Controller Board Mod for 6 ESCs


#1

Here are some photos and details of a controller board that I just modified for Kevin K to support vectored-thrust on his ROV. As some of you know, he’s building an open-frame ROV (similar to those made by Outland), and has had a running forum post about it for some time:

Though he’s currently working with three thrusters, the design really begs to have a full complement of six thrusters for vectored-thrust operation. That is, two forward thrusters, two aft thrusters, and two vertical thrusters. The fore and aft thrusters are placed at an angle to allow sideways (strafing) motion. The two vertical thrusters enhance the lifting capability of the vehicle, and cancel out torques caused by using a single vertical thruster.

Kevin had the foresight to pot a bunch of extra wires into his endcaps to support future expansion, so…it was time to get to work on implementation.

This spring I experimented with an OpenROV modified into a 4-thruster “vertrans” configuration to allow it to move laterally. Here’s a thread showing the modifications to the controller board:

For Kevin’s ROV, I just expanded this concept to the fullest extent- mount an additional ESC over top of each existing one. Here’s what it looks like:

And a closer view of the ESCs themselves:

Just as earlier, the power leads to the new ESCs are soldered into the spare pads that are on the board (they facilitate the use of 20A ESCs if there’s a supply issue with the stock 12A ESCs). I drill out the spare ESC output pads to allow the output wires from the 3 new ESCs to go to the bottom of the board. 2 of the ESCs (6 wires total) are routed this way, and connect to the 6 spare leads on the DB-25 connector:

I avoided using the existing test pads TP20-TP25 since the traces attached to those pads are not designed for ESC currents.

As the DB-25 connector now has all its wires used, the third extra ESC, as well as any external lights or external servo channels, will need to use and additional connector.

Here is how I have the ESCs connected:

Existing servo connections:
D6 - rear port thruster
D7 - vertical thruster - with Y-cable to drive both vertical ESCs
D8 - rear starboard thruster

New servo connections:
D9- forward port thruster
D10 - forward starboard thruster

The D11 channel is still used for the camera tilt.

The ESC output pins are wired as follows:

DB-25 pins 20-22: port forward thruster
DB-25 pins 23-25: second vertical thruster
Free wires, to be added to an additional connector: starboard forward thruster

A couple of additional notes:

The AFRO12 ESCs can be purchased from HobbyKing, but note that they need to be reprogrammed to provide forward/reverse behavior (they come shipped working in forwards only). There is a script in the OpenROV software to do this (search the forums- the issue has come up before)- just plug the 3 new ESCs into D6,D7, and D8 temporarily, run the script, then plug all the ESCs back into their desired servo channels.

Note that this is not a long-term solution for big vectored-thrust vehicles. The total power-handling capability of the controller board has not changed, so in reality the power capability of each thruster has been cut in half. But what it will allow Kevin to do is to test the vectored-thrust software and mechanical layout, while working on a long-term solution for the power supply (probably using ESCs built-in to the Blue thrusters)

Hope this helps people with ideas for their own ROVs. Let us know if any of you do something similar!

-Walt


Two controller boards
Work Class OpenROV
#2

Big thanks to @Walt_Holm, @badevguru and @Eric_Stackpole for the help and support on this one!

Walt is exactly right, this is not a long term solution, but it is the path to larger vectored thrust ROV’s where the power is scaled up proportionally. The 6 vectored thrust combination is the epitome of ROV design and any further thruster additions are simply adding more power to a given axis. This 6 thruster combination allows for stability in the vertical axis with the counter-rotating thrusters, the ability to strafe (constant heading but with lateral motion), and better yaw control.


Seabotix vLBV300

Although we won’t be getting into advanced power systems for awhile, this is still an excellent opportunity to get the controls set-up and working. Power increases will be incremental after this.

Additionally, I have some questions of my own that will need to be answered on the hardware side:

  1. What is the optimal degree of offset for the horizontal thrusters?
  2. At what height on the vehicle should the horizontal thrusters be mounted? Close to the Center of Buoyancy (Top) or Close to the Center of Gravity (Bottom)?
  3. How much strafing ability will we get?

Using 4x 2.7 battery tubes, I should have enough power not to blow out the polyfuses, but the limits of the controller board are still a concern.

Here’s to not blowing up the board!


#3

Thanks for the write up Walt


#4

So I took out my ROV frame and began modifications for 6 thrusters. So I didn’t have to do a complete redesign I am basically making it taller to accommodate two more thrusters forward and below the E-tube. This is by no means final, but should allow for adequate and even testing. The aft thrusters are angled outwards at about 30 degrees right now.

I ran out of parts for the vertical thruster mounts, so back to Amazon!



#5

This is looking terrific! Can’t wait to see more. Do you have a BOM made up as you go?


#6

I’ll make up a Google Doc one of these days, but here is where I am at so far. Everything will be shown in the coming days as I finish the upgrade and install:

Frame:
2x Makerbeam Starter Kits. Yes, I used them both. You can’t order more of the smaller beams on Amazon yet.
6x 600mm Makerbeams. These are what make up the skids.
10x 300mm Makerbeams.
1x Pack of T-Brackets
2x Pack of Straight Brackets
1x Pack of 12mm square bolts. These are for mounting the thrusters.

Electronics:
1x Completed 2.7 E-Chassis. Took this one off my #1790.
1x Modified 2.7 controller board with 6 ESCs
2x Battery Tube Replacement Kit (for 2.7)
1x Replacement Cast Acrylic Main Tube
2x Endcap Replacement Kit (for 2.7). I made an extra thick Port endcap to pot additional servo wires.
2x Battery Tube Adapters for Li-FePO4 Batteries
2x Rechargeable Li-FePO4 Batteries (6x) - Continental US Shipping
1x OpenROV IMU/Compass/Depth 2.0 Module
6x 24" Servo Extensions
1x Logitech C930e Camera. New Primary Camera.
1x SEA CON MCIL2F Micro Wet-Con Inline Connector - Female 2 Sockets. This plugs into my existing tether connection and allows me to swap ROVs easily. Optional.

Thrusters:
6x BlueRobotics T-100 Thrusters

Float:
20x 8x4x.5" General Plastics R-3312 submersible foam
20x 1" M3 standoffs for mounting to frame
20x plastic M3 washers

Misc:
5x Sea to Summit Alloy Buckle 3/8" Accessory Straps


#7

Great stuff! I’m SO making one of these myself :slight_smile:
Off to Amazon :wink:


#8

Do you have to add the leads to the DB-25 for the 2.8 Controller Board?


#9

I’m not exactly sure what you’re asking, but if it’s just about the compatibility of the 2.8 controller board:

You can do the same modification for the 2.8 controller board, but you have to unhook some things first- the 2.8 controller comes from OpenROV pre-wired to put external lights and an external servo onto the auxiliary pins. So if you’re going to use those auxiliary pins for ESCs, you need to disconnect the external lights and external servo channel.

I’d get the 2.8 schematic off of our GitHub site for reference. You need to desolder R73 and R74 (located on Sheet 6 of the schematic) and R70, R71, R72 (located on Sheet 2 of the schematic). These resistors are under the pre-mounted ESCs, so you might have to partially unmount the ESCs to do this.

Hope this helps.

-W


#10

Kevin, I am working on my first ROV now and wondered how your 6 thruster unit worked out? Can you share your experience with the build and what you would do differently next time?


#11

Hi Jim, thanks for your interest, the finished vehicle can be found here: Work Class OpenROV

The ROV works great and was used in February to discover a previously unknown shipwreck: https://www.openexplorer.com/expedition/southerncalifornia/view/4115229

The end results are a bit mixed. On the positive side, the ROV is super stable and is an excellent camera platform. It also had like a 3 hour power endurance time and could carry at least a 5lb payload. On the negative, it was underpowered and was a bit slow compared to a BlueROV2. The strafe never worked properly at full thrust. The compass also acted up from time to time, so navigation was somewhat difficult.

What would I do differently? Honestly, I’m not sure, I managed to meet most of my objectives and there isn’t another 6 thruster OpenROV derivative out there yet. Walt and I basically pushed the ControllerBoard as far as it could go, probably further than it should be. It’s still a good vehicle, but it would need to be upgraded to used higher power voltages and current.