Way to go Captain Obvious. So instead of a pointless comment, please explain what I did wrong and what I could/should do. Or are you just a drive by troll?
Thank you. I will check it out and try to apply these steps once I have repaired my aircraft.
- copying parameters from another drone
- using a automatic flight mode (land) on an untuned copter
a bit of a personal opinion, but also:
- flying a Tarot 650
Folding arms pressed into plastic C-clips = vibration galore
The C-clips also end up breaking. The motor mounts end up rotating (causing motors to be off axis) unless you put a long bolt through the whole motor-mount/arm assembly.
Edit: but yes, using the methodical configurator is the best way to make sure you setup your drone correctly
2 Likes
This script can make tuning a lot less risky:
1 Like
It is actually. It’a fine lesson in how not to configure and tune a multirotor that new pilots, should they come across this post, can avoid.
I see I’m not the 1st to be promoted to Captain. But to stay in character, what to do? That was obvious when you were correctly posting in unsupported.
2 Likes
At last update, this user was still using APM hardware on 3.x series firmware. Scripting will not be an option.
He seems aware of the risks at hand and obvious benefits of updating the entire system.
Unfortunately, the decision to stay on deprecated versions makes support quite difficult.
2 Likes
Yes, I used an APM controller. The APM was deprecated when I first built quads. I made it clear that my needs were modest and saw no advantage to going to a Pixhawk mainly because I am using pwm radio gear and didn’t want to add an encoder to satisfy the S-bus requirement.
To date, none of you have given me a reason not to use an APM other than it is no longer supported. If I can use an older version of Mission Planner to load firmware and configure an APM, what is wrong with that? Are you suggesting that prior to the Pixhawk it was impossible to build, tune and fly a quad using APMs? Spare me your judgemental remarks and offer some useful advice or don’t bother.
I am willing to change my mind regarding my choice of f.c. but tell me why I cannot use legacy h/w with compatible legacy s/w and I’ll consider changing.
First and foremost, I made no judgmental remark. My reference to your hardware/firmware choice was an acknowledgement of it and nothing more. It was an important distinction to make for the reasons stated below, as well as in response to the scripted tuning suggestion that will not function on APM boards.
To try and address your questions:
The tuning links presented above all apply to far more recent versions of firmware than the APM hardware supports. You can certainly apply the basics to older firmware, but many parameters have been renamed, newer ones don’t exist at all, and nearly all of the filtering and sensor fusion algorithms have changed for the better, while deprecating the ones that existed some years ago.
The methodical configurator (linked above) developed by @amilcarlucas has become the de facto standard for new multirotor configuration. However, that method will simply not work for APM hardware, since it is entirely parameter file driven, and too many of the critical parameter names do not exist in older firmware. You’ll spend more time fighting it than actually generating valid, useful results.
If you share a log to try and get further help with issues like the one stated above, many (or even most) of us will have a hard time making sense of it to a degree that will be helpful. I won’t even attempt it, because it would be a disservice to us both. I entered the community long after APM had come and gone.
Finally, indulge me in a little philosophy regarding your “last build” mindset: I think you’d find the path to success a lot less challenging and time consuming by updating to supported, recent versions of hardware and software, thus affording you more time to enjoy this project rather than staying mired in the challenge of maintaining deprecated versions and garnering support for them. Just an opinion, and you know what the say about those, I’m sure.
Nothing above says “cannot” regarding your present approach. But I’ve tried to make a reasonable case for “should update.”
9 Likes
I build my first tricopter on a KK board with piezo gyros. Did it fly? Yes! Did I upgrade it to a Openpilot CC board as soon as it arrived from Australia? Yes!!!
I always tried to go with the time, be it FCs or other tech and I find it most respectable how the devs and forum experts go out of their way to help people with old and unsupported hardware. Expecting to get help here for those old APMs and getting mouthy and demanding about it, is not the way to go.
4 Likes
“Copying parameters from another drone”.
The default parameters from Misison Planner came from an original 3D Robotics drone - a 10" machine as I recall. So what was wrong with using PIDs from another Tarot Sport quad identically equipped as a starting point?
