HELP! quadcopter build has me stumped! homebuilt wont fly right!

hello, thanks for looking. i built a quadcopter for a friend, and it wont fly right. here are the specs:
- frame= homebuilt pvc, 730mm diagonal, X config.
-motors= emax 450mm quadcopter como from readymade rc, 2213 935kv
-ESC's= emax 450mm quadcopter combo from readymade rc, emax simon series 20A
-props= included in combo, 10X4
-flight controller= CC3D with openpilot
-RX= orange RX 615 6ch dsmx
-TX= little non computer eflite remote included with sport cub s (also have my dx6i)
-battery= venom 5400 mah 3s 20c
-lights= getfpv.com led strips
some cheap little power distribution board
all up weight= 1.6kg

the build: motors are screwed to the pvc, ecs's bullet connected on, esc's soldered to the power distribution board, one of the BEC's from the ESC's is powering the FC, the others are connected only via ground and signal. rx is plugged into FC via pwm wires, fc is mounded almost directly on top of the power dist. board, lights all wired to the power dist. board,.


ok, the problem: it flies OK until i add a yaw input. pitch/role is a little wobbly, but ok. right yaw is very slow and causes some pitch/role drift. left yaw causes a severe pitch/ role reaction to the front left, but does slightly yaw. also, sometimes such severe oscillation get started, all i can do is drop the quad to the ground at the right time. i at first suspected the flight control board, a CC3D "evo", as it acted a little odd in the programming stage (saved settings through a upgrade/erase cycle, etc.) i ordered a new FC, and have the exact same problem. i also suspected that i was picking up interference from the motors, as the FC was in perfect plane with them. i put a little box under the FC and had the same problem. i tried a bigger box, separating the RX and FC, and some aluminum foil over the power distribution board. i suspect maybe it could be magnetic interference off the power dist. board? aluminum foil as it is non ferrous would have no affect against this??? i also (with the original FC) tried my dx6i, as well as a different RX.:confused: please help me, i made big promises about this quad, but it looks like i might be keeping it, (and paying for the parts:() i can not think of what the problem could be, so i home someone reading this can tell me, or give me things to try. oh, i did try plugging the ESC's directly into my RX, and the motors seem to spool up normal. thanks for the help,
Leo

P.S. while i can keep this thing in the air, it will be going to a total beginner, so anything but perfect flight is unacceptable.
 

Revere

New member
PVC tends to flex which is why carbon fibre is the current big thing, especially if there is any damage, like tiny cracks caused by cutting the length of PVC to size with the wrong kind of sawblade (been there, done that), if its tubular pvc the cracks can be inside the tube and pretty much invisible until enough force is applied to open them up. First thing I would do is mount the main body firmly to something (or get someone to hold it) and then gently try and twist/flex each arm. If you notice an arm that flexs more easily than the others, replace it and check again. If there is no flexing problem then start with rechecking your electrics, all your solder joints, look for nipped wires etc.
 
Last edited:

Darkback2

New member
I tried building my own frame once using similar materials. I found that the material I was using flexed enough to amplify any bad tuning, so a wobble due to bad tuning would be transferred to the flight controller which then created a feedback loop leading to much stronger oscillations. I remade the frame using MDF that I bought online and aluminum square tube from a local hardware store. The stiffer materials worked to settle down the feedback loop. Later I replaced the MDF with carbon fiber scraps from another project.

You may want to try using some different materials for the arms. Wood square tube for example you could also try MDF for the body...You may be best looking at /copying a frame like the electrohub...

At this point we are all guessing though, posting a video might help.

I hope this helps!

DB
 
I appreciate the help! The frame is 3/4" pic, and no cracks, and is pretty stiff... As for the solder, that's what I'm thinking right now, and I will replace the power distribution board with a mess of wire;)
 

Snarls

Gravity Tester
Mentor
Are you sure it is not just bad tuning? Also make sure that the flight controller is not receiving too many vibrations.
 
It's just out of the box tuning. I don't know much about tuning, but I figure pid's would cause a little wobble, but the yaw? Also, the original FC had many things not set to default, even though I never touched them. Could bad tuning/pid's cause this? The FC is pretty vibration isolated.
 

Snarls

Gravity Tester
Mentor
Make sure the ESCs are properly calibrated. Shortening the arms might help if the problem is due to lack of rigidity in the frame. You can also try flashing the CC3D with cleanflight to see if those stock PIDs work better. Pics/a video would be nice too. If its a hardware problem you can try moving motors and ESCs around and seeing if the problem follows.
 
ok, i also tried adjusting the yaw relative to the board. i set it to 180, thinking that then the front would become the back, right? it still reacted like the front was the front when i tried powering it up and adding a pitch/role input. however, when i tried taking off it would keep flipping over in random directions. then i tried setting it to -180, and had the same bad results. when i set it back to 0, i tried taking off again, and it became un-flyable, it would just keep flipping over! i managed to get it a couple feet up, but then it flipped into the ground.
 
i think the ESC's are properly calibrated. my frame is very stiff, and i doubt that is the problem. the problem only seems to affect yaw. could it be that one of my motors might not be perfectly straight? i can post a pic of the build, but will have to wait for unlimited internet for the vid.
 
IMG_2683.JPG IMG_2763.JPG IMG_2769.JPG
 
so, i went to shorten the arms, and after i took the landing gear/prop protectors off, it flies pretty darn good! a little wobbly, but with some pid tuning.