varying motor torque

s0071904 · ‎2010-08-25

Posted on August 25, 2010 at 17:45

#h-bridge

bille.stello · ‎2011-05-17

Posted on May 17, 2011 at 10:20

Hello Andrew,

I believe you're talking about brushed DC motors, is that right?

From a control point of view, the brushed DC motor torque is proportional to the current flowing into the rotor winding. All what you need is a closed loop system able to adjust the PWM duty cycle applied in order to make the real current follow a target current (a current feedback trough a shunt resistor is required). Of course, the target current (torque) will have to be the output of another PI controller processing the real grasped slip object (output of your sensor) and the target slip (0). For sure this can be implemented on MCUs both on STM8 or STM32 (it depends whether your application needs to do other tasks then motor control).

Regards,

Billino

oinklett · ‎2011-05-17

Posted on May 17, 2011 at 10:20

Solution to H-Bridge Question.

Varying the PWM of a MOSFET H-Bridge to power your robot gripper will control the DC Motor torque as you will be controlling the average current to the motor. I strongly suggest you use an H-Bridge Driver Controller like the ST L9903 or Intersil HIP4081A. They have been around for decades, are extremely rugged and predictable and will more than meet your requirements. I also strongly suggest you use an Atmel AVR ATmega88 or ST 8/9 microprocessor. I have used both in military avionics systems because of their rugged and predictable behavior, execution. Also the Atmel AVR Studio 4 and ST development systems are solid develope friendly mature environments to develop and debug your code as you will need to write simple drivers in Assembly, note very easy to learn with reduced instruction set, and very easy to compile and debug. These development systems are free from Atmel and ST. They beat anything from a third party that you will have to purchase.

First you need a Benchmark system that you can learn and validate your generic system design to. You can verify the voltages and logic are correct on your system verses the benchmark system that already works with a known functional behavior. You can then program your microprocessor and debug it, incrementally.

I found a website [url]www.edemoboards.com[/url] that has a complete Demo/Evaluation Board for a complete H-Bridge system with 0.001 Ohm MOSFETs, H-bridge controller-ST L9903 or HIP4081A, with/without an Atmel AVR Atmega88 microcontroller, current sense amplifier, and all of the circuit protection components, at a great price. One of their H-Bridge Demo Boards will be your benchmark system. The demo boards that have the Atmel Atmega88, you can connect an optical or Hall magnetic sensor servo to the demo board for full motor feedback and control. Check it out as it will greatly accelerate your learning and troubleshooting effort. Otherwise you will be bungling around just to get an H-Bridge to work to control torque.