Bad Twitching, how to diagnose?

ArduCopter Copter 4.0
1

First, here is my log.
2021-03-27 11-33-47.zip (717.8 KB)

I am finishing build and doing some test flights, I think I’ve eliminated vibration to the point that’s acceptable.
But I’ve noticed really bad twitching that almost made my crash, kind of like one of the motor was loosing connection ?
How can I diagnose that from the log ? I can’t see anything obvious wrong.
I know battery was not fully charged, but I’ve seen that even when it was fully charged.

Side questions, Why I wasn’t able to switch modes to GPS modes? I am assuming it was due to GPS not having correct positions, but then, is there a way to prevent arming, until GPS is ready for those?

2

FENCE_ALT_MAX,120 <- set for safety and local laws
FENCE_ENABLE,1
FENCE_RADIUS,600 <- set for safety and local laws
FENCE_TYPE,3
This will make you wait until home position can be set before arming.

1 Like
3

This craft is underpowered/overweight and when your battery has sagged to <10V there is insuffient thrust to fly with all motors commanded close to or at Max.

RCout
RCout3367×1009 249 KB

Graph GPS NSats and Hdop for the answer to this question.

1 Like
4

That makes sense, I think I have enough power, it was just that battery wasn’t fully charged, I guess I have to make battery low limit higher.

From the graph above ( or other graph) how can I see/check that I have enough power when fully loaded ?

5

It should hover with an average commanded motor output of ~1500us or less.

If this is a 4S battery as your parameters suggest you will kill it discharging it this low.

Or is it a 3S battery and these are set wrong?
MOT_BAT_VOLT_MAX,16.8
MOT_BAT_VOLT_MIN,13.2

6

it is 4S, and you are right, I’ve changed failsafe to 12V.
Thank you for your help

7

So I’ve charged battery and did another fly, and it twitched so bad so it crashed.

I think it’s only 1 motor ( second) it’s loosing power and that corner goes down.
here is the log
https://drive.google.com/file/d/1X48p_-Mpgm1cCbOOPdKyCg2HCVYfz25D/view?usp=sharing

8

What is the weight of your aircraft? And what is the discharge C rating of your battery?

9

4S 70C, weight is around 1500g.
I have 600KV motors with 10" props

10

A photo of the drone would be helpful! The battery C rating is more than enough!

11

Right, it does seem to be an intermittent problem with Output 1. which with your motor order is motor 4. Then the opposite arm motor is dropped to compensate.

Out
Out3316×1001 224 KB

12

Oh yeah, make sure that the connections to the motor (soldering) is okay and no cables are faulty!

13

While you were typing, I just discovered that ground came out from connector to ESC but on motor 3.

I have different ordering of servos vs motors I think.
But order on the graphs should still apply as
4 1
3 2
Or not ?
Either way that seems like it could be an issue, but it doesn’t really match ordering.

1 Like
14

You mean this?

image
image573×758 52.1 KB

15

Re-solder all of your esc cables, because on an aircraft that weights that much, if one motor disconnects it would not be able to compensate and will straight fall to the ground (happened to me personally on a drone that weights 2kg).

16

Yes, so using Motor test in mission planner, that’s my order
But to achieve that I had to swap outputs.

Yes it’s scary when it does that, it goes right to the ground.
I will recheck everything

17

You do have different ordering but not a big deal. I suppose you did that because of how the ESC is oriented with the outputs to arms? Many 4in1’s for example have Betaflight motor order.

1 Like
18

Not sure why is that, as I have Kakute F4 V2 AIO, it already forces order of ESC’s, so I did it that way, but that seems to be different to what’s programmed
image

Actually when I look at the drawingand naming, it makes sense now

19

Not a big deal, if the order would be wrong, it would not take off, it’s just Mission Planner messing it up.

20

well I guess before I reprint my legs again, I will re-check and re-solder all wires again :wink:

1 Like