Please help with shakey QAV250

Hey bud I just got home and will check into this and let you know what I find....thank you!

So I am trying to look into this, how would I go about checking into this, I dont see it on any of the menu windows. Is it something to be done in the CLI?

Also, make sure if you do the ESC cal that you have no expo on the sticks.

I don't believe I do, everything on the taranis and cleanflight are at stock settings only thing I have changed is turn on the rx ppm option to run through ppm and the oneshot option, because these are oneshots. other than that stock everything. Does the serial receiver provider need to be something other that spectrum 2048?

I am about to try calibrating the esc's, gonna void these esc's but this all seems like crappy calibration.....
 


on the flip I have all esc's hooked up signal and ground except for esc #1 has power, it was hooked up all powers, grounds and sig but everything seemed crazy over powered and messing with video so took out all motor 2-4 powers. Would adding a seperate BEC be something that could fix things? Just shootin in the dark at the moment trying to find anything I can. would a video be something that would help with whats up? I can do some dropbox links or upload video to youtube if it would be beneficial.
 

Motopreserve

Drone Enthusiast
The Taranis has expo in it - if you haven't adjusted it - you should be fine.

You don't need the power from all the ESCS, only 1 to run the FC/Rx. If you want to eliminate the BEC in the last ESC - you'd need to run a separate one to power the FC/Rx. Im not sure what power inputs you have to the board. Multiwii had some extra pins that you could use - but these things are small.

I do t think anyone would ever to be able to tell you calibrated the ESCs unless they really dug deep with testing. Give it a shot. You can always send them back of they do t work.
 

Bartman

Welcome to MultiRotorForums.com!!
Hey Bartman, when I go past 30% the shakyness just gets worse and when I give just enough throttle to spin the motors only a motor or two will spin up and the other two wont spin up till more throttle is given.

i had a very similar problem with the CC3D board on the 250 racer I built. the motors didn't start up evenly but once they were spinning they were ok. DanM diagnosed it as a much too high P value in the PID's and he's got it now and said it's flying much better.

having said that you seem to have a different problem but it's hard to diagnose without knowing if the problem is there even when you run the motors one by one using an RC receiver both with and without a prop. flight controllers can introduce problems into the motor/control systems so using a receiver allows us to see if the motors and ESC's are ok if given a clean PWM signal for control.
 

i had a very similar problem with the CC3D board on the 250 racer I built. the motors didn't start up evenly but once they were spinning they were ok. DanM diagnosed it as a much too high P value in the PID's and he's got it now and said it's flying much better.

having said that you seem to have a different problem but it's hard to diagnose without knowing if the problem is there even when you run the motors one by one using an RC receiver both with and without a prop. flight controllers can introduce problems into the motor/control systems so using a receiver allows us to see if the motors and ESC's are ok if given a clean PWM signal for control.

So I went against what Lumenier said and performed an ESC calibration and things look to be better, still running odd on the motor start up but not nearly the craziness that it was doing. I am going to guess that the rest is P settings. Is there any way to adjust the PID's without having to land and adjust thru the computer every time?
 

Motopreserve

Drone Enthusiast
Nice work! Glad you're gaining ground.

Unfortunately there is no way to adjust the PIDs mid-flight, unless you have Bluetooth (if these smaller boards even have that option).

With Multiwii, the Bluetooth connection made it FAR less frustrating to tune :)
 

okay guys I have tried and tried with this and have tuned this shake out of the quadcopter but now its real unstable and doesn't fly stable anymore, it's give or take in this scenerio....I'm really having a hard time with this build. I've never had so many issues with a build before. I don't see why a motor would be bad since they all fire up, they are really a dumb device that takes orders so I am figuring that they are all good.

The FCB functions just fine and tuning and all options, accelerometer and all work so FCB failure seems not an option.

props being unbalanced seems like it would make smaller vibrations not whole frame shaking to bits, so Im ruling out prop balancing, but I did order a balancer.

Only thing I can boil it all down to would be either an esc issue...ie..one is bad or they are all just junk due to that I have altered loop times and changed looptimes, so buying one esc seems to be a waste since your not supposed to calibrate or change anything on them which I have since I had no other options. This is the only option I can come up with unless someone can recommend something else, but the plan is to buy all new esc's. Only issue is I'm a tad unsure if I really want to mess with lumenier anymore and these "plug and play esc's" but they are at getfpv for $20 a pop right now. I am looking at the 20amp esc's at massive rc right now that have a 22amp burst and are only $15 at the time....any advice or direction on that or if you think it's even an esc issue...( I conclude it to be since the shake cant be tuned out completely and motors all start at different times and rpm's)...massive rc has always given me good equipment and I've never had an issue and these can be "calibrated" normally.

