Help Deciphering A Crash - GPX File

I'll have a look at it now Colin. Sorry to hear about the damage. Can you double check on the FC board that:
1. The air pressure sensor legs are all intact.
2. There is some hot glue below the air pressure sensor and the FC board to stop it vibrating.

I'll post again in a few minutes after I've examined the GPX file.

Andy

 

I used MK_GPXTOOL to examine the GPX file.

The only apparent error (apart from a Low Battery warning at the end of the flight, was that the compass calibration was off.

Also motor #3 is pulling way more current than it should be. Could be the prop was loose, or the motor's bad, or it's having to work overtime to counteract one of the other motors not being vertical -- I realize you cannot check that latter possibility now.

The conclusion: it looks like the battery voltage decayed away and the copter could no longer maintain AH. Bear in mind:
1. The GPX log only samples once a second....so you can't see what's going on in between.
2. The last few lines of the GPX file show that it was improperly terminated -- the NC board had just started to write out a new "trackpoint" (the data point that it writes once a second) and then the GPX file terminates -- so there is missing data at the end -- and that might show why the copter descended those last 8 meters so quickly.

Here's the summary from MK_GPXTOOL (with some interesting data in red):

MK Version: FC HW:2.1 SW:0.88e + NC HW:2.0 SW:0.28h​

​

Flight time: 10:50:54 AM - 10:51:55 AM (61 secs, 00:01:01)​

Total: 62 secs, 00:01:02​

Elevation(GPS) : 0.775 10.74 13.857 m​

Altitude(Barom.): -0.05 10.28 12.35 m​

Vertical speed : -0.77 0.17 1.79 m/s​

Max speed : 56.5 km/h​

Max target dist.: 2.3 m​

​

Sats: 10 9 10 (min/avg/max)​

​

Voltage: min. 13.5, max. 15.8 V​

Current: 25 44 72.7 A​

Wattage: 395 661 1009.58 W​

Capacity: 800 mAh​

​

Motor1: 1.2 9.1 14.4 A Temp: 22 42 52 °C​

Motor2: 0.5 2.8 4.8 A Temp: 22 38 47 °C​

Motor3: 1.6 10.8 19.1 A Temp: 22 45 57 °C (these are min, average, max)​

Motor4: 1.5 5.3 9.5 A Temp: 16 34 43 °C​

Motor5: 1.7 8.4 10.7 A Temp: 18 39 50 °C​

Motor6: 1.9 4.2 6.0 A Temp: 21 34 42 °C​

​

Magnet Field: 104 116 125 % <- check!​

Magnet Inclination: 56 57 62 deg​

​

Errors / warnings:​

Compass is disturbed by magnetic fields! ​

Corrupt GPX file end detected! (Loaded 59 tracks)​

FC-Flag "LowBat" (5) occured 2 times!​

​

I can see that you enabled Altitude Hold 12 seconds into the flight and it remained on until the last trackpoint was recorded 60 seconds into the flight. The altitude was holding around 10.9 - 11.1 meters until 58 seconds, then it drops to 10.552, 10.345, and 9.835.

You had a continuous GPS lock on 10 satellites (which is good).
You took off with GPS Hold on and disabled GPS Hold at second 30.

The last few seconds of voltage/current are telling:
Second Voltage Current
54 14.5 48.4
55 14.4 49.4
56 14.5 47.8
57 14.4 52.4
58 13.9 63.5
59 14.3 70.6
60 13.5 72.7

As you can see the voltage is decaying, and the current has jumped to 72.7 amps. This implies that the MK boards were perhaps struggling to maintain altitude pull trying to put more power to the motors, and pulling more current, which causes the voltage to sag....

At second 54 (which is the last "normal" looking reading), the motors were pulling 8.2, 3.3, 14.0, 5.8, 9.4, 3.3 Amps.
At second 60 (just before impact), the motors were pulling 14.4, 4.8, 19.1, 9.5, 10.0, 3.9.
Note how the odd numbered motors are really working much harder than the even numbered ones. I'm not sure why.

There are no I2C bus errors so that's not the problem.

Also, looking at the RC Sticks (that is to say, the position of the RC sticks) it appears that you had the throttle stick in its center position for almost the entire flight. Do you have the throttle stick self-centering? Just curious, as it starts out just slightly below the center position and after about 4 seconds goes to the center position and stays there for the rest of the flight.

There is nothing to suggest that the AH failed -- it was getting value altimeter readings that correlated to within a meter of the Elevation (which is what the GPS thinks you were at). So I infer you were flying pretty close to sea level.

I don't see anything else anomalous in the GPX file. No wacko readings of nick, roll angles, no major yawing and no control inputs from you that would cause the crash (nor from the MK boards either).

