Several times tonight we had my 3" quad suddenly fly up into the air for no good reason.
Looking back at the video there was no reduction in Link Quality. Nothing reported at all.
It seems to be happening when we do tight U-turns.
I responded in the video below by cutting throttle.
We’ve been running for the first time with Airmode on and I do wonder if that is causing it.
I don’t know what a failsafe is.
Thank you Crossfire 
lol
I don’t think it was Failsafe… nothing reported in the OSD at all.
Gyro error? Does it have a black box log?
I do have black box… I can’t for the life of me work out where in the logs that bit is
I haven’t done a lot of black box stuff myself, but I think there is a software you can download that gives you visual representation of the data. Something along the lines of Betaflight blackbox reader. I can’t remember it’s exact name.
Yes I’m using that
Yeah i get ***** when I’m close to failsafe, then “failsafe” on the warnings with betaflight OSD. Also more importantly on a failsafe it will just drop out of the sky, not fly up.
Yes, what I thought… it has to be something odd happening in the FC.
I think possibly because I’ve started to see some damage in the PP frame.
I was going to rebuild that anyway… it’s interesting to note that the cracks may be causing this.
Where do you have cracks? i would be apprehensive to fly if the arms are cracked or something tbh. Have you played with your PIDs at all? it very well could be airmode if you have some cracks and the frame is flexing under G force. Actually watching the clip again…how sure are you that you dont have horizon mode enabled?
I only noticed after this. The 3d printed layers are finally starting to split after a month of serious crashes by beginners. No idea if a carbon fibre frame would have broken under this stress.
Anyway, I have noticed that if I increase the flow rate in the printer I will get stronger prints with PP so I will be doing that with the new frame.
Cracks are right near the motor bases.
No.
Yes, I’m starting to think that myself.
Very sure. See it says AIR in the bottom right corner.
You can be in both airmode and horizon mode. Airmode simply counteracts any external forces on the quad…tbh i wouldnt fly without it there is little reason to.
From the sound of this its likely the motors flexing and airmode trying to counteract that.
A carbon frame would be an awful lot stronger than 3d printed…layer delamination really doesnt take much force at all, i can peel it apart with my fingers…ive put my carbon frame full speed into a wall and cant even tell where it hit.
There’s a separate position on my switch for horizon mode and then it shows HORZ or summat.
Agreed
Interesting… I have no experience with it.
I have just been given a 5" carbon frame which will be my next build. This 3" will stay as PP for now at least.
Just checking because it looked like it was trying to find the horizon basically.
Yeah so with 3d printed stuff the simple rule is that its only strong along the grain, across the grain its incredibly weak…I wouldnt 3d print structural components for a quad tbh its good for things like antenna mounts and gopro mounts but you want the frame to be nice and rigid for good performance.
I’m going to see how I get on with it for a bit longer yet before switching to a stock carbon frame. That seems like defeat to me. I’m having too much fun designing this all myself.
If you’re sticking with a pp frame and you get more runaways like this, try increasing the filtering. And/or reduce the PD gain. This should help a little with the vibrations that are translating back to the gyro.
I had to do this on 3d printed frames
Design carbon rods inside the arms to stiffen them up, print it 100% infill and consider using a stronger material you can get all sorts these days.
Sounds like a plan. I do have a black-box on this stack so I can do some proper analysis… should I ever work out how!
Though I think it only became an issue once the frame started to crack.
This PP is pretty good… and something to stiffen the arms would be good… maybe even just glued or tie-wrapped on.
I already print 100% infill with these. Actually I increased the horizontal and vertical shells.
I’ve increased the flow rate by 10% in this next one, which is printing right now.
My design already incorporates struts in order to try to increase the strength.
You could design it with a channel to take a carbon rod in the middle, pause the print at that point put the rod in then continue printing. I’ve seen a few people do things like that and its pretty effective.