Cogging Torque & Anticogging

Torque is the tendency of a force to rotate an object around a pivot point. Certain electric motors experience something called cogging torque (also known as detent torque), which results from the interaction between the permanent magnets in the rotor and ferromagnetic materials in the stator. Cogging torque ripple results in motor vibration that can be detrimental to the performance of a machine or vehicle. You can feel “cogs” by turning these motors with your hand. This jerky sensation occurs when the permanent magnets snap magnetically from tooth to tooth as the rotor spins around the stator. Cogging torque is always present, but it can be cancelled by applying the appropriate motor currents, a process called Anticogging.

Permanent magnet synchronous motors (PMSMs) are used in a wide range of applications. High power devices, like drones, use brushless DC motors, while position control devices, like 3D printers, use stepper motors. These motors are power dense, torque dense, and/or efficient, but they suffer from torque ripple, a variation in torque output that can partially be attributed to cogging. Cogging torque is a function of position and its frequency per revolution depends on the motor’s layout. In other words, a motor experiences different levels of cogging torque depending on its position. Cogging torque adversely affects PMSMs at all speeds. However, it is more noticeable at lower speeds. So, open-loop brushless DC motors have trouble with tasks that require precise position control (motors move slowly for this purpose). Instead, these tasks are handled by stepper motors, which operate in discrete steps and move/hold their position without any feedback sensor. While stepper motors are more precise, they have the tendency to get lost, what’s known as “skipping a step.” For example, if you bump into a 3D printer causing the motor to shift, it will continue to print from that offset point.

Ideally, we want a motor to spin as smoothly as possible, have an awareness of its position, and not compromise torque and efficiency. In hardware, the most effective way to reduce cogging torque is skewing the rotor magnets or stator slots. This design reduces cogging torque ripple but complicates the production process and increases the final price of the product. Also, the skewed design reduces the overall torque output of the motor, so the smoothness achieved from a skewed design comes at the cost of max torque and efficiency.

Using Vertiq’s Anticogging algorithm, it is possible to maintain the typical straight stack, straight magnet design that optimizes torque and efficiency, while also cancelling out the vibration in software. The above graph showcases the result of an Anticogging test run on a $21 motor using a force-torque sensor. The red waves indicate the cogging torque ripple RMS (vibration) inherent in the motor, while the blue waves indicate the Anticogged RMS. Across a voltage range from 0.52V to 2.0V, Anticogging is able to reduce vibration by about 90%. As a result, this inexpensive BLDC motor is now smoother, quieter, and more precise than motors that are 10x more expensive. Another interesting advantage to Anticogging is that it significantly reduces the motor's minimum speed, which can be important for high precision applications. On this motor, Anticogging reduced the motor's minimum speed by 83%, as shown in the top (normal) and bottom (Anticogged) graphs above.

Compared to stepper motors, Anticogged brushless DC motors have superior position control, positional awareness, and smoother motion. Furthermore, Anticogging technology allows cheap brushless DC motors to operate as smoothly as high end brushless DC motors. The Vertiq Anticogging algorithm works by mapping the cogging torque as the motor moves from cog to cog. This information is then used to determine how much current should be applied to overcome the effect of cogging torque. Smoothing the transition between cogs allows the motor to spin at a more constant speed and experience less torque ripple.

Previous
Previous

Position Sensing: Encoders & Linearization

Next
Next

Vijay Kumar joins Vertiq Board of Directors