An FG output seems simple, but it is only useful after its pulse definition and electrical interface are clearly understood. Many brushless motors with integrated drivers provide this signal, but receiving pulses does not automatically mean the controller knows the exact RPM.
To convert an FG pulse train into reliable speed information, a brushless motor controller must know the number of pulses per mechanical revolution, the signal's electrical interface type, and the measurement limitations at both low and high speeds.
To interpret the signal correctly, the controller needs to understand the pulses per revolution, signal frequency, output voltage, pulse detection method, and measurement interval. A simple wire labeled "FG" is not enough information for reliable system integration. The main engineering question is: how does the controller convert this stream of pulses into accurate speed information across the application's entire operating range?
What Is an FG Signal in a Brushless Motor?
FG usually stands for Frequency Generator and refers to a speed-related pulse output from a brushless DC (BLDC) motor, typically one with an integrated driver. In many integrated-driver designs, the FG output does not require an additional external speed sensor.
In many integrated-driver BLDC motors, the FG signal is generated from information the motor's internal controller already uses for commutation, such as Hall sensor transitions or rotor position estimates derived from back-EMF.
From an OEM engineer's perspective, its main function is to provide feedback for system-level control tasks. These include:
- Closed-loop speed feedback
- Motor speed measurement during valid feedback operation
- Rotation monitoring after successful startup
- Detecting abnormal speed conditions, such as unexpected slowing or overspeed
However, it is just as important to understand what a standard FG signal does not indicate. Typically, it does not provide:
- Absolute rotor position within a revolution
- The high resolution needed for precise positioning tasks
- Rotation direction (a single-channel FG signal cannot distinguish clockwise from counter-clockwise)
- Information about available torque or motor current
FG pulses also do not always confirm actual rotor movement. Depending on the driver design, the output during certain startup modes may reflect the applied commutation frequency rather than measured shaft rotation.
System-Level Observation: An FG signal provides information about the frequency of defined rotational or commutation-related events. Its usefulness as actual speed feedback depends on how the driver generates the signal and when that feedback becomes valid.
How Many FG Pulses Represent One Motor Revolution?
This is one of the first questions to ask a motor supplier. Before a controller can calculate RPM, it must know the relationship between the pulses it counts and one full mechanical revolution of the motor shaft.
Before calculating RPM, you must confirm the number of FG pulse cycles generated per mechanical revolution and exactly which signal edges (rising, falling, or both) the host controller is designed to count.
The number of pulses per revolution (PPR) can vary significantly depending on the motor and driver design. One mechanical revolution may generate:
- 1 pulse
- 3 pulses
- 6 pulses
- A number related to the motor's pole pairs
- Another manufacturer-defined number
For example, a motor with four pole pairs completes four electrical cycles during one 360° mechanical revolution. If the driver generates one FG pulse for every electrical cycle, the output would be 4 PPR.
However, this relationship is not universal. The driver's internal logic can multiply or divide the underlying commutation-related frequency, so you cannot assume the PPR based on motor pole count alone. Always confirm with the datasheet or supplier.
Furthermore, how the host controller counts these pulses matters. If it counts only rising edges, a 6 PPR signal provides 6 counts per revolution. If the controller counts both rising and falling edges, it may register 12 transitions per revolution.
This is single-channel edge counting, not quadrature decoding. Quadrature decoding normally uses two phase-shifted signal channels to determine movement direction.
The controller must distinguish between complete FG pulse cycles and detected signal edges. Otherwise, a mismatch between the PPR definition and the counting method can produce an incorrect RPM value, potentially by a factor of two.
Open-Drain vs Push-Pull FG Outputs: Why Electrical Interfaces Matter
You cannot assume an FG signal line can be connected directly to any microcontroller input. The electrical characteristics of the output driver stage must be compatible with the host controller's input. Two common output configurations are open-drain (or open-collector) and push-pull.
FG compatibility is both a pulse-definition problem and an electrical-interface problem. A mismatch in logic voltage, output configuration, or input requirements can produce unreliable speed readings or, in some cases, damage connected components.
