Recently I had a shoot where I needed to fly indoors. Never doing so I decided to fly my backup flat 8 with the U5 motors on it. This way if anything happened I had my main rig with the Tiger 4120's. I was 50-50 on whether or not to fly it with the new heavier gimbal which is about 2-3 pounds heavier than the gimbal I was flying it with. Anyways, I did a small up and down test with it before I tried to perform the shot inside. It was kind of struggling to get the copter off the ground but it did. So I go to fly the copter to perform the shot and as soon as I raise the sticks to get the copter of the ground "BOOM"........ INSTANT FIRE........ THis was not just a small fire either it was pretty intense as you can see from the pictures. The U5's were just drawing too much current because they were really undersized for the payload. I was attempting to lift 28 pounds with them. I really should have known better because early this year when I tested different payloads with just weighted plates the hover throttle jumped up considerable when switching from 10 to 12.5 pounds. I checked the motor that the BL was attached to and its fine just to eliminate the possibility of a motor going south. This was my main copter for most of this year with over 400 flights on it most of it carrying the red at 25 pounds. It became my backup when I built another flat 8 with the Tiger 4120 motors which flys 28 pounds perfect.
Dave: Many thanks for publishing this information. It's sad to see it happen, but, on the other hand, it's very generous to publish this kind of information so other people can learn what the limits are... When you have a failure like this, is it possible to get the board repaired or is the whole board toast (and overdone toast at that!)? My concern with the v3 boards, from the outset, has been that a single failure takes out the whole board -- which is, as you know better than we do, a pretty expensive proposition. Andy.
Had a similar issue with the 4012s no payload. Did a flight and noticed some in-stability in motor 8. Bench test and the ESC lit up like a sparkler. Was a Non-cool version.
This is very concerning from the point of the boards just catching fire! Isn't there supposed to be current limiter on them to prevent this kind of thing?
I have gone through two MK octo XL boards. First lit off, 2nd was a massive I2C 75 ft off the ground. The cool has been very solid with the tiger 4012's as has the BL3.0 hex with the same motors.
Sorry to hear Dave. I echo Andy's sentiment regarding all-in-1 controllers. Had that been individual controllers out on the booms a repair would take less than an hour and the fire would not affect other components. Obviously the expense would be much less as well.
This was for a client that I have been working a lot with and collobrating with him to promote UAV's so it was't too bad. At first it was pretty shocking to all of us but I remained calm. This was on the start of a full day of shooting and I pulled out my main rig with the 4120's and it went flawlessly for the rest of the day. The client was happy with the footage so all is good.
Basically any failure on these boards and its trashed. To be honest I really don't know if the board (specifcally the I2C buss) will remain in tact and communicate with a BL failure. I know that's what the redudancy is supposed to do. But you know what they say about assumptions.
The MOSFET drivers are current limiters but I think what in this design in the case of a big short, the excessive current has the potential to arch to other components. I still believe in these boards and the MK electronics. 1500 flights in 2.5 years and its been extremely reliable. In my opinion I brought this on myself and should have known better. I just though the worst case scenario would be higher hover throttle but I think I have come to the conclusion that at 25 pounds the U5's are really at the end of their limit of what they can take and I pushed it well past their limit. Looking at these motors now and comparing them to the 4120's and 4014's they really are a small motor.
I know. It's really a cloudy solution. WE really have 3 choices. 1. Make the MK electronics work within their vast limitations. 2. Use the MK electronics with the I2C converter and have efficiency losses and possible flight quirks. 3. Go with the DJI stuff and then "hope" you don't have any fly aways or any other serious flight issues. There seems to be 2 groups of people that use DJI stuff. One group consists of very experienced flyers that swear by it with no issues. The other group is the complete opposite with a ton of scare stories. So what does one do? I'm trying option #1 which is why I chose the 4120's which I feel is a very good choice for payloads of 25-29 pounds. I was able to do a max throttle test straight up and reached a max of 200 amps at 28 pounds so I feel this is well in the safe range. Howard I know you are a fan of #2, but on the same side of the coin there others that have commented to me a boat load of problems. We are all pioneers, and hoping that we chose the right path. All we can do is learn from each other as to what doesn't work for sure.
Unfortunately I had the navigation and GPS boards disconnected for the flight indoors. So no data recorded.
BTW: I didn't mention what I plan on doing now. I have a set of KDE 4014's that I want to try on the flat 8 and compare the efficiency to the X8 I had on them. I think the 4014's with 16 inch props will be enough, Holger thinks they will be as I sent him a motor to test and he seems to like it a lot. The reason why I don't want to use Tiger 4120's for this flat 8 is because I do a lot of jobs with the DSLR and flight time does go down a bit with the size of the 4120's thus I'm trying to balance the needs with having a backup able to carry the red with the heavier gimbal. Maybe not as efficient as the 4120's but hopefully they can avoid issues like this.
They do not. My first board that went would not communicate and even managed to put some feedback across the I2c into the FC.
The original 2XL Quadro boards did not, then some did, now all of the current ones going forward supposely do. I think they are on version 4 of these boards.
I do know I had one of the very first XL quad boards. I WISH I had the camera running for the bench test, but my friend who was watching jumped about two feet when it lit off.