State of the art in stable accelerometers?

Tauno Voipio wrote

I thought laser gyros were effectively strap-down gyros - something counts the pulses from the diffraction of the beams or whatever, and the pulse count is counted to give an integrated velocity value directly.

Sure, but we are still just talking about an artificial horizon, aren't we? The G1000 does no inertial nav whatsoever - other than pure GPS.

Funny that - I am an FAA CPL/IR but not CFII :) Always suprises me how many electronics / hardware / software people fly too. I will try to drop you an email.

There are also thermal accelerometers, which have a little bubble of fluid (air, probably?) with a heater surrounded by temperature sensors; the warmer (and therefore less dense) part of the fluid moves in the direction of acceleration. I don't know if these are any more accurate than the types you mention; I'm just mentioning them for completeness. :) I think their main advantage is shock/impact resistance.

I presume we're talking about an explosive-based mechanism to get to the 50000 g values.

Yes, you would likely rip everything out of the PCB with 50K G anyway. Ususally, you don't care much about the device after impact, as in ICBMs.

Everett M. Greene schrieb:

IMHO any X-g spec does not mean anything without a second parameter, like duration or way.

Imagine two pieces of extremely hard material touching eachother. Two bottles of beer, for example. Not much m/s, but a very small term 1/s, thus m/s^2 may reach a few thousands of g.

On the other hand, a 40tons truck smashing a compact car will hardly be exposed to more than 1g.

Falk

Artillery shells experience 10,000 to 15,000 Gs at firing, and they have put some pretty complex electronics in them.

Mark Borgerson

s;

. :)

drop a accelerometer 3 feet on a hard surface will get you 10,000+ g's and kill most mems accelerometers