Homebuilt ROVs Forum

Hi all,

Haven't posted in a while, been too busy trying to make a decision on every aspect of this ROV!
Anyway, finally decided just to bite the bullet, get cracking on it, and deal with the issues later, rather than try to predict every problem that might happen.

Anyway, of course, I hooked up the system and have burnt out (pop noise and smell of smoke) three ESCs. Can anyone please tell me what might be going on. A few points to note:

-this is not the final wiring diagram, it represents where the project is at the moment, except: ignore ESC 3, it is not hooked up. And ESC 4 only supplies power to the Rx, it is not hooked up to a motor at this stage
-motors R and L are my forward/rear motors. One has a left hand prop, the other is a right. They are wired to spin in opposite directions at the same time (ie hooked to the same ESC)
- the system had been running fine on motor R and L, but burns up ESC1 and 2 when I introduce motor three (ie say i push forward and left into the top diagonal, which should activate the two forward motors and motor 3)
-the ESCs are rated for 110A forward and 40A in reverse. Bilge pumps supposedly draw 3A
-the multicoloured wires represent an 8 core cat 5 cable, with a bus bar (made out of copper wire) at each end
-all these components are housed on the ROV. the tether is not shown. The multicore wire connects to the motor wires at a junction box. The multicore then runs to the ROV enclosure
Any help much appreciated.

Regards,

mg

The common return (through the tether) may not be up to the task of sinking all the current ...... I'd suggest driving the motors initially without the tether in the loop.

Martin

If the ESCs have reverse, which I assume they do, I think you are creating an open circuit from one ESC from one to another if one is in "forward" and one in "reverse". If one ESC is in reverse, it is actually sending positive current down the negative wire (blue in picture). That goes to the negative bus and into the negative (blue wire) of a forward ESC. There is negligble resistance along this and the current spikes to melting levels and poof!

If they are forward only, then ignore the above.

To correct this, each ESC needs a unique pair in the tether. The other option is to have the ESCs on the ROV and directly connect each motor. They could share a common connection to the 12V topside battery. You would need to bring the controller signal down the to ROV, but that requires much smaller wires.

Hope this helps,
Greg

great, thanks for all the tips...I should point out, all these components are on the ROV itself (ie none topside)...I have not shown the tether, as it is not relevant to my problem.

To explain, the motor wires run into a junction box, they connect to the multicore inside the junction box, then the multicore cable leaves the junction box to enter the ROV enclosure.

autonomous ROV, that may indeed be the problem. So you're saying, running the forward/back motors through the one ESC should be fine, but as soon as the third motor is engaged , in an opposite direction to the other esc, there is a problem...Sounds feasible

you said "The other option is to have the ESCs on the ROV and directly connect each motor" The ESCs are on the ROV, however to minimise cable (and complexities with potting multiple wires), I wanted to use common earths-looks like that may not be possible...

Any other ideas before I go off to make radical design changes?

Cheers

mg

If this is all on the ROV side, I think the bus bar is probally more complicated than not having one. Also, I cannot fathom the advantage of combining the two motors on one ESC. It cannot spin the ROV in the yaw direction.

My suggestion: For each ESC, pick a motor and have dedicated wires to it. Coupling wires/motors may seem advantagous, but it has many inherent issues and makes tracking down issues very difficult.

Good luck!

BAD idea with the wiring setup. I know this from RC Quad Copter etc.. never paralell any leg of ESC outputs. the inputs are fine and dont care. One ESC per motor keep everything isolated and seperated.

ok, sounds like I stuffed up! Thanks for all the help...

autonomous ROV and bikerbones, I put the two motors on the one esc so they would be controlled on the one channel. I don't want them for direction, I just wanted two motors for forward/reverse (ie extra thrust). I have a very basic RC tx; do you know a way to wire two motors to separate escs, but be controlled by one channel (ie if I push the stick forward it will activate both motors)?? Or is this a computer controlled RC mixing type situation?

Again, thanks for all the help, maybe the ROV is back on track!

Cheers

mg

NP just remember always use one ESC per motor and never parallel output wires of an ESC. You can parallel power feeds or inputs till the cows come home. On the controller wires from one channel I use a servo y cable (splitter) to drive two ESC's from one channel. ie two verticle motors. This is so they both run at the same time in the same direction with one stick movement.

just reread your post, I will draw a quick and dirty schematic. give me a few minutes.

Here it is. Hope it helps.
***NOTE the "DC Power In" (Power Inputs on the ESC's) can share common positive leads and common negitive leads. Tie all the positives and negitives together if you like and it will be fine. In this example lets use channel 2. And for all purposes its the rudder channel. This way the control stick movement to the left will operate both the ESC's and motors clockwise at the same time and the stick control to the right will operate both the ESC's and motors counter clockwise.