I'm experiencing a strange motor twitch problem with my cs8. The copter does some erratic yaw movements without command. I'm running all Mikrokopter electronics: gps/navi + FC 2.2 and BL control 2.0. It's still flyable but the pulsing motor sound and sudden yaw changes are really worrying and irritating. I believe my Cs8 started this behaviour after the new acc upgrade which went with no errors and was done by a professional. I fly with two 4s batteries in parallel using y-harness, usually two Quadrocopter 4s 6200mah 25C. Motors are Tiger 2814-11 710kv. The same behaviour occurs with both 0.90j and 2.02b firmwares. The new altitude hold works perfectly and positiong hold is solid. Really don't know what to do and it doesn't feel safe to fly anymore Payload is GH3 and CS 360 gimbal. Video downlink is 5.8ghz 500mw immersion rc. TX/RX is Graupner mx-20 & gr-24. Here's some recent footage with audio which you can clearly hear the pulsating motor sound in. https://dl.dropboxusercontent.com/u/12790890/Motor_noise.mp4 GPX log from the same flight Pictures of my CS8 setup: http://sdrv.ms/19OmHiD http://sdrv.ms/1f7d3aK Thanks!
Do you have your flight control board on any sort of vibration isolation? I wonder whether the board may be vibrating or getting some kind of jolt, which might cause it to think it has to correct. There are gyros and other components on the flight control board, and they are prone to problems if the board gets subjected to vibrations.
Lauri Make sure all your motors are completely aligned. If one motor is just a little out the rest can be working harder to compensate. It's very easy for one of the motors or booms to move just a hair with the plastic clamps. I use a bubble level and also eye ball it to double check. It's easy to see visually with the prop adapters on it by referencing the battery tray. One motor might also have its bearings starting to wear down. Those are the 2 top two things I can think of.
Another suggestion with your yaw problem where you flying near metallic objects - bridges, power lines, cranes etc. Turn off your compass effect to 0
One other possible cause for this kind of twitching is a bad connection on the I2 Bus that runs between all of the MK boards, Lauri. Can you connect up with Mk Tool and see if you're getting an increasing number of I2C bus errors -- I'd even go so far as to remove the props and run all the motors under RC control just as if you were "flying." If you have the NC board and insert a 2GB microSD card (formatted FAT16) you can even get a flight log and check for I2C bus errors. I had a similar situation (uncommanded flight "twitches") that instantly went away when I replace the six wire micro-Molex cable between the GPS and Navi-Control board. The other symptom I saw was, with the Smart On Screen Display, the Flight Time clock would get reset to zero each time the copter twitched. That's not to say you're definitely having the same problem, but it is to say that it's one possible cause. Try disconnecting all of the micro-Molex cables and reconnecting them up. Andy.
Thanks all for your great suggestions! I've checked and double checked my motor aligments and that must not be the reason. Also my flight control board rests on rubber isolation so that's unlikely the problem but will check that too. I've flown my copter in various different environments, sometimes in the city, in a football field and even in the forest yet the same problems remain. Also tried setting compass effect to zero as Jason suggested but that had really small influence if any :/ So it seems it indeed could be a bad connection on the I2 bus. I've connected with Mk tool several times during past weeks and had zero errors, but will try "fly# without props when connected to Mk tool and see if I'll get any errors then. I'll also try reconnecting the micro-molex cables and maybe order couple of new ones just to make sure. Thanks again for the fast help, I'll share my findings!
I found the same thing a while back. Turned out the screws on the hub plate got loose enough to allow a boom to turn in flight. This changed the thrust vector enough that the predictive control system had to compensate for. I had five flights whee this was occurring. Finnaly I did a good tug on one of the booms and found that with enough force I could move it. Since then when I re-tightened things it went away.
Shaun: thanks for the good idea, have to double check all my hub screws and motor mounts if there's anything loose. But I doubt that is the case because my copter has had this same motor switch strange noise -thingy going on for couple of months and I have disassembled and put it all together several times :/ Strange thing is that this twitching behaviour seems to get better after being couple of minutes in the air, not sure though :O I've ordered two 6pin molex -cables and also both the ribbon cables just to make sure it's not some bad connection as Andy suggested. I also did a "bench" flight with navi connected to Kopter tool and got zero I2C errors. Is there anything else that I can do to check the BL controls than checking all the leds are solid green? All my BL controls have heatsinks both sides so I can't really visually check if the fets look good. But I suppose I wouldn't getting all green if there was a serious problem. One more thing, how critical is the ACC calbrations accuracy, shouldn't it only have an influence how stable in place the copter hovers? Thanks again for the support!
Just a really long shot: could the parallel battery setup have anything to do with this problem, I'm using a EC5 y-harness? Next week I'll get couple of 10000mah 4s Gens Ace batteries so will test out the flight behaviour with single battery. Of course two batteries in parallel = half the load for each so it really doesn't make this seem to be the problem.
Lauri Not too sure I recommend using 2 10000 batteries in parallel on the 4S system because they are heavy beasts. I think it will be too much weight for the 2814's.
Dave, the 10ah packs are only for single use, not gonna fly two of them, definitely not efficient on my system as you said. AUW with one 10ah pack and my new Bmpcc is about 13,4lbs (6,1kg) so I should get pretty good flight times.
Its possible that a Capacitor might be going bad and not getting a good charge on it when it is cold and the BL could be browning out, its being deprived of instant power power that the CAP would provide. After a few minutes the CAP starts working like it should. If it goes away after a few minutes, then thee cap may be working more efficiently. This is just hypothesis and may not be the case. But it could explain the symptoms. A way to check visually is to see if there is a cap that is bloated at the end of the "can" compared to the others. Maybe bech test like andy suggested but pick the copter up and move it around aggressively to see if you can induce the behavior. The ACC wont do this if its not calibrated well the copter just wants to fly in one direction with the sticks centered.You may see it tilt in that direction a little.
Another thing to try is to tether bench test outside with the props on connected to K-tools and monitor the motors. Without the the props on the motors are not really exerted even at full throttle because you don't have the drag from the spinning props. This is where you can at least see if you can narrow down the trouble spot BL.