ROV shutting down/crashing mid-dive: 2.5 with beta software


#1

I'm not sure if this is a software or hardware problem, but my OpenROV 2.5 with the beta software stops working after being in use for 40-50 minutes.

I have left the ROV on for many hours, so I know it only happens while it's in the water and being used. Both times that I tested it in the pool, around 40 minutes in, the ROV just completely stopped working and the cockpit stopped updating any information. The second time, I remember the battery voltage was around 11.5 and when I "rebooted" the ROV by disconnecting and reconnecting the USB cable, only the "power" led on the control board lit up.

I can get the ROV to boot up properly again simply by opening the battery tubes, closing them again, and then reconnecting the USB cable to the laptop. This makes it seem that as long as any power is connected to the ROV, the un-bootable state is retained.

Could it be that the battery voltage dropped too low for a second and caused one of the boards to automatically shut off? Maybe some hardware started to overheat? Any other thoughts/suggestions?


#2

I forgot to mention that when I tried to SSH in after "rebooting" the ROV, terminal returned "ssh: connect to host 192.168.254.1 port 22: Host is down".


#3

Hi Franton:

I would guess that at some point in your driving, you are tripping the overcurrent protection in the batteries.

First I would check to see that there are no wiring shorts in your motors. See this blog post.

Then, I would pay careful attention to the battery current telemetry shown on the right hand side of the cockpit screen, mixed in with all the other telemetry. Offhand, I believe the two battery currents are BATI1 and BATI2 or something like that. After you do a fresh reset of the ROV (and the battery packs), the two current numbers should be roughly similar, indicating that the ROV is pulling out of both battery packs equally.

If, at some point in time while you are driving, one of the battery currents goes to zero, then that means you tripped one of the protection circuits in the lithium battery cells. At that point you are running on one pack only, and need to drive gingerly to avoid the second battery pack shutting off as well.

Once the protection circuit for a battery pack has tripped, the only way to reset it is to open up the battery tube, completely disconnecting the cells from any current draw.

If you find that this is what is happening with your ROV, try to drive it a bit gentler, or otherwise figure out what is happening to cause the first battery pack to trip its protection circuit.

We are experimenting in the lab here with other battery choices that don't have this issue with nuisance tripping, and are close to making a new recommendation. Stay tuned.

-W


#4

I have exactly this same problem with my ROV version 2.6. Can someone confirm that the batteries were actually the problem? I was wondering if there was some other protection circuit on the controller board.


#5

Hi Jeff:

Read the last couple of responses on this forum thread.

Pretty much all the cutout problems were caused by the combination of the trustfire batteries, and the way the software runs the motors for yawing motions. If you have the Trustfire batteries and don't want to buy new ones, then it's critical that you apply this patch to your software. That should minimize the problem, although in our experience the only way to totally get rid of the nuisance trips on the Trustfire batteries is to switch to the LiFePO4 batteries.

Let us know if that works for you.

BTW, the controller board has no over-current protection built into it for the motor circuitry.

-Walt


#6

Thanks Walt. I also saw this post (http://openrov.com/forum/topics/how-to-remove-protection-circuit-from-trustfire-batteries) and followed those instructions. I suspect everything will work fine now... just need to figure out how to use a poly fuse...


#7

I've got to say, I'm not a fan of removing the protection circuit. It's there for a reason.

If you pull out the specs for LiMnNi battery cells, you'll see that they're not rated for as high a current draw as are the LiFePO4 cells that we are now using. If you pull off the protection circuit and go driving around at throttle setting 5, it's anybody's guess what's going to happen. Maybe nothing at all, but maybe something bad....

-W


#8

Hey Franton,

Did the suggestions below help?