T-Rex 450 loiter
APM 2.0
MediaTek MT3329 and barometer
Now winter has passed in New Zealand, so I went to make loiter working on the T-Rex 450.
After playing a bit with the PIDs it worked pretty well as you can see in the video.
There was a sea breeze going but the heli did very well holding altitude (only barometer) and position.
Thank you very much DIY DRONES Team.
The next nice day I’ll get navigation a try, some waypoints and RTL. I’m sure it will work the same, if I get the setup right.
I’ll keep you posted.
Comments
Hi Hein
Sadly I have not progressed in flying any waypoints, because a few days later I tried loiter again and it was disappointing all over the place. I have not changed anything, except a newly charged battery.
The fact is, a good loiter depends entirely that you can produce a stable hover in stabilise mode without any shifts in attitude. But sadly this is not constantly the case with my heli. I have times he hovers for quite some time very stable and then all of a sudden he starts to shift its attitude.
It’s like a little devil in the system, suddenly changing your trim.
And of course if the heli leans to one side you have an asymmetric system, which is impossible to control via a PID controller, hence a very bad loiter etc.
I know this is caused by vibrations, but every heli vibrates and I tried everything to minimise this.
It’s actually the frequency of the vibrations and the relationship to the sample frequency of the controller called aliasing causing the problem. A minimal change in head speed can cause more or less attitude shifts.
A few days ago I changed the heli back to be controlled by “Heli Command “, and the hover was very nice and stable all the time, no little devil in the system. It would be great if APM2 could produce the same result.
I don’t know many if any Heli’s flying really successfully on APM2.
I know it works with Quad’s very well; they have surely a differed vibrational footprint and the APM is placed in the symmetry centre of the system which makes a big difference.
I don’t give up yet; I try to suspend the APM2 with rubber bands like a trampoline..... have still to test it, but the weather is bad at the moment.
And if it’s still doesn’t work I may fly the heli with Heli Command and build a Quad for my APM2.
Hi Manfred,
Any updates on waypoint navigation? Did you do some Auto flights? Would like to hear about it.
Greetings,
Hein
Thanks Ravi
Yes a Heli is much more delicate to tune, because it’s much more reactive compared to a Quad. And you see the tiniest wobble in attitude created by the APM much more than on a Quad.
Compare a Quad to a T-Rex 450 it’s like comparing a Bus with a super bike.
heli's are complex to tune them mechanically for loiter. but APM2 is very powerful code so may be one day APM2-heli would make it easy. but for the present APM-heli code great achivement by Manfred.
Nice work !! I love traditional helis !
Great stuff and love the video!
This PIDs might not fit to every T-Rex 450,
it depends very much also on your Headspeed, I fly with about 3000rpm
if you change Headspeed you may have to change some PIDs
Hi Kenneth thanks
I have posted some pics in this link :
http://www.diydrones.com/xn/detail/705844:Comment:959115?xg_source=...
Hi Ravi thanks
I use APM2 for everything, no sonar.
The PIDs
amazing! i never got my align 450 to loiter like that, just few questions. r u using the APM gyro for yaw or external gyro did u use sonar. can u share the PIDs. or can u share param file. wonderful. i used the first batch of APM2.0
Manfred, great result considering the MediaTek GPS performance. That's about as good as it gets.