We've been flying and testing v2.74b and v2.76 in a number of aircraft over the last three months namely two X8 flying wings, a Skywalker Condor 1880, a RvJet, a Multiplex Easyglider v2 and a 1.7m span twin boom own design pusher.
We have problem with the X8's and the RvJet with loss of altitude in turns. I had quite a bit of correspondence with Paul Riseborough one of the developers, to try get to the bottom of this and we fixed a number of issues but our problem never went away.
The graph attached from a flight with the Easyglider illustrates what I believe to be the problem, what I can't understand is the seemingly conflicting information the graph shows
.
It seems to me that the plane is [i]climbing[/i] according to the increasing altitude and reducing altitude error but it is [i]pitching down[/i] according to it's attitude.
Am I seeing things, reading the graph wrong or is there really a problem here? The graphs from the other aircraft are show a similar pattern.
The Easyglider being a very forgiving airframe seems to handle this issue OK whereas the X8 particularly struggles to the point of either exceeding "VNE" or descending dangerously low.
Please can someone assist in getting to the bottom of this?
Tlog is here: https://dl.dropboxusercontent.com/u/21460892/2013-12-06%2010-40-15Easyglider.tlog
Droneshare log here: http://www.droneshare.com/view/d7tar4n
Replies
I'm having trouble as well maybe I don’t understand TECS all that well but it seems there is an couple of people battling with this, I did all the basics on Roll and Pitch PIDS etc. and yes maybe my Pitch PIDS is not all that perfect but speed control on my Phantom flying is an disaster.
My TRIM_ARSPD_CM is at 17m/s FBW, ARSPD_FBW_MIN = 15 and ARSPD_FBW_MAX = 22 and it is just not maintaining those speeds. Altitude control with TECS_SPD_OMEGA set at default of 1 as per first graph is not that bad but as per 2nd graph I pushed it to 2 with more error on alt as expected but airspeed is still not good.
I noticed that during an climb in Auto mode in an straight line the airspeed would rise initially very high and then towards the target alt would drop to well below FBW_MIN. I know I can set the FBW MAX lower but that would not be in favour of Auto take off or raise TECS_CLIMB_MAX but that is not what I want either.
APM2.5 controller
Mediatek GPS
MPXV 5004 Airspeed
Flight with TECS_SPD_OMEGA at default = 1
OMEGA at 2
Sorry I lost the RLOG but here is the TLOG and Parameters.
2013-12-21 16-44-29.tlog
2013-12-21 16-44-29.param
Similar problem with two Skywalker, 168 and 1900.
OK, I think I may be reading the graph wrong :oops: . But we're still getting these large variations on height and pitch during the flight. Here's a smaller log from a Skywalker 1880 in a loiter in which there is a 22m variation in altitude during the loiter.
2013-12-18 14-10-37Easyglider_loiter.tlog