Please take a look at this videos samples at 10% speed and please, please don't tell me it's a FC gyro issue or ESC as it's the third one  I soldered. I hope it can be solved by BF settings, or worst case changing 4 motors...It would be exotic. 







I will do the short story ( liar   ), after changing all the hglrc stack ( gyro issues out of the box, misfire and runaway, and a smoked motor on bench, but as the motor was  still running and looking good when I opened it I kept it, but I recently changed it too to be sure ) by the same F760 stack, I decided to changed again the stack by an other different brand one. Why? because, and because i have some suspicions on the last 60A hglrc according to little web birds.

-new stack : MAMBA F722 DJI v2+ 50a ESC BLHELI32.




Now this is where I am, I tried it yesterday and I thought it was allright compare to isues I usually had after taking off ( don't know english words, like woobling, angry motors !) because while I still had some vibration, twitching shaking (propwash?) above all when rising the gaz again after a gaz off descent, I had this way more annoying roll issue after doing a flip: A perfect 360° one, or some kind of 180 sometimes...

1. I TESTED THE CONTINUITY between motor bolts and esc before and it's ok. I tested the esc between pads and XT60 it's ok.
   
2. The tx16S is calibrated, and no roll parasites appearing.
   
3. BF motor tabs test is allright ( except with DSHOT 600 8K/8K I have errors, whereas some don't have with 4S version as seen with other customers, sector 5 issues are with the stock 6S, but as I changed the stack I'm lost )
   

To be honest I don't rember the PID I used during the biggest issue, because I was short flight testing and changing on the field with BF APP without noting, but anyways I never gone above a combination of what follow:

• Actual rates 960 roll, 760 pitch ( before this last stack change, the flip was VERY VILOLENT at 960, so I decided to lower it and the difference is huge since, I don't know if it's normal)
  
  
• UAVTECH pid sliders preset...
  
• ...or even BF stock pid.
  
• slider FILTER stock,  or up to 1.1 /  1.3max ( to see if it could do something, no ) 
  
• ESC updated, stock, 16 motor time, 24KHZ
  

---

I first tried :

• DSHOT 300
  
• 14 magnets
  
• 8K/4K
  
• rpm ON
  
• Dmin ON or OFF I don't remenber which one during the roll issue.
  
• air mode always ON of course
  

I stopped because I have kind of stuttering during a vertical full gaz test. It's pretty hot where I live but I had hot motors too, without beeing crazy on stick ( I even can't )  with the rpm ON and filters stock...


After the issue, I rember to have finished  the lipo ( 3mn) whith everything stock, ( except dyn notch filter that still was UAVTECH numbers )
-D SHOT 600
-8K/8K
-RPM OFF
-Slider filter and PID

And it was some kind of OK, I managed ( was afraid by the previous rolls ) to do a pair of front flip ( delayed ) without unwanted rolls, I don't know if it was because they weren't done in one stroke...

just a thought
check that no wire is passing in top of he gyro in the FC
most of the time, those strange behaviors are caused by it

If you have changed out the flight controller and ESC and tried new ones with stock settings and it still does this, then it is another part causing the problem. First I would put everything back to stock BetaFlight config, it is good for a 5" quad. There is no point in trying to tune while it is having problems.

The video looks like it is also losing connection with your Rx or something, like the stick inputs go down to 0, even on pitch. And for a moment the throttle goes up to max, right before the sticks both go to 0. Very strange. Also while the quad is still after the roll, the motors black box shows 1 motor at full throttle, even though it is still in video.

If the radio is fine, and the FC and stack are fine the only parts left that could interfere with flight are the receiver or motors.
Or as rolo said above something else is interfering with the gyro. Something loose or touching it.

Sorry I can't be more help, I have never seen that before.

that looks like a motor desync, it could come from excessive noise, from a faulty motor if its allways the same axies where it flips....

check the following:
-everything fits well - nothing loosen vibrating?
-force the frame by hand to check the stiffness, if one arm is less stiff as example
-check your capacitator

to solve it, the following could help:
-try an increased motor idle speed value
-set your motortiming to 23 in blheli suite
-decrease rampup power

to avoid that zero throttle doesnt help, if you stay a few percentage on the throttle, it would be like increased idle.
try if you get it with airmode disabled, if so it could be a sogn for a boise issue - dont decrease to zero throttle for acromode without airmode feature anyway ;-)

hello,
I don't have anything touching the gyro. I do my test with the minimum, no gps, no buzzer...