Understanding the difference is important for hardware design.
| Interface Parameter | Open-Drain / Open-Collector FG | Push-Pull FG |
|---|---|---|
| Logic High | Established by a pull-up resistor connected to an appropriate logic supply. | Actively driven high by the output stage. |
| Logic Low | The output transistor pulls the signal line low. | Actively driven low by the output stage. |
| Pull-up Resistor | Required for normal logic operation. An internal MCU pull-up may be sufficient if its electrical characteristics meet the application requirements. | Normally not required because the output actively drives both logic levels. |
| Voltage Matching | The pull-up voltage must remain within the FG output's rated voltage and current limits while meeting the controller's input requirements. | The output's high and low voltage levels depend on its logic output stage and must be compatible with the receiving input. |
| Rise Time | Influenced by pull-up resistance and total line capacitance. | Typically faster, although actual timing depends on the output stage and connected load. |
| Main Concern | Pull-up selection, allowable output voltage, sinking current, and signal integrity. | Logic-level compatibility, output loading, and input protection. |
For an open-drain interface, the pull-up resistor deserves particular attention. A weak internal pull-up may work at a low FG frequency but produce excessively slow rising edges when cable length, capacitance, or operating frequency increases.1
A stronger pull-up can improve rise time, but it also increases the current that the FG output must sink when the line is low.2 Both the resistor value and the pull-up voltage must therefore be checked against the driver specifications.
For push-pull outputs, the logic-high voltage is not necessarily equal to the motor's main supply voltage. The driver may use a separate internal logic supply or output-voltage domain.
Common Integration Problem: An FG signal may appear to be missing or unreliable because an open-drain output has no suitable pull-up resistor, the output voltage is incompatible with a 3.3V controller input, or the wiring introduces excessive signal distortion. These electrical conditions should be checked before assuming that the motor or driver is defective.
FG is only one part of the motor-controller interface. Other signals, such as PWM, direction, braking, and enable, must also be defined according to the driver's electrical requirements. For a broader explanation of these connections, see our BLDC motor wiring guide.
How to Calculate Motor RPM From FG Pulse Frequency
Once the pulse-per-revolution count is confirmed, converting the measured FG frequency into RPM is straightforward. This calculation is the basis of using an FG signal for speed measurement.
The rotational speed in RPM is calculated by multiplying the FG signal frequency in Hertz by 60 and dividing by the number of complete FG pulse cycles per mechanical revolution.
The basic relationship is:
RPM = (60 × f_FG) / P
Where:
RPMis the mechanical motor speed (revolutions per minute)f_FGis the measured FG pulse frequency (in Hz)Pis the number of complete FG pulse cycles per mechanical revolution
Example Calculation:
Assume a motor datasheet specifies an FG output of 6 pulses/revolution and your controller measures an FG frequency of 150 Hz during valid speed-feedback operation.
RPM = (60 × 150) / 6 = 1,500 RPM
The calculated motor speed is 1,500 RPM.
This relationship also allows you to calculate the expected frequency for a target speed. For the same 6 PPR motor:
| Motor Speed | Expected FG Frequency |
|---|---|
| 60 RPM | 6 Hz |
| 300 RPM | 30 Hz |
| 1,500 RPM | 150 Hz |
| 6,000 RPM | 600 Hz |
| 18,000 RPM | 1,800 Hz |
This table highlights an important engineering challenge: the same FG output creates different measurement requirements across the operating speed range3.
At 60 RPM, pulses are relatively far apart, making rapid speed updates difficult. At 18,000 RPM, pulses arrive much more frequently, requiring the controller's input circuit and measurement hardware to handle the increased edge rate.
The calculation also assumes that the FG frequency corresponds to actual mechanical rotation. During startup or certain fault conditions, the driver may not provide valid speed feedback, so the same formula should not be applied without confirming the FG operating state.
