you are doing well and you have an excellent teacher in Shawn no’s arducopter very very well
I am very thankful for his help. So much
1 Like
This thread for that error, Ardupilot determined an IMU was not operating correctly and reset it.
Could be an anomaly or bug, or sign of a hardware issue.
Roll is tracking extremely closely to Desired Roll (and same for pitch) so I wouldn’t change any PIDs yet.
Your RC1 channel (roll, left/right) has a bit of jitter but it doesn’t look very bad. Are you sticks spring-centered?
You could try increasing RC1_DZ,20 to 30
or lower these a bit
ATC_ACCEL_P_MAX and ATC_ACCEL_R_MAX to about 70000, they dont need to be very precise.
1 Like
I am using a Futaba T14SG.
Here is a video of it.
I will make those changes and see how it goes.
The mission worked well. The drone was pointing to the middle and collision detection is working well also. I will change my I2C bus this week to get power from the VTX 5V. I want to add more sensors as the money comes in.
I called my drone Demon and Meteor, but I need to rename it now because @xfacta “The Exorcist” has cured my drone. Maybe the new name can be Chopper since it will chop you into pieces if it gets too close 
The flight today running a mission
Avoidance working well
Changed RC1_DZ from 20 to 30
Changed ATC_ACCEL_P_MAX from 79171.95 to 70000
Changed ATC_ACCEL_R_MAX from 79171.95 to 70000
This was a quick test before the flight, it still bounces. Not sure if this is normal for this kind of wind.
This was the flight recorded using my goggles. The quality isn’t great, but maybe it will show something useful. I configured brightness and contrast, but seems when recording it ignores that. I am using the legal 2mW power on the VTX and I didn’t actually fly with the goggles.
1 Like
It’s an altitude shift up or down rather than a overshoot/undershoot in the angles and rates. So PIDs are still good, don’t change any of that technical stuff yet.
ATC_ACCEL_P_MAX and ATC_ACCEL_R_MAX could go back to how they were after Autotune, or leave them like they are now for testing. It’s not critical.
I think the barometer is affected by prop wash during those leans and velocity changes. You have quite large props for that sized frame.
Open the flight controller and check there is enough foam around the baro, even change it or add more if you have to. The baro is on the underside of the board and looks like a tiny steel box on the circuit board. There’s usually a moulding in the case to hold the foam in place.
EDIT: the pic shows a small amount of foam that may not even completely cover the baro - more is more!
If that doesnt work you will probably need to put a cover over the whole flight controller/top plate area with only the GPS poking out. It doesn’t need to be sealed, just to shield the FC from direct prop wash.
Think upside-down take away food bowl
1 Like
I will do a basic test with something over it today. Will it help if I have a downward sensor?
1 Like
I taped a plastic bag over the drone to stop wind from hitting it directly. Not sure if it did much, but I think I will still print a casing for the controller when I get my new body. Anything that might help.
1 Like
I swapped motors 2 and 4 (motor 2 = problem). I still have a problem. The motor shakes a lot when I am running at a low RPM. I think the motor is not happy or maybe not balanced. I have one in stock that I will try. The blades are balanced. I just swapped the motors.
Here are the logs of today. The wind was crazy on the mountain where I was filming so it is probably a bit difficult to read the logs. I didn’t fly with the plastic bag over the Pixhawk today. I will open it up this week and see if I can add that foam insulation
https://www.dropbox.com/sh/24sgijgljhb3p85/AAC1ZaEByL5m9LuFggYAU4s5a?dl=0
Would a downward facing lidar help more to keep the height constant or is there an external barometer that people add to their systems to make it better?
At some point, the range finder just didn’t work. The Pixhawk complained it couldn’t detect it. I moved wires around, pressed in connectors and nothing worked. After a restart, it worked. I wonder if it is because I am still running it off the Pixhawk power. I will make that change soon after I fixed the retractable legs. I have to print parts for them because the microswitches were soldered in the wrong spots so they don’t activate. These Tarot retractable legs are really not great.
This was my first flight away from the oval and it was scary. The video isn’t wonderful. I was above the landing zone most of the time, but at least it is a start. This was without a gimbal. Just the GoPro 10 cable tied to the base. You will see how the wind is bending the trees. Seems the drone is fine in this kind of wind.
I am not sure if I need more powerful motors, seems fine. I am using the 360kV at the moment. I started at 530kV, went down to 260kV and I have been using the 360kV for some time now. The drone will get lighter when I convert to carbon but then it will gain weight again when I add sensors.
1 Like
Today the crash demon struck again 
This was a bad one.
I need to replace the body after this one so no flying again until I get my carbon body.
I tested a few fast forward and backwards without loiter on and a few fast up and down. I guess this caused something to give in. It looks like the one motor went into override. This was the problem motor that I moved from position 2 to position 4.
This is the logs
https://www.dropbox.com/s/y80gr12nfrlhlc8/00000190.BIN?dl=0
This is the last few seconds
1 Like
Yes motor 4 commanded to maximum due to it not providing enough thrust, then chaos.
1 Like
Sorry about your crash Kobus but your enthusiasm is so inspiring obvious you are enjoying the Arducopter journey not the crashes keep up you are doing great
Thank you. Yes I learn by doing and I keep on going until I get something working or I run out of money 
1 Like
I just tested the motors. Two of the good motors are bent. The third cuts out. So seems I need a new batch or I need to take the old motors apart. I don’t know if I can just use the shafts of the 260kV and 530kV motors that had their wires ripped out. Now I am wondering if I should go back to the 530kV or stick with these 360kV motors. If I go to 530kV I can just change a setting in Ardupilot to make lifting off easier with an overpowered drone and then I have spare power in case I need it.
The motors seem to be running mostly max 1650 if I am flying in loiter. When I pushed it to max speed the motors hit almost 1750. If I am correct the drone was doing about 16km/h in both directions. This is probably not enough power for a very windy day. According to the weather website, the wind was about 27 km/h today (moderate).
Do you guys use parachutes? I think I want to install one before I fly the car in version of the drone because that is when crashes will become expensive
New motors are in the post and carbon body is in the post. Hopefully I can order my parachute next week.
I decided to go back to 530kV motors. I can’t find the setting that reduces the throttle middle point. It isn’t the one that is around 0.2. It is the one that is around 500 that I need to reduce to probably 300.
1 Like
I haven’t tested their parachute yet, but these guys were super helpful with a project I was working on and really went out of their way for me.
1 Like
Thank you I will have a look at these guys also. I was going to go with Fruity Chutes because they just use a spring. So cheap to reload
1 Like
The actual parachute used by Skycat is made Fruity Chutes. Skycat makes the launcher, and it’s also spring powered.
1 Like
Seems you need to buy fuses, not a big issue because you get 20 with the package. If you fall 20 times you have bigger issues. I have not even fallen 20 times
Looking at this one
Costs about USD 500
The Fruity Chutes is around USD 350