You can use 2 more of these sensors and look up and down as well. Now
the software for that is not complete but that gives you a pretty good 3
axis position.
Ardupilot did use 3 axis sensors and the really good thing about it is
it did not require recalibration to re-aquire position after a fault. It
would be easy to do this in the gen one Ardupilot (I don't think you
even need to rewrite any code to do that). Which is very desirable.
"sun sensor and a vector magnetometer definitely suffices" You know I've
read this and I've heard it talked about but why have I never seen
amateurs use this?
There is no need for that now though with these Ir sensors and some
modern programming on a simple chip this thing should cost 50 bucks.
Get one and try it on a little rocket. It will knock your socks off.
I think we started testing it on a 2 1/2 inch sugar motor rocket.
I never tried the pitch control (tilt table) but it would only work
without a mag for a few seconds and not very well probably. But you can
use that gen one Ardupilot and add the mag and better algorithms. That
one is DCM (Direction Cosine Matrix) Which works pretty good. But the
third-gen Ardupolit has Kalman filtering.
Merry Christmas to all.
Monroe
The big barrier for me with the Ardupilot is rewriting the code to use a
frame of reference other than that of earth and I'm just not up to that
myself (yet). But you can do almost everything else you could want with
it.
-------- Original Message --------
Subject: [AR] Re: Spin stabilized rocket
From: Henry Spencer <hspencer@xxxxxxxxxxxxx>
Date: Mon, December 24, 2018 7:15 pm
To: arocket <arocket@xxxxxxxxxxxxx>
On Mon, 24 Dec 2018, David Summers wrote:
With this and a magnetic direction sensor, you could probably get easy
instant direction fixes in 3 axis.
If you're launching in daytime, a sun sensor and a vector magnetometer
definitely suffices, at all altitudes -- a number of our satellites use
that combination for attitude control -- if you avoid times/places where
the Sun axis is lined up with the magnetic field. (Note that the magnetic
field in temperate latitudes generally is *not* horizontal -- normal
compasses are only set up to measure the horizontal component, but that
doesn't mean there isn't a vertical component too.)
Henry