Why FG Speed Measurement Becomes Difficult at Very Low and High RPM
A controller's ability to measure speed from an FG signal is not uniform across the entire RPM range. The most suitable measurement strategy depends on pulse frequency, timing resolution, and the application's required feedback response.
At low speeds, long intervals between pulses limit how frequently new measurements become available. At high speeds, the controller must capture faster signal transitions without missing or misinterpreting edges. Measurement accuracy and feedback latency must be evaluated together.
Low-Speed Measurement Challenges
At very low RPM, the time between consecutive pulses becomes long. For a 6 PPR motor running at 60 RPM, the FG frequency is only 6 Hz, meaning pulses arrive approximately every 167 ms.
If your controller uses a fixed 100 ms window to count pulses, it may detect one pulse in one window and zero in the next, resulting in coarse and fluctuating speed readings.
- Common Solution: Instead of counting pulses within a fixed time window, the controller can measure the period between consecutive pulses. This often improves low-speed measurement resolution but remains sensitive to individual pulse timing variations.
Period measurement does not eliminate the physical time required for a new pulse to arrive. A controller cannot obtain a new independent pulse-period measurement until another valid edge is detected.
For example, at 300 RPM with a 6 PPR FG output, the frequency is 30 Hz and the pulse interval is approximately 33.3 ms. A controller can update its displayed or estimated speed every 20 ms, but not every update will contain a new independent FG-period measurement.
This distinction is particularly important when an OEM application specifies both a short feedback update interval and a strict speed-measurement accuracy requirement.
High-Speed Measurement Challenges
At high RPM, the FG frequency rises significantly. For the same motor at 18,000 RPM, the frequency reaches 1.8 kHz, with a pulse period of approximately 556 microseconds.
A frequency of 1.8 kHz is manageable for many modern microcontrollers, particularly when hardware timers or counters are available. However, the actual measurement capability depends on the controller architecture, interrupt handling, input filtering, and the total number of signal transitions being monitored.
- Common Issue: Software-based edge counting can lose pulses if the processor cannot service events consistently.4 Electrical noise, inadequate rise times, or unsuitable input filtering may also cause missed or false counts.
High speed does not automatically mean poor FG measurement accuracy. The engineering task is to verify that the signal interface and controller can handle the maximum expected frequency with sufficient timing margin.
The Measurement Trade-Off
| Operating Condition | Main Challenge | Possible Approach |
|---|---|---|
| Very Low RPM | Long intervals between pulses; limited availability of new measurements. | Use period measurement and define an appropriate timeout. |
| Medium RPM | Balancing measurement resolution and response time. | Use frequency counting or a hybrid measurement method. |
| High RPM | Higher edge rates and possible signal-capture limitations. | Use hardware timers or counters and verify signal integrity. |
| Rapid Acceleration | Speed changes during the measurement interval. | Use shorter or adaptive measurement windows where practical. |
| Sensorless Startup | FG may be absent, invalid, or related to commanded commutation frequency. | Confirm when valid speed feedback begins and apply startup-state logic. |
Sensorless startup deserves special attention. Some driver ICs generate FG transitions during open-loop acceleration based on the applied electrical frequency, even if the rotor has not successfully started rotating.
Therefore, a changing FG signal during startup should not automatically be treated as proof of actual rotor motion. OEM engineers should confirm the driver's FG validity conditions and how the signal behaves during unsuccessful startup, stall, or fault recovery.
What an FG Signal Can—and Cannot—Do in Closed-Loop Motor Control
Connecting the FG output to a host controller provides a speed-related measurement input that can be used for closed-loop speed regulation. However, obtaining a usable feedback signal is only one part of implementing a functional control loop.
FG is primarily a speed-feedback interface. It can support speed regulation, but a standard single-channel FG signal is generally not a substitute for a positioning encoder when precise angular position or bidirectional position tracking is required.
In a typical system where the host controller manages speed regulation, FG feedback can help compensate for disturbances such as changes in mechanical load or supply conditions5.
The control sequence may look like this:
- The system defines a target motor speed.
