Motors twitching with transmitter, but not twitching when in Betaflight?

[Duddits4141]

Motors twitching with transmitter, but not twitching when in Betaflight?*SOLVED*

Hi everybody ,

I am building a FT 210 but have an issue with the motors ... When i throttle up with betaflight they seem to work as they should . But when I disconnect from betaflight and throttle up with my Transmitter they start to twitch(yeah don't know how to discribe it otherwise...) . First I thought it was a solder joint , but as they seem to work when i am connected with betaflight that doesn't seem to be the problem ...

I copied the settings from the build video , and i am using the "better" version of the powerpack .
I also updated the esc's to the latest version .
Any ideas ?

*update*
see 2nd page for sollution , Thanks for helping !!!

 

[Craftydan]

Duddits,

Smooth run from the motor tab followed by jittery run when under the TX control means your control loops are reacting (overreacting) to some mechanical noise in the environment.

Usual suspects for this kind of thing are . . . poorly mounted control board (room to wobble or shake freely, or not solidly mounted), poor vibration isolation (some frames don't need it, some do), Overtuned PID's, poorly balanced motors and poorly balanced props.

To know which it is, we'll need more info. See what you can find, and if the issue doesn't pop out at you, post some pics of your build and we can dig deeper.

 

[Duddits4141]

These are some pics , the FC is not wobbly ... PID's are not changed(still default) .
Never had any props on it , was stil in build when I found this problem .
Working on a stable table .
The only thing that I noticed was when I was setting the transmitter (DX8) in betaflight , I had to adjust my transmitter to +147% to get the 1000-2000 settings in betaflight . I also have the Versa copter , there i had to adjust to 127% ...

 

[Duddits4141]

This is how it reacts with transmitter going up to 15% of throttle .
When I connect it on betaflight and use the motor tab , no problem...

 

[PsyBorg]

Did you redo the esc calibration after flashing the esc's?

 

[Craftydan]

Looks like a clean build in progress.

ok . . . one quick-and-stupid why-didn't-I-ask-this-earlier test . . . pick up the quad as you run the TX test. If it goes away, I blame your hard table beating against the bolt-heads in your frame. Otherwise, carry on.


Again, looking at the symptoms -- Motors run smoothly when controlled ONLY by the configurator, but when controlled by the TX and control loops, the motors jitter.

You've only changed two things between the tests -- the transmitter, and the PID control loops. If you'd added props, the loading would come into play, but you haven't gotten that far yet, so at least that's not this problem. The TX is rarely the problem here, but to eliminate it, look on the RX page -- if you're not seeing much jitter in those signals, the noise isn't coming from there.

So let's look at the loops . . . it's either poorly tuned (stock isn't always good, but usually isn't horribly bad) or noise is getting injected into the loop. This could be electrical or mechanical, but in either case it should be visible in the plots from the motor tab. I'd recommend for the next step plug the configurator back in and run up each ESC one at a time and look at the gyro channel's plots -- hold the frame still as you do this. If you've got an imminent bearing failure in one of your motors or a motor bell WAY off balance, it should be plain as day in the gyro plots -- that boom should be MUCH higher in vibe. The noise from that would impact ALL of the other motors through the control loops.

If they all look fine, Cut your P and D gains in half and spin up the motors again with the TX (you're testing in acro mode, right?).

 

[Duddits4141]

Thanks Craftydan ,
looks got some testing to do later
I'll let you now what the results are !

 

[Duddits4141]

Did you redo the esc calibration after flashing the esc's?

Hi PsyBorg , yup , even tried a lower firmware (1.4.6) of the firmware and did a esc calibration every time ...
But thanks for the advise !

 

[Duddits4141]

This is how the motors react on a flat surface .

 

[Duddits4141]

Looks like a clean build in progress.

ok . . . one quick-and-stupid why-didn't-I-ask-this-earlier test . . . pick up the quad as you run the TX test. If it goes away, I blame your hard table beating against the bolt-heads in your frame. Otherwise, carry on.


Again, looking at the symptoms -- Motors run smoothly when controlled ONLY by the configurator, but when controlled by the TX and control loops, the motors jitter.

You've only changed two things between the tests -- the transmitter, and the PID control loops. If you'd added props, the loading would come into play, but you haven't gotten that far yet, so at least that's not this problem. The TX is rarely the problem here, but to eliminate it, look on the RX page -- if you're not seeing much jitter in those signals, the noise isn't coming from there.

So let's look at the loops . . . it's either poorly tuned (stock isn't always good, but usually isn't horribly bad) or noise is getting injected into the loop. This could be electrical or mechanical, but in either case it should be visible in the plots from the motor tab. I'd recommend for the next step plug the configurator back in and run up each ESC one at a time and look at the gyro channel's plots -- hold the frame still as you do this. If you've got an imminent bearing failure in one of your motors or a motor bell WAY off balance, it should be plain as day in the gyro plots -- that boom should be MUCH higher in vibe. The noise from that would impact ALL of the other motors through the control loops.

