Humbled...

Hi folks,

I apologize this is kind of a long read, but I wanted to finally share my story:

I'm a filmmaker and video professional that's been working in the industry for almost 15 years. I've been doing the UAV thing for just about 2 years now... It started out as a curiosity and hobby that I've been trying to integrate into my professional business. I built and upgraded my own F550 with naza v2 that I've logged tons of hours on, and so far has performed flawlessly, but in my line of work, go-pros usually don't cut it.

A couple partners and I were trying to plan out options for a heavy lift build, when I found what looked to be a fantastic deal on a used premade system... Specifically the insane looking Aerigon by Intuitive Aerial. It had taken a fall with a previous owner, but was repaired by a reputable source. I opened it up myself and poured over it, replaced a questionable esc and messed up gimbal motor, and made sure that everything looked to be intact and in working order. We proceeded to do many test flights with dummy weight, before moving up to flying a 5D to get the gimbal tuned. We had developed a long pre-flight safety checklist, always treating the machine as seriously as it deserved - especially since our end goal was to bring it on professional sets.
Anyway I'll get to the point. We had finally gotten up to the point of feeling comfortable enough with the system to put a "real" camera on it, and start fleshing out our reel, so we picked up a Red Scarlet and Red lens from local shop we are in good with, and set out to get some "cool reel shots". You can guess what happened next.

I really wish we had a third person filming the flight, so you could all feel the PTSD that I do. It was probably one of the most spectacular crashes you guys would have ever seen. It was the 3rd or 4th flight of the day, and about 5 seconds after takeoff it got a mind of its own and started bucking like a mad bronco. I dropped out of GPS into ATTI immediately, but it made no difference. Within about 15 seconds it was flying sideways at almost 90 degrees straight into a concrete wall at what must have been at least 30 mph.

Now first off I want to say I am NOT going to use this thread to present any sort of "review" of the Aerigon system, especially since it was used and unsupported. All I'll say is it's an X12 running T-motor 4014s and an A2 FC, with a not very user friendly gimbal all wrapped in a ton of awesome looking CF. If anyone has more specific questions about this relatively rare system, send me a private message. For this thread I'd prefer to focus more on the A2.

This all happened a few months ago, but I wanted to wait for all the insurance BS to be sorted before I talked about it. (Insurance is a whole other ****-storm story as well... but anyway). The main reason I'm laying this all out, besides needing a shoulder to cry on, is to get people's impressions about what potentially went wrong, and where one goes from here.

The best conclusion I've been able to come up with is compass interference. We were flying in a park, but the wall that it hit was the crumbling remnants of an old steel foundry. In retrospect I should have thought about the potential for there to be a ton of magnetic iron buried underground in the area. Before the flight we got an error from the A2 indicating a bad calibration, so we redid compass calibration and it appeared fine... until I took off. But is it possible for just bad compass data to make a copter fly around like a madman, turning sideways, probably trying to turn upside down - all in ATTI mode?

The broader question is about safety and reliability of these systems. People often say that 99% of crashes are due to user error, and it's possible I made an error in not checking the area for magnetic materials... is this something you guys routinely do? Or is the unpredictable A2 to blame? I hear a lot of horror stories about "flyaways" and such, but I've always chalked that up to the fact that there are simply more DJI systems out there than other brands, so you'll hear more about the fluke cases.

So sorry for the life story. Basically, I know i've less than 2 years experience, but I felt like I was attentive enough, thorough enough, and safety focused enough to be "ready" to move up to the next level. Was I wrong?

Once I recover financially and emotionally, I want to get back on this horse. I feel like I'll trust a system I build myself this time around. But are events like this just inevitabilities in our industry? Are there better, more redundant and more reliable FC solutions? Did I just screw up? The idea of not FULLY knowing if something like this is going to happen again seems like too great of a risk when you are planning on having almost 6 figures in the air.

So anyway, I'll include some happy before and sad after pics. Also my iOSD log of the flight, which I find hard to interpret. Please comment on all of this or none of it... thanks for listening.
 

