Cat5e as a tether - initial test results + leaks through tether


#1

Well as a few of you may know we have been extensively testing our OpenROV up at BC Robotics (#049)the past 2 weeks. We decided to go with a Cat5e tether on our build - i am not sure if anyone else has gone this route yet?

Our test tether is about 65' long , cat5e patch cable. One thing to note is that in the colder PNW waters the cable does get very stiff, and definitely affects the maneuverability of the ROV. The weight is also a concern as even a short run of this cable will weigh the ROV down. Floats attached to the line do help a lot!

Also another thing to note... We happened to have a small pinhole in our cable about 3' up the tether from the ROV .. on a dive to 15' depth and found a good 4 ounces of water in the tube after only 2 minutes!! The Cat5e will work like a pipe.

Standard wire will do this also if it is not solid core - be sure to allow solder to wick into the jacket and use heat shrink tube and liquid electrical tape!

If you have a leak in your ROV during testing and cant seem to find the source.. check the wires! And if you are going the Cat5e route - be prepared to fiddle with the flotation a lot!


#2

This is really valuable feedback. Thanks Chris! We should have the results of more tether tests up soon, too. I'm thinking we should create a dynamic document like the tether decision tree so we can collectively keep notes of our learnings. Thoughts?


#3

Inspired by a discussion with Tiziano, and Eric, I was doing some "dry" tests with a modified Cat5e cable, and I think what I found might be a good-enough solution for the first official OpenROV (and definitely the solution I'm going to adopt).

Instead of using a complete Ca5e cable, you rip it apart, remove the sheath, cut the 2 unneeded twisted pairs (100Base-TX runs on 2 pairs, not 4) and put some tape every now and then to keep the 2 pairs together and you are good to go. Maybe some float are needed to keep it from sinking the ROV too.

You don't have the problem of pinholes in the cable that leak water, you remove the rigidity given by the sheath and reduce weight a lot (50% cables and no sheath).

Ethernet runs fine of just 2 pairs of twisted cable (been running since a few hours using that cable). I have no ROV yet so someone have to test the hypothesis in real life :)

Simo


#4

i think the flexibility would be there - but at a sacrafice of durability - also POE becomes an issue

We have been looking at ways to have some sort of hybrid cable... thick and durable for the majority of the length with a short flexible section allowing free movement of the ROV.

The tether is deffinitly going to change over time!


#5

Personally speaking, PoE is a no go for where I intend to run the ROV (nowhere near a power supply) so 2 twisted pairs are enough :)

As with durability... a single twisted pair as the current solution has the same issue, and 100m of ethernet cable, if bought in a roll as the one used by cabling companies costs almost nothing: I found 300m of cable for 50USD, and with 300m you can make 6 100m long tethers


#6

Simone, very interesting. I like the idea. I wonder how it does in water. Testing in salt water has changed results in some situations, I’m eager to hear how this strategy does in that environment.