I’m setting up my X500 V2 drone with Mission Planner. I configured the autopilot with DShot300 and ran a quick Motor Test and all motors were running okay at 5%.
Continuing the test on the bench, I switched on my RC, armed the drone, and noticed that the motors are not working as expected in one of the following ways: a) not running, b) running but not synced where few motors run first and then the rest start, or c) running only when throttle is at 40% or higher. I thought maybe this has to do with the firmware on the ESC and updating it might do the trick. I switched to esc-configurator and I noticed that motor 1 (M1) is set to PWM Min of around 1400 which made me think that this might be an ESC calibration issue. I went back to Mission Planner and started a round of ESC calibration (ESC type was set to ‘Normal’), let the calibration finish, and then unplugged and reconnected the battery.
After connecting the battery, I noticed a spark and quickly disconnected the battery. To see where the spark came from, I had to connect the battery 2 more times until I noticed it came from M1 arm; there was a lot of smoke on the 3rd time. Not my proudest moment! Pictures of the toasted ESC are attached. After doing some research, I now know that DShot ESC’s do not require calibration.
Only M1 motor was damaged. I inspected the other ESCs and there is no sign of damage. However, to limit any more ESCs from getting toasted, I’m only working with 1 ESC at a time for now. This ESC is constantly beeping and I’m not sure how to move forward. I’m guessing that the ESC calibration has changed the values on the AP only (and not on the ESC itself) but I’ve had enough magic smoke fill the room for one day.
My questions are as follows:
What causes this behavior where the ESC gets toasted like that after ESC calibration?
Should I consider the motor damaged? I do not remember it moving when the spark occurred.
How can I “undo” the calibration (that is now causing the ESC to beep)?
The list of params is attached. params.param (18.1 KB)
That looks like a plain old fashioned electronics ESC failure. Just about nothing you can do with the autopilot can cause that, especially if the copter is not flying.
ESC calibration doesnt change anything in the autopilot, it just signals the ESCs to “understand” what the full range of PWM the autopilot will be putting out.
Agreed, electrical failure on input side. Maybe voltage rating exceeded, missing external capacitors, you name it. Happened while the motor was not actively controlled, otherwise you would see self-desoldered mosfets on the motor side.
I’m glad I’m not the reason for getting the 5th ESC fried yaayy!
But, honestly, this is concerning. I haven’t even ran a maiden flight and I’ve got all these ESCs toasted. Is this a build quality thing? Is it a power module thing? Are there other ESCs that will perform better that I should be getting? It would be quite embarrassing if, let’s say, I’ve got the people ready for a demo, I plug the battery, and a bit of spark followed by whole lot of smoke comes out all the sudden. I’m thinking that working on magician tricks as a contingency to perform after the smoke fills the air so the audience isn’t disappointed but I sure won’t have that as anything other than plan Z.
I can work with that. Price isn’t that far off. I take it the Random Self-Combustion feature on these ones is switched from factory (and, therefore, I don’t need to practice my contingecy magician tricks as much)?
Honestly, I think it’s highly suspicious that you are able to burn the ESCs if you use the stock ESC+Motor combo from the holybro kit.
Use a multi-meter and check for short circuits. A decade ago it was a common mistake to use the wrong length screws on the motors thus penetrating the windings and causing shorts, mainly on race quads with carbon base plates.
Yup, I’m using the dshot version of ardupilot. However, they’re set to DShot300 - I’m presuming that’s a safer number until the first flight and I can take it from there. Do you reckon DShot600 is worth it before I run the first flights?
Good point. I was careful about it. I hadn’t touched the motors when the first set of ESCs got damaged - I think it was more of me forgetting about not running motors unloaded at high RPMs. Also, I think I ran the calibration sequence wrong; the motors (all of them) were actually turning during the ESC calibration for some reason. Those motors and the ESCs are all in the scrap bin; I’ll see what I can do with those motors, if anything at all. These motors are all new as are all the ESCs.
I measured the lengths and I don’t think there’s any damage to the windings due to the screws cutting into them (or the enamel). They’re the one sent with the kit. Here are the measurements:
Mounting plate height (a): 1.9mm
Motor mount hole height (b): 3.2mm
Thread length (c): 6mm
Distance between end of thread on motor mount hole and the winding (d): ~1.9mm
a + b = 5.1mm
Remaining thread length (protrusion): 6 - 5.1 = 0.9mm < d
Thanks @Allister! Appreciated. I gotta spend some time and learn about the DShot and BDShot. I know they’re replacing PWM but know very little beyond that.
Alrighty! I just set up everything and it seems like things are working; all motors are running nicely at 5% and they can all run at the same time when I run ‘Test all motors’. Not going to tempt fate and I’ll leave it as is until I flight test it.
To wrap it up for now, I’d like to circle back to the main issue that led to this whole thing: the lack of sync of motors when controlling the motors with RC and/or that RC throttle needs to be quite high before the motors start spinning. Anything you’d suggest?
Thank you for all your inputs! I’ll share updates once I flight test it which I think would be sometime this April - gotta make sure I won’t get arrested when I fly as a hobbyist xD
