XAircraft First crash. Boo. Flipped on takeoff.

[mediaguru]

Sometimes on my lunch break I'll go out and fly my 960 hexa. Today I did the same but had NO fun. When I brought the throttle up to 50% the bird flipped over. Damaged some props, broke GPS tower. Need to do a closer inspection. Can't look at the log until tonight. Normally I'd think it is a prop being on the wrong way, but the props haven't been off for a good 50 flights.

I usually like to take off from grass, but naturally today I chose pavement, so all the props have some scuffs. Not sure if they're usable after some sanding but a few are trash now.

I do remember some takeoffs where it seemed that the right side was less responsive than the left at first and the bird would take off at a slight angle. Wonder if that's related

 

[mediaguru]

Ok check out the log and tell me if this is pilot error:

http://log.xaircraft.com/report.htm#2014_10_30_204038.250_32.1.00.02
http://log.xaircraft.com/#2014_10_30_204038.250_32.1.00.02

Is it possible I just got careless and went too fast with the throttle on takeoff?

 

[kloner]

the problem is solely relying on gps mode and most likely something to do with either solar flares or just a bad lock before taking off and it went to where it thought it needed to be. You should make a habit of leaving the ground in atti then flip to gps mode to prevent these kinds of things. I have landed an actor drone of ours with this thing and went to take off again and it pulled a similar stunt.

 

[mediaguru]

the problem is solely relying on gps mode and most likely something to do with either solar flares or just a bad lock before taking off and it went to where it thought it needed to be. You should make a habit of leaving the ground in atti then flip to gps mode to prevent these kinds of things. I have landed an actor drone of ours with this thing and went to take off again and it pulled a similar stunt.

Interesting. I had 9 satellites from the log. Let me ask you a question. I haven't discounted your theory yet, but I did mention a lean to the right on some previous takeoffs. I recently moved my gains from 1.2 to 1.1 and had a couple of successful flights.

I think it may be a throttle calibration issue. Last night I took off the props and started up the motors, running the throttle to full and then back down. In the log below you can see that there is a difference between the motor levels, especially the motors on the RIGHT of the bird 1, 5 and 6.

http://log.xaircraft.com/report.htm#2014_10_30_212022.500_34.1.00.10

Is it possible that could cause the flip, a quick throttle up with the 3 motors on the right of the hex having a lower throttle cal point? At one point there's a 14 point difference between motor 1 and motor 4.

 

[kloner]

that log kinda backs up my theory. you were in gps mode and no props yet it was drifting away from where it thought it was. The superX gps sucks on the ground,,,,, my only incident with it was similar to what you describe happening to you.

Did you do a full three point sat calibration?

Try taking off in atti and flip to gps as soon as it leaves the ground...... it should fix EVERYTHING. when you land get it close then flip to atti to land..... the heading magnetism showed an aweful lot of movement for it sitting still.... gps lock was not even close to solid in the last graph,,, it showed good on the first graphs that it crashed on.

is the gps puck kinda "springy" when you try moving it?

 

[mediaguru]

that log kinda backs up my theory. you were in gps mode and no props yet it was drifting away from where it thought it was. The superX gps sucks on the ground,,,,, my only incident with it was similar to what you describe happening to you.

Did you do a full three point sat calibration?

Try taking off in atti and flip to gps as soon as it leaves the ground...... it should fix EVERYTHING. when you land get it close then flip to atti to land..... the heading magnetism showed an aweful lot of movement for it sitting still.... gps lock was not even close to solid in the last graph,,, it showed good on the first graphs that it crashed on.

is the gps puck kinda "springy" when you try moving it?

The unit has been 3-point sat calibrated.

I should have mentioned that the props-off test was in my basement. I couldn't get a lock down there.

The GPS is on a GPS tower, perhaps not the most stable platform. I used to have it flush on a battery mount but had two incidents of GPS interference and that is why I went with the GPS tower. Here's the tower attached as an image. The tower is now broken, so I may have to go back to what I had before.

 

[kloner]

i have that tower on mine too....

 

[mediaguru]

i have that tower on mine too....

Well mine is now about an inch shorter after it snapped off. Still might not work.

So the difference in motor percentages doesn't stand out with you? I'm going to recal all my motors this weekend regardless.

 

[kloner]

i see that as it being in gps mode thinking it was someplace it wasn't, kinda like what i think happened on that og graph posted.

out of curiosity, before it flipped over, how long was it powered up on the ground prior to arming and trying to fly?

 

[mediaguru]

i see that as it being in gps mode thinking it was someplace it wasn't, kinda like what i think happened on that og graph posted.

out of curiosity, before it flipped over, how long was it powered up on the ground prior to arming and trying to fly?

Perhaps 30 seconds to a minute. I always take off in GPS mode (may not now), and wait for gps lock before flying (double green flashes). I was between a couple of buildings in an open area. I suppose there could have been interference from the buildings also.

