Implosion of my OpenROV 2.7


I have been asked by Open ROV to share my experience with my new OpenROV 2.7.

I assembled my ROV and vacuum tested it, -0.5bar for 30 min – All OK. After this I ran it in my bathtub. All worked well and as expected. I was unable to do more tests on functionality since I´m waiting for the battery chargers to arrive.

After seeing this video - and especially this one:, I felt it safe to try my ROV to 90m.

I decided to do a real life pressure test at sea, so I removed all electronics and batteries. I had 3x30m of line for this test. My dive computer and a led weight was attached to the hull. The led weights were needed to make the ROV slightly negative buoyant.

First test was to 30m. All well and no leaks. Second test was the full length 90m. I was very surprised when I pulled it back to see that is had imploded completely both main tube and both battery tubes. The implosion of the main tube was so powerful that it ripped apart the whole structure. Absolutely all acrylic part except from the battery end-caps with the O-rings was broken, including the ROV hull?!

I have taken some pictures of this. When reviewing the log on the computer I could see that the decent from the surface down to approx… 60m were slow and fine, but after that the speed increased rapidly. This indicates to me that the implosions happened at 60m.

The battery tubes weren’t actually imploded. They where squeezed flat and ruptured along the squeeze line.

I contacted OpenROV support, and they quickly replied –
“Sounds incredible! I’m guessing that the implosion of the battery tubes occurred around 60m which made the ROV sink faster to implode the main E-Tube. With batteries inside the battery tube, it’ll prevent the battery tubes from imploding at that depth and should be stable beyond the E-Tubes failure point. We would love to see and hear what you found out through this test on our forums. That being said, we can send you the replacement parts that you need in exchange for a forum post of your test! It’s always helpful to hear these things.”

Excellent support, thanks!

In the after math of this one could say that I should know that the battery tubes needed some support inside before this test could be done, but after seeing the last video I thought I was safe. In that test there is no extra support in the battery tubes. Is the video showing an earlier design with thicker walls in the battery tubes?




Thanks for sharing the results of your testing Skambest.

It’s an unfortunate thing to lose your rig but, seems many lessons were learned and we all get to benefit from your experience. Such is the beauty of this open community.

I’d love to see the pictures of the carnage if you get a chance to post them.

Your 2.7 rev 2 will be better than the first :smile:




This is a lesson learned in early camera housings, canister light and underwater DPV builds in the scuba diving world, and i imagine submersibles prior to that (meaning as much as I wish we were Marco Polo, we are not reinventing any wheels here). Bulkheads are key for structural support. Even the electronics tube will garner some benefit from the plexiglass structure of the chassis inside if/when its starting to compress. The early cave diving canister lights were made from PVC and acrylic, not so dissimilar from what we are doing here and needed the batteries (or proper size) in place so they would not leak or implode. Current ones are Delrin and have bulkheads built into the battery/battery housing. The same with underwater scooters of certain designs (when they started using large PVC tubes) needed a battery carriage that functioned also as an internal bulkhead structure so that the cylindrical hull would not deform enough to loose seal and flood.


In the second video, there were supports inside the battery tubes, but they were clear tubes so you can’t see them easily (or perhaps at all). I was expecting them to fail at some point since the test was very deep, so I didn’t want to put my batteries in. Sorry that I didn’t make that clear.The second camera died, or you would have been able to see the regular unsupported battery tubes bunch up in their elastic band, like a handful of pick-up sticks.

I have pondered why the endcaps did not fail until so deep (in a previous test, Endcaps, of a similar thickness but different layout, popped at around 100m). In this test I had o-rings sandwiched between the endcap and the rim of the tube. I Still don’t know, but I suspect that the o-ring may have allowed the endcap to flex and slide where a hard edge might have acted the a sharp cutting edge. Immediately before the cap finally fails, you can see the o-ring slip out around the cap. I think it could have gone further if the o-ring was held in place. Though I doubt much further. :smiley: