My ROV team has run into some trouble. We have just taken eight t200 thrusters with blue esc and stripped and soldered them to cables that will connect onto our ROV. The trouble we are having is that three of our thrusters are not be working any more. Here are some details that might be helpful.
We tested each motor with 16V using and arduino uno with i2c to control them. We soldered each of the thrusters onto cables first and then tested them. They were tested with the example arduino library written for the motors. We did not test the motors before we connected them to cables, but we assumed that each motor would work because we had not taken them out of the boxes before we tested them.
The odd thing is that some of the motors worked for a little bit and then later they suddenly stopped working. Whenever we power up the motors, the leds built into the blue esc still work and remain constantly on even though we cannot control the thrusters. This is leading me to believe that there might be an issue with the i2c lines that is causing the esc to not respond.
That was a mouthful so I will leave it at that for now. If anyone has any advice or sugestions for how I could potentially fix the blue esc’s, I would love to hear them.
I2C is still not working, but I was able to reprogram the blue esc’s. This means that the blue esc’s are working and that the pwm line and ground are not broken.
I would guess you’re right and the I2C communication is not making it through properly. The constant blue light means that it is powered on but not yet initialized.
What are you using to control the thrusters? If you have a lot of thrusters on the bus or you have long wire lengths it may be helpful to slow down the I2C bus rate to improve communication reliability.
When testing the thrusters, I am only controlling one at a time. Also, I am using a teensy 3.2 with no internal pullups. The only pullup resistors that I am using are the ones inside the bluesc.
Actually, I am on the Purdue ROV team. Bad news, we were not able to fix the i2c lines. We think the issue was that the blue esc experienced some kind of electrical back voltage or other disturbance that broke the i2c lines. A good way to test this is to put a multi-meter up to one of the i2c lines. It should read off 5 v because of the internal pull-up resistors connected to these lines. Ours were reading off much smaller voltages, like 3.5 V and 4.1 V, which is not enough to count as a logical one for i2c.
Interesting. How long are your I2C wires from controller to the ESC? Did you ever try external pullups?
I would also suggest looking at the I2C lines with an oscilloscope. The motor itself can introduce quite a bit of noise into the system and with long wire lengths that’s an issue.
We had 8 blue escs connected to one I2C line. We used a 3.3V teensy 3.2, which we turned off the internal pull-up resistors to and just used the pull-up resistors inside the blue escs. Each cable to the escs were only about foot in length.