If they all look fine, Cut your P and D gains in half and spin up the motors again with the TX (you're testing in acro mode, right?).

Hey Craftydan , thanks for youre time !

If I try the engines separetly with the motor tab , no problem at all. When I trow in the TX ... well I posted a video part 2 . Here You can see how the motor booms react when i throttle up .
Could it be the NAZE ??? It was a new one , but anything is possible ...
Like you said "Cut your P and D gains in half " , did that , didn't change a thing .


Grts Duddits .

 

[Snowblind]

Set your quad on a foam mousepad and try the transmitter to see if the table is resounding into the frame.

Credit to CraftyDan if it works.

It appears to twitch when changing RPM which makes sense to me if the table is the problem, but if you're not trying to yaw, those settings may need to get attention, it's yawing on the table when it twitches.

None of the quads I have controlled have liked being run up without being aloft, they try to continually correct in one direction,(acro or s-l) be happy you're not chasing your tail over that one!

 

[PsyBorg]

Hey. I looked at the build pics again.. I think I see antenna wires but don't see the receiver. What receiver are you using and did you by chance remove it from the case and some how mount it in the stack with the PDB and the FC? That may produce signal noise. Usually the receivers get mounted on the base plate in the rear away from the FC. I had twitching issues with my KISS set up until I moved the signal wires off to the side instead of a straight run over the top of the FC and that seemed to clear it up for me.

It is probably not the issue but worthy of a move to check the odd chance it is catching signal injections from being too close to other components.

 

[Bluto78]

I had the exact same issue. I put mine together similar to the way you have yours. I changed the way I mounted the boards. I mounted the PDB like they did in the build video, double sided tape onto the base plate. I then put on the stand offs and used the O-rings on the top and the bottom of the Naze. Twitch is gone. I got so confident after this that I tried a flip. I completed the flip but lost the quad on a side slip. Disarmed it for safety and the battery saved the quad as it came down on pavement. Dal props and the rest of the quad is just fine. The battery gave us a good smoke signal.

 

[Duddits4141]

Set your quad on a foam mousepad and try the transmitter to see if the table is resounding into the frame.

Credit to CraftyDan if it works.

It appears to twitch when changing RPM which makes sense to me if the table is the problem, but if you're not trying to yaw, those settings may need to get attention, it's yawing on the table when it twitches.

None of the quads I have controlled have liked being run up without being aloft, they try to continually correct in one direction,(acro or s-l) be happy you're not chasing your tail over that one!

Hi Snowblind , thanks for thinking along !

I placed some foam spacers on the boom's (you can't see them on the foto's...) .I am almost sure that the table isn't the problem . It's a solid wooden table about 6cm thick , and I also tried the floor , same story there ...

 

[Duddits4141]

Hey. I looked at the build pics again.. ***---*** components.

It's a Spektrum Quad Race Serial Receiver , it's mounted on the back on the top plate of the quad . Maby on the picture it's more clear .

 

[Duddits4141]

I had the exact same issue. I put mine together similar to the way you have yours. I changed the way I mounted the boards. I mounted ***---***

Bluto78 , I didn't use the O rings (oh man...)
Thanks for the tip , going to try this later !

If this works I'll by you a beer !

 

[Bluto78]

Bluto78 , I didn't use the O rings (oh man...)
Thanks for the tip , going to try this later !

If this works I'll by you a beer !

I just backed the screws out and pulled out the spacers. Carefully put double sided tape under the PDB as there is not a lot of room to move it. Put the screws back, put stand offs on topside of PDB not the bottom. Then Orings on each side of the Naze. Tightened it all down carefully. I can get you a picture later if you want.

 

[Duddits4141]

It WORKS!!!

Bluto78 ! you did it !

The only thing i didn't do was the O-rings . I just placed O-rings on both sides of the Naze and it is looking good so far !!!
It seems that there was a small electrical contact between the usb port and the screw .


Thanks my friend !

send me your adres in a pm and ill try to send you that beer i promised !

 

[Bluto78]

Bluto78 ! you did it !

The only thing i didn't do was the O-rings . I just placed O-rings on both sides of the Naze and it is looking good so far !!!
It seems that there was a small electrical contact between the usb port and the screw .


Thanks my friend !

send me your adres in a pm and ill try to send you that beer i promised !

Sweet. I do not drink beer, but I am glad I could help.

I flew my first real FPV flight today.(More than hovering and going back and forth.) Freaking awesome. Now just to do it on ACRO mode and learn some flips, rolls and dives. So happy.

Changed Angle mode out to Horizon mode to take another step forward.

 

[PsyBorg]

Good job on solving this Bluto. Also congrats on your first FPV flight. You are now hooked for life hehe. You will dream in FPV.

Duddits make sure ya get us some video of your flights at some point so we see how well the solution worked.