Other issue it could be is that my 2208-9 2000kv motors have to much torque with 6045 gemfan props and it's causing the arms on the G10 frame to flex. Sounds like a shot in the dark but the G10 frame is crap and is flimsy as hell, mine is already cracked and I havent even flown really....so a CF frame is in the near future too. My buddy had the same G10 frame is crap issue and got the CF frame today and it looks WAAAY more solid than the G10 is...

What you guys think? Thanks for everything this is all much appreciated.
 
Last edited by a moderator:


crayfellow

Member
Man, I dunno... I have had a "cheap" G10 250 frame and a benchmark CF frame and I got by with the G10 just fine, to be honest. I can't imagine it's flexing and causing what you are seeing. Even if it were, it would be much higher frequency. QAV250 has a great reputation.

It just sounds like you're down the rabbit hole, really. When that happens to me I have to just start over. I've only ever used cc3d on 250's but I saw all sorts of funny things until I found a preset tune to start from that was close to the mark.

I have a feeling if you go about changing out ESC's and completely rebuilding the frame using the CF one you're going to see the same exact thing.

Can you start over, erase any EEPROM config, and start from proven PID numbers other QAV250 (or Blackout, or FMX, or ZMR, ...) folks are using?
 

Man, I dunno... I have had a "cheap" G10 250 frame and a benchmark CF frame and I got by with the G10 just fine, to be honest. I can't imagine it's flexing and causing what you are seeing. Even if it were, it would be much higher frequency. QAV250 has a great reputation.

It just sounds like you're down the rabbit hole, really. When that happens to me I have to just start over. I've only ever used cc3d on 250's but I saw all sorts of funny things until I found a preset tune to start from that was close to the mark.

I have a feeling if you go about changing out ESC's and completely rebuilding the frame using the CF one you're going to see the same exact thing.

Can you start over, erase any EEPROM config, and start from proven PID numbers other QAV250 (or Blackout, or FMX, or ZMR, ...) folks are using?

Thats what my issue is, I have tried so many PID config's and nothing...so I went solo and figured out PID's on my own and nothing....PID's are simply killing me. I am debating swaping to the cc3d atom in hopes for easier tuning possibly....and yes this rabbit hole is getting crazy deep....thats why I am asking on here....to many tunnels to go down to figure all this out...lol

Thx
 

crayfellow

Member
Well if you do grab a cc3d it might be an interesting test just to take the flight controller out of the loop.


Sent from my iPhone using Tapatalk
 

Motopreserve

Drone Enthusiast
One thing that may be worth a try, before pouring more money into new FC/ESCs, and getting deeper down the rabbit hole, is solder up a servo squid - 1 into 4 (you may have this from calibrating the ESCs?).

Hook up all ESC/motors (no props!) to it and plug it directly into the throttle channel output of your Rx. See if they respond to throttle stick input the same, or if any lag behind. This would rule out the ESCs and hone in on the FC.
 

One thing that may be worth a try, before pouring more money into new FC/ESCs, and getting deeper down the rabbit hole, is solder up a servo squid - 1 into 4 (you may have this from calibrating the ESCs?).

Hook up all ESC/motors (no props!) to it and plug it directly into the throttle channel output of your Rx. See if they respond to throttle stick input the same, or if any lag behind. This would rule out the ESCs and hone in on the FC.

sorry if this sounds noobish but whats a servo squid? are you saying tie all of the esc's to only 1 esc channel and run them all off that together rather than on all 4 servo channels?

thx
 

Motopreserve

Drone Enthusiast
Never be sorry for asking questions. That's what we are here for. Everyone was a noob at some point - and we all needed help!

A squid is just a multi-wire harness (like the tentacles of a squid) ending in a single connector on the other end.

The one I am referring to would have a single servo connector and wires on one end (the end that connects to the throttle channel of your receiver - typically channel #3), the other end would fan out to have 4 sets of wires and servo connectors that mate with the servo leads coming from your ESCs. If plugged into the throttle channel directly this way, when you hit the throttle, not just one but all 4 ESC/motors should react.

Does that make sense?
 




Motopreserve

Drone Enthusiast
If you plug this into the receiver directly ((D4R) - throttle channel, power the quad with battery, you'll have direct access to the throttle and motors only. It's how you would have calibrated the ESCs. If you calibrated differently - they are not calibrated at all :)

See if the motors all start at random times, or if they fire together. If they do fire together, it's safe to say the FC is doing something to the throttle command signal.
 


Hey bud your the man! The squid cable worked well! I plugged the esc's into the channel 1 on the d4r and throttled the motors, everything started perfectly so this rules out the esc's right? Im onto the FCB as a virdict too. I have an extra flip32 board so ill swap that out. I normally wire motor 1 with power and the other 3 have ground and signal, would it be better to eliminate all power wires and go with a bec? Or is what I'm doing okay? This is a brand new fcb so its wierd its acting like this, also I have my fcb taped down with some black scotch brand 20lb double stick tape....is this tape maybe conductive and making the fcb freak out? Ever hear of that?

Thx
 

Top