Strange Issue with IMU Related to RC

[stbuebel]

Hey everyone,

I'm an undergrad student studying engineering, finally building my first racing drone, and have run into a bizarre issue:

When I connect my drone via USB to Betaflight, I can see the IMU output and everything works as expected. When I turn on my RC transmitter, again, everything works fine, and roll, pitch, yaw seem acceptable. If I turn off the RC and plug in the flight battery, everything is still fine. When I plug in the flight battery and turn on the RC transmitter, I can read the RC signals, but the IMU bugs, with strange drift in roll, pitch, and yaw. It doesn't seem like a typical mag interference problem as I have moved the FC away from high current wires and I have to conclude that it is caused by the combination of RC transmitter and flight battery.

Here are my drone's specs:
-Frame Anniversary Special Edition Martian 215 215mm Carbon Fiber RC Drone FPV Racing Frame Kit 136g
-Flight Controller Micro 20x20mm Betaflight Omnibus STM32F4 F4 Brushless Flight Controller Integrated with BEC OSD
-ESCs Racerstar RS20Ax4 20A 4 in 1 Blheli_S Opto ESC 2-4S Support Dshot150 Dshot300 for RC FPV Racing Drone
-Motors 4X Racerstar Racing Edition 2205 BR2205 2300KV 2-4S Brushless Motor Black For 210 X220 250 280 RC Drone
-Transmitter FlySky FSi6 (cheap one)
-Battery AHTECH Infinity 4S 14.8V 1500mAh 85C Graphene LiPo Battery XT60 Support 15C Boosting Charger

Has anyone heard of anything like this?

Thanks,
Spencer

 

[tevek]

Not sure if you calibrated your accelerometer in Betaflight to ensure your quad is correctly aligned and level but you can also trim a lot of your drifting issues with your controller. Should work with Flysky. Courtesy of Joshua Barswell:

 

[stbuebel]

Thanks for the reply. I have calibrated the accelerometer, and thought it might have been a trim issue. The drift is not your typical drift, (not steady, slow), it jumps every 1.5ish seconds, about 10 degrees. Again, this doesn't happen when the battery or RC is not connected, but only with both. Is it possible that I have a bad FC?

 

[tevek]

You may want to save your current FC profile (CLI dump and save in Betaflight), and flash new FW for your target FC. You’ll have to remap your receiver settings and check the other tabs as you will have lost your original settings. I have done that in the past when I’ve had stability issues. The other obvious thing to check is that your props are in good condition and tight.

 

[stbuebel]

I'll try it - sounds like a good call. Haven't really tried to fly much since the IMU readings are garbage and I don't love crashing so I'll worry about props next.

 

[stbuebel]

I'll try it - sounds like a good call. Haven't really tried to fly much since the IMU readings are garbage and I don't love crashing so I'll worry about props next.

Update - flashing firmware again did not fix it

 

[tevek]

Hmm
Have you run your motors up in the motor tab of Betaflight (props off). They should every spool up between 1000 min and 2000 max. Mid point throttle for your sticks should close to 1500 when on the receiver tab. There is a command in CLI to adjust all these if you are not able to do so with your Flysky Tx. Nowhere near my computer to show you a screen dump at the moment. There are also settings in the config tab that may be causing you issues such as loop time but again I’m at work and can’t oblige with a picture.
Check the props to make sure they are tight and spinning the same way (should be ok otherwise you would be all over the shop).
I’ll tag @HighTechPauper and @RENOV8R to see if they can drop by and help out. They are much more advanced at this type of troubleshooting than I am.

 

[stbuebel]

Hmm
Have you run your motors up in the motor tab of Betaflight (props off). They should every spool up between 1000 min and 2000 max. Mid point throttle for your sticks should close to 1500 when on the receiver tab. There is a command in CLI to adjust all these if you are not able to do so with your Flysky Tx. Nowhere near my computer to show you a screen dump at the moment. There are also settings in the config tab that may be causing you issues such as loop time but again I’m at work and can’t oblige with a picture.
Check the props to make sure they are tight and spinning the same way (should be ok otherwise you would be all over the shop).
I’ll tag @HighTechPauper and @RENOV8R to see if they can drop by and help out. They are much more advanced at this type of troubleshooting than I am.

Yah, motors work and I don't think it's a motor or prop issue (not even flying when the issue presents itself, just looking at the Betaflight tab which shows roll, pitch, yaw). I tried lowering the PID loop frequency thinking my FC might be overwhelmed with the addition of editing the OSD image and reading RC signals (since it only breaks down with battery connected, so video TX, and RC connected). This didn't help either. I would immediately assume the issue lies in the IMU or FC, but the fact that they work without RC connection makes me think this is unlikely and is rather some weird bug with the IBUS interface or with the way FC is being powered by the battery versus the USB connection.

 

[HighTechPauper]

Sorry folks, I have been away and am just now seeing this.
The description here doesn't really give enough information to take a guess as to what is going on. Can you please post a couple screenshots or a short video of what you believe is a good indication and what you believe is the bad indication? I assume you are looking in the Betaflight Configurator to see these indications? There seems to be some misunderstanding from what I am reading, these quads generally do not have "magnetic interference" issues since they do not have a compass.

Have you tried to hover test this quad?

 

[stbuebel]

Was just trying to get a video demonstrating the issue and then heard a popping noise and my FC will no longer power on... Seems like this was indeed an issue with the FC, but still not really sure what was wrong (guessing something related to the power supply because I was hot swapping battery power and USB, which likely isn't a good idea).

I had hover tested in the past, and it led to a very short flight and then the roll/pitch drifted to the point that I couldn't bring it back to level (basically just kept rolling right until it crashed, regardless of RC input).

 

[HighTechPauper]

From your description it sounds like the FC had a bad gyro, I have seen it a few times in my own quads, once when the FC was brand new out of box and the other after a crash. A new FC should fix the issue.