Torque Sensor Technical Information

Principle of Operation
All torque sensors manufactured by PCB are strain gage based measuring instruments whose output voltage is proportional to applied torque. The output voltage produced by a resistance change in strain gages that are bonded to the torque sensor structure. The magnitude of the resistance change is proportional to the deformation of the torque sensor and therefore the applied torque. The four-arm Wheatstone Bridge conguration shown in Figure 1 depicts the strain gage geometry used in the torque sensor structures. This conguration allows for temperature compensation and cancellation of signals caused by forces not directly applied about the axis of the applied torque. A regulated 5 to 20 volt excitation is required and is applied between points A and D of the Wheatstone bridge. When torque is applied to the transducer structure the Wheatstone bridge becomes unbalanced, thereby causing an output voltage between points B and C. This voltage is proportional to the applied torque. Series 2300 reaction torque sensors have the wiring code illustrated in Figure 2. Series 4100 rotary transformer torque sensors have the wiring code illustrated in Figure 3. Series 4200 rotary transformer torque sensors have the wiring code illustrated in Figure 4.

Highlights
I

Introduction to torque sensors Glossary of terms

Axis Definition Figure 1.

Wheatstone Bridge

PCB torque sensors comply with the Axis and Sense Denitions of NAS-938 (National Aerospace Standard-Machine Axis and Motion) nomenclature and recommendations of the Western Regional Strain Gage committee. Axes are dened in terms of a right-handed orthogonal coordinate system, as shown in Figure 5. The principal axis of a transducer is normally the z-axis. The z-axis will also be the axis of radial symmetry or axis of rotation. In the event there is no clearly dened axis, the following preference system will be used: z, x, y.

Figure 2.
Series 2300 Reaction Torque Sensor Wiring Code

Figure 3.
Series 4100 Rotary Transformer Torque Sensor Wiring Code

Figure 4.
Series 4200 Rotary Transformer Torque Sensor Wiring Code

Figure 5.
Right-handed Orthogonal Coordinate System

PCB Piezotronics, Inc.

Toll-Free in USA 888-684-0004

716-684-0001

www.pcb.com

6.1

Introduction to Torque Sensors

The principal axis of a transducer is normally the z-axis. The z-axis will also be the axis of radial symmetry or axis of rotation. In the event there is no clearly dened axis, the following preference system will be used: z, x, y.
Figure 6 shows the axis and sense nomenclature for our torque sensors. A (+) sign indicates torque in a direction which produces a (+) signal voltage and generally denes a clockwise torque.

Torque Sensor Structure Design Figure 6.

Axis and Sense Nomenclature for Torque Sensors

Torque sensor structures are symmetrical and are typically manufactured from steel (SAE 4140 or 4340) that has been heat-treated Rc 36 to 38. Common congurations are solid circular shaft, hollow circular shaft, cruciform, hollow cruciform, solid square, and hollow tube with ats. The solid square offers advantages over the solid circular design, especially in capacities greater than or equal to 500 in-lb (55 N-m). The solid square offers high bending strength and ease of application of strain gages. Torque sensors with capacities less than 500 in-lb (55 N-m) are usually of the hollow cruciform type. The hollow cruciform structure produces high stress at low levels of torque, yet has good bending strength. Common congurations are shown in Figure 7. A variety of end congurations are available, including: keyed shaft, ange, and spline. (See below).

Keyed Shaft

Spline Drive

Reaction Torque Sensors

Typical reaction torque sensor applications include: I Bearing friction I Starter testing I Stepping switch torque I Automotive brake testing
I

Axle torsion test

Reaction torque is the turning force or moment, imposed upon the stationary portion of a device by the rotating portion, as power is delivered or absorbed. The power may be transmitted from rotating member to stationary member by various means, such as the magnetic eld of a motor or generator, brake shoes or pads on drums or rotors, or the lubricant between a bearing and a shaft. Thus, reaction torque sensors become useful tools for measuring properties such as motor power, braking effectiveness, lubrication, and viscosity. Reaction torque sensors are suitable for a wide range of torque measurement applications, including motor and pump testing. Due to the fact that these sensors do not utilize bearings, slip-rings, or any other rotating elements, their installation and use can be very cost effective. Reaction torque sensors are particularly useful in applications where the introduction of a rotating inertia due to a rotating mass between the driver motor and driven load is undesirable. An example of this can be found in small motor testing, where introduction of a rotating mass between the motor and load device will result in an error during acceleration. For these applications, the reaction torque sensor can be used between the driver motor, or driven load, and ground. An added benet is that such an installation is not limited in RPM by the torque sensor. PCB manufactures reaction torque sensors with capacities ranging from a few inch ounces to 500k in-lb (56.5k N-m), in congurations including keyed shaft and ange.

