VESC Field Oriented Controller for T500

Hi all,

i’m currently testing a hi performances FOC controller with the T500. ( with pretty good result at this time … )

To get the exact rpm of the thruster ( hard to measure underwater … ) , i need to know the number of poles of the T500 motor.

Anybody knows ?

Thanks in advance ! :grinning:

@luc-rossi That’s awesome! Would love to hear more. The T500 has 14 magnetic poles, just like the T200.

-Rusty

Thanks Rusty !

It’s about a very exciting ROV project were i need precise control at low speed…

I’m using an ESC based on VESC project.
In addition to fine control at slow speed, there’s many benefits :

  • the current control loop allow limitation of torque, preventing damages to motor or ESC if stuck
  • you know in real time speed, current, power, …
  • ppm, serial and CAN interfaces
  • and many more ! :grinning:

I’ll probably post something at the end of my tests …

Hi Luc.

This sound very, very interesting. I have done some testing with these solutions also. But when we run it at full speed it only delivers 80 % of the force at the same current as the Bacis ESC, do you have any experience with that?

Which VESC are you using?

Best Sigurd

Hi Sigurd,
i’m using VESC 6 EDU with latest software.

Since our interest is in low speed and high efficiency, i didn’t tested the thruster at full speed. Anyway, i’ll be very surprised that it perform worse than the basic ESC !

I think it’s just a tuning problem. There’s MANY parameters in VESC tools that you can tune to optimise ( or affect ! ) thruster performances !

HI Luc.

Sounds reasonable, do you have a tuning profile for the T500 you can share? I am using it for an ROV which needs to clean, which means that I’m using a lot for power.

I have not tried with the newer firmware, I will do that tomorrow.

Sigurd.

Hi Sigurd,

I have not the hand on the VESC at this time, but i can share one of the setup i tested in august. ( probably not optimum for full speed )

Let me know if it’s better …
t500_test4_vtool4.xml.hex (10,2 Ko)

( i added the .hex extension to be able to upload it, but it’s an xml file )

Hi Luc,

Thanks, that seems to work well. But I can not go any higher than 89% input. I have tried to adjust the different parameters as current and duty cycle, but I can not find the limiting factor at the moment.

If you would like to share some more information, please reach out to sk@subcpartner.com.

Sigurd

Hi Sigurd,

I should play again with VESC in January, i’ll see if i can optimize the settings …

Hi Luc,

I would also love to get in contact with you. We are also trying to read real time data such as rpm, current, power consumption from our motors using a vesc based motor controller. We currently have the Pi+Navigator setup in our electronics enclosure on our ROV. Your knowledge on this topic would be greatly appreciated, please reach out to my email konstantin@remorarobotics.no.

Hi Konstantin,

what kind of informations are you looking for ?

Hi,

I’ve read up a bit on VESCs now and I’m interested. Mainly for the low speed performance but also the CAN bus diagnostic part.

Anyone set them up for CAN bus thruster speed control from the Navigator?

Any tips or ideas, I am using 6S batteries and T500s so I guess the 6 Mk VI is suitable.

Thanks

Hi @IntoTheBlue -

The Navigator doesn’t support any CAN bus protocols! A USB adapter could be used perhaps? Controlling the ESC via CAN from the autopilot would require quite a lot of development - maybe easiest to control via servo PWM signal like any other ESC, but log the telemetry data via a custom extension?

If using 6S with the T500, remember you need to limit the amount of time you run at full throttle!

You can use a serial port to control the first VESC, and chain all others in CAN. The first VESC of the chain is going to work as a serial/CAN bridge.

Hi @luc-rossi -

That’s a good point! However you’d still need to write some software to convert the autopilots outputs into whatever serial protocol is necessary to control all the ESCs… in a way that minimizes latency, so likely modification of the autopilot firmware directly?

Would you say that performance in low revs makes it worth trying he VESC out?

For a start they can be run simply with PWM…

Real speed ! :wink: