Quadcopter Altitude Gain in Altitude hold

quadcopter gained altitude as soon as i change the mode to altitude hold, i reduced the p gain of velocity controller and pos controller as stated in the ardupilot documentation but still the altitude gained to 15 m from 2m in 2 seconds, in stabilize mode the fligght is good enough and i didnt observe any oscillations as well. thats why i moved to altitude hold. even in logs the desired velocity down is positive and the the actualy velocity down is negative means controller is not correcting the velocity down.
drone specs: 5055 props. 3 bladed, f405 speedybee fc, battery 1300mAHs 6s lipo 100C

6 1-1-1980 5-00-00 AM.bin (604 KB)

i am sharing my logs below, can anyone help me figure out this problem?

What firmware version?

Did you set the altitude acceleration P and I gains like explained in the wiki and in AMC step 20?

1 Like

version 4.6.3 stable release. i didnt follow methodological approach rather i followd ardupilots documentation

can you see the log file and let me know what might be the issue

The outputs are oscillating badly. The Rate Pitch & Roll D gains are still at default, lower them. And MOT_SPIN_MIN is at default also. Follow the Wiki for setting motor ranges.

I’ll take this opportunity to say that the text in the Wiki here
Mandatory Hardware Configuration>Configure Motor Range (Optional)

Is not Optional at all IMO for this and many other high thrust/weight craft.

2 Likes

I will adjust the parameters. Is there anything else that can be changed? Did you see the vertical velocity graphs? They are quite strange.

Yes, enable the arming checks.

2 Likes

The rate pitch and roll D are zero

The ATC_RAT_PIT/RLL_D are at default .0036 You dropped the P&I terms but not D. Cut those in half.

1 Like

ATC_THR_MIX_MAN is still at 0.1
PSC_ACC_I is set to 0.046 instead of 0.46
PSC_ACC_P is set to 0.023 instead of 0.23

Yes, it is very clear that you did not follow methodological approach and rather miss interpreted the ardupilot wiki.

1 Like

I suppose he tried to make some hover throttle sense out of this Ctun>Tho data:

1 Like

The version i am using is 4.6.3 so i assumed that the pSC_D_Acc_I for this firmware will be 0.2 times of MOT_THST_Hover.

It is mentioned here that ATC_Thr_Mix_man should be at 0.1. that’s why i didnt change this paramter.

Nah bro. I didnt look at this. I saw PSCZ graphs where i could clearly see the desired Vz and actualVz diverging and it didn’t make any sense

No, that’s for 4.7

2 Likes

Why not? Hover thrust is Tho at hover.

2 Likes

I didnt know this that’s why. Last i made a copter it was 2 years ago and I didn’t face these issues. 8 had the same motors and props and bettery. But this time things are different

I did try the parameters suggested for 4.6 and belo as well but the result was same. Si i assumed i am using wring parameters for the current version i am using

I just read this.and this is saying ThH is the calculated hover throttle while thO is only throttle output

i looked at the graphs now for CTUN.ThO and CTUN ThH. could you make some sense out of it