- The host controller calculates the difference between target RPM and measured RPM.
- The controller adjusts the supported speed command, such as a PWM duty-cycle command.
- The motor driver changes the motor operating condition.
- The FG signal provides updated speed-related feedback.
- The controller repeats the process according to its control strategy.
This arrangement depends on the motor driver's architecture. Some integrated drivers already provide internal speed regulation, while others accept an external speed command without independently closing the speed loop. The role of the host controller must therefore be confirmed during integration.
It is also a mistake to assume that a higher FG pulse count automatically guarantees better closed-loop performance. Speed stability depends on several interacting factors:
- FG Output Resolution: More pulses per revolution can reduce the interval between measurement events, particularly at low RPM.
- Speed Measurement Accuracy: Timing resolution, pulse jitter, noise, filtering, and the measurement algorithm affect the calculated speed.
- Feedback Latency: Averaging and measurement-window selection determine how quickly changes in speed become visible to the controller.
- Control Loop Dynamics: Controller tuning, update rate, motor torque capability, mechanical inertia, and load disturbances influence the actual speed response.
FG Output Resolution ≠ Speed Measurement Accuracy ≠ Closed-Loop Speed Stability. A higher pulse count cannot compensate for poor signal integrity, inadequate measurement timing, or an unsuitable control strategy6.
For OEM engineers, the objective is not simply to maximize the number of FG pulses. It is to obtain a feedback signal that provides sufficient speed information within the time and accuracy limits of the intended control system.
How OEM Engineers Should Specify and Validate an FG Interface
To avoid integration problems, OEM engineers should treat the FG signal as a defined electrical and measurement interface rather than an optional motor feature. This requires coordination between the motor supplier, driver configuration, and host-controller design.
A successful integration starts with clear operating requirements. Define the minimum and maximum speed, acceptable measurement error, feedback timing, electrical interface, and expected startup or fault behavior before selecting an FG configuration.
The following checklist can be used when specifying or validating an FG interface:
| Specification Item | What the OEM Engineer Should Define or Confirm |
|---|---|
| FG Pulse Count | Number of complete FG pulse cycles per mechanical revolution (PPR). |
| Edge Counting Definition | Whether the host controller counts rising edges, falling edges, or both. |
| FG Output Circuit | Open-drain, open-collector, or push-pull configuration. |
| Logic Voltage | Compatible logic-high and logic-low levels, output voltage limits, and input thresholds. |
| Pull-up Requirements | Required or permitted pull-up voltage and resistance, where applicable. |
| Speed Range | Minimum, nominal, and maximum operating RPM. |
| Frequency Range | Minimum and maximum expected FG frequency in Hz. |
| Measurement Method | Frequency counting, period measurement, or a hybrid strategy. |
| Update Interval | Required controller update rate and maximum acceptable age of speed-feedback data. |
| Accuracy Requirement | Maximum allowable RPM measurement error under specified operating conditions. |
| Startup FG Validity | When the FG signal begins representing valid rotor-speed information. |
| Stall and Fault Behavior | How the FG output behaves during stall, failed startup, shutdown, and fault recovery. |
A well-defined inquiry provides more useful information than simply requesting a motor "with FG output."
For instance:
"Our application requires a motor that operates from 300 to 18,000 RPM. Our controller uses 3.3V logic with an internal pull-up, needs a speed update every 20 ms, and must maintain an RPM measurement error of less than ±2%. What FG configuration can you provide to meet these needs?"
This is a useful starting point, but the supplier and OEM team still need to verify whether the requested measurement performance is achievable.
For example, if the proposed motor provides 6 FG pulses per revolution, the pulse interval at 300 RPM is approximately 33.3 ms. The controller may produce a new speed estimate every 20 ms, but it cannot obtain a new independent pulse-period measurement at that rate.
A different PPR configuration, measurement strategy, or definition of acceptable feedback latency may therefore be necessary. The required ±2% measurement accuracy must also be evaluated under clearly defined speed and operating conditions.