Figure 7.
Common Torque Sensor Congurations

6.2

PCB Piezotronics, Inc.

Toll-Free in USA 888-684-0004

716-684-0001

www.pcb.com

Introduction to Torque Sensors

Rotary Torque Sensors

Typical rotary torque sensor applications include: I Chassis dynamometer I Clutch testing I Engine dynamometer I Blower or fan testing I Efciency testing I Small motor / pump testing Rotating torque sensors are similar in design and in application to reaction torque sensors, with the exception that the torque sensor is installed in-line with the device under test. Consequently, the torque sensor shaft rotates with the device under test. In PCB Series 4100 and 4200 models, the rotating torque sensor shaft is supported in a stationary housing by two bearings. Signal transfer between the rotating torque sensor shaft and the stationary housing is accomplished by means of rotary transformers.

Figure 8.

Rotary Transformers
Rotary Transformers provide a non-contact means of transferring signals to and from the rotating torque sensor structure. Rotary transformers are similar to conventional transformers, except that either the primary and secondary winding is rotating. For rotating torque sensors, two rotary transformers are used. One serves to transmit the excitation voltage to the strain gage bridge, while the second transfers the signal output to the non-rotating part of the transducer. Thus no direct contact is required between the stationary and rotating elements of the transducer (see Figure 8).

Figure 9.

Rotary transformers are made up of a pair of concentrically wound coils, with one coil rotating within or beside the stationary coil. The magnetic ux lines are produced by applying a time varying voltage (carrier excitation) to one of the coils (see Figure 9).
Figure 10 depicts a typical rotary transformer torque sensor:

Transmission of energy through any transformer requires that the current be alternating. A suitable signal conditioner with carrier excitation in the range of 3 to 5000 Hz is required to achieve this.

Mechanical Installation of Keyed Shaft Torque Sensors

Proper installation must be observed when assembling a torque sensor into a driveline. Careful selection of components must be made so that problems are not created which could lead to part failure or danger to personnel.

Shaft misalignment Figure 10.

Rotary Transformer Torque Sensor Diagram

Provision must be made to eliminate the effects of bending and end loading on the torque sensors shaft due to parallel offset of shafts, angular misalignment, and shaft end oat. The proper use of couplings can reduce these problems to a negligible level. All shafts must rst be aligned mechanically, as accurately as possible, to lessen the work the couplings must do. Alignment within 0.001 inch per inch of shaft diameter is normally satisfactory, however, for some critical applications such as high speed, this level of alignment is not acceptable, and a tighter tolerance must be achieved. Please contact our factory, or your coupling vendor, for information regarding your application.

Torque sensor with foot-mounted housing installation

A foot-mounted torque sensor has a plate on its housing, which can be securely attached to a machine base or bedplate. This installation reduces the mass in suspension on the couplings and can increase the shafts critical speed, if the torque sensor is within its speed rating. Normally, if both the driving and load sources are fully bearing-supported in foot-mounted housings, and the torque sensor housing is foot-mounted, double-ex couplings should be used on each shaft end. Double-ex couplings provide for two degrees of freedom, meaning they can simultaneously allow for angular and parallel misalignment, and reduce the effects of bending on the torque sensor shaft. Half of each coupling weight is supported on the torque sensors shaft, and the other half is carried by the driving and load shafts.
PCB Piezotronics, Inc. Toll-Free in USA 888-684-0004 716-684-0001 www.pcb.com

6.3

Introduction to Torque Sensors

Torque sensor with floating shaft installation

A oating shaft torque sensor does not have a foot-mount plate on the housing, nor is the housing afxed to a bedplate in any other fashion. It depends on being carried by the driver and load shafts for its support. The housing, which is meant to remain stationary and not rotate with the shaft, must be restrained from rotating with a conductive exible strap. Tapped threaded holes are provided on the side of the housing for this purpose. The other end of the strap is bolted to a bedplate or other stationary-grounded member, which will electrically ground the torque sensor housing to the electrical system ground. Therefore, with the oating shaft, there is just one degree of freedom between each shaft end of the torque sensor and the adjacent mating shaft, which is bearingsupported (driver and load shafts) on the bedplate. Consequently, a single ex coupling is required at each end of the torque sensor.