Attachments

  • before1.jpg
    before1.jpg
    106.1 KB · Views: 453
  • before2.jpg
    before2.jpg
    84 KB · Views: 423
  • after.jpg
    after.jpg
    104.9 KB · Views: 431


Motopreserve

Drone Enthusiast
So sorry you had this experience. It's always a hit to the wallet, ego and confidence.

I think there are many things you can do to avoid crashes like this, but ultimately, sometimes they "just happen." I have nothing that nears the investment you have, but I learned fairly early that I needed to be willing to lose the investment I had in the craft, because I can only cover so many things before I need to trust in the technology.

I think this is why you see the "pros" having backup rigs. Might not have a disaster as you have, but something can always happen, regardless of how thorough you've been.

Glad to hear you're working toward getting back on the horse.
 

Pumpkinguy

Member
I had a similar experience with a new heavy rig I built running wookong and lost control of everything except throttle. I had to force it into a roof to save potential damage to property or personal injury. This was on its 3rd flight and the payload was 10 pounds of steel rather than 10 pounds of camera and lens.
It sucks big time.
Did you buy that rig from Canada? I am from Toronto and saw the exact rig in the local classifides. It had a repaired leg and the guy wanted 10% of retail value.
 

dazzab

Member
What a horrible experience to go through. You know, people always say it's pilot error - until it happens to them. It certainly sounds to me that you did everything right. Did you send the logs to DJI? It would be interesting to see what they have to say. Honestly, I'm amazed someone hasn't been injured by one of these events.
 

Pumpkinguy

Member
I don't see anything out of the ordinary in the data log. The whole thing happened real fast eh?

When I had my crash dji had no interest in looking at the logs. They started off saying no one in customer service had the program to read them. I gave up after 5 or 6 back and forths with them.
 

dazzab

Member
Would a kill switch independent of the FC have been useful in this situation? I'm fitting one to my SkyJib right now as I simply don't trust any of these flight controllers and want to be able to kill the motors if anything goes wrong. It's not ideal but other than a self destruct or blowing it out of the sky I don't know what else can be done.
 

I did have what appeared to be control of the throttle, but nothing else. Sometimes my stick inputs worked for a second, then did the opposite of what I told them to (and no I didn't lose orientation until the very end)... In retrospect probably could have killed the throttle and dropped it, but instead I climbed as an attempt to regain control. It was just over too fast. It hit the wall about 30 feet up and fell.

I understand there's always a risk of loss of investment. That's why backups and insurance exist, and yeah, it's always gonna be a risky endeavor. I'm actually a little more shaken up by the fact that we were getting shots of a dancer at the time, and if this went differently, she could have been hit with almost 40 pounds. Also, the batteries went up immediately after impact and we had to run to stomp them out. Financial risk is something I can stomach. But is the technology REALLY mature enough to feel secure about not causing property damage, injuries or worse? As terrible as this experience has been, I'm eternally grateful no one was hurt.
 

Pumpkinguy

Member
I am by no means an expert at UAV's but I am an expert at crashing. Lol
When I first read this thread and looked at your log I thought "mechanical failure". I was thinking it was very similar to my crash which was deemed by me to be a motor problem.

You have 12 x 4014's. They are good for about 1000 grams of lift each with 16" props. Let's take away 10% efficiency loss due to coax configuration. That only leaves you with a safe load of around 24 pounds. You were close to 40. Ouch.
 

Motopreserve

Drone Enthusiast
If the craft seemed to operate opposite of the stick control, and sometime no stick control at all (except throttle), could this mean Rx connection/data failure?

The A2 has the internal Rx? Is there any RSSI logging?
 

It was running with a futaba Rx rather than the internal. I don't think there's any logging for that?

Pumpkin, it was running 18" props. I'll admit I was skeptical about the company's rating of max 50lbs considering the specs, but I figured they knew more than I did? Like I said, we tested it with almost this much dummy weight multiple times before with no problems. They claim the gimbal can take up to 20lbs of camera. The thing retails over $60k new! Admittedly the only reason I even considered buying it was because I got it for MUCH less.