---

check the following:
-everything fits well - nothing loosen vibrating? If it is, I didn't find it.
-force the frame by hand to check the stiffness, if one arm is less stiff as example: Already check, I think It's ok.
-check your capacitator : Hard for me to test it without unsoldering it from the FC...Wich I'm not very good at it when it's soldering directly under the XT60 pad...I took me ages last time I did it, I Just tried some test with my tester with everything mounted, but not sure it's accurate. Any case I see numbers moving.

to solve it, the following could help:
-try an increased motor idle speed value : For now I'm at 5.5, was at 4 tuned by HGLRC.
-set your motortiming to 23 in blheli suite : I'm in AUTO
-decrease rampup power : Already tried, in fact I followed advices for desync from UAVTECH , I dont rember if I'm at 20 or 25 for now.

---

I could change the 4 motors...And the CRSF receiver, so everything would be changed except the dji vista lol.  

But as I've already change one motor after having magic smoke from it, on the bench after an arming test  ( I don't remenber the number, the one on bottom left side ) and despite this I've seen it clipping at 100% on a BB log. I'm not sure, do you think change all the motors could change anything? I tested the motors with BF and nothing is strange at full throttle,  sound, vibration, errors etc... ( I even opened them to check)

the cap is still connected thats test enough test :-)

i would go for fixed a motortiming value befor auto timing, but i didnt any research.

you can additionly test the motorwires for any issue, no motorscrew to long-touching winding?

if it is a motor issue, even if the whole batch is faulty, one motor should be the worst and it should flip over this axie all the time... as long as it doesnt freakout over random axies, the barticular motor which receive 100% power command by the fc in the log is the faulty one or the worst.
by two motorsets and an additional frame xD maybe it forced you to something nice :-) two good builds thisway xD
another motorset is always nice, the emax ecos would be good inexpensive ones :-)

as you told you dont like soldering; the gear does 90% of the job, maybe you need to improve yours?
show us a picture pls, a bad solder connection can cause issues

(30-Jul-2021, 11:22 PM)hugnosed_bat Wrote: you can additionly test the motorwires for any issue, no motorscrew to long-touching winding?
I had one, so since I carrefully test them. Now the wire end on the the fc pads are hard because it's thrid time I soldering unsoldering them, but I wooble them and test the continuity and it seems ok.

if it is a motor issue, even if the whole batch is faulty, one motor should be the worst and it should flip over this axie all the time... as long as it doesnt freakout over random axies, the barticular motor which receive 100% power command by the fc in the log is the faulty one or the worst.
Yes it seems to flip on the same direction, wich in a certain way is a good thing, but my fear is that in a log even the motor I changed ( AEOLUS 2306.5 / 1900 ) was at 100%. 


by two motorsets and an additional frame xD maybe it forced you to something nice :-) two good builds thisway xD
another motorset is always nice, the emax ecos would be good inexpensive ones :-)
I have only 2 spare AEOLUS, I could buy 2 more OR mount the 4 Iflight xing e pro 2207 1800 I have in spare for my rock solid nazgul 5 v2. 

as you told you dont like soldering; the gear does 90% of the job, maybe you need to improve yours?
show us a picture pls, a bad solder connection can cause issues I will as possible.

Since original post day, I kept the mamba esc but switched the Mamba FC by my previous brand new HGLRC ZEUS f722. To be sure, for a cross test.

Differences :

-I had more hard wooble/propwash during punch out / and above all trembling when rising the throttle after a zero gaz vertical descent, whereas I ( think) managed to get rid off it by rising the filter up to 1.5 with the mamba : with the ZEUS I didn't managed to do it even up to 1.6.  As I live in a windy area, I'm not sure the wind force was the same.


