicators subject to gimbal lock?

With all the shiny new glass cockpits it would seem that the days of the spinning mechanical gyro (and associated tumbling due to gimbal lock) should be over: Sparing everyone the boring math it should suffice to say that solid-state gyros can be engineered and built in such a way that gimbal lock is impossible, but I'm not certain that's how they're actually designed.

Do modern AHRS systems with solid-state gyros (or replacement electronic horizons like the RC Allen 2600 series) still suffer from gimbal lock, or do they provide true 3-dimensional freedom?

I'm interested primarily in answers from a light General Aviation standpoint, but answers about electronic gyros on commuter and transport category aircraft would be interesting too.

-

AHRS cannot suffer from gimbal lock (as there are no gimbals that can share axes):

Conventional gyros are also susceptible to gimbal lock under certain conditions. The AHRS is an all-attitude system and is free from such problems. — from A Layman's Guide to AHRS

That said, AHRS systems can still be tumbled - I know that for a while an air show routine was flown in a (then) Columbia 400 with G1000; the pilot remarked that the system Xed out only briefly while executing inverted maneuvers. In comparison, the Avidyne AHRS at the time required a full 2+ minutes of straight and level flight before resetting itself.

-
The G1000 has built-in attitude limitations beyond which it will not read correctly. I will try to look them up next time I'm near a technical manual. But like you said, since they can't tumble it's entirely possible to make an AHRS capable of accuracy at any attitude. –  StallSpin Feb 11 at 18:53
Yeah, I don't recall what they are either - I'll try to take a look at the manual I've got at home this evening. –  egid Feb 11 at 19:26
Specific attitude limitations would lead me to believe they're using X/Y/Z (Euler) coordinates directly instead of deriving them (so "It's not gimbal locked" (because there are no gimbals), but the effect is largely the same (big red X on the attitude display when you exceed the limitations until the gyro can be reset)) –  voretaq7 Feb 11 at 20:23
Hm, I still don't understand how they get the limits. The gyros are always orthogonal, so the input has the necessary information at any attitude. Rotating the accumulated vectors is possible in any representation given correct formulas and with incorrect formulas (e.g. just adding the pitch axis to the inclination directly) will noticeably loose precision even at shallow bank angles. So it seems strange to me they could get away with incorrect formulas (and the "correct" formulas are not that complicated anyway). –  Jan Hudec Feb 11 at 20:35
@JanHudec the G1000 AHRS also uses the magnetometer for some information (it's part of why the in-flight realign takes so little time) so it's possible that outside of a certain realm the data isn't verifiable between the sensors. A dual AHRS system might be able to compare the two, but what most GA aircraft have maybe cannot? –  egid Feb 11 at 23:57