Roboclaw 2*15A Motor Controler advice

Hello,
I am currently in the process of constructing a BallBot. The project requires a fairly precise control of brushed DC Motors. However I am not sure wether the robowclaw suits my project or even worse I am not sure about the functionallity itself. So hopefully you can give me some advice.

1:
The data sheet of the controller states:
Minimum operating voltage: 6 V
Maximum operating voltage: 34 V
which means that the supply voltage can be cosen in a range from 6V-34V. Now I assume that the supply voltage is also constrained by my motor choice.
In case I want to use one of Polulus 12V brushed motors (Stall Torque 5A), the maximum supply voltage for the controller board should be chosen 12 V and if I choose a higher supply voltage, I might harm the motors but not the controller board. Am I right here ?

2:
The data sheet also states following control option:

[quote]
Drive Motor With Signed Duty Cycle
Drive Motor using a duty cycle value. The duty cycle is used to control the speed of the motor
without a quadrature encoder. The command syntax:
Send: [Address, CMD, Duty(2 Bytes), Checksum]
The duty value is signed and the range is ±1500[/quote]

Here my question: Assuming I choose the duty value 1500: Does this mean that the duty-cycle is 100% or in other words the motor will be powered at my chosen board-supply voltage with a 100% active period. Would this be equal to directly plugging the motor into a 12V battery?

Thank you very much in advance !
Florian

Hello.

In general, the maximum operating voltage of the driver is the highest voltage you can safely supply to the driver. The voltage rating of a motor is the recommended voltage for that motor. Most motors can be run at higher than the rated voltage, which will increase its performance (speed and torque), however doing so could drastically decrease the life of the motor. If the motors you are using are rated for 12V, you can safely use a supply voltage of 12V with a motor driver rated for much more than that (such as the 34V maximum for the RoboClaw).

As for your question about the duty cycle, at 100% duty cycle, the motors will be receiving a constant voltage that is equal to the supply voltage you used. In this case, the motors should respond as if they are being connected directly to the 12V source.

By the way, a ballbot sounds like a neat project, but is really ambitious as well. Based on the questions you are asking, it sounds like you might not have a lot of experience with these kinds of systems. I would suggest breaking the project up into smaller parts and building toward the larger project. I am not sure how you plan on controlling the motors; if I knew more about your system and what you want to do, I might be able to tell you if the RoboClaw seems appropriate for your application.

-Brandon

Hi Brandon and thanks for your reply,

at the moment Im in the step of designing the actual robot.
Control approach:
In order to control the the robot on the ball, we need to balance out desired angles. The desired angles (roll, pitch yaw) are measured using a raspberry pi and your MinIMU v2. At the moment I am using a complementary filter, as soon as I have my system set and hence can model the overall system, I will switch to a Kalman filter approach. For the actual control I will probably try different approaches. One could try to have a PID controller for each vertical plane in a inner control loop and a third PID controller to keep a stationary position.Another option is optimal LQR control, as were operating in a small range around the unstable equilibrium. In order to verify the controllers, I have modelled the mechanical design in Solid-Works and exported it as a Matlab SimMechanics model (please let me know if its not ok to use brand names here). The controllers work just fine in Simulink although I am of course only assuming a “literature” transfer-function for the speed control of the motor as well as a noise affected model for the IMU. After searching around a lot I fell over the roboclaw and thought it could be very useful. However I am fairly stuck at the moment. The overall system dynamics very much depend on the motor choice and the controller board choice, as it will also determine my power supply and thereby important things like the COM of my ballbot. Hence unless I’ve finally chosen and verified my actuator setup, I cannot continue designing things like the Kalman Filter or finding the LQR gains.
If you can recommend a different kind of controller or if you would suggest a different approach, please let me know.
Regards Florian

Hello.

It sounds like you are planning on doing some PID speed control. The RoboClaw might work for your application, but it is hard for me to say for sure without knowing all of your requirements. If you have more detailed questions about it, you might consider contacting the manufacturer, Orion Robotics.

-Brandon