-Still the more annoying problem of roll, but I didn't have a perfect 360 this time. In fact except for the issue mentionned above, I thought it was ok excpt for the bounce back too.
I began to tweak slowly BF to try to deal with the vertical propwash ( never happened) and the bounce back , and when I tried flips and rolls I had the issue again coming back , I would say less pronounced but still there, can't tell what was the PID*, but nothing crazy for sure. And hot motors too.

*PS : I need to check twice but unfortunaltly my 16mo BB didn't catch the more important logs, as I first tried to deal with the propwash trembing with punch in/out/in first...


Thank you for your help every one.

100% throttle does only mean it loosed control over the motor, its just a sign it didnt what it should...

change the motor which was forced to do 100%, if you got damaged winding on it, it should solve it :-)

   
   

(31-Jul-2021, 08:44 AM)hugnosed_bat Wrote: 100% throttle does only mean it loosed control over the motor, its just a sign it didnt what it should...

change the motor which was forced to do 100%, if you got damaged winding on it, it should solve it :-)

Is it possible to test this kind of motors/wires issues  with betaflight ? I did idle, 1/3 gaz, and a quick full gaz on each one for a second and it seems smooth and ok.


I'm in a bad spiral, after FC, ESC,STACK, I changed the smoked motor, BUT the long bolt was with this same motor if I remenber well ( I found it before flying while testing on BF because of a little strange noise )

I had, first out of the box, very bad trembling grumpy rusty motors sounds on arming, the BF motor tab level was dancing like EQ music, and gyro graph was high, until magic smoke from on motor, the number 3: It stoped when I changed the FC, you know the rest,  but maybe others motors have been damaged...and/or I added problem by myself elsewhere.


I'm going to watch my log to see wich motors are at 100%, but again from my souvenirs I'm afraid it's not all the time the same, if not each of them alone or by pair:
With only 2 aeolus home stock I'm not sure wich motors I should change, or buy two more and change all, or runaway from aeolus and try the 2306 from iflight. 

"Excuse my bad english, and use your imagination to understand me when you see strange words/sentences!"
same from my side xD


betaflight motortab doesnt help for a desync issue


just change one at first and give it a try, step by step wont be fast but it bring us ahead and more usefull informations.
i would suggest to give your two spare motors a try and go for another one if the stay as possible source for the issue.

don't loose your head on this one
get 4 new motors
problem solved

(11-Aug-2021, 03:11 AM)rolo95 Wrote: don't loose your head on this one
get 4 new motors
problem solved

-I changed one motor since my last post, and still the problem, and today I changed the last two one at once to gain time.
I did 3 or 4 lipos tonight, and didn't notice the issue ! I prefer to wait before opening a bottle of champagne but I'm happy.

Still bounce back, hot motors without going crazy on filters and PID D, *and above all a lot, a lot  of wobble when rising the gaz during a descent ( after a vertical full throttle/gaz off ) but for me it's the first time I was managed to fly this F...G quad after months !



-I have first followed the UAVTECH tuto for the desync issues, so I wonder if I should undo it now in the ESC ? If it would fly the quad better?!
actually : 48KHZ and THRUST LINEAR = 25  ( in the cli ), DMAG= HIGH, RAMPUP POWER = 22 , MOTOR timing = AUTO




*PS: the motors are 1900kv 2306.5 and I have S5 ethix props, and no go pro, I wonder if it's relevant...about the descent/gaz trembling...

i would go for 24khz, demag low and fixed motortiming of 22 as a try, but if it flies nice?... dont open the explore box if you arent ready to get into issues again :-)

(11-Aug-2021, 09:38 PM)hugnosed_bat Wrote: i would go for 24khz, demag low and fixed motortiming of 22 as a try, but if it flies nice?... dont open the explore box if you arent ready to get into issues again :-)

I'm starting reviewing logs, is this motors up/down spikes graph normal during a simple Left Roll ? ( by the way I have a lot of similar )
And is the clipping 100% motor 4 freshly changed also ok?
I have bounce back but this roll, and the entire flight was ok.

What is your target firmware for HGLRC and MAMBA FC?

I had an issue out of the box with the wrong BetaFlight Firmware...

Also, do you have another receiver