There are two main applications with dynamic torque excitation:
Impulse wrenches are fastening tools for screw connections in industrial assembly lines. These screw connections need to be fastened quickly and – especially for safety relevant connections – to a known and traceable torque. Impulse wrenches operate by applying a sequence of short impulses (with durations in the range of milliseconds). These impulses are generated by releasing a pre-tensioned hydraulic fluid. Impulse wrenches offer short fastening times, good health and safety for the worker using the tools and good reproducibility.
In research and development, as well as for homologation, the mechanical output power of combustion engines or electric drives needs to be determined. It can be measured by means of torque and rotational speed measurement following, .
The measurements are carried out to determine mechanical output power ratings and – far more critical in terms of demand for low measurement uncertainties – power efficiency measurements. This application has gained importance due to the efforts to further improve the fuel efficiency of vehicles.
The torque output of electric machines and even more of combustion engines can be highly dynamic in comparison to the mean torque. The frequency content extends up to the kilohertz range.
Although there are applications with dynamic torque excitations, currently no procedures or standards exist for a dynamic calibration of torque transducers.
The measurement principle of the majority of torque transducers is based on the measurement of torsion, which is typically carried out by means of strain gauges, but other principles are used as well [1]. For industrial applications and dynamic measurements strain gauge transducers are far more common than those based on other measurement principles.
In a strain gauge transducer, the applied torsional load elastically deforms the mechanical structure of the torque transducer. These deformations are sensed by several strain gauges bonded onto the surfaces at specific locations, in particular at spots of strong and uniform strain. The electrical resistance of a strain gauge changes proportionally with its mechanical elongation. Usually, four strain gauges are electrically connected to a Wheatstone bridge circuit to supply an output signal voltage that is proportional to the applied torsional load.
At present, the calibration of torque transducers used for the above-mentioned applications can only be carried out statically. A static calibration of torque transducers is possible over a wide range of torque levels and with low uncertainties. However, the use of a statically calibrated torque transducer for dynamic torque measurements may lead to incorrect results. From experience with force transducers which are related in mechanical design, measuring technology and mounting situation, there may be measurement deviations due to frequency-dependent sensitivity. Because of the fact that torque transducers are always coupled on both sides to their mechanical environment (as with force transducers), their dynamic behaviour may vary depending on the coupled components.
The strain gauge measurement principle results in a characteristic mechanical design. In general, transducers are designed to have a high torsional stiffness, but the structural components on which the strain gauges are applied have to exhibit a sufficiently high compliance.
To be able to characterise the dynamic behaviour of a torque transducer and its effect on the coupled components in the drive train, the transducer is described by an appropriate model and the model parameters are identified during the calibration process [2].
The dynamic calibration of torque transducers is still under research and development; at present no final procedure exists.
The calibration itself is carried out by applying periodic torque excitations in a broad range of frequencies. For this purpose, a method for a primary realisation of dynamic torque was proposed [3] and a measuring device was developed. The measurement principle is based on Newton's second law; the acting torque is measured by means of a mass moment of inertia and the angular acceleration , giving
The model parameters of a torque transducer under test are to be identified from measurement data. To be able to identify these parameters, the model parameters of the calibration device itself need to be identified first [4]. With this additional information, the properties of the transducers can be identified [5].
With the model parameters of a transducer determined, the dynamic behaviour of such a transducer can be predicted with sufficient knowledge about the system/set-up of the actual application. Dynamic influences can be corrected with respect to the measurement uncertainties assigned to the transducer’s model parameters.
[1] Haim H. Bau, Nico F. de Rooij, Benjamin Kloeck, Wolfgang Göpel, J. Hesse and J. N. Zemel, Publisher. Mechanical sensors. Weinheim [u.a.]: VCH, 1994.
[2] Leonard Klaus, Dynamic Torque Calibration – Necessity and Outline of a Model-Based Approach in 5th COOMET Competition for Young Metrologists - Competition Papers, ISBN 978-3-944659-00-8
[3] Thomas Bruns, Sinusoidal Torque Calibration: A Design for Traceability in Dynamic Torque Calibration in Proc. of XVII IMEKO World Congress; 2003, Dubrovnik, Croatia
[4] Leonard Klaus, Thomas Bruns, Michael Kobusch, Modelling of a Dynamic Torque Calibration Device and Determination of Model Parameters, ACTA IMEKO 3 (2),pp. 14 – 18, 2014
[5] Leonard Klaus, Barbora Arendacká, Michael Kobusch, Thomas Bruns, Model Parameter Identification from Measurement Data for Dynamic Torque Calibration, in Proc. of Joint IMEKO International TC3, TC5 and TC22 Conference; 2014