A puzzle for the electronics types

[Monkeyleg]

I have a DJI Phantom 2 Vision + drone quad copter. I had an earlier version and sold it to buy this one. I bought this used, and the owner gave me a demo of it.  Everything worked fine. However, I did not feel the motors to check their temperature.

I found out that this version of the Phantom had an issue with the ESC ('electronic speed control"), a circuit board, one for each motor, that controls the motor's speed. The ESC's could fail, and the copters fall out of the sky.  I ordered the latest ESC's and installed them. Some soldering is required. I'm not the best at that, but not the worst.

Yesterday I took it out for the first time. The Phantoms are equipped with internal compasses, GPS and loads of other goodies. When you take off, the system uses the takeoff point as the home point and measures everything relative to that with GPS. If you hover, it should stay within a couple of feet, even if there's some wind.

Mine was drifting ten or more feet, and was also ascending and descending, none of which it should do. I thought it might be because the wind was 15 mph+. When I landed it, I felt the motors, and they were as hot as a cup of coffee.  At least 160 degrees. They should not be that hot.

I opened the unit up to look for any obvious wiring problems, but couldn't see any. I turned the motors by hand, but couldn't feel anything unusual. One or two motors didn't spin as freely, but it was a very subtle difference. I checked the rpm of each prop using a strobe. At idle, they're all pretty close. At idle, the motors don't get hot. It's only when they're under load.

The Phantom system is designed so that, if one motor is weak, the others work harder to keep the craft in its GPS lock.

Something is making these motors really hot, and it's making all of them really hot. If there were a bad motor, it probably wouldn't be hot.

I should mention that each ESC is connected to the power supply using a ~ 1.5" run of coax cable. This is to keep noise from the circuit from disturbing the GPS. I wasn't able to get the exact coax used by the factory, but it was close. 18 AWG tinned copper center conductor, tinned copper braided shield.

Any idea what generally or even specifically might make the motors too hot to touch?

[AmbulanceDriver]

Monkeyelg, my neighbor builds and sells an assortment of multicopters - I'll try to ask him next time I see him.

[DustinD]

For the drifting, did you fully calibrate it after putting it back together?

I would try the DJI forums for the motor question.

[Ben]

Not an electronic type, but I have stayed at a Holiday Inn flown a drone before. When I flew the little RQ20s, wind was definitely an issue. They were fixed wing, but it will apply (somewhat differently) to rotor aircraft as well. Was it a steady 15mph (and was it mph or kts?) or was it gusting? If you're getting gusting winds and variable directions, you're going to get some movement of the UAV.

We often bagged precision flying if the wind was doing variable 25KT+ with gusts. We would also see LOTS of 3D movement flying in places like shorelines with adjacent vertical surfaces. When the UAV was set to laptop control, you could watch the battery drain in those instances as systems kept trying to adjust to shear, etc., and doing that will also definitely heat up components.

Before you do anything else, try it out on a windless day or else inside somewhere if you have a larger enough area to check stability in controlled conditions.

[DustinD]

The Phantoms can buck wind pretty well, it takes sharp gusts to push them temporarily off their spot. The electricity usage is about the same in gusts as in calm air because keeping the drone in the air is what takes the energy.

[mangle]

160F isn't cause for concern with motors or electronics unless it's drastically different than what you were seeing before.  Given the situation, I'm inclined to say that there's no issue - You're just observing it for the first time.

I would imagine that the motors in the quadrotor are simple brushed DC motors, so there's really nothing the ESC could do that would cause them to get hotter than normal.  (That said, there is commutation timing in both a brushed and brushless system, so if the motors are turning opposite their intended direction, they may be less efficient than normal.  I highly doubt that this is a problem in your case.)