Posts: 155 Threads: 29 Likes Received: 36 in 28 posts Likes Given: 68 Joined: Apr 2017 Reputation: 0 Hi guys, It started today, I take off, everything looks good, and after few seconds I see the drone keep trying to flight backward, at the beggining is easy to compensate with pitch, but after a minute or so, it gets really difficult to keep the drone stable, it I release the pitch forward, it comes backward really fast. Any idea what could be causing this behavior ? My acelerometer looks good, properly calibrated. Thanks • Posts: 1,590 Threads: 89 Likes Received: 1,283 in 768 posts Likes Given: 1,274 Joined: Jan 2017 Reputation: 31 So it sounds like you're flying in a horizon/self leveling mode right? I am guesing this because you say Quote: I release the pitch forward, it comes backward really fast Have you tried re-calibrating the accelerometer? Maybe something went wonky with it Other than that I might check settings / trim on your radio... That's probably where I'd start, you may have tried all 3 but I figured I might as well just throw it out there just in case. carl.vegas Current Quads: Operational: Diatone GT2 200 In need of repair: Bumble Bee, tehStein, Slightly modified Vortex 250 • Posts: 2,416 Threads: 51 Likes Received: 1,861 in 1,175 posts Likes Given: 3,315 Joined: Mar 2016 Reputation: 74 You had trouble in an earlier thread with proper motor direction and prop mounting. Are you sure you have those correct now? Can you post a photo of your quad with a clear view of the propellers (not spinning)? Posts: 155 Threads: 29 Likes Received: 36 in 28 posts Likes Given: 68 Joined: Apr 2017 Reputation: 0 (02-May-2017, 02:04 AM)Carl.Vegas Wrote: So it sounds like you're flying in a horizon/self leveling mode right? I am guesing this because you say Have you tried re-calibrating the accelerometer? Maybe something went wonky with it Other than that I might check settings / trim on your radio... That's probably where I'd start, you may have tried all 3 but I figured I might as well just throw it out there just in case. It was all done, in theory is not about acelerometer. Radio trim was also reviewed. And yes, I´m using self leveling mode • Posts: 155 Threads: 29 Likes Received: 36 in 28 posts Likes Given: 68 Joined: Apr 2017 Reputation: 0 02-May-2017, 02:36 AM (This post was last modified: 02-May-2017, 02:37 AM by flabombarda.) (02-May-2017, 02:26 AM)sloscotty Wrote: You had trouble in an earlier thread with proper motor direction and prop mounting. Are you sure you have those correct now? Can you post a photo of your quad with a clear view of the propellers (not spinning)? Hi sloscotty, Yes, I can post a photo as you suggest, I reviewed that also, and motors look good. It was working normally yesterday. Nothing was changed at configuration level or hardware level. btw, the front is the right of the picture. Hope it helps ! • Posts: 1,773 Threads: 30 Likes Received: 1,199 in 755 posts Likes Given: 714 Joined: Oct 2016 Reputation: 45 Check the receiver tab and make sure your pitch (elevator) channel is at 1500 with stick centered. If you've tried to adjust with the trim on your transmitter (which you shouldn't do) you may have changed this Posts: 2,286 Threads: 38 Likes Received: 1,527 in 995 posts Likes Given: 1,881 Joined: Apr 2016 Reputation: 72 Drifting in self levelling modes that gets worse and worse as the flight progresses is normally caused by excessive vibration as this confuses the accelerometer readings. However, it normally affects a random axis rather than just one and in one direction. It's hard to say much more without a black box log though. Posts: 947 Threads: 66 Likes Received: 350 in 248 posts Likes Given: 164 Joined: Apr 2016 Reputation: 34 What fc? I had a diatone naze32 rev6 clone FC once that did something similar, and that board had a known fault with accelerometer drifting. I could plug it in to betaflight and in the illustration quad/brick in betaflight it slowly kept spinning faster and faster Posts: 155 Threads: 29 Likes Received: 36 in 28 posts Likes Given: 68 Joined: Apr 2017 Reputation: 0 (02-May-2017, 07:48 AM)oyvinla Wrote: What fc? I had a diatone naze32 rev6 clone FC once that did something similar, and that board had a known fault with accelerometer