A successful build of my heavy lift quad copter.
Thanks especially to William Cox and Steve Keep - it is now flying a lot smoother.
Attached is my parameters. I read a lot about people not sharing parameters. And these need a lot of work, but here they are... shared...
In the next 4 weeks it should be flying down in Antarctica doing some radio tests to wave buoys and photos of sea ice floes.
Can't figure out how to get a picture loaded.
We are taking losses with coaxial props, but we have removable arms and can pack in shipping box.
We wanted to keep the size down.
AUW to date is ~ 45 lbs No issues on motor heating.
On the throttle we have found that we can't fly at empty payload weights. Throttle too low for control. Ok
add more batteries.
BREAKING NEWS - Based on some good advice I fought I would try 3.2 RC12 today (yes it is T-2 days for packing, but living on the edge) - AND WOOOOWWWWWW ... Rock solid Loiter. Just hung there in the air ! Now I do have to prefix that with no wind today, but still (and... still !). My understanding for this great improvement is: improved filter fusion; use of both GPS (I have two now); use of both Magnetometers (which are now both calibrated - must use Mission Planner for compass cal). Probably lots of other reasons.
right now i was to going to test loiter wiht your params in v 3.1.5, but after reading this... maybe I should try with the 3.2...
Yes 3.2. Yes same parameters (change GPS_TYPE_2 = 1; recalibrate accels and both comapss) however I did not enable any of the EKF filters - will look into that tomorrow.
but I think filter fusion improvement depend on AHRS_EKF_USE = 1
My experiences today with loiter:
- Scott parameters.
- No too windy but variable.
- When no wind it seems stable, keep position quite well.
- When it need to fight against wind it destabilizes quickly.
V3.2 beta RC12:
- Scott parameters.
- Zero wind.
- Keep position quite well and seems stable (remember, no wind).
- Keeping altitude and moving stick I can desetabilize it. generally can recover after 2 or 3 oscilations.
- Moving in one direction and stopping suddenly causes a strong instability (oscillations of more than 60º), the crash was very close, I was able to recover with throttle up and waiting...
- I think it is important to note that it was too light (only one 8000 mah 6s battery, no payload, no landing gear). probably a more realistic weight mean better performance due to higher RPMs.
- I did not find differences enabling or disabling EKF.
Attached log file.
In my next attempt I will add weight and try to tune loiter params (PIDs, maximun angle rate, ....)
I've been reading the whole conversation,so did you find any difference in loiter about adding more weight?We have a quite similar configuration than Scott's,with 170 kv u8 motors and don't understand too much why low kv motors have more problems to react when they turn at low rpms. currently we are testing with no payload (weight is around 5.5 kg) with:
Loiter PID P=0.9
Rate loiter P=1.25
Rate loiter I=0.7
Rate loiter D=0.05
and it makes turns around the point (not to big,less than 1 m radius) so maybe we will have to low loiter P
Thank you in advance,
I found with my 170Kv U8's I needed to add more weight before the system reacted appropriately. Is your craft a quad or octo?
It's a quad,there's any difference with yours?
I have an X8 that weighs about the same, I ended up having to add 7kg to make the system stabilize correctly. Try adding some extra weight to your system.
But what happens if you want to use it with a small payload (like 1 kg).You must be able to have a solid loiter with that.Not all your operations will be with 7 kg payload,but thanks by the way for the idea
If you're only running a 1kg payload, you're severely overpowered. My plan with this craft is to decrease extra weight during tuning, until I get to my bottom end. Depending on your payload, you'll have to re-tune depending on your AUW. What size props are you running?