Misha,
Similar to the other discussion, you need to try to raise the pitch and roll P gain (ATC_RAT_XXX_P) as high as possible. Have you tried to raise them yet? I would think you should be able to raise them higher without and bad oscillations. Start raising them in 0.02 increments until you see the aircraft oscillating in pitch or roll, typically it will be in roll axis first. Do you know how to tune using channel 6? Continue raising the P gain until you see oscillations or get to 0.2 or 0.25. If the aircraft feels too responsive, start reducing the ATC_RAT_XXX_VFF. I think you should be able to almost make that gain zero once the P gain gets above 0.18.
Also you should not require much more than 0.2 for the pitch and roll I gain (ATC_RAT_XXX_I). It looks like you have your I gain values at 0.6.
Getting the P gains above 0.15 will go a long way toward fixing the toilet bowling effect that you are seeing in loiter but it all depends on whether you can get the gains that high without seeing oscillations.
Hi Misha,
Yes, please follow Bill’s advice. I knew I could fix your collective jumping problem with the ACCEL_Z_P parameter. But the default rate PID settings will only provide a basically flyable helicopter. They need to be tuned for your particular heli. Please read that thread that Bill posted the link to. It explains what needs to be adjusted. And do not use Loiter flight mode with GPS assist to adjust it. You must fly the helicopter in Stabilize flight mode to tune PID’s.
Thank you very much for your help, today got a lot of flight parameters, the funnel was removed, it became much less to spin around its axis. I managed only to adjust the parameters today (ATC_RAT_XXX_P) I did not have much time to do this.
To date, I received the most:
(ACCEL_Z_P) 0.28
(ATC_RAT_RLL_P) 0.12 If you raise higher, then the oscillations begin.
(ATC_RAT_PIT_P) 0.06 If you raise higher, very strong fluctuations begin.
Tomorrow I will occupy and configure (ATC_RAT_XXX_I) and I will watch the result.
By the way, 9Dsmart P9X is glued on foamed, double-sided adhesive tape, 3 mm thick. 9Dsmart P9X gives small vibrations on a classic helicopter and do not need any additional vibration shunting, I really liked it.
Thank you again for your help.
I have to say this - that’s a nice looking little flight controller. And it runs ArduPilot and all the connectors are on the ends and none on the top of it? Never seen one before now. But I like the looks of it.
The 9Dsmart P9X has very convenient connectors, standard ones like the Pixhawk, everything fits the praxes from Pixhawk and I think it’s very good, you do not have to search for connectors like Pixhack
I will be engaged soon in setting up my T-Rex 500, the results will be written.
Thanks for the info on it. I looked at your BIN file and for such a simple mount it looks like the internal damping in that flight controller does an amazing job of keeping vibration to manageable levels in a heli.
Today I continued to set up my T-Rex 500, but there was not much success.
(ACCEL_Z_P) 0.28 today did not touch.
(ATC_RAT_PIT_P) 0.07 higher raise is not how it turns out, there are strong fluctuations in RAT, and if you raise it even higher, for example, 0.1 is resonant and very difficult to stop in the air, you have to make a dangerous landing.
(ATC_RAT_PIT_I) 0.200
(ATC_RAT_RLL_P) 0.13 higher today did not.
(ATC_RAT_RLL_I) 0.18
(ATC_RAT_ PIT_VFF) 0.05 installed as recommended.
(ATC_RAT_ RLL_VFF) 0.05 installed as recommended.
I also noticed that when there is a complete calm, there are fewer variations.
When descending, the helicopter begins to fluctuate severely
In general, I do not zayu what to do next (((, please help
Misha,
I’m sorry I haven’t responded sooner. You caught me off guard posting in Russian. I had to use google translate so I think I understand most of what you posted.
So it sounds like you reached the limits of the P gain for roll and pitch. If you are seeing strong oscillations when making sharp inputs or when it is windy then you may have to reduce the P gain values a little more.
As for the FILT parameter, I don’t think you want to raise it to 6 because that would let oscillations less than 6 hz pass through. Do you have a log I can look at where the helicopter experiences these strong oscillations? I want to determine the frequency of the oscillations in both the pitch and roll axes.
So if you reduce the P gain until strong oscillation are no longer present when making sharp inputs, how does the aircraft fly in loiter? Sorry I don’t have any immediate answers. Just more questions
Sorry for the language, already fixed. And thanks for the answers.
Today I continued testing and adjusting. I think that the mechanical characteristics of my 500 P value do not increase, the reason is a small servo speed, or wear and tear, and backlash mechanics, since the helicopter is already 4 years old, and I did not do major overhaul of the mechanics.
I think that for me there will be a limit of PIT_P) 0.09 and RLL_P) 0.15
I do not have a flight log today, I’ll do it tomorrow.
Today has dropped (ATC_RAT_PIT_IMAX) 0.5 and (ATC_RAT_RLL_IMAX) 0.5
The first time in the stabilization got very good results, quickly raised to the top, briskly down, very good, even with the wind the helicopter in the stabilization behaves well.
There was another problem, in GPS mode, a very large drift, the helicopter swims the horizon to the left and right, about 4-5 meters, I noticed a problem appeared when I changed, PIT_IMAX from 1 to 0.5 and RLL_IMAX from 1 to 0.5
What do you think?
Send me your log file. If you don’t have LOG_BITMASK = 393214 then set it to that value and fly again to show me the problem. I need you to use that LOG_BITMASK because it gives me the PID controller data. Thanks
Fly in stabilize and raise your Pgain and show me the strong oscillations in pitch and roll. Then set them to values that caused this issue in loiter with the aircraft drifting.
Misha,
Did you conduct a flight with parameters settings where the aircraft drifted in loiter? I haven’t looked at the logs yet but it looks like you just showed me the strong oscillations in pitch and roll.
Hello, I’m on the first video, I added RLL_P, and on the second video PIT_P. Today there was a breeze, and on the second video there was a big PIT_P and had to be planted, since the control was not stable.
Misha,
I looked at your logs. I see your strong oscillations in pitch and roll are at a frequency of approximately 5hz. So the low pass filter (FILT=4) is helping but you won’t be able to raise it any more unless you try to change your set up to try to reduce the vibrations causing this response. Or is a notch filter to try to stop the bad frequency from getting through.
thanks for answers
My helicopter T-Rex 500 PRO with a shaft on the tail, and gives a vibration. It will, of course, be better if there was a belt on the tail)
I most likely will not be able to remove more mechanical vibrations, many vibrations are transmitted through the wires.
I did not understand you, the notch filter, how to call this parameter in the settings menu?
so I think I understand most of what you posted.