drifting. I could plug it in to betaflight and in the illustration quad/brick in betaflight it slowly kept spinning faster and faster This is Naze32 Rev5. When I connect it on Cleanflight/BetaFlight it doesn't spin or do anything abnormal. What I'm going to do here is erase the board and configuration and flash it again. Is a simple test and worth a try. Thanks all ! • Posts: 34 Threads: 11 Likes Received: 55 in 22 posts Likes Given: 34 Joined: Mar 2017 Reputation: 0 I have a Naze32 Rev 5 that's been in use for a year on a couple of quads - I think it's a clone. In horizon mode the quad stays perfectly level, but if I do some aggressive flips/rolls it starts to drift in a random direction as you describe. I tried calibrating, checking for excess vibration etc but I've had to put it down to a dodgy board. I only fly in acro mode now and the quad flies beautifully so it's not a problem. Fyi - the graphic in Betaflight configurator was always perfect and stable, and the quad would initially fly great and level. Posts: 2,286 Threads: 38 Likes Received: 1,527 in 995 posts Likes Given: 1,881 Joined: Apr 2016 Reputation: 72 With the older F1 boards, this seems to happen quite a lot with recent firmware. The latest Betaflight/Cleanflight reads the accelerometer at a much slower rate than it reads the gyro and I think this leads to greater accumulation of errors in the system. Maybe even of aliasing noise due to the lower sampling rate. With older versions of Cleanflight - 1.13 and earlier - this was never a problem, especially if you used the MWRewrite integer PID controller. Posts: 155 Threads: 29 Likes Received: 36 in 28 posts Likes Given: 68 Joined: Apr 2017 Reputation: 0 (03-May-2017, 10:42 AM)unseen Wrote: With the older F1 boards, this seems to happen quite a lot with recent firmware. The latest Betaflight/Cleanflight reads the accelerometer at a much slower rate than it reads the gyro and I think this leads to greater accumulation of errors in the system. Maybe even of aliasing noise due to the lower sampling rate. With older versions of Cleanflight - 1.13 and earlier - this was never a problem, especially if you used the MWRewrite integer PID controller. Guys, That was really weird, I've flashed firmware again (latest Naze version, erasing all configuration and the problem seems resolved. It's not clear to me what caused the issue. About my board, it's for sure a clone of Naze32 Rev5. At some point I'll start flying in Acro mode, and my guess is these issues might go away since acelerometer is no longer as a source of information for autolevel. Am I right ? Thanks, • Posts: 947 Threads: 66 Likes Received: 350 in 248 posts Likes Given: 164 Joined: Apr 2016 Reputation: 34 (03-May-2017, 02:44 PM)flabombarda Wrote: Guys, That was really weird, I've flashed firmware again (latest Naze version, erasing all configuration and the problem seems resolved. It's not clear to me what caused the issue. About my board, it's for sure a clone of Naze32 Rev5. At some point I'll start flying in Acro mode, and my guess is these issues might go away since acelerometer is no longer as a source of information for autolevel. Am I right ? Thanks, Yup, the accelerometer is not used when flying in acro mode Posts: 2,286 Threads: 38 Likes Received: 1,527 in 995 posts Likes Given: 1,881 Joined: Apr 2016 Reputation: 72 You are correct. In acro mode, only the gyro is used and it has no concept of 'level'. It's really odd that reflashing it made the problem go away though. Ghosts in the machine. • Posts: 155 Threads: 29 Likes Received: 36 in 28 posts Likes Given: 68 Joined: Apr 2017 Reputation: 0 (03-May-2017, 02:49 PM)unseen Wrote: You are correct. In acro mode, only the gyro is used and it has no concept of 'level'. It's really odd that reflashing it made the problem go away though. Ghosts in the machine. I agree, especially because I did not change any setting in the past week or so. Maybe is a corruption on the board memory. Before flashing I was trying to connect on Cleanflight to take screenshoot of all my settings, but when I tried to do that, after a reboot, the board "bricked" (happened before already). I had to put a jumper on the BOOT pins, and reflash it erasing the whole board, resolving the problem. • |