On 6/25/19 11:33 AM, Keith Packard wrote: > That's not what we're waiting for -- we're just compensating for the > inaccuracy in computing apogee based on acceleration alone. We don't > have a lot of precision in measuring acceleration because we need such a > large range, that means there's a significant error in each measurement, > leading to (potentially) large errors in speed computation and missing > apogee by many seconds.
I have seen this concern before and I find it puzzling as it isn't really a problem. The now ancient BlackSly AltAcc used an 8 bit ADC and worked quite well. Digging out an handy data set I see it reporting 3.1387 GHarrys / G. (GHarry being an ADC count.) > >> What about Drew's concern over the lack of integration over the three >> axis of accel? > > Yeah, having a full 6dof state computation would be 'better' in some > way, but we still can't get much precision in the z-axis acceleration > value, so I don't know how much more accurate apogee detection would be. > If there is considerable error in alignment between the sensor axis and the vehicle axis then you might want to do this. Or just align them. (Made harder by sensor to package and package to PCB alignment errors.) >> Can we assume that as the rocket arc's over the change in acceleration >> will become detectable and thus trigger apogee detection? Is there a >> set threshold for change in acceleration for the algorithm to detect >> apogee? > > Uh, I think you're missing something here -- there's no change in any > acceleration values across apogee; the rocket trajectory is ballistic > precisely because the only large force acting on it is gravity, which is > constant. > Accelerometers cannot measure gravity because it works on all parts of the rocket/sensor equally. On the pad you see the acceleration from the ground holding the rocket up. Near apogee there are only a couple of things it could measure. One is aerodynamic forces and the other matters only if the rocket is rotating. -- https://web.archive.org/web/20190214181851/http://home.earthlink.net/~david.schultz/ (Web pages available only at the Wayback Machine because Earthlink terminated that service.) _______________________________________________ altusmetrum mailing list [email protected] http://lists.gag.com/mailman/listinfo/altusmetrum
