1. How do dynamic torque sensors work?
Dynamic torque sensors (also known as torque meters) measure the torque of rotating shafts by detecting the resulting mechanical strain. The sensor consists of a rotating shaft and a stationary housing. The sensor’s shaft is mounted in-line with the machine shaft, so that the torque passes through the sensor’s shaft, placing it under mechanical strain. The strain is measured by strain gauges (explained below) attached to the torque sensor’s shaft, which output a small voltage signal proportional to the mechanical strain. The sensors calibration sheet provides the relationship between strain and torque.
The stationary housing houses the electronics and provides the user with a stationary base onto which the electrical cable (for carrying the supply voltage and the output signal) can be attached. The housing is mechanically coupled to the sensor shaft, via bearings. The electrical connection between the housing and the shaft (which contains the strain gauges) is made either by graphite brushes (similar to those found in DC motors) or wirelessly via a rotary transformer. Brushes are less prone to electromagnetic interference than rotary transformers but undergo mechanical wear and must therefore eventually be replaced.
A strain gauge is a thin laminated foil pattern used to measure mechanical strain. Stretching the strain gauge causes its length to increase and its cross sectional area to decrease, thereby increasing its electrical resistance. By measuring the change in electrical resistance, it is possible to determine the amount by which the strain gauge has been stretched. Most torque sensors include 4 strain gauges, configured in a full bridge Wheatstone bridge formation, providing temperature compensation, common noise cancellation and a high signal to noise ratio.
2. General characteristics of dynamic torque sensors
Dynamic torque sensors are highly accurate, but high cost sensors. They can measure a wide range of torque magnitudes, up to about 300,000 Nm, at rotational speeds of up to 50,000 RPM. They must be mounted in line with the existing machine shaft, effectively becoming part of the shaft. Electrical contact between the sensor shaft and housing is made by either graphite brushes or a wireless rotary transformer. The rotary transformer does not experience any mechanical wear and therefore improves the lifetime of the sensor. Dynamic torque sensors provide an mv/V output signal, though often include a built-in amplifier that provides a 0-10 V output. Dynamic torque sensors generally have poor ingress protection, though a limited number of models are available with an IP67 rating.
3. Input and output signals
Dynamic strain gauge torque sensors require fixed 10-30 Vdc supply voltages. The strain gauge elements are arranged in a full bridge Wheatstone bridge formation, providing temperature compensation, common noise cancellation and a high signal to noise ratio. The sensors output signal takes the form of millivolts per volt of supply voltage. Many torque sensors include built-in amplifiers to amplify the signal, which is then converted to a standard form such as 0-10 V or 4-20 mA. If the torque sensor does not include a built-in amplifier then either the process controller must contain one, or an external amplifier must be used.
4. Applications of dynamic torque sensors
Dynamic torque sensors are used for measuring the torque of rotating machine shafts and motors. They are the most popular torque sensor for applications in which it is desired to measure torque without absorbing the shaft power. They are often used in test rigs and laboratory applications for measuring power. They are less common in commercially available machinery because of their high cost and the fact that torque can be estimated by measuring motor electrical power and speed.
5. Typical specification
|Measurement range||0-300,000 Nm|
|Sample rate||0.1-5 kHz|
|Ambient temperature||-40 to 110 °C|
|Supply voltage||10-30 Vdc|
|Output voltage||4-20 mA, 0-10 V, mV/V|
|Ingress protection||IP40 typical, IP67 available|
|Passive / active||Active|
|Contact / non-contact||Contact|
6. Purchasing tips
- Brushed vs. brushless: Voltage must be sent back and forth between the rotating strain gauges and stationary housing. This is performed either wirelessly or via conductive brushes. Brushes are less susceptible to background noise but are prone to mechanical wear and so are found predominantly on lower speed sensors (<5,000 RPM).
- Mounting: Several shaft mounting configurations are available, the most common being smooth shaft, keyed shaft and square shaft. The choice of configuration depends largely on the configuration of your existing shaft. Avoid using smooth shafts for high torque applications.
- RPM: Almost all dynamic torque sensors can operate at speeds of at least 3,000 RPM. For higher speeds refer to the product data sheet, as maximum RPM varies significantly between different models.
- Built-in amplifier: The lowest cost torque sensors output a non-amplified mV/V signal. However, the signal must then be amplified by either the process controller or an external amplifier. Many sensors include built-in voltage amplifiers and can therefore provide 0-10 V output signals.
- Built-in encoder: Dynamic torque sensors may be purchased with optional built-in angular encoders, providing relative shaft position data and enabling shaft speed to be calculated.
- Built-in display: Dynamic torque sensors may be purchased with small built-in screens displaying measured torque in real time without the need to connect the sensor to a HMI.
7. Advantages of dynamic torque sensors
Dynamic torque sensors:
- Are highly reliable and simple to use.
- Have a high measurement accuracy.
- Have very good temperature compensation owing to their full bride Wheatstone bridge.
- Do not consume shaft power.
8. Disadvantages of dynamic torque sensors
Dynamic torque sensors:
- Must be mounted in-line with the existing shaft, effectively becoming part of the shaft. This may not be desirable in some applications e.g. because it adds to the effective length of the shaft.
- Have a lower Ingress protection rating, a shorter lifetime and a higher cost than static torque sensors.
- Must be accurately aligned with the existing shaft to prevent high amplitude vibrations from occurring.
- Contain graphite brushes (except for telemetry based torque sensors). The brushes undergo mechanical friction and therefore limit the life of the sensor.
9. Application tips
- Bearing moment: The bearings of most dynamic torque sensors are not intended to be bear a large mechanical load. If the existing shaft requires additional support, then separate bearings must be used without relying on the torque sensor for support. Failure to adhere is likely to result in damage to the torque sensor.
- Calibration: It is advisable to occasionally re-calibrate torque sensors, particularly if they have been subjected to high amplitude vibrations. In most cases, calibration can be performed by statically securing one end of the shaft and attaching a lever of known length to the other end. By applying a known weight to the lever, it is possible to map the voltage – torque relation of the sensor.
- Vibration: High vibrations, typically caused by poor alignment, will reduce the life of the bearings and in extreme cases, the life of the strain gauges. Vibrations can be minimized by using the correct shaft couples and by performing proper shaft alignment using a dial indicator.
- Very high RPM applications: Dynamic torque sensors are limited to speeds of 50,000 RPM. For higher speed applications it is possible to substitute a dynamic torque sensor for a static sensor which measures the reaction force. For example, instead of mounting a torque sensor to the shaft of a high-speed motor, mount it to the motor body so that its reaction force travels through the torque meter (see illustration below).