Issues with my Green Cube Solo and ArduCopter 4.0 & Open Solo 4.0

teo58 · Jan 21, 2020

I have been flying my 2 Solo birds successfully for more than a year. Both my Solos have mRo GPS modules, with one Solo (OS-A) with Pixhawk 2.1 (Green Cube) & WiFi mod and the other one (OS-B) with Pixhawk 2.0 and standard configuration.

Recently, I decided to update both to ArduCopter 4.0 & Open Solo 4.0. The installation process was very smooth, without any problems. After completing the calibration (IMU & Compass) the first flights were perfect and I started to enjoy the extra features, especially with my Pixhawk 2.0 – Solo, smoother flight characteristics. The Green Cube Solo was a bit quieter, but the rest was almost the same as with Open Solo 3.0.0 and ArduCopter3.5.4, until the prearm issues started to pop up…

When I start my Green Cube Solo and I arm immediately, after acquiring the minimum number of satellites, everything is 100% and can complete my flight without any issues. If I wait more than 20 – 30 seconds (as I usually do, to get more satellites), Solo cannot be armed anymore and the Solex App displays a flood of prearm error messages: “ekf3 yaw inconsistent by xx deg; gyros inconsistent; gps is not healthy; etc. If the throttle stick is kept bottom right continuously, an error beep will sound every second or so and the angle of the ekf3 yaw will increase with a few degrees every time. Sometimes after 40-60 beeps Solo will arm, like nothing was wrong… Please see pics and clip attached.

Does anybody have any idea what is wrong with my Green Cube Solo? Thanks.

Please note that my other Pixhawk 2.0 Solo, with exactly the same firmware & software, has no issue at all and flies beautifully, much better than before. Many thanks to Matt & all the people involved in development, who give us this wonderful new life for our Solos!

just_bruce · Jan 22, 2020

That sounds like a hardware issue methinks.

teo58 · Jan 22, 2020

just_bruce said:
That sounds like a hardware issue methinks.

It could be, but I’m not convinced… Are you flying a Green Cube Solo with ArduCopter 4.0 & Open Solo 4.0? I would like to hear from people flying the same configuration Solo (hardware & software) as I do. Thanks.

fvincent · Jan 22, 2020

teo58 said:
It could be, but I’m not convinced… Are you flying a Green Cube Solo with ArduCopter 4.0 & Open Solo 4.0? I would like to hear from people flying the same configuration Solo (hardware & software) as I do. Thanks.

Have you tried calibrating the controller?

teo58 · Jan 22, 2020

fvincent said:
Have you tried calibrating the controller?

Which controller?

DJMc · Jan 22, 2020

A stick calibration on the controller unit itself - the hand held bit, there's a thread on here...

teo58 · Jan 22, 2020

DJMc said:
A stick calibration on the controller unit itself - the hand held bit, there's a thread on here...

Thanks, I found the thread and did the remote controller calibration. Unfortunately, the problem was still there ☹

It appears the issue is probably more with the software (firmware) rather than hardware. I went back and re-installed Open Solo 3.0.0 (Controller & Copter) and all my prearm issues disappeared. I armed 10-15 times and had a short flight without any problems. I’ll keep testing and let you know.

Are there any Green Cube & Open Solo 4.0 - Solos flying without any problems? It would help me a lot to know if my Solo is different from the rest…

Pedals2Paddles · Jan 22, 2020

A stick calibration is not going to fix a gyro error. lol.

This is actually probably due to temperature causing gyro drift. Immediately upon boot, it's fine. Then over time the gyros are heating up from the heater causing drift. If you reboot the copter, that will recalibrate the now-heated gyros on boot and you'll be fine.

This is especially prevalent in the winter outdoors since it can start out very cold and heat up to very warm. Best practice in cold weather with the heated green cube is to let it warm up on the ground for about 5 minutes, then reboot the copter so it recalibrates hot.

teo58 · Jan 23, 2020

Pedals2Paddles said:
A stick calibration is not going to fix a gyro error. lol.

This is actually probably due to temperature causing gyro drift. Immediately upon boot, it's fine. Then over time the gyros are heating up from the heater causing drift. If you reboot the copter, that will recalibrate the now-heated gyros on boot and you'll be fine.

This is especially prevalent in the winter outdoors since it can start out very cold and heat up to very warm. Best practice in cold weather with the heated green cube is to let it warm up on the ground for about 5 minutes, then reboot the copter so it recalibrates hot.

You are right, it didn’t fix my problem, but I’ve learned something new – how to calibrate my sticks…

Temperature causing gyro drift is logical and could explain my problem, but, there are still a few questions I cannot answer:

When the flow of prearm errors starts, I get not only gyro problems, but also compass, GPS and other issues. Could the temperature be the only reason for all?
Here, in NZ is middle of summer and the temperature is between 19C–25C (66F-77F), which I don’t think qualifies for a cold start.
As I mentioned, I went back to Open Solo 3.0.0 (Controller & Copter) and all the prearm problems disappeared. Everything was working 100%. Then I installed Open Solo 4.0 and prearm issues appeared again. Are the prearm checks different?
Today I installed the old ArduCopter 3.5.4 Firmware and everything looks fine, so far. Can I use AC3.5.4 and Open Solo 4.0, or I have to go back to Open Solo 3.0.0?

Thanks.

Pedals2Paddles · Jan 23, 2020

Lots has changed in the two years from 3.5.4 was high tech. So I would suggest simply redoing your level and compass calibrations and see if the errors settle down.

teo58 · Jan 23, 2020

Pedals2Paddles said:
Lots has changed in the two years from 3.5.4 was high tech. So I would suggest simply redoing your level and compass calibrations and see if the errors settle down.