That being said, it didn't SEEM like a motor failure... but I could hear them revving up and down like crazy while I was barely moving the sticks.
 

stevemaller

Heavy Lifter
You have 12 x 4014's. They are good for about 1000 grams of lift each with 16" props. Let's take away 10% efficiency loss due to coax configuration. That only leaves you with a safe load of around 24 pounds. You were close to 40. Ouch.
My reading of the T-motor 4014 props with 16" props shows 1920g of thrust at 65% throttle. What are you reading? See here: http://www.rctigermotor.com/html/2013/Navigator_0910/40.html
Interestingly enough, the 4014 motors don't appear to be spec'd to run 18" props.
There are numerous stories of A2 systems flaking out, although there are also hundreds of them running heavy lifters on a regular basis.
Without a detailed look at the logs, you shouldn't jump to conclusions. There are so many possible causes.
 

The only other thing I was thinking was possible besides compass, was I was technically operating between 2 parallel concrete walls, about 35 feet high, 15 feet thick, and maybe 100 feet apart. Due to the wall thickness, is it possible it created a sort of "echo chamber" for my Tx signal, confusing the copter?
 

As far as the future goes, are there any FCs that are considered to be more reliable than the A2 because of their FEATURES or SPECS? Not just people's anecdotal evidence? On the surface, A2, WKM, MK, SuperX, Hoverfly, etc all kinda look the same.
 

Benjamin Kenobi

Easy? You call that easy?
Hey mate. So sorry to hear about your crash. I feel your pain.

I honestly don't think the walls or structures were the problem. I've flown in places like that before such as in the middle of the city and up the side of a skyscraper without incident.

It was a second-hand aircraft so there could of been a gremlin in the works somewhere.

Alas all you can do is build another one and move on. :)
 


I'm still struggling with the underlying philosophical question of the thread... Let me explain with an analogy:
If you had been learning to fly an actual helicopter for example, and had a huge crash where you barely escaped with your life, where the cause was vague - and there was a chance that it may have happened because "helicopters just do that sometimes" - would you ever willingly get in a chopper again?
Yes there probably was a tangible cause of my crash - maybe it was avoidable if I had caught something beforehand. While the general idea that "eventually you're gonna have a crash" might be ok in the small hobby world, but in the large scale production world I'm having a hard time reconciling that. For the most part I won't be shooting pretty landscapes in the middle of nowhere. I'll be shooting actors and stunts, often in urban areas, often lifting almost 40 pounds, often lifting other peoples equipment. The risk to people's property and health is too great to have a "**** happens" attitude.
Planes and helicopters still crash sometimes. Sometimes the causes are unknown. But it's very rare. If an A2 is hovering with 40 pounds directly above Tom Cruise, what are the odds it decides today's the day it goes rogue and drops on him? 1%? .1%? .01%? Are those numbers significantly different with Mikrokopter? SuperX? Hoverfly? My career and more depends on those #s. These are the things I'm trying to calculate before I drop another 5 figures +.
 


fltundra

Member
fltundra, those numbers make it seem like the 4014s are ok? 56% hover... They only get slightly in the red at max output...

No experience with your motors.I originally purchased MN3110-26 470kv's and within 10 hrs. had bearing play. Changed over to KDE 515kv's over 90 hrs ago, and the difference in quality was drastic. And we are only talking a 2400g quad! That's just my experience with tmotor navigators. If I were you with just having crashed, I would call Andrew at Hexacrafter and build an x8 or whatever you need to lift with the SuperX FC, KDE motors and some large Castle esc's, which have logging built in. And go from there. I am a firm believer in plenty of headroom (amp wise)when it comes to esc's. Take a look at Aerial Mob video's... Steve's moved over to SuperX after flying the A2 and losing confidence as I have with any DJI FC.

The X in action:
 
Last edited by a moderator:

stevemaller

Heavy Lifter
Quite frankly this thread is a good illustration of why I've decided to invest in the ALTA from Freefly Systems. They're the only one who's building their own frame, motors, props, ESCs and flight control. About the only thing they're not building are the LiPos.

The buzz from those who have tested it is that their system is going to create a new standard in flight stability and reliability.

I have faith in Freefly, and the copter's fairly priced. It isn't a panacea, but it'll be perfect for me. If you're trying to lift 10-15lb camera packages, you're S.O.L. but for those of us who are flying DSLRs and REDs and the like on MōVIs, it's gonna crush the competition. And any DIY, too.
 

Top