“Using a automatic flight mode (land) on an untuned copter”.
Point taken. I will resolve not to do that in the future. Still, even for an untuned aircraft, I don’t understand why it simply shut down when I switched flight modes.
“Folding arms pressed into plastic C-clips =- vibration galore”
So what is your basis for this assertion? Do you have any factual data to support this claim or is this just your opiinion? I would be interested in knowing. Since I don’t plan to fold the arms, sould simply cementing the arms in the clamps solve the problem?
Regarding the motor mounts, I check the alignment before each flight. The only time I experienced any rotation was during a hard landing when the props struck the ground. Again, I think I can mitigate this by applying cement to the mounts.
Finding a methodical process for tuning my Tarot has been a challenge. One of the recommendations was to follow “How to methodically configure and tune any Arducopter”, dated Dec. 2023. He starts of fine with getting the hardware right - steps that I already taken. But then he suggests using ecalc. Really? Is that really necessary? I was unaware of the “MethodicConfigurator” - I will look into that and see how I can use it.
An open question for me is if I use an APM on my Tarot once it’s repaired, can I use these tools to arrive at an initial set of PID parameters that will enable me to start off with a stable initial flight. Once I get to that point, I feel confident that I can tweak the PIDs via subsequent test flights.
If someone can provide a compelling argument against using an APM I will surrender and buy a Pixhawk/pwm encoder combo. But don’t just argue that the controller is obsolete and incompatible with the latest Mission Planner. Give me a compelling technical reason why I can’t use a legacy copy of M.P. Given that I have successfully loaded the firmware and configured the imu, compass, radio, ESC, etc. is there any reason why I can’t use the current version of M.P. for PID tuning, mission planning, etc?
So, you say it’s a fine lesson in how not to configure and tune a multirotor and yet you offered no explanation as to what was wrong with my configuration. In fact you offered no advice at all whatsoever. Pehaps you should be demoted?
Replace the Flight Controller. Any advice I might have is not relevant to AC V3.2.1 and the APM. Or any of the Tuning Tools that predominate today.
1 Like
Thank you for your reponse. You are the first to offer good reasons for making the change and I will make the switch. You are correct in that it has been a challenge - searching for legacy versions of M.P. that had the functionality I needed, getting them to run on my computers etc. I also had the challenge of trying to configure the Tarot gimbal since it’s s/w is also really old.
At any rate, I thank you for your advice and I will go ahead and buy a Pixhawk and s-bus encoder.
Once you have it flying somewhat stable, measure the vibrations. If they are within limits you’re good, if they are excessive look at reinforcing the clamp attachment.
1 Like
Very few if any of the process’s in this configurator will be available to you. Unless you replace the Flight Controller that is. Nor any script like Quick Tune. And if you do have borderline bad vibration levels there is no Notch Filter available to help with the resulting tuning issues.
How about the Online tools? WebTools Nope.
1 Like
I have decided to switch to a Pixhawk. I’m considering a PX4 PIX 2.4.8 from Amazon for $115 plus an 8CH PPM Encoder&Pixhawk I2C Splitter Expand Module, also from Amazon. Any thoughts/recommendations? I won’t get it right away as I have to order some other parts to make repairs to my Tarot.
For that price you can also get a Matek H743 WING / H743 Slim or Holybro Kakute H743 WING. Sure, they don’t have a nice enclosure but are much more capable boards overall.
4 Likes
Agree with the others. That one is overpriced for its capabilities, is also quite dated, and likely a cheaply made clone of a board that is EOL.
Of the alternatives suggested, the MatekH743 series is the most performant and supports scripting. It does lack an onboard compass, but it’s usually best to use an external one anyway (often present in compatible GPS modules).
If you prefer a form factor that requires less soldering, the QioTek ZealotH743 or Holybro Pixhawk 6C are usually well priced for their capabilities.
5 Likes
Hi FossilRider,
First, never copy PID parameters because they differ a lot from a drone to another (even if the have the same radius, they can have different weight and mass distribution!).
Second, you can follow the steps in this video, its so helpful!
1 Like