Major compass issues

[htotheunit]

Hello,

After my incident last week, I fixed my hexa replacing all but one of the arms, two motors and the rest of the broken bits.

However I now have some major issues with my compass. I remounted it where it used to be on the top plate which resulted in green/yellow flashing. After successfully calibrate it several times I moved it up onto another standoff plate well above the rest of the electronics. This appeared to work for a lengthy hover test and around 10 flights afterwards. Then I began getting the green and yellow lights flashing again, with the hex flying in atti mode periodically regaining compass data before losing it again.

A couple of weeks earlier I did notice some odd behaviour before the crash flight, The copter appeared to fly forward before quickly levelling off even though my stick was forward before continuing on its path. On these occasions the light was again yellow and green, is this normal should the compass data be lost during movement?

Is there any way in which I can test the compass for faults?

Here's a picture of my setup, apologies it's sideways I would fix it but I have to go somewhere in a minute!

View attachment 16786

 

[kloner]

when you crash enough, the flight controller will beging to break down... there is fragile chips, really small soldered in connections, alot of room for failure. only way i know to test parts is to swap em out for new ones..... some shops might let ya come by and do such. most dealers will sell the parts individualy

 

[htotheunit]

That's true, Is it possible to connect the compass to a different port on the WKM to test whether is it the MC or the GPS module

 

[kloner]

not that i'm aware of, when i've had this i went to a dealer, they broke out a wk and we tried each piece,,,,, my gps was similar to yours and it was my main flight controller that was damaged.... when it starts malfunctioning theres no make since of it all enough to troubleshoot it

 

[htotheunit]

Thank you! I'll see whether any dealers round here will let me drop by. May I ask, how much did you spend replacing the flight controller or did you have to buy the entire Wookong kit?

 

[kloner]

i was scared to use any of it so bought a new a2..... didn't know what else had happened, she registered a 10g whallop of a crash and came up no worky worky

 

[wearypanda]

I have had similar issues and ended up replacing the compass/GPS sensor. One thing I noticed before it was replaced was that when I looked at instrument panel in the Ground Station the actual compass reading was about 30 degrees out from the actual heading. The compass rose on the GS was behaving normally when I rotated the hex, but the actual value was all wrong. Other symptoms were strange flying characteristics, bad position hold etc. The symptoms gradually became worse over time up to the point where in 5.36 I cold not invoke GPS Atti at all and the compass became increasingly difficult, and then impossible to calibrate. I suggest that you borrow or otherwise acquire another GPS and try it out. Hope that helps a little.

 

[htotheunit]

Just a quick update on this. I replaced the compass and moved it back down to the level at which I used to have my compass and got the same issue. I then moved the compass up another level (as in the first picture) and it's working OK.

The compass issue (from what I can see) not only occurs if I climb quickly or fly quite fast. This would maybe suggest that it is indeed interference from somewhere on board only coming to light when the current draw from the batteries is increased.

I have also installed 2 power cables from the main power distribution up to the Wookong M and IOSD. (4 cables from power board merging into a deans connector, then split one each to Iosd and Wookong PMU) Would this potentially cause any undue interference?

 

[R_Lefebvre]

The main cause of compass interference is inductive loops. All power wiring should have both positive and negative wires run along exactly the same route, and tied together so they have no separation between. Twisting them is even better. The next best solution is simple separation between the mag, and all the power wiring. Also, some Lipos generate a magnetic field internally.

Does DJI give you any ability to view the mag field strength? This could help track down problems.

 

[htotheunit]

I think I have found a major contributing factor, I'll downgrade to 5.22 and see if the issue eases.

http://www.multirotorforums.com/sho...-Firmware-5-24&highlight=yellow+green+flashes