The 1st harmonic, or primary noise from the motors is so low frequency that any notch filter is very close to the Gyro filter and is causing large phase changes.
One option is disable the notch filter since noise is not very bad anyway, and slightly lower the Gyro filter.
The other option is run with a notch filter, this is the best I could come up with, giving the least adverse effects:
in this ins accel x is haveing oscillation
should i now proceed with roll and pitch autotune
ok i will go ahead with the second option and share the log in next 5 mins
Use my latest notch filter settings then try pitch and roll autotune. Do a small test flight before hotting Autotune just to make sure everything is OK.
doign a test flight with new notch setting then will share the log post you check it only then will proceed with autotubne
That’s working great as a filter:
There are still big phase changes - so just be cautious.
This can cause PIDs to be slow to react and instability.
If stability is an issue just try INS_HNTCH_ENABLE,0 to see what the difference is.
so i shall do a roll and pitch autotune ? and if there is any problem in stability that time then i just land and [quote=“xfacta, post:48, topic:134200”]
INS_HNTCH_ENABLE,0
[/quote]
and then takeoff again check stability and if fine then again start autotune for roll and pitch
@xfacta The 0.7 FM_RAT reduces the attenuation of the notch filter linearly between 0.7 and 1 of the min frequency, ensuring a non abrupt filtering of the data, but a gradual increase in attenuation.
Wy are you suggesting that it is not meant for ESC telemetry? It makes sense for all RPM sources, as the Noise amplitude also increases with the frequency between 0.7 and 1 of the hover freq.
Yes, or if you notice your pre-Autotune flight is unstable then try INS_HNTCH_ENABLE,0 but the log you sent looks reasonable for pitch and roll. Yaw is still a little poor but that can be sorted out later.
These copters with very big props can be difficult to tune without some payload to make the motors work a bit.
should i make changes in this and with the same notch filter setting do a flight ?
If you use 0.7 then you must also set
INS_HNTCH_FREQ,30 # the peak of the first harmonic during hover flight
Thanks for pointing that out, I’ve only ever used INS_HNTCH_FM_RAT,1 with ESC telem data.
I just checked these logs and a couple of others - I could not see any noticeable difference between INS_HNTCH_FM_RAT 0.7 or 1.0 , which doesnt mean there isnt a difference somewhere, just that I couldnt see it.
So if 0.7 seems to give more stability then definitely use it.
ok ,
right now as per shawn we had made other chanegs also in this should we get that also back to
ins gyro filter 20 (shawn 15)
ins hntch freq 30(20 shawn)
ins hntch bw 11(5 shawn )
this is what i am getting shall i set these and do a altmode flight
The reason I specified these is because the ESC RPM data will drive up the frequency and bandwidth from these values to whatever is required. If you set them too high, like right at the hover frequency, then the notch filter will never go below that.
Just so you know the reason for those changes 
ok so now i have set the values and will now do a autotune in roll and pitch winds are calm
With the FM_RAT of 0.7 it will go down to 70% of the hover frequency. And scales all accordingly.
OK I see what you are saying. Thanks again.
I’ll test that out some more too.
