Sensorless BLDC Motor Control via BEMF Voltage Comparison

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing sensorless driving methods for Brushless DC (BLDC) motors face challenges in accurately detecting rotor position without position sensors, leading to reliability concerns and increased costs due to the use of Hall effect sensors and additional hardware, which can affect high-speed performance and motor efficiency.

Innovation Solution

The method involves detecting crossover times from back electromotive force (BEMF) voltage by comparing instantaneous and average BEMF voltages, using a reduction factor to reduce noise sensitivity, and requiring only minimal CPU processing and three pairs of resistors for voltage sensing, eliminating the need for pseudo-neutral voltage sensing.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If Hall effect sensors are used to detect rotor position, then rotor position detection accuracy is improved, but system cost and complexity increase

Engineering Contradiction:
Improverotor position detection accuracyVSAvoidsystem cost and complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent extracts the position sensing function from physical Hall effect sensors and implements it through sensorless BEMF voltage detection. By measuring the back electromotive force voltage on the floating phase terminal during no-switching periods, the system determines rotor position without requiring additional sensors, wiring, or mounting hardware, thereby reducing system complexity and cost while maintaining position detection capability

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The motor's own BEMF voltage is utilized as the sensing signal for position detection. The system leverages the motor's inherent electrical characteristics during normal operation to generate the necessary position information, eliminating the need for separate sensing components and their associated complexity

Inventive Principle:
Principle #25Self-service

2Reliability

If RC filters are used to filter PWM noise from BEMF voltage, then measurement reliability is improved, but sensing delay increases and high-speed performance deteriorates

Engineering Contradiction:
Improvemeasurement reliabilityVSAvoidhigh-speed performance
Core Design Contradiction:
ReliabilityVSSpeed

Solution Approach 1:

The patent applies partial filtering by selectively removing only the PWM switching frequency noise components from the BEMF voltage measurement while preserving the essential position information. This is achieved through targeted noise rejection techniques that filter specific frequency ranges without requiring heavy RC filtering that would cause sensing delay and degrade high-speed response

Inventive Principle:
Principle #16Partial or excessive action

3Measurement precision

If conventional BEMF measurement methods are used, then rotor position detection is achieved, but additional hardware components are required

Engineering Contradiction:
Improverotor position detectionVSAvoidhardware components
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The existing power semiconductor switches and phase terminals of the motor are made multi-functional. The same switches used for motor commutation also enable BEMF voltage measurement during their off-states, and the phase terminals serve both as current injection points and voltage sensing points. This eliminates the need for separate sensing hardware and reduces overall system complexity

Inventive Principle:
Principle #6Universality (Multi-functionality)

Solution Approach 2:

The motor's own electrical structure and operating characteristics are exploited to provide position sensing. The BEMF voltage naturally generated during motor operation is directly measured at the phase terminals without requiring external sensing components, making the motor self-sufficient for position detection

Inventive Principle:
Principle #25Self-service

Applied Scientific Principles

This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.

Function Achieved in This Case

This approach enables efficient high-speed operation with reduced CPU processing, minimal hardware requirements, and improved reliability, while maintaining cost-effectiveness and efficient motor control.

Implementation Method 1

A common way of estimating rotor position without position sensors is by measuring back electromotive force (BEMF) voltage. BEMF voltage of a motor varies according to rotor position.

Methodology Applied
Scientific EffectBack electromotive force (BEMF): Electromagnetic Induction

Implementation Method 2

An instantaneous BEMF voltage and an average BEMF voltage are compared to detect the crossover time, which can be used to change the commutation switching sequence.

Methodology Applied
Scientific EffectVoltage comparison:

Data Source

PatentUS8558495B2Sensorless BLDC motor control by comparing instantaneous and average BEMF voltages
Publication Date: 2013.10.15 ATMEL CORP
  • US8558495B2 patent drawing
  • US8558495B2 patent drawing
  • US8558495B2 patent drawing

AI summary

Sensorless driving of a brushless DC (BLDC) motor includes detecting a zero crossing time from back electromotive force (BEMF) voltage of the BLDC motor. An instantaneous BEMF voltage and an average BEMF voltage are compared to detect the crossover time, which can be used to change the commutation switching sequence. Since the average BEMF voltage differs for odd and even steps of the commutation switching sequence, average BEMF voltages are calculated separately for odd and even sequences and compared to instantaneous BEMF voltages to detect crossover points for the odd and even sequences. The times to commutations for the odd and even sequences are averaged to provide an average time to the next commutation cycle. The average time can be scaled by a reduction factor to reduce the effects of measurement noise.