IMU says "moving" while standing still after two calibrations

[modbuilder]

Have a recently acquired P4A with a IMU issue. PIcked it up from another P107 pilot thinning the backup equipment and am pretty sure he accurately described it a "low hours". Before discovering the IMU issue, I had replaced the the GL300C RC with a GL300F from a downed P4P with no problem to get increased range. IMU said it needed calibrating, so I did. After calibration, it shows "moving" when I check the health of sensors even while sitting static on a table, and when I power it up static in the driveway with no props, the location shown in Litchi slowly drifts even with a fairly steady GPS location. Did all the IMU re-calibration again as well as compass, and IMU still says it's drifting while it's static on my desktop. Is this likely to respond favorably if I keep trying to re-calibrate it, or have I got a repair issue?

Thanks for your thoughts.......... Bob R.

Didn't expect it to make any difference, but just to make sure, I re-linked the original GL300C RC, and the problem was the same with both controllers. I'll recalibrate the sticks just to insure there's not a stick input being received at the bird, but wouldn't expect a stick input to be considered in an IMU output.

 

[BudWalker]

Have a recently acquired P4A with a IMU issue. PIcked it up from another P107 pilot thinning the backup equipment and am pretty sure he accurately described it a "low hours". Before discovering the IMU issue, I had replaced the the GL300C RC with a GL300F from a downed P4P with no problem to get increased range. IMU said it needed calibrating, so I did. After calibration, it shows "moving" when I check the health of sensors even while sitting static on a table, and when I power it up static in the driveway with no props, the location shown in Litchi slowly drifts even with a fairly steady GPS location. Did all the IMU re-calibration again as well as compass, and IMU still says it's drifting while it's static on my desktop. Is this likely to respond favorably if I keep trying to re-calibrate it, or have I got a repair issue?

Thanks for your thoughts.......... Bob R.

Didn't expect it to make any difference, but just to make sure, I re-linked the original GL300C RC, and the problem was the same with both controllers. I'll recalibrate the sticks just to insure there's not a stick input being received at the bird, but wouldn't expect a stick input to be considered in an IMU output.

Could you be a bit more specific? Are you looking at the accelerometer data?

 

[Capt KO]

Have you flown it yet? Does it truly drift or wonky flight airborne? I fly many P4 models but haven't seen the problem..yet

 

[modbuilder]

I was looking at the screen that shows IMU health and provides the software button to start IMU calibration. The IMU health usually shows the word "bias" with a number where smaller is better. But instead of 'bias', mine said "moving".

I never got a chance to check whether the bird actually would hold station or drift, because I got the IMU to accept calibration. In browsing the internet, I came across a guy who insisted that the IMU needed to be calibrated "cold". He preferred that the bird actually be cold when IMU calibration is initiated, which is easy to achieve in the winter. But he insisted that at least the calibration should not be started when the bird had been in use and was at operating temperature, which I had been doing. I couldn't see how that could make any difference at all, but I tried it. The calibration went much slower than it had previously, and after it went to completion, the IMU showed health status in the green with "bias' and a low number. The drone's location in Litchi no longer showed the bird drifting while sitting static on the ground. After also re-calibrating the compass, I took it out to the local AMA field and put it through its paces. Flew great! Booted up quickly with no error messages, held hover beautifully, and did a perfect RTH.

I'll be the first to say that the good outcome of cold IMU calibration is probably random reinforcement. But I'm reporting the whole thing back to the forum. Call it what you want. You can use it or lose it. Up to you.

Thanks All............ Bob R.

 

[modbuilder]

Capt KO -- I think this bird had legitimately been not used in a long time, and I hear that the IMU can suffer from not being used. This IMU acted like it was stuck with some fixed output and not tracking movement.

 

[modbuilder]

Bud Walker -- Yeah, that was the accelerometer

 

[BudWalker]

I was looking at the screen that shows IMU health and provides the software button to start IMU calibration. The IMU health usually shows the word "bias" with a number where smaller is better. But instead of 'bias', mine said "moving".

I never got a chance to check whether the bird actually would hold station or drift, because I got the IMU to accept calibration. In browsing the internet, I came across a guy who insisted that the IMU needed to be calibrated "cold". He preferred that the bird actually be cold when IMU calibration is initiated, which is easy to achieve in the winter. But he insisted that at least the calibration should not be started when the bird had been in use and was at operating temperature, which I had been doing. I couldn't see how that could make any difference at all, but I tried it. The calibration went much slower than it had previously, and after it went to completion, the IMU showed health status in the green with "bias' and a low number. The drone's location in Litchi no longer showed the bird drifting while sitting static on the ground. After also re-calibrating the compass, I took it out to the local AMA field and put it through its paces. Flew great! Booted up quickly with no error messages, held hover beautifully, and did a perfect RTH.

I'll be the first to say that the good outcome of cold IMU calibration is probably random reinforcement. But I'm reporting the whole thing back to the forum. Call it what you want. You can use it or lose it. Up to you.

Thanks All............ Bob R.

Makes perfect sense. The accelerometer sensors will always have some bias which translates to some movement even though the P4 is at rest. If that bias is excessive then the P4 uses the "moving" label instead of the actual bias.

The IMU has a heater that brings the temperature to, and then maintains 65° C. The IMU needs to be at or above the temperature when it was calibrated. Calibrating at a cold temperature will reduce the amount of time required to bring the IMU to that calibration temperature. Is it possible that the "moving" indication was occurring before the IMU had reached the calibration temperature?