I’ve never tried such a test, 0% → 100% throttle.
In reality the worst case scenario is going to be MOT_SPIN_MIN to MOT_SPIN_MAX
which equates to about 10% → 95%
Also can you check with Sine Modulation enabled. I think this will lock you out from changing the PWM frequencies. I’m curious to see if it’s any different, and also I dont have anything set up where I can test at the moment.
EDIT:
I’ve always used Sine Modulation, but I’ve also only ever used what most would consider LowKV motors.
We’ve experienced extended flight times (up to 50 mins) and excessive payloads that were beyond expected limits - Never yet had a desync with a BLHELI32 ESC.
Yes, agree. I would suggest just continuing on with the setup/configuration. As I mentioned I have used default parameters (except for low RPM Power Protect off on low kV) from 320kV to 3800kV and the desync bogeyman has never arisen in-flight. I have forced it to happen on the bench at Dshot1200 and madly jerking the throttle but that doesn’t count for anything. Think about what the motor can actually do with a prop mass attached…
Tried 10-95% Again almost guaranteed desync…
Even 80-95% is desync ! However 0-90% seems to be ok, It looks like it happens only when trying to reach over 90.
I am trying to be extra safe because I already had the first desync in flight after 12 years ! (Last time that happened we were using turnigy plush escs with SimonK)
I cannot trust it still.
I’m out of ideas at the moment… Only thing left would be trying a different motor at same kv. .
As for sine modulation : Tested it and seems to be slightly worse.
@dkemxr That’s a first for me too… Have you by any chance used the same motors? Tmotor 4004 400kv?
Are you certain you’re actually using DSHOT? I’ve experienced poor motor behavior when accidentally allowing the autopilot to revert to PWM output (though I’m not sure I’d characterize it as desyncing).
I was going to ask what motors you are using.
I would say those MN4004 Antigravity motors are made specifically for trying to achieve long flight times, and you can guarantee they will have poor reaction to disturbances like wind - either because of some inherent motor design issue (unsure), or often because people pair them with props that are much too big (chasing that elusive long flight time)
They may work better with some specific motor timing and demag settings, but I’d say that may be masking a performance or design issue.
Tomorrow I’ll try a couple of tests when I get organised. I dont have those motors, but 390kv should be near enough.
I used ecalc to compare our two different motors, which might be a bit meaningless but can be a reference anyway. They give surprisingly similar performance for the same inputs.
I have escs connected to 9-12 outputs. And MOT_PWM_TYPE = 6. So the FC better be sending dshot.
@xfacta Yeah I’ve used their cousins the, mn4006 380kv with great results and was able to achieve a very good tune.
I tried all possible combinations (demag high, timing set to low or high, max acceleration etc ) and nothing seems to get it better than it already is.
I think this might be an inherent issue with those motors, see here:
Seems like only some esc FW is able to correctly run them at high rpm.
Since changing to non 4-1 escs is impossible we may need to swap motors, @xfacta do you have any tested reccomendations for 13inch props?
On hover tests those motors were able to give about 7A with 1302 props and 1.8kg AUW.
EDIT: I’ve seen ecalc being terribly wrong before so I don’t use it at all. got an octo with 60min flight time and ecalc said it flies for 32mins
Pretty sure the motors on the Hexsoon 650 are 4004’s. I have these with 15in T-Motor props and Holybro BLHeli ESCs and have never had a problem. I do have a genuine 4004 as a test motor as well since I seem to remember someone else complaining about these motors a while ago. I run with pretty high filter settings and everything is very stable.
It seems that the issue occurs when the motor tries to reach a certain rpm, so maybe the 300kv version does not exhibit this problem.
If you do have the 400kv version could you please run a test with a full battery?
Just a 10% to 100% in motor test and see if it desyncs.
ecalc is usually fairly accurate, maybe a bit conservative in that you could get a bit more flight time than it calculates.
The trick is the battery weight - ecalc uses a per cell weight, so if you are entering some custom battery pack you would divide the total battery weight by the number of cells and enter that into the cell weight in ecalc.
Otherwise you can enter the total all up takeoff weight regardless of individual components, which is OK if you already have all the components or know the final weight
I didnt get to test my 390kv motors today, but I might tomorrow.
I tried rampup power 15 and 20% with no luck it seems that this has not effect in higher rpm where the problem is.
I was using dshot 600 though when I tested that though…
Tried dshot 300 with the recommended settings and the behavior is the same. The only thing I have not tried is testing it using pwm control. Just for the sake of it.
This seems fishy to me. Typically with desyncs you lose the motor and its usually associated with large throttle swings. Why do you think this is a desync - your video doesn’t sound like desync to me?
What you mean is not very good for the motors ? Rapidly going from 10-100?
Desyncs from my experiece sound like that, these grining noise as the esc tries to regain sync and the almost complete loss of power.
I aggree about the large throttle swings, But maybe the ESC is barely staying in sync with the motor at that RPM and any change from that is enough to throw it out of synchro - just an assumption.
Aditionally if I very gradually raise the throttle it works fine.