Error Analysis
PCB typically supplies accuracy information on its products in the form individual errors. They are non-linearity, hysteresis, non-repeatability, effect of temperature on zero unbalance, and effect of temperature on output. The customer can combine these individual errors to establish the maximum possible error for the measurement, or just examine the applicable individual error. If the temperature remains stable during the test, the temperature related errors can be ignored. If the sensor is used for increasing load measurement only, ignore the hysteresis error. If the load measurement is near the full capacity, the linearity error can be ignored. If the capability exists to correct the data through linearization-t or a look-up-table, the error in the measurement can be minimized. A sophisticated user can get rid of all the errors except for the non-repeatability error in the measurement. Often overlooked by the customer is error due to the presence of non-measured forces and bending moments. Even though the single axis of measurement sensors are designed and built to withstand these non-measured forces and bending moments (extraneous loads), the errors due to them are present. The user can design the set-up to eliminate or minimize these extraneous loads. However, if these extraneous loads are present, the errors due to them should be considered.

Application Questionnaire Determine the capacity required

A. What is the maximum expected torque, including transients? B. What is the minimum expected torque? C. What is the typical expectedtorque? D. What are the dynamics of the system, (i.e. frequency response)? E. What are the maximum extraneous loads to which the torque sensor will be subjected?

How will the torque sensor be integrated into the system?

A. What are the physical constraints, (e.g. length, diameter)? B. Will the torque sensor be foot-mounted or oated? C. Couplings, torsionally stiff, or torsionally soft?

What type of environment will the torque sensor be operating in?

A. Maximum temperature? B. Minimum temperature? C. Humidity? D. Contaminants, (e.g. water, oil, dirt, dust)?

What speed will the torque sensor be required to rotate?

A. What length of time will the torque sensor be rotating, and at what speed?

6.4

PCB Piezotronics, Inc.

Toll-Free in USA 888-684-0004

716-684-0001

www.pcb.com

Glossary of Terms

Accuracy
Stated as a limit tolerance, which denes the average deviation between the actual output versus theoretical output. In practical transducer applications, the potential errors of non-linearity, hysteresis, non-repeatability and temperature effects do not normally occur simultaneously, nor are they necessarily additive. Therefore, accuracy is calculated based upon RMS value of potential errors, assuming a temperature variation of 10 F ( 5.5 C), full rated load applied, and proper set-up and calibration. Potential errors of the readout, cross-talk, or creep effects are not included.

Creep Recovery
The change in no-load output occurring with time, after removal of a load, which has been applied for a specic period of time.

Natural Frequency
The frequency of free oscillations under no-load conditions.

Nominal Load Limit Capacity Cross-talk

With one component loaded to capacity, and the other unloaded, the output of the unloaded component will not exceed the percentage specied of its full-scale capacity. It is the designed normal maximum capacity of a transducer. Output sensitivity of the transducer is based on this capacity unless specied.

Non-linearity Deflection
The change in length along the primary axis of the load cell between no-load and rated load conditions. The maximum deviation of the calibration curve from a straight line drawn between the no load and rated load output, expressed as a percentage of the rated output and measured on increasing load only.

Ambient Conditions
The conditions (humidity, pressure, temperature, etc.) of the medium surrounding the transducer.

Drift
A random change in output under constant load conditions.

Output
This signal (voltage, current, etc.) produced by the transducer.
Note: Where the output is directly proportional to excitation, the signal must be expressed in terms of volts per volt, volts per ampere, etc. of excitation.

Error
The algebraic difference between the indicated and true value of the load being measured.

Ambient Temperature
The temperature of the medium surrounding of transducers.

Output, Rated Excitation, Electrical

The voltage or current applied to the input terminals of the transducer. The algebraic difference between the outputs at no-load and at rated load.

Calibration
The comparison of transducer output against standard test loads.

Overload Rating Fatigue Capacity

Capacity as percentage of the nominal load limit capacity, and based on 100 X 106 cycles (minimum) from zero to full fatigue capacity and 50 X 106 cycles (minimum) from full fatigue capacity tension to full fatigue capacity compression load. The maximum load in percent of rated capacity, which can be applied without producing a permanent shift in performance characteristics beyond those specied.

Calibration Curve
a record (graph) of the comparison of transducer output against standard test loads.

Combined Error (Non-linearity & Hysteresis)

the maximum deviation from a straight line drawn between the original no-load and rated load outputs expressed as a percentage of the rated output and measured on both increasing and decreasing loads.

Primary Axis
The axis along which the transducer is designed to be loaded; normally its geometric centerline.

Hysteresis
The maximum difference between the transducer output readings for the same applied load, one reading obtained by increasing the load from zero and the other by decreasing the load from rated load.
Note: Usually measured at half rated output and expressed in percent of rated output. Measurements should be taken as rapidly as possible to minimize creep.

