Lost Heartbeat on System

I am receiving an error code “MYGCS: 255, Heartbeat lost”
This has happened a few times now.
Can anybody tell me what this means and how to fix it?

Thanks

Hello,

This means that the autopilot on the ROV has not received a HEARTBEAT message from QGC for 2.5seconds. QGC emits heartbeats at 1Hz.

If you see this message it indicates there was some transient failure in communication between the autopilot and QGC.

You should make sure that the tether conductors are well seated in the interface boards, and try a different power supply for the topside board.

Thanks Jacob. I updated Pixhawk and it seems to have fixed the problem.

Hi Jacob, hi Blueys,

I have seen the same old error code a lot lately, and checking the topside connection and trying different power sources has not helped. Is there anything else I can try, before I open my ROV (which I have not done in years) to check in there? Or can I change the heartbeat check to be less sensitive? I do not notice any problems in the controlling of my ROV, it seems to reconnect seamlessly. However, it disarms automatically, which is annoying.

However, the good news is that the switch from MacOS 10 to 11 seems to be unproblematic, at least in the office. Any idea if GroundControl will run on a M1 MacBook as well?

Best greetings from home office in Northern Germany,
stay healthy,
Kai

Hi @KaiH ,

If you are really seeing many “MYGCS: 255, Heartbeat lost” messages, I would guess there’s something wrong with your bandwidth.

What is your setup like? Which versions of QGC, ArduSub, and Companion? Any non-stock components?

Hi Willian,

and thanks for the quick response. No non stock components, I am using (or rather non-using, as my ROV has not seen any work in more than a year) the same basic set-up I started with in December 2017. I faithfully upgraded the software, now it is QGC 4.0.5 with the fitting upgrades to ArduSub etc. (cannot check right now, as my son ist still homeschooling on the laptop). My computer is an older MacBook Pro running the current MacOS 11 Big Sur.

I hope the fix is in some (to me) obscure communication settings. Hardwarewise it could be the top-side companion, which is in a rather crude box, but I did check the flimsy wires. Maybe the usb to ethernet adapter? Or is it likely something could be wrong with the tether itself? Also, I noticed that the time lag seems to be getting worse, but still acceptable.

Cheers,
Kai

Well. If your video is smooth, that probably means that the bandwidth is fine.
When you get back to your laptop, check what is in http://192.168.2.2/mavproxy, as recently we had mavproxy consuming too much CPU and we dialed down the telemetry rates to take care of it.
If you don’t have any unusal configuration, restoring the mavproxy settings in this page coud help.

I missed this question. I heard one user report that it worked very well under Rosetta and had no compatibility issues. Hopefully we get some more reports soon.

On it.
Seems I wasn’t too faithfully updating:

Companion software is 0.0.18 - updating now - got 0.0.25
ArduSub 3.5.4 - updating now (stable) - 4.0.2

The heartbeat lost messages stopped

Found the mavproxy page at http://192.168.2.2:2770/mavproxy and restored the settings

Now I get „critical: Lost manual control“ every 30 seconds and the ROV disarms
And the video is not smooth at all.

Do you think it’s the physical connection after all?

Experienced the same. We had power issue.
I hope Bluerobotics rethink the way they are powering the camera and all other boards.

Hi BR team,

One of my client is getting the same problem right now during a mission abroad. They checked the power supply, reset the mavproxy settings, all versions are up to date.
They have a heavy BR2 with a 200m slim tether and a GETAC S410 surface computer.

I am out of ideas, can you help?

Hi!
Next tip, I would check bandwith from Raspberry to topside/laptop.
This can be done from http://192.168.2.2:2770/ in old companion system
Bo

I would also try disconnecting the camera from the raspberry.
And look at the voltage value measured by the pixhawk. Hw_state I think. Is it 4.6v or a nice 5v?

Hi @Stevenlb,

On top of @Boko and @Charles’ suggestions, some of our troubleshooting steps may be helpful. If this is still an ongoing issue though it’s likely worth contacting our support email (support@bluerobotics.com) to explain the issue and what’s been tried so far.