(12-Sep-2020, 10:14 PM)the.ronin Wrote: Try reseting just the filtering again makes sure you have a good starting point. Also take a real good look at the motor mounting make sure screws or anything else are not touching the windings. Other than that, the only real explanation is noise on your craft - or a faulty FC.

Hmm Mamba. Not the most reliable but personally, I've had a few that flew fine.

Next time I'm at the field I'll completely reset the filters(after backing up everything of course) and the PIDs and see if it's any different,
I'll also unscrew the motors and screw them back(maybe with another set of screws?) to make sure it's not that, the motors are not touching the windings.

I guess it could be a faulty FC, or maybe noise that I just can't detect.

about my Mamba, my VCC pads were recently torn after a crash, so I soldered the battery lead directly to a negative point and the shunt resistor on the lower side of where the VCC pads used to be, if it has anything to do with it.
My VTX and receiver are mounted on the back with hot glue, it's not pretty but it's really strong and it's not moving at all.

(12-Sep-2020, 10:28 PM)bLoWsMokE Wrote: I have similar issues with my 5 inch....... it just flies weird..... I found out that lowering the I to like the mid 30's helped for me..... The only thing I can't get rid of now is a dip when i cut the throttle fast after a punch out...... Nothing I do can get rid of that. I fear this FC's software is wonky because for one betaflight alerted me saying that I had to use a custom file or the thing wouldn't work.... sure enough I said no and betaflight didn't even register the gyro. I ended up having to get it though.I don't like putting D term much higher than 34. I try and use the safest route possible in anticipation of a crash (or many) ;D

lowering the I (was 70 on default, wtf?) helped me too, but only to a certain degree, we can't have it all especially on my first build haha.

Hopefully next time will go smoother, btw my frame is TBS Source One if it has something to do with anything.

Wait. could it be related to the capacitor I put on where the battery lead goes to?
it's not tied to anything, could it be it?

(12-Sep-2020, 10:28 PM)bLoWsMokE Wrote: I try and use the safest route possible in anticipation of a crash (or many) ;D

That video I posted, in addition to flying on rails after the initial rubiks cube crash in the video itself, was after I bashed that Acrobrat over 3 packs practicing the juicy flick. I always leave headroom for unanticipated noise. I actually use beat up props to tune to make sure. Aggressive tuning does not necessarily have to mean reckless tuning.

(12-Sep-2020, 11:23 PM)SaHaRzZz Wrote: about my Mamba, my VCC pads were recently torn after a crash, so I soldered the battery lead directly to a negative point and the shunt resistor on the lower side of where the VCC pads used to be, if it has anything to do with it.
My VTX and receiver are mounted on the back with hot glue, it's not pretty but it's really strong and it's not moving at all.

You might want to post pics of your FC for others more learned in that area to comment on if there is something off. Personally, I couldn't find my way around a PCB if my life depended on it lol.

(12-Sep-2020, 11:23 PM)SaHaRzZz Wrote: lowering the I (was 70 on default, wtf?) helped me too, but only to a certain degree, we can't have it all especially on my first build haha.

You don't want to mess with I until after you've got your P-D ratio down solid. Generally, I've left I default. It is not meant to be first on scene and messing with it could alter that role.

(12-Sep-2020, 11:28 PM)SaHaRzZz Wrote: Wait. could it be related to the capacitor I put on where the battery lead goes to?
it's not tied to anything, could it be it?

Doubtful since I have not heard you indicate any video noise which is where insufficient capacitance often first rears its ugly head. But just in case, what is the rating on the cap?

(13-Sep-2020, 12:39 AM)the.ronin Wrote: That video I posted, in addition to flying on rails after the initial rubiks cube crash in the video itself, was after I bashed that Acrobrat over 3 packs practicing the juicy flick. I always leave headroom for unanticipated noise. I actually use beat up props to tune to make sure. Aggressive tuning does not necessarily have to mean reckless tuning.



You might want to post pics of your FC for others more learned in that area to comment on if there is something off. Personally, I couldn't find my way around a PCB if my life depended on it lol.


You don't want to mess with I until after you've got your P-D ratio down solid. Generally, I've left I default. It is not meant to be first on scene and messing with it could alter that role.


Doubtful since I have not heard you indicate any video noise which is where insufficient capacitance often first rears its ugly head. But just in case, what is the rating on the cap?

It is a 35v 1000uf esr.

Ive messed around with I as a 'last resort' in the field to try and get a different output, generally I know it should be messed with after the P and the D are set.

I'ĺl post photos of my FC later when I get home, maybe its something simple that I cant see or understand

Yea that cap should be sufficient.

I took some photos of the FC and the quad:
https://imgur.com/a/TlqZV3O

Yea I personally couldn't find anything that really stood out to me.

