Motor start question

[Kisssys]

Hi Glen,

I was testing a repaired AQ6 board and thought I would demonstrate what your copter should do. It is powered up in the tilted position, I take off slowly or fast and land and takeoff again with no problems.


Cheers
Steve

[LPR]

Glen

I have some of the same problem the you're describing. All of my motors always start at the same time even if the octo is tilted but I often have motors that don't start correctly. I've learned to watch the motor carefully to make sure all 8 are running correctly. I have seen a number of times that one motor will run the wrong direction slowly or stop.
On a test stand the motors have the same behavior if a try to start them with a fast change of the throttle. Changing the motor timing In the ESC32 might help. I have not tried it. Other motors I've used with the ESC32 have not had starting problems

I think the motors are RC Timers 4215- 650 KV and the props are APC 12 x 3.8. I'm using 5s A123 batteries.
I just tested my octo and if it's level while starting the motor all start well. If the octo is tilted the lower motors start at a higher RPM and about one out of ten times I try starting them, one of the lower motor will have a starting problem. A motor will turn slowly and then stop or start backwards or not start at all. If the octo is level all motor will start at slow RPMs.

Both quads I built with SimonK ESC never have had a motor starting problem.

Larry

[Max]

That does sound like an ESC/motor combo problem. I've seen this many times in the last 10 years of electric RC. Some combos will start smoothly, some won't, and some will reverse the rotation, often depending on how quickly power is applied. The problem got worse when Chinese motors starting hitting the market (for a while cheap Chinese ESCs were smoother than expensive Castle ones). Some earlier versions of Simon's firmware had the same issue with my Tiger motors, but this got fixed with a later release. Advance setting does seem to have an effect on this (I played with it in WiiESC once, before Simon updated his fw, and it helped smooth things out), but I'm no expert in such things.

The bottom line is that you want your motors to start smoothly every time, regardless of how quickly (or slowly) power is applied. Anything else will be frustrating, at best.

-Max

[SuperSharpShooter]

Mine seems to be behaving after the last 4 or 5 flights. It looks like the dampers holding plate that the FC was on were slightly too soft, allowing it to wobble a bit on spoolup, causing the flip. I've replaced them with four threaded m3 rubber dampers, which look to be performing better, but I still think they are slightly too hard (my motor logs still aren't as clean as I believe they should be). I'll try it with some dampers that are somewhere in between once they arrive. Hard mounting the FC to the frame will generally cause the ACCs to get swamped, causing fly-aways.

I'm running simonk F-30A ESCs and sunnysky 2814-11 motors. On a slow takeoff from a tilted surface, I get that same bouncing behaviour from the low side that you can see in your video Steve. Once it's in the air it looks to be performing well.

[Kisssys]

[SuperSharpShooter]

I spent some time tuning the PIDs yesterday, and everything seemed to be going well, but then I took it out today, and it flipped 5 times in a row on takeoff from a relatively flat surface. After a power cycle it took off normally and was behaving well.

Looking at the logs, there are some weird spikes on some spoolups and not others. There looks to be a lot of noise on only my Z gyro too. Is it possible that the z gyro is being incorrectly initialised sometimes which might cause this behaviour until rebooted? I've found that if it flips once, it will keep doing it until it's power cycled. Other logs I have where it flips show noise on only this gyro, whereas on successful takeoffs, this noise looks to be absent. I could be interpreting the logs incorrectly, so if someone else could take a look it would be appreciated.

Log 373 shows flips (ends up as being tilted onto its side on each takeoff attempt) and log 374 is the subsequent successful flight.



Thoughts?

Gyro log:


Motor logs - All 5 takeoffs


Motor logs - First 2 takeoffs

[chschmid]

What function do you have on Radio channel 5 (4 in the Log).
I cant get rid of thought that you take off in PH. That would explain the random flip and your strange log.

Cheers
Christof

[SuperSharpShooter]

Hi Christof.

Channel 5 is the flight mode switch. The logs show that it's low (-863) and in manual mode for all of the times the copter flipped on takeoff in log 373. Log 374 shows takeoff in manual mode, and then position hold was engaged for the remainder of the flight.

I've never taken off in anything other than manual, so that's definitely not it.

[LPR]

Glen

The one thing I did when I first started with the AQ is I adjusted the DIPs like the AQ was a normal FC. After way too many crashes I went back to all defaults and I found that defaults work best. If you have too much vibration lowering tilt rate D usually helps other wise all defaults work best on my 9 lbs octo and two quads under 2.5 lbs.

Larry

[SuperSharpShooter]

Hi Larry, I did start with default PIDs the other day when tuning, but reduced Tilt Rate D and Tilt Angle D slightly which gave me much smoother performance. I also reduced Tilt Angle P to 50 from 60.

I think this intermittent flip is caused by something other than badly tuned PIDs. When it takes off it flies well. It's just when it flips, there's something weird going on which will make it keep flipping until it is power cycled. I would like to either confirm or eliminate a hardware issue as being the cause of this, but I lack the expertise to make a definitive diagnosis.