Here's the parameters file from my X-5. Note that there is a cross-track velocity error gain which is used by the SW mods that i am currently testing.
Cheers,
Paul
Tags:
Permalink Reply by paul riseborough on February 19, 2013 at 3:27am It works without the mod on 2.9, but there is a bit more overshoot on track capture.
Permalink Reply by Richard on March 20, 2013 at 2:34pm Hi Paul - how are you getting on with the x5? I am going to try your parameters file. I know a lot of people are experiencing nasty flight characteristics with this plane, have you experienced any?
Rich
Permalink Reply by paul riseborough on March 21, 2013 at 6:31pm It's not a nice flying airframe as it doesn't have enough weathercock stability (not enough vertical fin area). I have improved this my adding a fixed vertical fin made out of Depron which is fixed on top of the motor mount. It has the advantages of being cheap, compact and robust, but it doesn't like being flown too slow as it can stall one wing and roll out if you try slowing down with too much elevator. I've set the minium speed at 12 m/s and the maximum at 25 m/s and use 16 m/s for normal stooging around.
I would fly it in manual first and fly constant throttle circles and check your data logs to make sure that the airpseed is sitting in the middle of the ground speed (which will go up and down as it flies into and out of wind). You will then need to adjust the airpseed scaling parameter to calibrate out any airpseed error. It is also important that you fly in FBW-A before going auto and check that you get a resonable roll response without gaining or losing too much height. Your airframe will have different servo throws to mine so you will still need to fly it in manual first, get your trims and throws sorted, and then setup-up the radio throws and offsets in the MAV setup menu before trying FBW-A or AUTO.
Permalink Reply by Dwgsparky on April 3, 2013 at 5:25pm
Hi Paul
Will I need to modifiy this file for a standard X5 ina APM 2.0? or is it a good starting point for a new plane (X5)
Dwgsarky
Permalink Reply by paul riseborough on April 3, 2013 at 10:21pm You'll have different throws and different airspeed cal so you will need to modify. Follow the tuning process on the wiki -short cuts aren't advised.
Permalink Reply by Richard on April 8, 2013 at 1:39pm
Permalink Reply by Richard on April 9, 2013 at 12:31am Hi X5 params group - I am having trouble tuning my airframe for good performance in wind, and without wind. Here is a video to show the shocking performance I am getting (sped up by 5x).
If anyone can look at the PID file or perhaps give me some idea of what is wrong, then that would be great!!
I have also attached the .tlog that has a few FBW-A bits in i. (PWM 1500 is fbw-a)
Any help greatly appreciated!
Rich
Permalink Reply by Richard on April 9, 2013 at 12:34am Another .tlog with some fbw-a for analysis! If anyone can shed some light on what I am doing wrong i'll buy em a beer!
Permalink Reply by paul riseborough on April 9, 2013 at 1:25am You're trying to violate the laws of physics with your missionplan. At a ground speed of 25 m/s which is what you are flying at on the downwind turns, the absolute minimum turn radius achievable for a 45 degree bank angle is 64 metres,which means you should be spacing your waypoints 128m apart If you want it to capture the next track without excessive overshoot.
Also your model is out of trim in auto (demanded bank angle offset from measured bank angle) so you either need to enable auto trim or recalibrate the RC channel trims using the MAV facility. One way to improve this is to use a small amount of integrator gain (say 0.05) in the roll PID and allow 10 degrees of integrator action. This will enable the control loop to progressively compensate for the offset)
The tele rate is too slow to make an assessment of your roll PID tuning - you need to enable the ctun logging onboard your APM and pull the log files off afterwards.This will give you demanded and achieved roll data at 10Hz.
You also have a tuning problem with your speed and height control loops. Before you start tuning those it is important that you adjust your thrust line so that when flying in manual mode with the model trimmed, if you advance the throttle, the models climbs without picking up a lot of speed (you want throttle to change climb rate, not the speed which is different to the way a lot of RC models are set up). The X-5 is very sensitive to thrust line.
Permalink Reply by Richard on April 9, 2013 at 4:46am Hi Paul, cheers for the detailed analysis and information. I have read through it a few times, and I think I understand what to do in order to help solve the problem.
I will begin the changes this week, and hopefully get some better results! I have a learnt a lot from the information you provided,
Owe you a beer!
R
Permalink Reply by Richard on April 9, 2013 at 3:15pm Violating the laws of physics is never a good thing! Having read your information in more detail, I have a few questions..
I am using manual level in the missionplanner. It is much easier to level the plane on the dining table than it is to level at the field. I am a little concerned about the fact that manual level says 'for advanced users only'. Is it needed? or can APM determine the optimum angle the plane needs to fly at? It is difficult to know when the x5 is level.
Regarding thrust line, I used the included wooden washers on the top two bolts of the motor mount, which gives the plane some down thrust. But it seems to want to climb no matter how much throttle i apply, Even when i set the reflex to 0. But I this sounds like a good thing. How much downthrust have you added?
Regarding the out of trim issue. the plane hasn't been crashed badly and the airframe is in very good condition, the horizontal balance of the plane is perfect, all the equipment balances in roll. Could this be because i am using manual level?
I am going to do the following to try and get this thing on rails...
The info on integrator gains and using mavlink for rc channel trim went over my head a little, so i'll try auto-trim to start with.
I am unable to download the logs from the APM. they are on there, typing logs in the CLI mode shows there are 8 logs in the APM. But the download logs button doesn't work. Apparently this is a problem with Arduplane 2.70.
Cheers again for the help, It is much appreciated!
Rich
Season Two of the Trust Time Trial (T3) Contest has now begun. The fourth round is an accuracy round for multicopters, which requires contestants to fly a cube. The deadline is April 14th.9 members
1289 members
140 members
313 members
44 members
© 2013 Created by Chris Anderson.
Powered by
