Ok,
Such a small benefit isnāt worth it here. This is just a test aircraft. Better to save processor capacity.
Iāll post my test logs here tomorrow hopefully!
Ok,
Such a small benefit isnāt worth it here. This is just a test aircraft. Better to save processor capacity.
Iāll post my test logs here tomorrow hopefully!
Sadly enough it has been raining here all day so tests have to wait.
Hopefully tomorrow I will bebable to do the testflights
looking forward to your update
Do you have the latest test results?
Sorry, itās been raining here the past days, Iāll try and get some flights in today if possible.
All good Axel, you can only do what you can do. We appreciate your help!!
I was able to fly and gather at least some data. I had to make it bit quicker than I initially thought since itās still some rain in the air. I put some tape over the ports on the cube and went anyways, worked out nicely since there was so little rain.
Here is the 4.2.3 flight:
https://drive.google.com/file/d/1CR3XepdM5hAS3QIXnNaqVLUTlWv_zbIO/view?usp=sharing
Here is the 4.3.0 flight:
https://drive.google.com/file/d/1_WUj0Argpfc0-kjKD6MnNHwEBKTxnhvZ/view?usp=sharing
It felt like it flew better on 4.2.3 than on 4.3.0. Donāt know it if thatās actually how it is but maybe you could take a look at that when you look over the logs, maybe some filter thing or PID thing, should not be PID though since all is same but it might just be me overthinking.
We (mostly Leonard and Tridge) have a number of bug fixes and enhancements that we hope will resolve the internal errors and control issues caused by overloaded CPUs here and in other threads (discussion1, discussion2).
Below are links to firmwares for a few different boards. In each download are two firmwares, the ā432rc1ā firmware includes simpler bug fixes while the ā432rc1-withdtā additionally includes Leonardās control timing improvements. Iām happy to pre-build firmwares for other boards upon request.
FYI @Axel1, @xfacta, @A_manafi, @TilCreator, @kalai1219
Hi Axel,
Thanks for all your help with this problem. I think we have fixed a range of issues and the binaries that Randy link to will improve the handling of these problems dramatically.
Both of your longs are very similar with just a little more load on 4.3 as we expect. I am not seeing anything to suggest that 4.2 is flying better than 4.3 now you have backed off the load. However, have a look at the binaries that Randy links too and see if you feel any improvement there if you have time.
Hi,
Iāve experienced a minor incident so Iām waiting on some new propellers.
No issue with my vehicle, just with me, was over 2 years ago since I had an incident so I feel kind of dumb nowā¦
Although I will test the new firmware (the one with the new controllers) once I get my new propellers and have conducted a full hardware inspection on my craft. Everything seems fine so Iāll probably be able to test as soon as I get propellers, but I havenāt done all the system and hardware tests yet so might have to swap something moreā¦
Thanks!
On another note, Iām trying to learn as much as possible about tuning and I just wanted to ask you if you got a log from a perfectly tuned aircraft I could look at as a benchmark for good tuning.
I know you work with this and that you wonāt be able to just post logs for vehicles you build but if you have any log you can post It would be appreaciated!
I might be able to help with that. Iāve got a couple of logs of different sized quads I can send to you
I PMāed you a couple of links and photos
@Axel1, just curious, in your original post you mentioned that takeoffs were āsportyā:
āMy quad makes like a jump into the air. I raise the throttle slowly like I always do.ā
I believe you ultimately concluded that this was not an issue in your use case, but Iām curious if addressing the internal errors helped mitigate it, or if it is just characteristic of 4.3.x?
I also noticed this when I switched to 4.3.0 on a Orange Cube; it was inconsistent, but compared to 4.1.5, it was startling. I experimented briefly with the TKOFF_SLEW_TIME, but did not observe any change in behavior. While it isnāt problematic on smaller aircraft, doing a full-power āsendā for the first couple feet of takeoff on a big ship would beā¦ Terrifying, haha!
Regards,
-Luke
Hi Luke,
I did experience the jumpy takeoff and still do on 4.3.0. I havenāt tested 4.3.1 yet so donāt know if itās still there. It did go away and came back each time I switched between 4.2.3 and 4.3.0 during my testing. Currently, Iām on 4.2.3 and donāt experience the jump on takeoff. Although I will soon test the 4.3.2 firmware including Leonards new control loop mechanics.
I came to the conclusion that itās an intended maneuver done by the drone to get some height right after takeoff. I thought it might had something to do with the slew parameter since it was the only new parameter listed regarding takeoff, but I might have been wrong.
Maybe @rmackay9 knows if there has been any changes that would result in this, and if there is any parameter to make it smoother again.
You could make a new post about it if you want to discuss it with the developers. I bet all feedback regarding new flight behavior is welcome, and maybe others also have some feedback to give as well.
@Luke_Ionno and @Axel1
Could you post logs and observations about the takeoff behaviour in 4.3 here so I can go over the logs and see if there are any issues or if I can help you.
Hi Leonard,
Iām sorry for the delays, Iām still waiting for partsā¦
Iāll get to testing as soon as Iām in the air again. If the testing is still relevant than.
I havenāt forgot about it!!!
Version 4.3.3 fixes this problem.
Iāve finally got it flying again.
Although, it might not be needed to test this anymore. I think I saw your new control loop management integrated recently.
Iām happy to help in the future though!