So I've just walked back from testing my new custom quad. This was the second quad that I'd built; I'm relatively new to multirotors, but have spent a number of years with fixed wing systems.
Anyway, the specs of this new quad:
Carbon fiber folding frame (this one; other frames share the same name).
2x 5000mAh Multistar 4S 10C LiPos
4x Multistar 4225 motors (610kV, max 330W)
4x Carbon 13" x 5.5" props
4x HK Red Brick 30A ESCs
Pixhawk with typical accessories
Here's a picture of it being built:
Note that the ESCs for all 4 motors are coexisting in the frame, under where the Pixhawk is sitting.
As I do with any multirotor, I tied it down to a heavy grate and then performed a low power motor & control test, keeping power at <20% power and just gently moving the controls around (this is all with the PID settings modified to have zero integral gain).
Unfortunately after about 15 seconds of running it at ~20% throttle and some gentle pitch / roll, it decided it'd had enough and with an eruption of smoke, a juttering of props and a very brief fire, the test came to a halt. I immediately cut power to the system and then physically removed the batteries (which requires the cutting of 4 zipties) from the quad to prevent them from getting any hotter.
Here's a picture of it after it's maiden ground test run:
As you might be able to see, the red ESCs in the center there have burnt up.
Here's a closer look:
Now, I know that Hobby King equipment gets a bad rap by some people, and at first I had thought that perhaps the ESCs had underperformed, as even at max throttle they were only expected to run at about half their max sustained current rating. It was only when I got back to a computer and went to double check their specs / what eCalc theorised the current draw was when I realised that I'd screwed up. If you look up the store page for those red brick ESCs, you'll notice something that I didn't at the time; they're only rated for 2-3S batteries; and I was using 4S...
A small mistake, but with relatively big results.
Well, I haven't had the time to check my other components but from what I've seen so far:
- Motors and props are fine. Some cables are singed but still appear sealed and intact.
- Batteries need to be tested; one looks fine, the other has some slight discolouration which may be from the smoke, or it may be from heat damage. I'll probably run them both through a charge and discharge (hooked up to an appropriate ESC and motor) cycle to see if they heat up or swell at all.
- Pixhawk status is unknown, but it was flashing lights normally as far as I can tell during the accident, so I think it's alright. Its 3DR power supply is a bit burnt, but I don't know if that was due to its proximity to the ESCs or not; either way I'll replace it.
- Frame looks alright, but either I have some molten plastic on it or the resin partly melted from the heat. I think I should be fine to use it again.
With the ESCs, I'll obviously have to replace all 4, this time, with ones that are 4S compatible. As those ESCs only just fit in there, and because their proximity certainly didn't help keep them cool, I'll be mounting the next set of ESCs probably under the frame.
So there you have it; always double and triple check your components, because the smallest difference of a number can mean the biggest variation from safe flight.
Comments
Can't really see what your battery setup had, and the ESC-to-motor lead length isn't as harmful, but as I have preached a million times: Battery/ESC feedback pulsing is a real serious killer!
The rest of the copter should be fine. You killed the FETs on the ESCs from too high of voltage.
The worst is when you Measure twice, three times and it was wrong each time. Damn.
I have been using cheap HK and Turnigy ESCs for years without problem but usually leave a wide margin between my motor's expected and the ESCs rated current draw.
Couple of weeks ago, I was flying my quad (Turnigy Plush 4 x 25A ESCs, total draw for all motors rarely if ever exceeds 20A, huge margin) when I noticed thick smoke pouring out when I landed. I examined the ESCs but couldn't see any obvious signs of damage. Took off and hovered low and could see continuous sparks in one of the ESCs. Changed batteries and tried again, same thing. Removed the ESC but still could not see any obvious signs of damage. Tried reproducing the same behaviour indoors but could not get the same fireworks going. Still, currently replacing ESCs to be on the safe side.
@Joshua,
Unfortunately this is another example of the old adage: "Measure twice, cut once."
Regards,
TCIII AVD