At BODENMOTION, our engineering team can support OEM customers in reviewing motor specifications, electrical interface requirements, and available customization options. Confirming these details early helps identify potential integration limitations before prototype testing or production.
Conclusion
An FG signal allows a brushless motor controller to estimate rotational speed, but its usefulness depends on how the signal is generated, measured, and interpreted. Higher pulse counts can provide more frequent speed information, yet they do not automatically guarantee better measurement accuracy or closed-loop stability. Low-speed pulse intervals, high-speed signal capture, and startup or fault behavior must all be considered when evaluating whether FG feedback meets an application's control requirements.
For OEM engineers, the goal is not simply to obtain an FG output, but to ensure that its pulse definition, electrical interface, and feedback timing match the control system. For projects requiring suitable motor and speed-feedback integration, explore BODENMOTION's DC motor solutions or contact our team at info@bodenmotion.com to discuss your application requirements.
FAQ
Q1: What does FG mean on a brushless motor?
FG typically stands for Frequency Generator. It is a speed-related pulse output provided by certain brushless motor drivers. During valid speed-feedback operation, its frequency is related to the motor's rotational speed according to the driver's defined pulse configuration.
Q2: How do I calculate BLDC motor RPM from an FG signal?
Use the formula:
RPM = (60 × FG Frequency in Hz) / Pulses per Revolution
The pulse count must refer to complete FG cycles per mechanical revolution, and the measured frequency must correspond to valid motor-speed feedback.
Q3: Is an FG signal the same as an encoder signal?
Not usually. A standard FG output provides a single channel of speed-related pulses. A quadrature encoder typically provides two phase-shifted channels (A/B) that allow the controller to determine movement direction and track relative position. Encoders designed for precise positioning commonly offer higher resolution than standard FG outputs.
Q4: Can FG feedback be used for closed-loop speed control?
Yes. A host controller can use valid FG speed feedback to compare actual speed with a target and adjust a supported motor speed command. The achievable regulation performance depends on feedback resolution, measurement latency, controller tuning, motor dynamics, and the driver architecture.
Q5: What is the difference between open-drain and push-pull FG outputs?
An open-drain output pulls the signal line low and requires a suitable pull-up resistor to establish a logic-high level. Depending on electrical and timing requirements, the pull-up may be external or integrated into the host controller.
A push-pull output actively drives both high and low logic levels and normally does not need a pull-up resistor. In both cases, the FG output voltage and current characteristics must be compatible with the receiving controller.
Total Phase's “What Are I2C Pull-Up Resistors and How to Calculate Their Values” explains how pull-up resistance and line capacitance affect signal rise time in open-drain interfaces, and why rise time must meet the timing requirements of the digital signal. ↩
Texas Instruments' “Choosing an Appropriate Pull-up/Pull-down Resistor for Open Drain Outputs” explains how pull-up resistance affects output sink current and how parasitic capacitance influences signal rise time. It also provides methods for selecting resistor values within the output device's specified electrical limits. ↩
Texas Instruments' “Monitoring Fan Tachometers Using the UCD908x Family” demonstrates how tachometer pulse frequency affects measurement-circuit requirements. It shows that adjusting pulse timing can extend the usable frequency range and that component tolerances influence measurement accuracy. ↩
Espressif's “Pulse Counter (PCNT)” documentation describes a hardware peripheral that counts rising and falling edges, supports pulse-frequency measurement, and allows the processor to read accumulated counts without handling every input transition through software. ↩
BODENFLO's “Why Does Speed Feedback (FG Signal) Matter in Your Precision Micro Pump Application?” explains how mechanical load changes and supply-voltage fluctuations can affect motor speed, and how FG-based closed-loop control allows the controller to detect speed deviations and adjust the motor drive command. ↩
Bill Dehner's “Efficient Controls Require Feedback,” published in Control Engineering, explains that effective closed-loop control requires appropriate feedback sensor resolution, measurement response time, and correctly configured controller parameters. ↩