SUNNYSKY X3508S 700KV about tgy HOT 18 OPEN

Hmm, looks pretty simple. I'll buy some and see if I can reproduce any problems. When you say sync issue, what are you seeing exactly? Can you make a video of the issue? DJI Naza should have no problems, though you might want to calibrate the throttle range if you are having problems with it idling a bit hot.

from tgy.

Thank you for the quick reply.
My problem is when I put fast yaw stick the left side, so copter tilt to the left side, And in this moment some ESC blinking RED light and sound of motors sounds like out of sync.

from tgy.

Copter hovering well. When turn with yaw stick slowly so everything work fine. To the right side work without problem. If turn fast left will cause problem. I tested 3 firmware 2014-3-6, 2014-1-19, 2013-9-20, from http://lazyzero.de/ KKflash tool repository.

from tgy.

Hmm, ok. Did it also not work with stock firmware (did you test it first prior to flashing)?

Yaw stick is a good way to accelerate motors from slow speed, since rocking from one side to the other can change motor speed from quite low to high power. Timing issues most often occur in this case, where current draw is highest.

I am not sure how to help you until I can also try the same motors. Do you have development experience? If you could build the assembly code, you could perhaps try to change some parameters. If not, you may have to wait or try a different ESC in the meantime.

from tgy.

I have development experience unfortunately :/ Stock firmware stock firmware also does not work. After I test stock firmware I try another one. I seen this video when is 2013-09-20 release https://www.youtube.com/watch?v=iXC9SuuEtqo . It is too 3508 700kv sunnysky. My motors have maybe slightly different windings. I must wait for newer firmware :/ Thank you for your response.

from tgy.

Sim: You know you recommend a firmware that would theoretically could go? Thank you.

from tgy.

Having the same issue with turnigy multistar 3508-640kv motors/ afro 30amp esc´s with APM 2.6 running arducopter 3.1.1, works perfectly hovering in stabilized mode, but goes out of sync in the alt hold or loiter modes, then back in stabilized mode it is stable until you give long or big yaw inputs, then it losses sync again until yaw input stops. Any help will be very appreciated!

from tgy.

Through NAZA me to do the same. Through the receiver engine works perfectly. Slow and fast throttle going well. Now we can buy again other new controllers and these Afros put to the box.

from tgy.

I have T-motor 3508-700Kv and tried different firmwares. Always had sync
issues at high throttle or at very fast speed change.
Then I tried this
http://0x.ca/~sim/tmp/tgy_testing_2013-10-29_8c2cf42_comp_pwm.zip and never
had any more issues.
Not changing until somebody with similar motors will report full success
with latest firmware

As I understand that commit (8c2cf42) never went into git, but it is the
only one that worked for me.
Sorry, but I have not tried the latest master yet because I have no more
motors to test, and don't want to sacrifice my X8 :)

Cheers,
Emile

2014-03-19 9:53 GMT+01:00 sqo3d [email protected]:

Through NAZA me to do the same. Through the receiver engine works
perfectly. Slow and fast throttle going well.

Reply to this email directly or view it on GitHubhttps://github.com/sim-/tgy/issues/36#issuecomment-38028806
.

from tgy.

Today I tried the firmware and still no success.

from tgy.

emile: just tried the afronfet hex from 8c2cf42....it did not work.

from tgy.

BTW the firmware release 2014-03-06 has the same issue

from tgy.

I tried to order some SunnySky X3508S 700KV from himodel to debug the problem you are seeing, but of course they ran out of stock right after I placed the order. I am trying to see if they can ship just one. You may have better luck if you enable COMP_PWM, or if you adjust MOTOR_ADVANCE or TIMING_OFFSET, but this really should not be required to make it work -- it should just be an efficiency change at most.

from tgy.

I do we have to wait on the resulting firmware. I do not know try to change the settings. My engines SUNNYSKY X3508S 700KVA (4S) will work with turnigy plush 25A that I had ordered as a plan B.

from tgy.

Hello! I have tested this motor and at 4S, with and without COMP_PWM, I am not able to reproduce any issue (2014-03-06). This motor is very low resistance and in a high current jump test, it does significantly exceed the current rating of the ESC, even though we do not lose timing:

http://0x.ca/sim/esc/timing_loss_fix/Afro_HV_SunnySky_X3508S-16_700KV_4S_1084_to_1900_step_phase1_phase2_Vbatt_10mVperAmp_split.png

I have another report of issues with the Afro 30A, and in that case, the red LED comes on very early on power off while the motor is still spinning. Owen was able to fix this issue by swapping the ATmega8 with a new one. I am wondering of some of the Afro 30A shipped with an ATmega8 containing a faulty comparator unit or similar.

Could you confirm if your red LED light turns on early on power off? What I mean by this is that if you spin up the motor (enough to make it take a second or so to stop spinning if you turn off the power), the red LED should not light except for a short time as the motor comes to a stop. If it lights right away on power off or shortly after, the ESC is having a problem sensing the motor rotation even when no power is involved, which is a sign of a hardware problem. See this video which demonstrates differences between ESCs/motors: https://www.youtube.com/watch?v=9xWrTMmFH5w

Do any of you have a similar situation, by chance?

from tgy.

Hi Simon, I sent Gavin's parts back and have heard that the oct seems to have no motor issues with the *tc module I built. That is a quite standard Afro module with COMP_PWM.

I have ordered another Afro30, so I will have two spare, and I will select the two with longest and shortest red time from my quad + 2 spare, so that I have a pair with different behaviour for testing.

Re paranoia, it is easy to think that some cut price deal offered out of spec chips... and I know that in production there will be devices that fail tests... but I am inclined to think that most reputable makers do not brand bad chips.

Owen

from tgy.

Thanx Simon, looks i won the lottery with bad chips. Ran the the early power down test and they go red immediately and stay red for a while even after props stop, then after a while they go green. Are they a lost cause, or are they worth repairing? Does HK know about this issue? Thanx in advance for any advice!

from tgy.

Forgot to send the email (multitasking fail!), so sent it a few hours ago. Waiting for reply.

Meanwhile, can you send a high res picture of your ESC? Put it on a scanner or take a picture with a macro or similar, with the heat shrink removed. I'm interested in the markings on the ATmega8.

Or, you could send me your worst one, for science. I'd like to test the comparator offset voltage to see how it lines up to the datasheet. I'll probably be away on vacation by the time it arrives, though.

from tgy.