After the helicopter landed, the collective pitch was still pulsating

[Flying2015]

The firmware version of flight controller is 4.2.3.

After the helicopter lands, the integral control items of roll and pitch remain zero, but the integral control of altitude control still fluctuates and does not remain unchanged to the minimum, which leads to the collective pitch still jumping and even causes the helicopter to jump.

This is the parameters file and log file.
After the helicopter landed, the collective pitch was still pulsating

Is it possible to keep the integral term of altitude control unchanged as the minimum value after landing, so as to prevent the helicopter from jumping on the ground?

Thanks!

[Leonardthall (Leonard Hall)]

You need to turn fast logging on:

Vibrations are crazy on this aircraft but with this logging we can’t see what is happening.

[Flying2015]

Thank you very much for your reply.

@Leonardthall

Yes, and that is different to what I said we needed to fully understand what is going on. We need Fast Attitude turned on.

OK, and we will get the new log data as soon as you suggest.

@Leonardthall

This is the single worst acceleration vibration I have ever seen in any (flying) aircraft.

Yes, this vibration is really big, and it gives us a lot of trouble. But it’s very hard to reduce the vibration of this oil-powered helicopter, it’s a very good effort.

@Leonardthall

And the key part of my advice is below:

Yes, your advice was critical and made a big difference. We’ve made the INS_ACCEL_FILTER a lot smaller, but I’m worried that if this is too small, it will affect the response speed of the helicopter.

@Leonardthall

This is being done in the height control, but we also need to restrict the values to the maximum you allow the autopilot to use. Your frame vibration noise is higher than these maximum values.

When the helicopter is on the ground, the integral terms for roll and pitch control remain zero, but I don’t understand why the integral terms for altitude control are not constant at the lowest point, but jumping constantly.

[Leonardthall (Leonard Hall)]

@Flying2015

This is the log bitmask interface and the vibration data from my flight today.

Yes, and that is different to what I said we needed to fully understand what is going on. We need Fast Attitude turned on.

@Flying2015

As you can see, the vibration is not alarming.

This is the single worst acceleration vibration I have ever seen in any (flying) aircraft.

Your acceleration noise or vibration is 2x your maximum. When on the ground exceeds your maximum and this is why you see the problem. With the faster logging I may be able to get an idea where this is coming from. It may be a higher frequency signal that looks like a low frequency signal due to the slow sample rate.

The VIBE message shows higher frequencies that are difficult to see in the controller logging. Your vibrations appear to be lower in frequency and dominate your controller logging. Due to the low update rate of this logging it is hard to know exactly what the frequency is.

@Flying2015

This problem is kind of similar to a problem that we had earlier.“How to turn off the cyclic blade pitch and collective pitch when the helicopter is”

And the key part of my advice is below:

@Leonardthall

Now this does not fix your problem as it is fundamentally caused by stupidly large vibrations getting to the position controller. These changes will help reduce your problem a little but the real fix is to:

1. Reduce the vibrations getting to your autopilot
2. Reduce the magnitude of the vibrations getting to the controller

I can’t help you with 1 but I can help you with 2. Your current setting of the acceleration filter is very high:
INS_ACCEL_FILTER 20 Hz

It looks like your acceleration noise is approximately 5 Hz. You could reduce your INS_ACCEL_FILTER setting to 0.5 to 1 Hz to greatly reduce this noise at the controller level. However we need to ensure that the lower controller response does not make Alt_Hold unstable. If so we may need to reduce the PSC Z gains.

@Flying2015

Can this design be applied to height control as well? When the helicopter is on the ground, the integral term of the altitude control is kept at the minimum.

This is being done in the height control, but we also need to restrict the values to the maximum you allow the autopilot to use. Your frame vibration noise is higher than these maximum values.

The real solution is to remove the 1g vibration that you are constantly feeding to the autopilot.