Rated Capacity (Rated Load)

The maximum axial load that the transducer is designed to measure within its specications.

Compensation
The utilization of supplementary devices, materials, or processes to minimize known sources of error.

Repeatability
The maximum difference between transducer output readings for repeated loading under identical loading and environmental conditions.

Creep
The change of transducer output occurring with time, while under load, and with all environmental conditions and other variables remaining constant.
Note: Usually measured with rated load applied and expressed as a percent of rated output over a specic period of time.

Insulation Resistance
The DC resistance measured between the transducer circuit and the transducer structure.
Note: Normally measured at fty volts DC and under standard test conditions.

Resolution
The smallest change in mechanical input, which produces a detectable change in the output signal.

PCB Piezotronics, Inc.

Toll-Free in USA 888-684-0004

716-684-0001

www.pcb.com

6.5

Glossary of Terms

Sensitivity
The ratio of the change in output to the change in mechanical input.

Temperature Range, Usable

The extremes of temperature within which the transducer will operate without permanent adverse change to any of its performance characteristics.

Application Notes and Technical Articles

To order copies of the following application notes, call PCB toll-free at 888-684-0004.

Shunt Calibration
Electrical simulation of transducer output by insertion of known shunt resistors between appropriate points within the circuitry.

Application Notes Terminal Resistance

The resistance of the transducer circuit measured at specic adjacent bridge terminals at standard temperature, with no-load applied, and with the excitation and output terminals open-circuited.
AP-1001 Extraneous Loads AP-1002 Equivalent Force of a Falling Object AP-1003 Mechanical Installation of PCB Torque Transducers AP-1004 Installation of PCB Driveline Torque Transducers AP-1007 Dynamometer Installation of PCB Model 1401 Load Cell AP-1008 Spline Lubrication PCB Model 4115A & K, Preliminary Release AP-1009 Explosive Environment AP-1011 Effects of Thrust and Bending Moment on The Torque Output of Torque Disk. Model 5304-101-01 AP-1012 Grease Lubrication AP-1013 Effects of Thrust, Lateral, Loads and Bending Moment on the Torque Output. Models 5307-01 & 5307-02 AP-1015 Effects of Extraneous Loads on TORKDISC Series 5308 and 5309 AP-1016 Shunt Calibration of a Strain Gage Sensor

Shunt-to-load Correlation
The difference in output readings obtained through electrically simulated and actual applied loads.

Standard Test Conditions

The environmental conditions under which measurements should be made, when measurements under any other conditions may result in disagreement between various observers at difference times and places. These conditions are a follows: Temperature 72 F 3.6 F (23 C 2 C) Relative Humidity: 90% or less Barometric Pressure: 28 to 32 inch Hg

Terminal Resistance, Excitation

The resistance of the transducer circuit measured at the excitation terminals, at standard temperature, with no-load applied, and with the output terminals open-circuited.

Terminal Resistance, Signal

The resistance of the transducer circuit measured at the output signal terminals, at standard temperature, with no-load applied, and with the excitation terminals open-circuited.

Static Extraneous Load Limits

Static Extraneous Load Limits are calculated such that only one extraneous load (Fx or Fy or Mx or My or Mz) can be applied simultaneously with 50% of the nominal load limit applied.

Traceability
The step-by-step transducer process by which the transducer calibration can be related to primary standards.

Zero Balance
The output signal of the transducer with rated excitation and with no-load applied, usually expressed in percent of rated output.

Technical Articles
TA-1001 What is a Transducer? TA-1002 Cross-talk in a Multi-Component Sensor TA-1003 Accuracy

Temperature Effect on Output

The change in output due to a change in transducer temperature.
Note: Usually expressed as a percentage of load reading per degree Fahrenheit change in temperature.

Zero Return
The difference in zero balance measured immediately before rated load application of specied duration and measured after removal of the load, and when the output has stabilized.

Temperature Effect on Zero Balance

The change in zero balance due to a change in transducer temperature.
Note: Usually expressed as the change in zero balance in percent of rated output per degrees Fahrenheit (change in temperature).

Zero Shift, Permanent

A permanent change in the no-load output.

Zero Stability
The degree to which the transducer maintains its zero balance with all environmental conditions and other variables remaining constant.

Temperature Range, Compensated

The range of temperature over which the transducer is compensated to maintain rated output and zero balance within specied limits.

6.6

PCB Piezotronics, Inc.

Toll-Free in USA 888-684-0004

716-684-0001

www.pcb.com