Limit on single recording

Is there a limit on single recording session other than the limit of the USB stick? We were doing a long job today and after about 3 hours or so of hot-swapping batteries and leaving the recording running it stopped at some point during the flight and had a “recording service is not available” error. Is there some limit where every x number of minutes you should stop and start a new recording?

I’ve done a 3.45hour single session before but would not recommend it. The file sizes are enormous and very unwieldy to work with later on. Our SOP is to now break it up into 2 hour sessions to make sure we dont have issues.

But the 3.45hour session did give us data that we could work with without errors in logging or PCMaster/Painter

After getting the data to the office - it was 4:12 before it stopped. No service.log created obviously. It was powered on, but not taking pictures and unable to bring up the status page. WiFi connected but no webservice running.

Copied data off, formatted usb, started new mission fine and after 25 min flight when landed, same thing. Lights on, can connect but no web interface.

Restarted drone and Lidar again, started fine and flew the last mission without issue.

3rd time data stopped recording on this unit and we had a previous unit that was replaced where It happened once on it.

Pretty frustrating as you have to refly the whole mission since you don’t know where it stopped.

Yes, we’ve had that happen too some time back. I honestly can’t remember what the solution was then, but I think it was something to do with the USB drive formatting not being done on the unit (as opposed to on a PC). But I’m not saying that that is the solution…

I agree with you that there is nothing more frustrating than the drone returns to land and you don’t hear the camera clicking away!

Today on power up the camera was not showing at all. Had to power off and back on before the camera showed as present in the system.

@iaerobotics how hot was it outside when you were doing your scan? Could overheating be a culprit?

Hey Alex. It was actually not bad. We were at 4500’ so it was in the 70’s. We had the same issue happen on the very next flight after a reboot as well. I zipped and sent matt a link on the rock cloud of the raw data so he can send off to have looked at. He also mentioned a good rule of thumb is 1hr (2 flights on m300) per project recording outside of what might be going on with our unit. It took about an hour and 1/2 to fully process the 4hr dataset where an hour dataset is about 15 minutes so in the end it will speed up the overall process and limit our exposure to lost data when these failures re-occur.

To answer your specific question @iaerobotics, there is no specific limit to recording time. We have bench tested very long recording time (24 hours). We have also had customers routinely scan 1-3 hours at a time. We personally have scanned 3+ hours.

But, your point is very valid. Processing time can be substantially longer the larger your single dataset is. (Processing runs into processing computer hardware limitations and slows down).

All that being said, we are looking into your specific dataset.

Another quick question, how were you formatting the usb?

We format through the R2A web interface.

Sorry, one more question, what version of the firmware are you on?

Dont have it in front of me, but the x.6.0