Uncertain about what to buy? Check out THE ULTIMATE FPV SHOPPING LIST: https://www.fpvknowitall.com/ultimate-fpv-shopping-list
SUPPORT ME ON PATREON at: http://bit.ly/2Ep5Sia
THIS IS MY FULL TIME JOB. Here are other ways that you can support me: https://www.fpvknowitall.com/support-me/
—
Betaflight 3.3 has a feature called gyro overflow detection. It detects and prevents a bug in the gyro that causes your quadcopter to yaw-spin to the moon (also known as YSTTM).
If you aren’t expecting this feature, it can cause you to think your quadcopter is defective. The problem that led me to this discovery is a person reporting their quadcopter would keep doing backflips even though it wasn’t commanded to.
Fortunately, you can turn this feature off if you want to.
I’ll show you what this feature looks like when it kicks in and how to turn it off if you don’t want it any more.
A few comments from one of the developers: If you have an ICM-206xx gyro I would not recommend turning off gyro_overflow_detect. These are the gyros that have the overflow/inversion bug and you will have no protection from a flyaway event. This feature was developed in direct response to these events and people losing their quads rather than just as a hypothetical “this may happen” kind of decision. The overflow can be quite easily triggered due to a gate strike or crash where the quad is spun through outside forces. If you have a non-affected gyro (like MPU6000) then you can generally safely turn it off. But there are still cases where a flyaway could occur due to multi-axis coupling and motor saturation where the spin is tumbling. However these events are more rare and will usually self-correct but there still will be a scary flyaway-like event and most people will disarm before waiting for it to stop on its own.
The detection starts when the gyro rate exceeds 1950 deg/sec and disables the PIDs. The PIDs remain disabled until the gyro rate drops below 1850 deg/sec for at least 50ms.
For those interested, the overflow event can bee seen in this image from a blackbox log of an ICM-20608G gyro showing raw data from the yaw axis as the spin rate exceed the overflow threshold.
Hi Joshua, I hope you don’t mind me asking here for another topic.
There is some discussion about prop efficiency. So I understand that 5 inch props on a 6 inch frame do better. So is there any disadvantage? Do we loose the agility?
Would be great if you could comment this.
Thank you Joshua.
Again really interesting video. Thanks
Hi Joshua I have been scanning your somewhat prolific output to try and find a cure for what I guess/hope is probably a simple problem but have yet to come across another example explained.
So here goes I have obtained a King Kong Hex 300 and decided to swap the flight controller for an SP Racing F3 a fairly simple task!!! I only have for reasons I wont go into Betaflight 3.3 on my computer so I flashed the controller with the latest firmware and proceeded to configure the controller.
All was fine I can look at the receiver input set the upper and lower thresholds etc the little picture of the copter spins and gyrates according to the appropriate stick inputs the little red light comes on when arming so then I look at the motor screen and find there is no input to the motors at all I cannot even drive them via the on screen controls I double checked the config and discovered that it should be in OneShot 125 and select it and click and save at this point the motors immediately went to full power still no stick control or onscreen control!! disconnect battery reconnect back to square one ??????
Is there anyway this would intefere with crash prevention? Or have they done away with that feature? I really like crash prevention especially on my 3″ set up, im still on 3.2.? And am about to upgrade and want to make sure that wont affect that feature. Thank you sir
the yoda of quads!!! it will be a sad day the day you quit quads!!! we will be lost!!!
I know that this isn’t related to this video, but i have a problem. I got new quad with TGY-IA6C and tgy Evolution and i want to set up RSSI on osd. Does anybody know how to do that? Please help im a newbie ?
Hi Joshua, I need some help!! I flashed Betaflight 3.3 onto a Furious F4 OSD board and couldn’t get it to arm, plus the Accelerometer won’t come on!!
Hmm. I have a new 2.5″ build (massive droner) and I was doing some crazy moves trying to test myself, I yaw spinned clockwise upright, then flipped upside down and yaw spined counter-clockwise inverted, it then dropped out of the sky- no FC reset, no range issues, it was close to me, I don’t think it was a brownout as I’m well below the 3A limit.
I have to try tomorrow and replicate it.
Kinda sucks as I fix my random dropping out of the sky due to a defective RX with one build, then the next acts up.
I have experienced YSTTM- it was scary as hell as it was in the front yard right over my car, I yaw spinned it LOS as fast as I could, it then yaw spun full HAM then shot up, I hit the failsafe and it bounced off the roof into the backyard it was like a ufo firework.. with blades- always wear safety glasses LOS!!!!
Thanks Joshua, that was it!!
hi, I just got this new quad with T-motor v3 *2600kv* motors and the Star f4s. So far i’ve had a bunch of problems and the gyro overflow was one of them. Another problem which I cant seem to find a fix for is whenever I fly above mid-throttle my quad starts to shake like crazy! the shakes are mostly on the yaw-axis, but it’s also on the roll and pitch.. Help?
Thanks Joshua, Noticed I did some extra backflips too.. Seems to be same problem. My rates are not that high either so you are right on that one. Thanks! never would have figured this out on my own.
Thank you so much Joshua, my blackbox looks the same of that – I tried a lot of solution and changed some parts ( like 2escs and a motor…) – Now I just need the rain who stop and a test!!
Good info. Indeed many would think they will never run into this issue because of not using ultra high rates, but clipping a gate or something can indeed cause a momentary situation where it does spin fast enough for this protection to kick in. Nice detective work. π
Is it possible that high frequency oscillation in a sharp 180 into prop wash might also trigger the PID to stop is detection is set to all?