Unexpected motor glitches (speed up) with Tri Hex and Octo APM 2 & APM 2.5 using 2.73

It seems since going to 2.7 and up to 2.73 I am having uncontrollable glitches with the motors. I think I have isolated everything, no bullet connectors, checked the amperage draw of the 880 motors to see if bearings could be the problem.
I have 2 units, TriHex & Octo with 1 inch carbon fiber tubing for the arms, exhibiting the same glitches, I don't know what's causing it. This morning I flew the TriHex for about 10 seconds and about 5 feet high and suddenly it rolled completely over upside down all within less than a second and broke 5 props, anybody else having this problem or a solution. I'm not flying anymore until I have a fix.....I see 2.74 is in the works but don't see anything that would fix this....Don

Views: 3839

Reply to This

Replies to This Discussion

If the repairer is thorough they should find it, and I am assuming the 12FGH is the same as the 12FGA I have here with a module on the rear, it is usually a weak or cracked solder joint on the pins in the Tx board that the module plugs onto.

The glitch's are not always evident as I flew some indoor models on my 9CAP that I know IS glitching the APM. 

"Sounds like you have RF interference" - called it.

 Try and ground that foil.

I spoke too soon, the glitches are back. I did notice on the triHex, the left front glitches most, not sure which motor, top or bottom, wonder if it isn't EMI, or an esc.......Don

Hi, I have the intermittent ESC blips on my APM2 (on current quad firmware 2.7.3) too.

They are quite easy to reproduce during low throttle, (make small cyclic stick movements esp. with roll&elev together, i.e. diagonals.) and seem to almost never happen during higher throttle levels.

The blips / twitches / glitches did occur in flight too, the quad would make a quick bank and quickly back to level again.

I will try to shoot and post videos when I can.

I have tried to check if it was a motor/ESC problem, so I tried connecting each of four ESC-motors one-by-one. And for each ESC-motor, I tried it in all 4 motor outputs of the APM2, the result is that I could get the blips in ALL cases. This tells us it is very unlikely to be ESC/motor related.

Then I recorded some telemetry, and graphed the "servo outputs" and "rc inputs". (see attached tlog file) And checked for spikes. The largest spikes were not commanded by the RC.

My APM2 (in plastic case) is simply cable-tied on to my GF frame with a little grippy-rubbery mat. But I don't think my vibrations are too bad, I did statically balance my props and I get nice stabilization and flight characteristics, except for the blips. I will test the possibility of vibrations causing the blips by removing unfixing the APM2 from the frame and maybe handholding it while running my low-throttle tests. If the blips still happen, then I think I can rule out vibration as a cause.

Attachments:

Here are videos showing my APM2.0 with 2.7.3 quad firmware, glitching:

On the ground: http://www.youtube.com/watch?v=_8Wsnhleqwc

In hover (1 of 2): http://www.youtube.com/watch?v=ZOhoo9j2How

In hover (2 of 2): http://www.youtube.com/watch?v=c8ezlKMyBak

With the APM2 isolated from frame vibrations: http://www.youtube.com/watch?v=AUot6Mzz4q8

This is evidence strongly suggesting that the twitching CANNOT BE BLAMED ON VIBRATIONS.

RsdPalm,
I think I experienced something similar to what you are seeing. With 2.81 I have about 8 flights all in pretty calm weather, winds 10mph or less, no problems, no noticeable glitches until today. It was very windy and gusty and it was glitching like crazy, kept it faily high, didn't want it flipping over to the ground. I had to use a lot of cyclic controls and that's where I noticed the majority of the glitches. It seemed while in loiter and nminimal cyclic control, I did not notice any glitches........Don

Now that Arducopter 2.81 is out, I decided to take this chance to document the glitching again on 2.73, then upgrade to 2.81, then check for glitching again, for the sake of comparison.

Unfortunately, (by my unscientific tests), the glitching problem appears to be the same or worse in the new 2.81 compared to 2.73.

Video of the glitches on 2.73: http://www.youtube.com/watch?v=pPqUlgg4CJo

Video of the glitches on 2.81: http://www.youtube.com/watch?v=J6HVBRuqYtU

Actually, you'll have to listen to the abrupt blips in the motor whine. The glitches aren't actually visible.

 

Attachments:

matching rlogs in this message.

Attachments:

Hi Don,


Had exactly the same behavior.
Tried to narrow it down by taking my quad apart completely as I needed to find the cause.

My setup:

Board: APM2.5 (2.8.1 firmware)

RX: AR8000 (DSM2)

ESC: 20A Hobbyking blue series (SimonK flashed)

Motor: 2212 - 850KV

Lipo: 3S 2200mAh

Bec: 3A Turnigy at 5V

Took out the board, hooked up to RX and one motor to ouput1.
To make sure there was no vibration as cause the board and motor were separated.

Even used separate power source to power the board and motors
Even with all taken apart still glitches.
So as the RX was the only thing left and possible cause.

RX setup set as described in the manual for basic channels

ch5 - Gear (AR8000) mode

ch6 - Aux3 (AR8000) knob to be able to tune in parameters

ch7 - Aux1 (AR8000) function to be defined

So I removed the connection between aux3 on the AR8000 and the ch6 of the APM2.5

The glitches were gone!!!!!!! Very, very strange.

I tried it at least 5 times to be sure.
Haven found this I started looking further and found that when aux3 (AR8000) is at maximum you will have glitches.
When you put it to minimum, no glitches.

It is very unfortunate not being able to use aux3 (knob) on the DX8 as it is very handy to dial in the tuning values.
To be sure nothing is wrong with the RX I hooked up another AR8000, same behavior.

At this moment I do not have an explanation for this but if someone who has the same behavior could test this, it would be nice and if someone finds a cure for this it would even be better :-)

Additional tests...

I removed aux3 (AR8000) and hooked up aux1 and aux2 of the AR8000 to ch6 and ch7 --> No glitches
Put aux3 (AR8000) to ch7 and turned aux3 tot maximum --> glitches.

Is there something wrong with the ppm encoder on the APM2.5?

I also have an APM1 where I use the same receiver and same setup (aux3 to ch6) without any problem.

Interesting,
I had a problem with all four motors pulsing and then it went away. The only change I made was to remove ch 6,7 and 8 connection to the RX.
I haven't had a chance to investigate further but this makes sense.

Randy

Randy,

I can confirm that the glitches are on all four motors. I just hooked up one motor to test.

What brand of RX did you use?

Jan

Reply to Discussion

RSS

Groups

Season Two of the Trust Time Trial (T3) Contest 
A list of all T3 contests is here. The current round, the Vertical Horizontal one, is here

© 2019   Created by Chris Anderson.   Powered by

Badges  |  Report an Issue  |  Terms of Service