 

[mediaguru]

As a test, I did a few seconds starting up in manual. There's quite a difference in throttle between motors:

http://log.xaircraft.com/report.htm#2014_10_31_131443.250_35.1.00.30

 

[mediaguru]

Just did a throttle cal. Guess what? Looks the same.

I've checked out the bird and will be trying a test flight shortly.

 

[mediaguru]

I've done a couple of flight now after the booboo. I put my gains back to 1.2 (except yaw). I've noticed that once every 4-5 launches, the lean right happens. I'll work on launching in ATTI mode.

 

[fltundra]

I've done a couple of flight now after the booboo. I put my gains back to 1.2 (except yaw). I've noticed that once every 4-5 launches, the lean right happens. I'll work on launching in ATTI mode.

Try lowering the height gain if your running it at 1.2

 

[mediaguru]

Try lowering the height gain if your running it at 1.2

The crash occured a couple of flights into running it at 1.1. I had 100+ flights at 1.2, but did notice the occasional right lean on takeoff then as well.

 

[Pumpkinguy]

Since I'm doing a build with superx I've been doing a lot of reading. Do you mind doing a follow up post? Did you come to any conclusion as to the cause and have you rectified the problem? Also, what is the purpose of taking off in atti? Steve says it sucks on the ground? Do you flip back to gps once you take off or after you gain quite a bit of elevation? Don't really understand why taking off in GPC could cause a flip. Thx

 

[PMaughan]

Don't really understand why taking off in GPC could cause a flip. Thx

I don't use superX but it's the same for DJI controllers. GPS is not pinpoint accurate and pinpoint accuracy is what you need for take off. While your machine has been sitting there waiting for take off, powered up in GPS mode, it's taken a fix and decided where it is in the world, a few seconds later and it's fix point has wandered half a metre so when you go to take off it wants to get to where it thinks it should be which can result in a flip if the gear catches. Similarly for landing, altitude accuracy is equally effected so best to land in Atti too so you don't have a sudden drop or lift when you're not expecting it.

Pete

 

[X650Mouse]

I don't use superX but it's the same for DJI controllers. GPS is not pinpoint accurate and pinpoint accuracy is what you need for take off. While your machine has been sitting there waiting for take off, powered up in GPS mode, it's taken a fix and decided where it is in the world, a few seconds later and it's fix point has wandered half a metre so when you go to take off it wants to get to where it thinks it should be which can result in a flip if the gear catches. Similarly for landing, altitude accuracy is equally effected so best to land in Atti too so you don't have a sudden drop or lift when you're not expecting it.

Pete

LOL sorry that is not the case with SuperX, best you get SuperX then comment. most of the time I take off in GPS mode as the vehicle stays in one position with Att mode even a slight wind can set the vehicle in motion, in GPS mode you can use NO sticks after the first 50% with thrust it simply stays in position with Att mode you must correct position with the stick however it will maintain height.

The issue of tipping has been mention at times, in all cases it has been various ESC's and suspect setups. few if any use the auto take off mode of SuperX

X

 

[PMaughan]

LOL sorry that is not the case with SuperX, best you get SuperX then comment. most of the time I take off in GPS mode as the vehicle stays in one position with Att mode even a slight wind can set the vehicle in motion, in GPS mode you can use NO sticks after the first 50% with thrust it simply stays in position with Att mode you must correct position with the stick however it will maintain height.

The issue of tipping has been mention at times, in all cases it has been various ESC's and suspect setups. few if any use the auto take off mode of SuperX

X

Well there you go, learn something new everyday, but all sounds familiar.

I don't know much about Super X at all but know a fair bit about GPS and it's foibles and differences in accuracy depending on your location. Does the controller in that case suspend it's GPS position acquisition once it has taken a fix until in the air? If it does then that's clever, if it doesn't then are you 100% confident that the GPS control won't decide that to be in position it has to drop half a metre, or move a metre or so just when you're not expecting it? Do you know the stated GPS accuracy of Super X?

Genuinely interested in Super X so good to know.

Pete

 

[X650Mouse]

Well there you go, learn something new everyday, but all sounds familiar.

I don't know much about Super X at all but know a fair bit about GPS and it's foibles and differences in accuracy depending on your location. Does the controller in that case suspend it's GPS position acquisition once it has taken a fix until in the air? If it does then that's clever, if it doesn't then are you 100% confident that the GPS control won't decide that to be in position it has to drop half a metre, or move a metre or so just when you're not expecting it? Do you know the stated GPS accuracy of Super X?

Genuinely interested in Super X so good to know.

Pete

Suggest you watch some YouTube videos for SuperX, many have been made by users showing the accuracy of position and maintaining station in all types of wind, needless to say many users are from the DJI camp and won't go back.

SX has a logging system called black box that records all data from the FC this can then be played back for problem solving of reliving the flight here can also be seen, to 6 decimal places the GPS coordinates of the entire flight.

Please do your research! it is not my thing to try to convince you, other than to say your comments were not related to the OP's question.

X