I’ve noticed a 30%-40% increase in the number of parameters for the new firmware, which is significant.
In the last few weeks, I have been calibrating, more than flying. Yes, after each (IMU & Compass) calibration the first 2-3 flights are Ok and then, the prearm problems start to show. It “feels” like some info. Is changing by itself, or decaying after a certain number of cycles. There is also the possibility that my Solo has a unique hardware problem that creates all this ☹
It encourages my old mind to think more, and I should be happy for that… ?

Pedals2Paddles · Jan 24, 2020

Well that does sound like a hardware issue. The firmware certainly isn't decaying or changing itself.

teo58 · Feb 1, 2020

UPDATE: I have been testing & flying a lot this week and I reached a few conclusions about my Green Cube Solo prearm problems:

Most probably my Green Cube has a consistent start-up hardware issue which generates all those prearm error messages if running latest firmware, but not if using the old 3.5.4 version.
You were absolutely right about the boot gyro drift, and after heating up, getting a perfect re-calibration on the 2nd boot.
I will follow your advice and stick with the latest Open Solo 4.0.0 (26/01/2020) and ArduCopter 4.0.1 Green Cube, even if I will have to boot twice and be in the air 3-4 minutes later compared to my other Black Cube Solo.

Thanks for your help & patience ?

user096 · Feb 11, 2020

Same issue here with green cube, open solo 4.0. First flight was beautiful, didn't think I had an issue. Then failed on updating a second solo. Went back to the first and, surprise! Around the fifth or sixth attempt hitting the fly button it armed and started. During that flight an error popped up and said something, yaw reading perhaps, was reset and message went away. Never an issue with open solo 3.0. Will continue to monitor.

Pedals2Paddles · Feb 12, 2020

user096 said:
During that flight an error popped up and said something, yaw reading perhaps, was reset and message went away

That's not an error, and is normal operation. After you takeoff, it's detecting a change in magnetic fields, usually due to magnetic interference in the ground where you took off from. Once you're in the air and turn a bit, the compass will realize this and realign itself. That message is telling you such. It would have said something along the lines of ground mag anomaly, compass realigned.

user096 · Feb 12, 2020

Two for two. Green cube Solo #2 exhibiting the same yaw error message not allowing arming. Seems that once the error drops to 20 degrees or less arming and flight is allowed. Seem to be able to arm and fly if I hit fly as soon as searching gps clears and hit fly message appears. Neither of these were loaded with the green cube specific firmware. May try that next and see what happens.

User96

user096 · Feb 14, 2020

Pedals2Paddles said:
That's not an error, and is normal operation. After you takeoff, it's detecting a change in magnetic fields, usually due to magnetic interference in the ground where you took off from. Once you're in the air and turn a bit, the compass will realize this and realign itself. That message is telling you such. It would have said something along the lines of ground mag anomaly, compass realigned.

Didn't necessarily mean to imply that the yaw reset message was an error. Certainly did not prohibit flying. Just an observation that may further indicate something going on with the compass. Solo #1 still exhibited the yaw inconsistent by ## degrees which prohibited arming and takeoff initially. And now my second green cube solo is exhibiting the same issue. I'm still pondering what the issue is. Could it be compass drift? To produce a "yaw inconsistent by XX degrees" assuming that's a meaningful message, would imply comparing two yaw readings that do not match. Is it comparing compass heading readings, initial to a "warmed-up" reading that has drifted? The drone has not moved or even armed before the reading error message occurs. Is it implying a heading reading from the GPS warmup drift and comparing that to the compass heading? Unfortunately I simply don't have the experience with this system to have a full understanding, or perhaps even enough understanding, to understand whats going on to cause the "inconsistent" message that prohibits arming and taking off. I'm not even certain what question to ask at this point. Being unable to arm and fly is an unacceptable error, although we seem to be able to work around it so far. I will continue to examine and troubleshoot and post any other relevant observations I can come up with.

Thanks,

User096

teo58 · Feb 14, 2020

user096 said:
Two for two. Green cube Solo #2 exhibiting the same yaw error message not allowing arming. Seems that once the error drops to 20 degrees or less arming and flight is allowed. Seem to be able to arm and fly if I hit fly as soon as searching gps clears and hit fly message appears. Neither of these were loaded with the green cube specific firmware. May try that next and see what happens.

User96

Thanks for letting us know your Green Cube Solo problems. It appears we have similar issues which, brings me back, to square one. I was almost sure, my Green Cube has one or more hardware problems, which generate the prearm messages on the first boot. Maybe we should wait to hear from other Green Cube Solo pilots flying with Open Solo 4 & ArduCopter 4...

user096 · Feb 14, 2020

Just another observation at this point, Mission planner is notifying me arducopter 4.0.2 is available. Under bug fixes from 4.0.2-rc3 I see the line: "d) EKF only uses world magnetic tables if COMPASS_SCALE is set." Can't really say if it is relevant but it raised a flag for me anyway. Is it okay for me to load 4.0.2? Anyone??

Thanks,

User096

teo58 · Feb 15, 2020

If the ArduCopter firmware notified by MP is for Solo, it could be our solution, if not, we’ll have to wait a bit more. As I mentioned above, I flew my Solo several times successfully, without any prearm or any other issues, with Open Solo 4 and the old firmware for Green Cube, ArduCopter 3.5.4