<#part sign=pgpmime>
On Sun, 01 Apr 2012 20:18:12 -0700, DK Duncan <[email protected]> wrote:

> Right, but the telemetrum has an accelerometer as well as a pressure
> transducer.

TeleMetrum logs all of its raw sensor data, which means we can use that
to check the algorithms on the ground. We used all of this data to
verify that the Kalman-filter based mach-inhibit algorithm correctly
detects when the airframe is experiencing mach transition effects.

So, in effect, "TeleMini" has flown faster than mach dozens of times,
but the code was run on recorded data and not in actual flight.

What the flight computer does is compare the barometric data with the
obvious physics model for a rocket in ballistic flight by looking at the
error term in the Kalman filter. When they agree, the computer arms the
apogee charge. When they don't agree, the altimeter inhibits the apogee
charge. This error term varies by a factor of 100 between ballistic and
non-ballistic flight, which gives us a strong signal for this technique.

> I'll be pushing mach 1.5 this summer and I want redundant altimeters.
> The telemini should be a good way to do that provided it will handle
> mach transitions well.
> 
> I'm hoping someone here has tested it in flight at those velocities.

I've got two near mach flights, one to 317m/s (as measured by Tm), and
one to 325m/s (as simulated; the rocket was lost due to a power failure
at apogee). Tm worked flawlessly in both cases.

We'll be doing a more testing this year, to see if there's any issue
with the Kalman filter not detecting mach transitions correctly. I just
need to make some time to build more little airframes for testing with
TeleMini hardware.

We could obviously add a mach delay setting to Tm's configuration; I
suspect it might make people more comfortable than trusting some fancy
computer software. 

-keith
_______________________________________________
altusmetrum mailing list
[email protected]
http://lists.gag.com/mailman/listinfo/altusmetrum

Reply via email to