Hi Rudderrudderrat
You make your point well; I was really just pointing out a rather interesting anomaly regarding the architecture of the ADIRU, rather than perceived question regarding reliability. (My head is swimming a little with your maths, but I think I see your point 
). I agree that the chances of a failure of two same axis RLGs OR accelerometers is extremely remote; I was merely making the point that CERTAIN dual component failures could in theory render the ADIRU useless. BUT, there still maybe something that Honeywell are not telling us, regarding RLG or accelerometer design that blows my argument completely out of the water. But at the moment, as I read it we still have 2 X, 2Y and 2Z axis RLGs/accelerometers; as I said before, I may be missing something here. 
Last edited by M2dude; 28th Jan 2010 at 15:50.