New guy, trying hard to learn, had a fly away

ArduCopter Copter 3.1
1

Hello everyone!

I am trying my best to learn and understand and absorb the encyclopedia of information about Arducopter, APM, Mission Planner, etc. But it’s kinda hard for this old man, lol…

I have a Quanum Nova RTF aka Cheerson CX-20 with the APM open software. I have had 10 very successful flights and then the 11th flight ended up with an out of control quad and a crash.

The difference was that I had tried my luck using APM and calibrating the compass, the accelerometer, and the Tx. I followed the tutorials and I thought I had performed the cal’s correctly. The reason I even bothered to do the cal’s was because in “Loiter Mode” the copter would not actually loiter. It would wobble and drift around a lot. Versus “Stable Mode”, where it could be held quite steady if there was no wind.

Prior, I had used successfully the Loiter mode, and RTL mode. Both worked just fine, except for the wobble/drift in Loiter. The RTL seemed to work quite well.

Yesterday after performing the calibrations I flew the copter in my backyard, my normal flying site so far. I took off in Stable mode and it flew nicely as before. I then switched on Loiter mode and it very swiftly started to fly backwards and disappeared from my view when it went over the roof of my house. I switched into RTL and I do not know of it actually attempted to RTL but it did not return. I heard loud buzzing noise and spotted the copter in one of my trees merrily chopping up the leaves. I was in a bit of a panic but then remembered to switch back to Zero and the buzzing stopped as I ran to the front of my house. By then the copter was on the roof and I was able to recover it.

Amazingly it only suffered a broken landing skid. Getting back up on the horse, so to speak, I then flew it again in Stable mode which was successful. But I am now afraid to try Loiter Mode until I sort the issue out.

I read the tutorial on downloading logs in Mission Planner and I was able to save three logs, which was everything in the onboard memory. They are huge, 2.7MB each. Could that be right?

Anyway, I tried looking through them and tried to figure out how to use the graphs, analyzers, etc but it’s currently WAY over my pay grade!

I suppose a simple question would be, is it possible I goofed up the compass calibration procedure? Should I perform it again? Does it matter that I was inside my home when I did it? Does it actually require being outside Like I saw in one video with the gent spinning around with his quad?

I mostly fly it in Stable Mode, but I would like to be able to use a nice and steady Loiter Mode from time to time.

Well, thank you if you have read all the way through this enormous post. :slight_smile:

2

I decided to perform the calibrations again, this time using Mission Planner. The compass values were similar.

I also verified my compass is not 180° off. I used an app in my phone to scan GPS satellites and give me an accurate pointer towards North, and then I set the copter down pointing in the same direction. When the flight data came up on MP it was showing the compass (yaw) reading correctly.

I also discovered a new button in MP. :smiley:

It is the “log analyzer”.

3

Can anyone tell me why the log analyser says the duration of the flight was almost FOUR days? :question:

4

Is there anyone here who helps newbies?

5

Did you also check that when you pointed it west it was showing west. Just don’t check north. Check north south east and west. If you rotate 180 degrees after doing your compass calibration you have to redo it.

Wow your hdop value is way to high. It should be less then 2. How long did you wait for a gps fix.

6

Thank you for the reply! :slight_smile:

No! I did not check E-W-S. I only checked N. :astonished:

I shall check the other compass settings now.

I waited about 1-2 minutes for the GPS fix.

7

The log you posted isn’t very helpful. Based upon the altitude it was recording, this log was only recorded the period after you crashed. To make matters worse, all your tuning and setup parameters are missing. It looks like they were truncated off the front end. There’s no indication of a RTL command being issued in this log.

You might try posting the log just prior to this. I’m not sure why there should be three logs as most arm-disarm sequences generate a single log file.

8

OtherHand, thank you for the reply! :slight_smile:

I will post the other two logs, just in case.

madhampster, I just checked all major compass points in Mission Planner and all were indicating properly. N-S-E-W

As a point of interest, in Mission Planner I have found two locations that show Frame Orientation and both of them show “X”

However, in APM2 the only Frame indication I found listed it as “+”

FWIW, I have changed it to “X” and I hope it does not create a problem for me.

9

I had a look at both the other logs you posted and am at a loss. They are pretty garbled up for reasons I don’t understand. None have the basic tuning parameters included at the front end and there are big jumps in the log time. For example in your -31609 log, your time goes up normally to 14 minutes, then your next time stamp is 2463 minutes. The logs just seem really messed up. I don’t know if it’s just the logs or a reflection of something going on in your APM. Sorry for the lack of help.

10

@OtherHand: Thank you for looking. You have actually helped a lot! Because I am clueless about them, and it least you pointed out the time discrepencies. It did seem odd to me that the “duration” of my log file was listed as 3 days, 23 hours.

So something ‘is’ going on, just what who knows.

Quick question again"

I noticed when looking at a log file there are hundreds of “lines”…

Approximately, how much “time” does a single line cover? I imagine it is a fraction of a second.

Thanks again! :slight_smile:

11

I’m not sure of the exact timing between lines but there are ways to get an idea. There is a newer “Time” field that gives you an interval. However I think that’s based off a time entry in the GPS row. That’s a time value in milliseconds, 1/1000 of a second. So if you want to know the time difference between two lines, look for the nearest GPS item values, subtract them and divide by 1000.

Were it me, I’d erase the whole thing and reinstatement the firmware fresh. If you save your parameters before doing so you can just reload it after the firmware reinstall. But not knowing exactly what is happening, it’s hard to say if a firmware reinstall would solve your problem,

12

I have not done a wipe and reinstall. To be honest, I am afraid to go that far.\

I have updated to Ver. 3.1.5, and flown a 5 minute flight. I once again downloaded the latest log, and once again it is enormous, over 2MB and when attempting to use Log Analysis I get an error dialog telling me it is no good and cannot be opened. Sigh…