So, as I say, it looks like you ran out of gas....

Hope this helps
Andy.

 

By the way, I agree with you -- it's odd that you didn't get a low battery warning.
Can you check that you do get a low battery warning either with a bench power supply or using a LiPo that has a low voltage?

Andy.

 

This is where the Xbee wifi setup allows you to monitor the yawing component of the copter in flight using the Balance window of MK Tool, so you can see whether there is a yawing component. I'm surprised that, following William Johnston's technique (formerly known as wgj000 on the forum), wherein his code analyzes the yawing moment in a GPX file, that there's not something calculating a realtime "odd motor currents versus even motor currents," so we could see this information on an OSD, for example.

That kind of ratio seems as interesting as the total current draw as it "telegraphs" a problem....

Andy.

 

Hi Colin:
1. Guinness please.

2. My unthorough analysis (which I didn't post) was: Hey, dude, motor #3 is naffed, and you crashed!

3. Compass errors: Check that you don't have heavy current carrying wires near the Navigation Control board. That's where the compass is located. Double check the orientation of the board too -- I suspect it's going to be correct as Really Bad Things happen when it's not. The most likely causes are either electromagnetically generated fields or a failed compass unit. You might also consider loosely braiding the wires inside boom #1 and see if that makes a difference. Don't braid them tightly as that shortens the wires a lot! Check which way your flight batteries are mounted -- do the battery leads come near the NC board -- if you rotated the battery 180 degrees would that move the leads further away from the NC board?

4. Motors #1, 3 and 5 always seem to be working signficantly harder don't they? Hard to know what's causing that. Can you verify that all the props are sufficiently tight that you can just barely move them if you hold the motor casing and pull on a blade -- you don't want them so tight they crack the prop, but they shouldn't slip that easily either. Another one of those "Goldilocks" settings (not too tight, not too loose).

I would run MK Tool motor test on each motor individually and listen to the sound the individual motors make. High pitched smooth whining is what you want, not high pitched coffee-grinderish noises. How's that for a technical explanation? If there's a raspy component, then you might be looking at new bearings. (Search the forum for Boca Bearings to find the relevant thread).

Hope this helps.
Andy.

 

Send us a picture of the center of your copter. Take it from left, right, back, front and from overhead. Usually as Andy says there is something close to the Navi that is emitting EFI. It can be the things he mentioned, positioning of an Xbee, position of a GoPro. That's why photos would help. And take them with the copter rigged as it was where you got the errors.

 

Colin do you have an Xbee installed? Carrying a GoPro? After a compass calibration you can take the copter inside and while watching MKTools or one of the MK data screens on the MX-20 you can sometimes see the compass information change as you power up/down various other devices on your copter. I had a problem with my Xbee at one point. Moved it over one boom space and compass issues went away.

Here's a couple of notes about the compass from MK

General Points

Caution: For use of the MK3Mag it is essential that the distance between compass and Buzzer minimum 10cm. It is best to mount the buzzer on the outer end of one of the frame arms.
The highest Interfering Magnetic field is generated by: Buzzer, Motors, Power Cables and Metal Parts (In approx. this order).
If problems with the MK3Mag are only noticeable at higher throttle, it is likely that the interfering magnetic field is generated by power lines. Ideally the individual DC-lines should have very few bends and Plus/Minus should be routed parallel as their magnetic fields are then compensated by each other (Tips and Video).

Operating Note: The Navi board trusts the compass data to be 100% accurate. Bad compass data will be reflected in a toilet bowling effect or worse case a flyaway.
The best thing to do is to turn off all navi functions and learn how to fly manually and regain orientation based on stick and craft movements. This is where simulator training comes in helpful.

 

Colin thats about as clean as you can get. I don't remember, but I know Andy will, as to where the compass is located on the board. The only thing you might consider is moving the receiver to the back boom. It sounds like you have had the compass reading since the beginning but do you get them if you don't have a camera on the copter?

 

Colin: Just a thought, but you might try twisting your Lipo leads or zip-tieing them together. In the image you link to above, they're separated and *might* be causing EMI with the compass. Not clear whether they would be, but it's a case of eliminating likely culprits.

Also, in that vein, is there any chance that you're compass calibration procedure is not as Casey shows in the QC video? (Albeit away from cars and metal railings! ). Not trying to be accusatorial, just trying to eliminate possibilities!

Andy.

 

Colin:
One other thing. That TM1000 is a transmitter and it's right under the GPS/Navi boards. Could you try relocating it? Just wondering whether the transmissions might be strong enough to cause problems -- again, this is baseless speculation, all we can do is try changing things until the symptoms change...

Andy.