I activate again oneshot 125 (type2) , I try two times, the secuence tones looks working but them, when I connect batt, two motors beep, after push button, then the other two motors beep, one first and few seconds later, the second one, not flying test yet, is night here, but not looks uniform yet.
Thinking about my issue, I have active breaking enabled that slow down motors faster, that can be the issue? I read at wiki that Brake shoud be =off but not shure if it’s the same.
Another thing I see after update rc3 is that Terrain flag appears yellow, I never see before.
Ok, I guess after you try again and provide more feedback we can look into it. I’m sure as a minimum we will need a dataflash log of the final result… it could be so many thing, maybe the deadzone on each ESC is a little different so MOT_SPIN_MIN and/or MOT_SPIN_ARM needs to be a little higher.
1 Like
Thank’s Randy, tomorrow I try to fly and provide log.
Have same problem. But what fly log if only 3 of 8 motors start good?
Some test outside, esc tones are not sync, I armed so motors spin but only tree spin,tried tree times, a wile later the stopped motor beep and them the four motors armed, I try to take off but motors sound badly (perhaps untuned for oneshot) and I abort. Here this log, tell me if you need another info.
(P.D: from other test I changed the FC antivibration pads for thinner ones to avoid upper frame touching the FC but I test and works well with type=0 )
Cala,
I haven’t had time to test , but this behaviour is exactly the same I got with rc2. Will test tomorrow and post the results.
1 Like
I return to type=0 and normal beeps return to my copter, the bad motor sound was a mistake I did with rate roll D term trying to tune pitch but I hope, no related with unsync esc…
M_S_MIN=0,15 and M_S_ARM=0,1 as default, change like 0.15 and 0.20?. Thank’s Randy
I try PX4 firmware and oneshot125 work good on all motors and fly.
But there, if I wait a minute, 6 of 8 motor do start beep one by one. When arm, spin only 6 motors too. If I give more throttle, another 2 esc go to recalibrate.
In different try have different order start beep escs. If wait 10 minutes after push safety - all escs start one by one by differen pause and when arm - all motors spin.
Disable safety switch and all escs start
At the same time? with PX4 firm? thank’s for comment your experience.
Yes. All at the same time. PX4LITE
1 Like
Cala2,
Have you tested the ESCs using the Mission Planner’s motor test page? If you use the Mission Planner’s motor test page you can test when each motor starts. You’ll see there’s a percentage field that you can change. Even after a calibration I find that the ESCs do not always have the same size deadzone at the bottom. They can still be off by 1% or 2% from each other. In these cases I set the MOT_SPIN_ARM just a couple higher.
http://ardupilot.org/copter/docs/connect-escs-and-motors.html#testing-motor-spin-directions
Edit: I think Tridge and Luis have found some further issues this morning with OneShot support that they’re investigating. I’ll post updates here when I have more info.
Thank’s Randy, no I don’t use motor test, it’s very noticeable time between the last motor armed.
Ok, I wait investigation, my copter flies well without oneshot so not hurry, great work, thank’s.
Randy and Cala,
Here is the current status with RC4 / BLheli 16.7:
mot_pwm_type=0
Everything OK
mot_pwm_type=1
Automatic esc calibration works fine. Esc arm at random time ( less than 2 min ) after unlocking the safety switch. Trying to arm without waiting for it ends up having a few motors dead ( of course haven’t tried to give throttle )
If you wait, the flight is OK.
mot_pwm_type=2
Automatic esc calibration works fine. After the automatic calibration , the motors started to twitch. After power everything down in a rush, I’ve find out that the twitching started right after unlocking the safety switch. Haven’t tried to arm or takeoff…
My parameters:
mot_spin_arm=0.035
mot_spin_min=0.1
mot_thst_expo=0.3
mot_thst_hover=0.3
in blheli all motors are around 1100 - 2000 pwm range, respecting the clock fluctuations.
Update:
A few test on the bench ( pwm_type=1 ) , it looks really really random the time, order and waves of activation the motors went. I take a look at the code and really have no idea what is happening. It is an organic behaviour…
I will keep using oneshot, i can live with this delay - better than disabling the safety switch.
I saw that LuisVale risen the issue in blheli thread:
http://www.rcgroups.com/forums/showpost.php?p=35665775&postcount=9848
I could reproduce the tests and it makes sense that this could be coming from blheli. Hope they fix it soon.
1 Like
Thank’s for report Fernando, mine, after every motors arm, then I can fly normally, I flyed like that with rc1 and 2, now I moved to type=0, in case.
Just so you all know this is being tracked at https://github.com/ArduPilot/ardupilot/issues/4789 and we are thinking this may not be an ArduPilot problem. @LuisVale has done an amazing work trying to debug this problem.
1 Like
Sure! I’ve just posted there that I was able to get rid of the twitch by decreasing a little bit the rc3_min ( thrust channel ) parameter. The mot_pwm_min didn’t helped me. It is still 0. So, the real output has decreased from 135 to 131. That margin was enough.
I’m ready to test that this weekend. I hope there aren’t any unwanted side effect 
The random start is still there … weird.
EDIT: Side effect found… Cannot arm. Throttle too high error. I feel that it will need more tuning in order to get it running. Will try to calibrate by setting the min value to something higher than 135… maybe that could be enough.