(13-Sep-2020, 05:44 PM)the.ronin Wrote: Yea I personally couldn't find anything that really stood out to me.

maybe it is just a bad FC/gyro or something,
anyways if I won't be able to figure it out, at least it flies decent enough to not crash haha.
Thank you for your dedication to help out !

Show us a video of how it flies with the tune you have that doesn't get your motors hot.

(13-Sep-2020, 10:04 PM)bLoWsMokE Wrote: Show us a video of how it flies with the tune you have that doesn't get your motors hot.

I will probably in friday,
I was trying to learn more about filtering because I think that's how I can fix my issues with the tune, this video:

seems very informative and helpful, I have assigned a switch to enable blackbox and I'll try to see how it goes and filter the noise

(13-Sep-2020, 11:07 PM)SaHaRzZz Wrote: I will probably in friday,
I was trying to learn more about filtering because I think that's how I can fix my issues with the tune, this video:

seems very informative and helpful, I have assigned a switch to enable blackbox and I'll try to see how it goes and filter the noise

Lowering the filtering will make your motors get hot with around the same ease as increasing D gains.... Make sure you are aware of that.

(15-Sep-2020, 03:53 PM)bLoWsMokE Wrote: Lowering the filtering will make your motors get hot with around the same ease as increasing D gains.... Make sure you are aware of that.

not for me I guess, after setting my filters like he described in the video, and then attempting to lower them did not make my motors any hotter than they were, I flew 2.5 packs(one wasn't fully charged haha) and recorded everything, I also activated blackbox in some of my flights to see where is the issue, I think something is wrong with my pitch vibrations, because the roll seems perfect but the pitch has vibrations in the lower HZ zone,
these are my blackbox files:
http://www.mediafire.com/folder/bckkkl3n55szs

I have named these files as to what I changed, first I tried flying with 0.75 filters on both gyro and D-term, and then I increased the sliders, it didn't make my motors any hotter, after 1.4 I couldn't figure out what was wrong because I always went to check on the data related to the roll, when I went to pitch I saw the vibrations, it didn't get filtered and the D-term is fighting the hell out of it, at least that's what I managed to understand from the graphs.
in the file named 'last_flight' it was after I lowered one setting which was related to the gyro filtering, I set the min to 90 manually and just raised the D-term slider 0.2 more to the right.
I flew a little bit and that's it.

The video will be available in 10 minutes here:


time stamps:

clip num 1: 0:00 recording first blackbox
clip num 2: 1:24 reducing filters
clip num 3: 2:41 reducing filters + crash clip
clip num 4: 4:11 reducing filters
clip num 5: 4:55 changing pids
clip num 6: 6:20 changing pids
clip num 7: 8:24 changing pids
clip num 8: 12:52 reducing filters
clip num 9: 13:20 recuding filters

About clip number 3, something weird happend to the quad.. it lost control?
could it be a failsafe? I'm using the r9 system and I wasn't far away at all.

Hopefully that's enough data!

(15-Sep-2020, 05:43 PM)SaHaRzZz Wrote: I think something is wrong with my pitch vibrations, because the roll seems perfect but the pitch has vibrations in the lower HZ zone,

And there you have hit the nail on the head. Not only is the vibration on your pitch axis excessive, it is also not being filtered effectively and is making its way through the filters to the PIDs. In the screen shot below, the upper graph is the raw gyro values and the lower one is the PIDs.




That's a truly horrifying amount of noise that the PID controller is trying to deal with!

So, the question is: can you see a mechanical reason for that vibration? Loose frame components? Loose motor? Flight controller not properly soft mounted? Out of balance motor? Out of balance propeller? Something has to be causing that noise.

You must get rid of that vibration, preferably by finding the source and dealing with it. It's possible you may be able to filter it out, but that will take some quite strong filtering which will add delay and at best only get it flying "OK". Solving the problem at its source is by far the best option.

Sorry, I forgot to turn off expo and smoothing in the black box log viewer. That often makes things look far worse than they actually are. In this case, it didn't help much but seems to show that there is still excessive vibration in the raw gyro data for the pitch axis, but it seems that the PID settings are actually amplifying the noise. So there's both a genuine vibration plus that your PID settings in the last flight at least, are making things worse.

(15-Sep-2020, 09:37 PM)unseen Wrote: Sorry, I forgot to turn off expo and smoothing in the black box log viewer. That often makes things look far worse than they actually are. In this case, it didn't help much but seems to show that there is still excessive vibration in the raw gyro data for the pitch axis, but it seems that the PID settings are actually amplifying the noise. So there's both a genuine vibration plus that your PID settings in the last flight at least, are making things worse.

II don't see anything problematic with the quad, all the screws are in there tight, maybe the right move is to take apart the whole build and redo it.
But something is bothering me, why is it that just the pitch axis is suffering? the roll axis is completely clean